European Federation of National Associations of Orthopaedics and Traumatology
European Instructional Lectures Volume 11, 2011
European Federation of National Associations of Orthopaedics and Traumatology Committees and Task Forces EFORT Executive Committee Executive Board Prof. Dr. Miklós Szendrői, President Prof. Dr. Pierre Hoffmeyer, Vice President Dr. Manuel Cassiano Neves, Secretary General Prof. Dr. Karl-Göran Thorngren, Immediate Past President Mr. Stephen R. Cannon, Treasurer Prof. Dr. Enric Caceres Palou, Member at Large Prof. Dr. Maurilio Marcacci, Member at Large Prof. Dr. Philippe Neyret, Member at Large Co-Opted Members Mr. John Albert Prof. Dr. Thierry Bégué Prof. Dr. George Bentley, Past President Prof. Dr. Nikolaus Böhler, Past President Prof. Dr. Karsten Dreinhöfer Prof. Dr. Klaus-Peter Günther Prof. Dr. Norbert Haas Ass. Prof. Dr. Per Kjaersgaard-Andersen Prof. Dr. Karl Knahr Dr. George Macheras Prof. Dr. Wolfhart Puhl, Past President Prof. Dr. Nejat Hakki Sur Prof. Dr. Dieter C. Wirtz
Scientific Coordination 12th EFORT Congress, Copenhagen 2011
Prof. Dr. George Macheras Prof. Dr. Maurilio Marcacci Prof. Dr. Phillip Neyret Prof. Dr. Søren Overgaard Prof. Dr. Miklós Szendrői Prof. Dr. Karl-Göran Thorngren
Standing Committees EAR Committee Prof. Dr. Nikolaus Böhler, Chairman Education Committee Prof. Dr. Maurilio Marcacci EA & L Committee Prof. Dr. Wolfhart Puhl Finance Committee Mr. Stephen R. Cannon Health Service Research Committee Prof. Dr. Karsten Dreinhöfer Portal Steering Committee Prof. Dr. Klaus-Peter Günther Publications Committee Prof. Dr. George Bentley Scientific Committee Prof. Dr. Enric Cáceres Palou
Task Forces and Ad Hoc Committees Chairman Prof. Dr. Enric Cáceres Palou, Chairman Scientific Committee Members Prof. Dr. Enric Cáceres Palou Mr. Stephen Cannon Prof. Dr. Benny Dahl Dr. Marino Delmi Prof. Dr. Benn Duus Prof. Dr. Lars Engebretsen Dr. Klaus Hindsø Prof. Dr. Pierre Hoffmeyer Ass. Prof. Dr. Per Kjaersgaard-Andersen Prof. Dr. Karl Knahr Prof. Dr. Rüdiger Krauspe
Awards & Prizes Committee Prof. Dr. George Bentley Fora Prof. Dr. Thierry Bégué Speciality Societies Standing Committee Dr. George Macheras Travelling & Visiting Fellowships Prof. Dr. Philippe Neyret Musculoskeletal Trauma Task Force Prof. Dr. Norbert Haas EFORT Foundation Committee Prof. Dr. Karl-Göran Thorngren
European Federation of National Associations of Orthopaedics and Traumatology
European Instructional Lectures Volume 11, 2011 12th EFORT Congress, Copenhagen, Denmark
Edited by
George Bentley
Prof. Dr. George Bentley Royal National Orthopaedic Hospital Trust Brockley Hill HA 7 4LP, Stanmore Middlesex, UK
[email protected] EFORT Central Office Technoparkstrasse 1 8005 Zürich, Switzerland www.efort.org
ISBN: 978-3-642-18320-1 e-ISBN: 978-3-642-18321-8 DOI: 10.1007/978-3-642-18321-8 Springer Dordrecht Heidelberg London New York Library of Congress Control Number: 2011925942 © EFORT 2011 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Product liability: The publishers cannot guarantee the accuracy of any information about dosage and application contained in this book. In every individual case the user must check such information by consulting the relevant literature. Cover design: eStudio Calamar, Figueres/Berlin Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Foreword
It is with great pleasure that I introduce this Instructional Lecture book for the 12th EFORT Congress. Since the EFORT Congress changed to take place every year, it has grown in both numbers of participants as in numbers of abstracts being submitted to be among the largest Orthopaedic Event taking place in Europe. This year the Congress is held in Copenhagen, the capital of Denmark, in a flavour of not only orthopaedics in Denmark – but with a scientific focus on main orthopaedic areas in all the Nordic countries. For this lecture book, the topics selected are all original and attractive and will predicate further lectures. The main goal of EFORT is to serve the European orthopaedics with the latest knowledge of diseases and trauma of the musculoskeletal system, and it is our hope that you through both attending the single instructional lecture, but also by reading the lectures in the present book, will increase your current knowledge within your fields of interest. During the latest EFORT Congresses we have seen more and more international colleagues form all parts of the world attending our meeting. A warmly welcome also to all of you to Wonderful Copenhagen. The Scientific programme this year again combines multiple aspects of Orthopaedics and Traumatology, with a specific attention to the use of registers to improve treatment of our patients, new approaches in cellular therapy to improve bone healing and fast track treatment and rehabilitation in variable orthopaedic surgical set-ups. Our Instructional Lecturers are from all over Europe, and present topics from several areas of interest. These lectures give you not only the opportunity to learn about various diseases but also to speak with colleagues with great experience based on their National philosophy, an unique chance to widen our European horizons. As the chairman of the Local Organising Committee, I thank all of our Lecturers for their excellent contributions for publication in this collection. My special thanks go to Professor George Bentley for organising this edition. EFORT should be congratulated for all its eforts in providing training material for all Orthopaedic surgeons, and particularly for this selection of Instructional Lectures for the Congress in Copenhagen. Copenhagen, Denmark
Per Kjaersgaard-Andersen Chairman, LOC Copenhagen 2011
v
Preface
The 11th volume of the EFORT European Instructional Lectures is a collection of all the Lectures to be presented at the 11th Congress in Copenhagen from 1st to 4th of June 2011. As always the topics are chosen to reflect some aspects of current Orthopaedic and Traumatology philosophy and practice by a group of specialists who also represent a variety of expertise which is unique to Europe. Particular thanks go to the authors, not only for preparing and presenting their lectures but also for other activities such as paper reviewing and chairing of Symposia and Specialist sessions, which are vital for the rich totality of the Congress programme. Preparation of the volume has been in the hands of Gabriele Schroeder and her colleagues in the Internationally-recognised Springer company to whom we are very grateful. My personal thanks go to Larissa Welti and the EFORT Central office staff for their expert and unfailing support, as ever. This volume is dedicated to all those who have contributed to the ever-expanding educational and scientific development of EFORT, to bring it to be the greatest Orthopaedic and Traumatology fellowship in Europe. Stanmore, UK
George Bentley Editor-in-Chief
vii
Contents
General Orthopaedics, Basic Science and Technology Bone Substitutes in Clinical Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jari Salo Epidemiology and Variability of Orthopaedic Procedures Worldwide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maria de Fatima de Pina, Ana Isabel Ribeiro, and Carlos Santos
3
9
Bone and Joint Tumours Cartilage – Forming Bone Tumours . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Antonie H.M. Taminiau, Judith V.M.G. Bovée, Carla S.P. van Rijswijk, Hans A.J. Gelderblom, and Michiel A.J. van de Sande
23
Paediatrics The Current State of Treatment for Clubfoot in Europe . . . . . . . . . . . . . . . Rüdiger Krauspe, Kristina Weimann-Stahlschmidt, and B. Westhoff
47
Polytrauma: Pelvis Management of Pelvic Fractures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Peter V. Giannoudis and Frangiskos Xypnitos
67
Shoulder, Elbow, Arm and Forearm The Reverse Shoulder Prosthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carlos Torrens
79
Spine (incl. Trauma) Spine Injury: Polytrauma Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Benny Dahl
87
ix
x
Surgical Management of Tuberculosis of the Spine . . . . . . . . . . . . . . . . . . . . Ahmet Alanay and Deniz Olgun
Contents
93
Hand and Wrist Scaphoid Fractures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Joseph J. Dias Hip Bearing Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Theofilos Karachalios and George Karydakis Hip Pain in the Young Adult . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Moritz Tannast, Christoph E. Albers, Simon D. Steppacher, and Klaus A. Siebenrock Bone Loss Around the Acetabular Component . . . . . . . . . . . . . . . . . . . . . . . 155 Jonathan Howell and Ben Bolland Knee The Uni-Knee: Indications, and Recent Techniques . . . . . . . . . . . . . . . . . . . 169 Sébastien Lustig, Gérard Deschamps, M. Alsaati, C. Fary, and Phillippe Neyret Osteotomies Around the Knee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Siegfried Hofmann, Philipp Lobenhoffer, Alex Staubli, and Ronald Van Heerwaarden Total Knee Replacement for the Stiff Knee . . . . . . . . . . . . . . . . . . . . . . . . . . 189 Philippe Massin Foot, Ankle and Leg Surgical Treatment of Displaced Calcaneal Fractures . . . . . . . . . . . . . . . . . 199 Zvi Cohen, Gershon Volpin, and Haim Shtarker Forefoot Deformities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Wolfgang Schneider
Contributors
Ahmet Alanay Department of Orthopaedics and Traumatology, Istanbul Bilim University Faculty of Medicine, Istanbul, Turkey
[email protected] Christoph E. Albers Department of Orthopaedic Surgery, Inselspital, University of Bern, Bern, Switzerland
[email protected] Maad Faisal Alsaati Pr Neyret’s Orthopaedic Department, Centre A Trillat, University Hospital, Lyon, France
[email protected] Ben Bolland Princess Elizabeth Orthopaedic Centre, Exeter, UK
[email protected] Judith V.M.G. Bovée Department of Orthopaedics, Leiden University Medical Centre, Leiden, The Netherlands
[email protected] Zvi Cohen Department of Orthopaedic Surgery, Souraski Medical Center, Tel Aviv, Israel
[email protected] Benny Dahl Spine Section, Department of Orthopaedic Surgery, Rigshospitalet, Copenhagen, Denmark
[email protected] Gérard Deschamps Centre Orthopédique, Dracy-Le-Fort, France
[email protected] Joseph J. Dias Department of Orthopaedic Surgery, University Hospitals of Leicester N.H.S. Trust, Leicester, UK
[email protected] C. Fary Pr Neyret’s Orthopaedic Department, Centre A Trillat, University Hospital, Lyon, France Hans A.J. Gelderblom Department of Orthopaedics, Leiden University Medical Centre, Leiden, The Netherlands Peter V. Giannoudis Academic Department of Trauma and Orthopaedics, Leeds General Infirmary, Clarendon Wing, Level A, Leeds, UK
[email protected] xi
xii
Ronald Van Heerwaarden AO Knee Expert Group, Department of Orthopaedics, Limb Deformity Reconstruction Unit, Sint Maartenskliniek, Woerden, Niederlande
[email protected] Siegfried Hofmann Head Knee Education Centre, Orthopaedic Department, General and Orthopaedic Hospital Stolzalpe, Stolzalpe, Austria
[email protected] Jonathan Howell Princess Elizabeth Orthopaedic Centre, Exeter, UK
[email protected] Theofilos Karachalios Orthopaedic Department, University General Hospital of Larissa, Mezourlo Larissa, Greece Orthopaedic Department, Faculty of Medicine, School of Health Sciences, University of Thessalia, Larissa, Greece
[email protected] Georgios Karydakis Orthopaedic Department, Faculty of Medicine, School of Health Sciences, University of Thessalia, Larissa, Greece
[email protected] Rüdiger Krauspe Orthopaedic Surgery, Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany
[email protected] Philipp Lobenhoffer AO Knee Expert Group, Sportsclinic, Hannover, Germany
[email protected] Sébastien Lustig Pr Neyret’s Orthopaedic Department, Centre A Trillat, University Hospital, Lyon, France
[email protected] Philippe Massin Service de Chirurgie Orthopédique, CHU Bichat Claude Bernard, Université Paris Diderot, Paris, France
[email protected] Phillippe Neyret Pr Neyret’s Orthopaedic Department, Centre A Trillat, University Hospital, Lyon, France
[email protected] Deniz Olgun Department of Orthopaedics and Traumatology, Hacettepe University Faculty of Medicine, Ankara, Turkey Maria de Fatima de Pina Serviço de Higiene e Epidemiologia, Faculdade de Medicina da Universidade do Porto, Porto, Portugal Divisão de Biomateriais, Instituto de Engenharia Biomédica – INEB, Universidade do Porto, Porto, Portugal Instituto de Saúde Pública da Universidade do Porto – ISPUP, Porto, Portugal
[email protected] Ana Isabel Ribeiro Serviço de Higiene e Epidemiologia, Faculdade de Medicina da Universidade do Porto, Porto, Portugal Divisão de Biomateriais, Instituto de Engenharia Biomédica – INEB, Universidade do Porto, Porto, Portugal Instituto de Saúde Pública da Universidade do Porto – ISPUP, Porto, Portugal
Contributors
Contributors
xiii
Carla S.P. van Rijswijk Department of Orthopaedics, Leiden University Medical Centre, The Netherlands Jari Salo Töölö Hospital, Helsinki University Hospital, Helsinki, Finland
[email protected] Michiel A.J. van de Sande Department of Orthopaedics, Leiden University Medical Centre, Leiden, The Netherlands
[email protected] Carlos Santos Serviço de Higiene e Epidemiologia, Faculdade de Medicina da Universidade do Porto, Porto, Portugal Divisão de Biomateriais, Instituto de Engenharia Biomédica – INEB, Universidade do Porto, Porto, Portugal Wolfgang Schneider Herz-Jesu Hospital Vienna, Vienna, Austria
[email protected] Haim Shtarker Department of Orthopaedic Surgery, Western Galilee Hospital, Naharia, Israel
[email protected] Klaus A. Siebenrock Department of Orthopaedic Surgery, Inselspital, University of Bern, Bern, Switzerland
[email protected] Alex Staubli AO Knee Expert Group, Privat Clinic Sonnmatt, Luzern, Schweiz
[email protected] Simon D. Steppacher Department of Orthopaedic Surgery, Inselspital, University of Bern, Bern, Switzerland
[email protected] Antonie H.M. Taminiau Department of Orthopaedics, Leiden University Medical Centre, Leiden, The Netherlands
[email protected] Moritz Tannast Department of Orthopaedic Surgery, Inselspital, University of Bern, Bern, Switzerland
[email protected] Carlos Torrens Hospital del Mar de Barcelona, Barcelona, Spain
[email protected] Gershon Volpin Department of Orthopaedic Surgery, Western Galilee Hospital, Nahariya, Israel
[email protected] Kristina Weimann-Stahlschmidt Orthopaedic Surgery, Heinrich-HeineUniversität Düsseldorf, Düsseldorf, Germany B. Westhoff Orthopaedic Surgery, Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany Frangiskos Xypnitos Konstantinopoulio Hospital, Agias Olgas 3-5, 14233 Nea Ionia-Athens
[email protected]
Part I General Orthopaedics, Basic Science and Technology
Bone Substitutes in Clinical Work Jari Salo
Bone Substitutes The problem of bone void has been known for a long time. Historical documents include e.g., the use of stone, wood, animal bones, corals and auto-, allo- and xenografts. Auto grafts or allografts still are regarded as the “gold standard” in the treatment of bone voids either in primary trauma, delayed bone grafting, non-unions, arthro-/spondylodesis or endoprosthesis surgery. Tumour surgery can have special indications where non-living material is preferred. Different indications also have different demands on fillers. One can create different kinds of classifications depending on each clinical situation, surgical hardware and the bone substitute used. The clinical goals for each procedure should be: (1) The final outcome is formation of good quality bone in desired extent, (2) The surgical procedure can most likely be done all at one operation, (3) The costs and morbidity of the procedure is tolerable. The structure, and handling, of bone substitutes varies largely. Some of the first generation tricalcium phosphate or hydroxyapatite-based materials are injectable, harden within the first day(s), and can be used in weight-bearing areas. Limited cohesion force can cause spreading of the material around the actual treatment site. Second generation materials are more easy to handle even in wet surroundings, but the principle in healing is the same. Limited sized voids are resorbed and remodelled in months or years whilst larger fillings risk being encapsulated and in that way
J. Salo Töölö Hospital, Helsinki University Hospital, PL 266, 00029 HUS, Helsinki, Finland e-mail:
[email protected]
become a dead tissue inside the bone. The newest materials still have the advantage of injectability, in addition they should become porous after injection. This is a property which is thought to support cell migration and growth of vessels inside large filling spaces. There also are extremely hard solutions, like old bone cement or a newer castor beanbased compound, for sites with a need for high compression strength. Other non-injectable materials include e.g. inorganic small porous particles, wedges or blocks. Materials vary from CaP/HA to bioactive glass, having differences in composition, microstructure or manufacturing methods. Depending on the product type they can have a limited to moderate compression strength. These materials are osteoconductive. Especially in this group it is important to estimate the surface area/volume ratio of the bone substitute. The ratio can have a remarkable effect on the remodelling speed and on tissue reaction at the filled site. Organic bone substitutes are mostly based on demineralised bone. They are commercially available as strips, putty, paste etc. They are of living origin and have a theoretical risk of transmitting diseases. After donor screening and heavy processing during demineralisation it can be assumed that this risk is far lower than in normal allografts. De-mineralised matrix-based products are osteoconductive, and some of them also have a limited osteo-inductive capacity. My personal opinion is that their most important feature is the wide [1], although mild, spectrum of natural growth factors (VEGF, IGF, BMPs etc.) promoting healing of mesenchymal tissues. Recombinant technology has opened a new era in osteoinduction. Although still very expensive, BMP-2 and BMP-7 are commercially available and can be used to kick-off bone formation in severe cases. There are several estimates and studies on the economical impact of these products if fracture healing can be achieved faster and more reliably. The risk is that molecules originally in high concentration are
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_1, © 2011 EFORT
3
4
J. Salo
rinsed away from the bone void area. The volume filled with the scaffold is limited, and the use of BMPs. has moved towards combination of recombinant BMPs. and allo- or autografts to fill larger defects. PMMA is still used in tumour surgery, and also as a spacer when a two-phase reconstruction is used. Then the initial, often critical size, bone defect is filled with PMMA and in the second operation at 4 – 5 weeks the biological membrane around the spacer is opened, re-filled with autograft (+/− bone substitutes) and preserved as a closed space for bone graft. The material itself (Fig. 1) has a remarkable effect on remodelling of the filled site. Bone substitute faces a healthy bone in which it should temporarily integrate closely enough to prevent fibroblast invasion. In optimal conditions material is then gradually resorbed and replaced by new osteoid and finally by mineralised bone (Fig. 2) If the surface area/ volume ratio is high, cells have a good possibility to rapidly remodel the bone substitute material. This can in some cases, and with some materials, be even undesirable if the resorption happens too fast or causes inflammatory reactions or local changes in pH. This kinds of ultra-porous materials are also limited in their compression strength. Totally solid material is the other end of the line, then a moderate to high compression strength is achieved, but the risk is that the final result is a dead piece inside living bone. Whether it is a risk or not can be discussed. The near future clearly offers some new combinations of familiar and novel materials and cell technology [2].
Rapid prototyping and manufacturing in large scale tissue defects, also in bone, can provide custom-made instrumentations [3] and scaffolds pre-loaded with cultured cells. These techniques are already available but the final clinical breakthrough is still to come.
Clinical Use of Bone Substitutes The conventional test setting often includes head-to-head comparison of a potential bone void filler against autograft. It has to be pointed out, however, that we then miss the other side of the coin. It cannot be assumed that one single graft would work in the same way in different patients, or even in different bones in the same patient. It is known from modern imaging techniques that some bones live with just a sufficient circulation and perfusion to keep the bone alive, even in young healthy patients. Combining smoking or other risk factors for circulation can turn this balance remarkably and lead to disturbance in bone regeneration. Seen from this aspect, we should remember that the normal reaction to bone fracture or void is proper healing. If this does not happen, we have some biological or mechanical problem. More attention should be paid to the environment in which the modern materials are inserted. An other interesting question is whether an autograft in patients over 75-years is sufficient. If we compare it to the bone graft in
Osteoconduction
Void filler ensures bony bridge formation in bony environment
d
eoi
ost
Fig. 1 The structure of bone void filler has a remarkable effect on mechanical strength and remodelling of the filled site
0, 2, 4, 6 wks
ler
d fil
Voi
ised
ral Mine
12 wks
e
bon
6 mths
5
Bone Substitutes in Clinical Work
Structure of void fillers
Surface area/volume Mechanical strength
Fig. 2 The role of osteoconductive bone void filler can be seen as a temporary scaffold preventing the invasion of fibroblasts to the bone defect area. Remodelling of the scaffold or autograft proceeds gradually, the degraded material being first replaced by
osteoid which then is mineralised to normal bone. This typically takes 200–220 days as a minimum, depending on the size and the properties of the filling material and on the function of the patient’s tissues
30-year old healthy patient it certainly is not of good enough quality, but is it worse than bone in the area where it should be grafted in such an elderly patient? Bone healing requires many other things than just proper scaffold or administration of local growth factors. Formation of bone and articular cartilage in adult skeleton share several features [4]. The relative amounts and time of appearance of different stimulants or inhibitors vary, but basically it can be generalised that the origin of cells and their biological surroundings is roughly the same. What does then cause formation of either bone or cartilage? Differentiation of these tissues is highly dependent on the pO2, perfusion and pH, along with the type of mechanical loading on the regeneration area. Continuous cyclical loading, low pH and low pO2 can turn bone formation towards non-union or cartilage formation (Fig. 3). One special, and often very complicated, question is filling of a bone void after deep infection. In these cases
laboratory tests can be clean, even cultures from biopsies can be negative, but still there is a risk of having a new infection if a large amount of foreign material is inserted in the bone to fill the cavity. The immune system can react very aggressively even without any living bacteria at the site. Toll-like receptors can recognise even some constructional components of bacteria, like lipopolysaccharides, and this can clinically mimic infection. The only bone substitute material at the moment showing antibacterial effects itself is bioactive glass. It has earlier been used in chronic sinusitis, but has now also successful according to preliminary data on post-infective bone defects [5]. Many patient-related factors have a known effect on fracture healing, e.g. smoking and some medications can disturb normal bone healing. Non-unions still are some of the most difficult bone voids to treat. It is not uncommon for one single fracture site which has been initially fixed in a reasonable position with stable fixation to need re-operation due to non-
6 Fig. 3 Bone grafts or substitutes are used in complex surroundings having partly known effects on regeneration. Much in this field is still to be discovered
J. Salo
Contact area
Stable Fixation Implants
Direct healing
No compression
Grafts
Mobilisation
No inflammation
Comminution
Scaffols
Weight bearing
Less scar
Gap
Active implants
Circulation
Patient rel probl
Soft tissues
Medicines
Periosteum
Nerves
Bone Healing
Nutrition
O2, pH, etc Scar
ExCorp stimulation
Cartilage
Vascular grafts
Bone
Bacterial Infection
Local factors
Cells Bone cells
Grafts
VEGF
BMP-2, BMP-7
Pericytes
MSCs
FGF
Coupling
Blood cells
Inflammatory cells
IGF
COX-2
Table 1 Some basic principles in selecting an appropriate bone void filler Clinical question
Mechanical properties
Biology
Price
Tibial plateau fractures
+++
+/−
++
Atrophic non-union
−
+++
?/−
Spondylodesis
−
++
++
Revision arthroplasties
+++
++
++
Intra-articular fractures
+++
+++
?/−
Bening cyst
−
+++
++
Old patient
?
?
?
Infection related defect
+/?
+++
?/−
union. In these cases it is good first to think what are the possible patient-related limitations or factors leading to impaired bone formation. We cannot overcome these limitations just by adding osteoconductive or osteo-inductive materials, both of which already were there prior to non-union in the form of osteogenic, host bone. It is also crucial that these additional materials or growth factors have cells to fill the scaffold or to be stimulated by the BMPs and other factors.
Product?
As mentioned earlier, every patient with a problem in bone formation has to be taken as a new clinical challenge, and every bone in that single patient should be thought of as an individual organ with it’s own circulation, function and environment. Some clinical problems are mentioned in Table 1 to stimulate thinking on how to select between different bone substitutes. There is not a single method to employ in all cases.
Bone Substitutes in Clinical Work
Conclusions/THM 1. A fracture is there to heal, but it can enlarge into a bone void – especially after repeated operations. 2. Autograft works well, but it can be successfully replaced with current bone void fillers but only if living cells are present. 3. Make revisions deep enough to get contact to healthy bone – applying dead material on dead bone will not work.
References 1. Bormann N, Pruss A, Schmidmaier G, Wildemann B (2009) In vitro testing of the osteoinductive potential of different bony allograft preparations. Arch Orthop Trauma Surg
7 2. Muschler GF, Nakamoto C, Griffith LG (2004) Engineering principles of clinical cell-based tissue engineering. J Bone Joint Surg Am 86:1541–1558 3. Björkstrand R, Tuomi J, Paloheimo M, Salo J, Lindahl J (2009) 3D-Digitalization of ankle movement and 3D-CAD method for patient specific external ankle support development and rapid manufacturing. 4th international conference on advanced research in virtual and rapid prototyping – VR@P 2009, Leiria Portugal, 06.10.2009-10.10.2009. Leiria, Portugal 2009, Taylor & Francis Group/CRC Press/ Balkema/Järj: Polytechnic institute of Leiria, pp 199–204 4. Caplan AI (1991) Mesenchymal stem cells. J Orthop Res 9:641–650 5. Lindfors NC, Hyvönen P, Nyyssönen M, Kirjavainen M, Kankare J, Gullichsen E, Salo J (2010) Bioactive glass S53P4 as bone graft substitute in the treatment of osteomyelitis. Bone 47(2):212–218
Epidemiology and Variability of Orthopaedic Procedures Worldwide Maria de Fatima de Pina, Ana Isabel Ribeiro, and Carlos Santos
v ariability in clinical criteria have been reported in Spain [19], despite another study from a hospital in Iowa which did not find such variability between surgeons [20]. Further increase Total hip arthroplasty (THA) and total knee arthroplasty in THA and TKA is predicted, making the projected increase (TKA) have been proven as efficacious and cost-effective more accentuated in TKA [21]. interventions in the treatment of osteoarthritis [1, 2]. Osteo Studies regarding THA revision and TKA revision arthritis remains as the main indication for those proce- trends have shown inconsistent results. Many studies dures [3–5], despite few Asian studies regarding THA reported that revision rates have been increasing [10, 13, reported otherwise [6, 7]. Also, THA and TKA are becom- 15, 22], even though some of them are not statistically siging safer, as mortality and complications rates, and length nificant and projections point that the revision burden is of stay in hospital decrease, despite the increase in co- expected not to increase [21]. On the other hand, Scandi morbidities in patients selected for those procedures [8, 9]. navian studies reported a decreasing in THA revision risk, The number of THA and TKA has been increasing [4, 5, mainly due to a decrease in aseptic loosening of both com8–16], but TKA rates have been increasing at a higher rate ponents [23]. The most common indications for THA revithan THA. As osteoarthritis affects more of the elderly, part sion are instability and/or dislocation, implant loosening of this increase may be explained by population ageing. and infection [23, 24], and for TKA are infection and There is a strong association between high body mass index implant loosening [25]. (BMI) and knee osteoarthritis [17]. As obesity becomes There are disparities in rates of THA and TKA between more prevalent, osteoarthritis rises and THA and TKA rates poorer and wealthier, with the wealthier populations showrise concomitantly. Moreover, association between high ing higher rates [10]. A study, including nine European body mass index and hip and knee arthroplasties has been Union members, showed that the mean cost of primary described [18]. Increase in THA and TKA can also be attrib- THA was 5,043 € in 2008, the type of implant and the ward uted to changes in criteria for selecting the patients for sur- cost being the most important cost-drivers. This study also gery. Better devices and better materials allow TKA to be showed that almost 80% of the explainable price variation increasingly performed in younger people [5, 9, 11], and between countries is explained by purchasing-power priaccount for the broadening of criteria and TKA rates increase. orities [26]. Not only criteria are broadening but also inter-hospital Arthroplasty register data can provide a crucial contribution for development of arthroplasties and quality control, allowing assessment of the number and epidemiology of procedures, rates of revision and corresponding causes M. de F. de Pina () of failure [27, 28]. The first arthroplasty register was creServiço de Higiene e Epidemiologia, Faculdade de Medicina da ated in Sweden in 1975. Since then, several national orthoUniversidade do Porto, Porto, Portugal and pedic societies have created their own arthroplasty registers Divisão de Biomateriais, Instituto de Engenharia Biomédica – and nowadays Sweden, Finland, Norway, Denmark, New INEB, Universidade do Porto, Porto, Portugal and Zealand, Hungary, Australia, Canada, Czech Republic, Instituto de Saúde Pública da Universidade do Porto – ISPUP, Romania, Slovakia, Slovenia, Portugal, Moldavia, Austria, Rua das Taipas nº 135, 4050-600 Porto, Portugal England and Wales have active arthroplasty registers. e-mail:
[email protected]
Introduction
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_2, © 2011 EFORT
9
10
Several other countries have projected, or have already established a pilot phase of arthroplasty registers [29]. Minimal datasets have been established by the European Federation of National Associations of Orthopaedics and Traumatology (EFORT). The European Arthroplasty Regis ter (EAR) is an EFORT project, created to co-ordinate the co-operation between the several Arthroplasty Registers in Europe. As far as we know there are no studies of geographical patterns of arthroplasty incidence rates between countries. Our goals are to analyze the worldwide geographic distribution of incidence rates of THA and TKA and to identify socio-economic and health determinants for such incidences.
Materials and Methods Data Procedures coded by the International Classification of Diseases, 9th Revision, Clinical Modification (ICD9-CM) for THA and for TKR, comprising both primary and revision procedures, were selected: codes 81.51 and 81.54 for THA and codes 81.53 and 81.55 for TKA. It was not possible to have disaggregated data for all the countries and therefore primary and revision procedures were analyzed together. Regarding THA, data from 31 countries were used while for TKA data from 28 countries were used. Data on knee arthroplasty procedures (number of inpatient cases in 2007, unless a different year is mentioned) for 23 countries – Australia (2006), Austria (2005), Belgium (2006), Canada, Denmark (2005), Finland, France, Germany, Hungary, Iceland, Ireland, Italy (2006), South Korea, Luxembourg, Mexico, Netherlands, New Zealand, Portugal, Spain, Sweden, Switzerland, United Kingdom and United States (2006) – and also for the 1990–2007 time interval, were obtained from the Organization of Economic-Cooperation and Development (OECD). For Romania [30], Czech Republic [31], Slovakia [32] and Norway [33], data about the number of knee arthroplasties came from National Annual Reports of the operating arthroplasty registers. Data from Slovenia was estimated based on the Valdoltra Hospital arthroplasties register, which accounts for 50% of all procedures in Slovenia. The database of the Hospital Admissions Authorization (Autorização de Internação Hospitalar – AIH), from the Health System of Brazil was used to identify primary and revision knee arthroplasty operations. The AIH is used
M.de F.de Pina et al.
nationwide in all public hospitals as well as in private hospitals that provide services to the national health system. In Brazil, the national system of health is universal and free for all the population, although, about 25% of the population above 40-years old has a private health insurance and goes to private hospitals. Hospital admissions, from private health insurances were not available to use in this study. Most data on hip arthroplasty procedures (number of inpatient cases in 2007, unless stated) also came from the OECD health data and was available for 26 countries – Australia (2006), Austria (2005), Belgium (2006), Canada (2006), Denmark, Finland, France, Germany, Greece (1999), Hungary, Iceland, Ireland, Italy, South Korea, Luxembourg, Mexico, Netherlands, New Zealand, Norway, Poland, Portugal, Spain, Sweden, Switzerland, United Kingdom and United States (2006). As mentioned above, for Romania, Czech Republic, Slovenia, Slovakia and Norway, data were retrieved from the Annual Reports of the national arthroplasty registers. From Brazil data refers to hospital admissions to public hospitals and from Slove nia data was estimated from the Valdoltra register of arthroplasties. Population data (denominator) was obtained from the European Commission Eurostat, U.S. Census Bureau, Statistics Canada, Statistics Mexico, Statistics of New Zealand and the UK Office for National Statistics. Socioeconomic data came from several sources: GINI coefficient, which measures the degree of inequality in the distribution of family income in a country, was obtained from CIA World Factbook [34]. Human Development Index (HDI) is a composite index that measures average achievement in three basic dimensions of human development – a long and healthy life, knowledge and a decent standard of living. It is produced annually by United Nations Development Programme (UNDP) and data used in the present work came from Human Development UNDP Statistics [35]. The following variables for OECD countries were selected: number of coxarthrosis and gonarthrosis hospital discharges; number of THA and TKA procedures; number of medical Doctors per 100,000 inhabitants; Number of medical consultations per capita; Number of hospital beds per 1,000 inhabitants; Perception of health system as good or very good by the population(%); Public Current Expenditure on Health per capita (US$); % of Public Expenditure on Health compared to the Total Expenditure on Health; Investment on Medical Facilities (% of the total current expenditure on health); Total Expenditure on Health, % of Gross Domestic Product; Total Expenditure
11
Epidemiology and Variability of Orthopaedic Procedures Worldwide
on Health per capita (US$); Percentage of overweight and obesity and Age-Standardized Prevalence of diabetes (%). From the EuroStat. and other official statistic sources the percentage of population over 65 years and the percent age of women among the population over 65 years were calculated.
Statistical Analysis In order to allow the comparison between countries, the Age-Standardized Incidence Rates (ASIR) for THA and TKA were computed using the indirect method and England and Wales (2009) as the reference population. This method comprises the calculating of the ratio between the observed cases and the expected cases, if the population of study had the same cases distribution of the reference population. This ratio is known has Standard Morbidity (or Mortality) Ratio – SMR and values above 100% indicate an excess of risk, while values under 100% indicate a lower risk than the reference population. The indirect method has the disadvantage of not being appropriate to compare between areas: rather, the comparison has to be done by pairs, each area being compared with the reference population. That’s the reason for being called the “indirect” method. Although less usual than the direct standardization, the indirect method is the alternative to be used when the number of cases for each age-group in the study areas is not available. In the present study, 2009 data from National Joint Registry (NJR) of England and Wales was chosen as standard population because it provides data with good quality from both National Health Service and private health-care sector. Furthermore, the National Joint Registry (NJR) of England and Wales allows extraction data by 10-year age-groups (<45; 45–54; 55–64; 65–74; 75–84; >85) in order to calculate the indirect standardization. Additionally, England and Wales have a quite numerous population, comprising 89% of all UK inhabitants in 2009. Data were selected from the 7th Annual Report of the NJR [36]. The ASIR of THA and TKA were calculated for the most recent available data. The annual percentage change in the number of arthroplasty procedures between 2000 and 2007 were estimated by linear regression. Multiple regression analysis was used to determine which variables were related to the ASIR of THA and TKA (the dependent variables). Sample size in each variable analysis varied due to missing data. A p-value < 0.05 was considered statistically significant. Geographical Information Systems (GIS) and Spatial Statistical techniques were used to analyse the data and
map the results. The Moran Index (I) was computed in order to measure the spatial autocorrelation, that is, the correlation between incidence rates in different countries [37]. First-order neighbourhood relation was defined by the sharing of common boundaries between countries. Moran’s I is a global indicator of auto-correlation and provides a single value for all the set of data. The interpretation of such an index is similar to the interpretation of the r in a linear correlation. If it is close to zero, it means that there is no auto-correlation, and the events occur randomly in space, otherwise, if it’s close to 1 or −1 it means there is a strong (positive or negative) auto-correlation and indicates that there is a spatial dependency in the occurrence of the events, meaning that what happens in one country is correlated with what happens in the surrounding countries and the events do not occur randomly. However, when dealing with large areas, it is possible that different spatial associations occur; therefore, one single value would not represent the underlying patterns. To deal with these different spatial associations the local Moran Index, known as Local Index of Spatial Auto-correlation – LISA was calculated [38]. The LISA indicates the presence of spatial dependence in some areas, that is, areas where the incidence rates are significantly correlated with the incidence rates of their neighbours. Statistical analysis was completed using the SPSS v17.0 for multiple regression analysis and GeoDa 0.9.5-i for spatial statistical analysis. ArcMap v9.3 was used to map the results.
Results The study included 31 countries with a total population of 1,197,214,619 persons and 1,422,046 THA, corresponding to a crude incidence rate in 2007 of 118.8 (118.4–119.2) per 100,000 persons-year. Regarding the TKA, the 28 countries included in the study had 1,198,148 TKA, corresponding to a crude rate of 104.3 (103.9–104.7) per 100,000 persons-year. Age-Standardized Incidence Rates (ASIR) and Standard Morbidity Ratios (SMR) for THA and TKA are presented in Table 1. Strong geographic disparities were observed. In Europe, the ratio between the highest and lowest ASIR (95% CI), per 100,000 inhabitants, was H:L = 7.5, with Austria [266.2 (269.7–273.3)] having the highest and Romania [35.4 (36.3–37.1)] the lowest ASIR. For TKA the extremes were again between Austria [183.6 (186.5– 189.5)] and Romania [5.3 (5.6–5.9)] but disparities were
12
M.de F.de Pina et al.
Table 1 Age-standardized incidence rates (ASIR) for total hip and knee arthroplasties Country
Total hip arthroplasties – THA/100,000 persons-year
Total knee arthroplasties – TKA/100,000 persons-year
SMR (95% CI)
ASIR (95% CI)
SMR_TKA
Australiaa
154.8 (156.5–158.3)
172.9 (174.8 –176.8)
136.3 (137.9 –139.4)
171.3 (173.2–175.2)
Austria
238.3 (241.5–244.7)
266.2 (269.7–273.3)
146.1 (148.5–150.8)
183.6 (186.5 –189.5)
Belgiuma
199.2 (201.7–204.2)
222.5 (225.3–228.1)
118.8 (120.6 –122.4)
149.2 (151.5–153.8)
Canadac
113.3 (114.4 –115.5)
126.5 (127.8–129.0)
114.8 (115.9 –116.9)
144.3 (145.6 –146.9)
b
Czech Republic
81.9 (83.6–85.3)
91.5 (93.3–95.3)
–
–
Denmarke
175.5 (178.8–182.2)
196.0 (199.7–203.5)
83.4 (85.6 –87.9)
104.8 (107.6 –110.4)
Finland
149.4 (152.4 –155.5)
166.9 (170.2–173.7)
125.0 (127.6 –130.2)
157.0 (160.3–163.7)
France
192.3 (193.3–194.3)
214.8 (215.9 –217.1)
85.9 (86.6 –87.2)
108.0 (108.8–109.6)
Germany
215.7 (216.6 –217.5)
240.9 (241.9 –242.9)
131.7 (132.4 –133.0)
165.5 (166.3–167.1)
a,d
Greece
53.5 (54.8–56.2)
59.8 (61.3–62.7)
–
–
Hungary
77.0 (78.6 –80.2)
86.0 (87.8–89.6)
31.3 (32.3–33.3)
39.4 (40.6– 41.9)
Iceland
160.4 (175.8–192.3)
179.1 (196.3–214.7)
99.1 (110.6 –123.2)
124.5 (139.0 –154.8)
f
Ireland
152.3 (156.3–160.4)
170.1 (174.6 –179.2)
45.8 (48.0 –50.2)
57.6 (60.3– 63.0)
Italyg
116.4 (117.2–117.9)
130.0 (130.9 –131.7)
60.3 (60.8– 61.3)
75.8 (76.4 –77.1)
South Korea
16.7 (17.1–17.5)
18.7 (19.1–19.5)
78.2 (79.0 –79.8)
98.3 (99.3–100.3)
Luxembourg
202.3 (215.2–228.7)
226.0 (240.3–255.4)
127.8 (137.5–147.7)
160.6 (172.8 –185.6)
Mexico
14.2 (14.6 –14.9)
15.9 (16.3–16.6)
6.2 (6.4 – 6.6)
7.8 (8.0 –8.3)
Netherlands
188.9 (191.0 –193.0)
211.0 (213.3–215.6)
98.0 (99.4 –100.8)
123.1 (124.8 –126.6)
New Zealand
149.9 (153.8 –157.7)
167.4 (171.7–176.1)
90.5 (93.4 – 96.3)
113.7 (117.3–121.0)
Norwayd
182.7 (186.5–190.4)
204.1 (208.3–212.6)
68.6 (70.8 –73.0)
86.2 (88.9 – 91.8)
Poland
32.6 (33.2–33.8)
36.4 (37.1–37.7)
–
–
Portugal
67.4 (68.9 –70.4)
75.3 (76.9 –78.6)
34.0 (34.9 –35.9)
42.7 (43.9 – 45.2)
Sloveniah
80.5 (76.8–84.2)
89.9 (85.8–94.1)
44.8 (42.3– 47.5)
56.3 (53.1–59.7)
Slovak Republic
88.6 (91.2–93.9)
99.0 (101.9 –104.9)
22.7 (23.9 –25.2)
28.5 (30.1–31.7)
Spain
86.0 (86.9–87.7)
96.1 (97.0 –97.9)
80.6 (81.3– 82.1)
101.3 (102.2–103.1)
Sweden
171.1 (173.6 –176.1)
191.1 (193.9–196.7)
80.5 (82.1– 83.7)
101.1 (103.1–105.2)
Switzerland
202.2 (205.2–208.3)
225.8 (229.2–232.6)
138.1 (140.5 –142.9)
173.6 (176.5 –179.5)
United Kingdom
165.5 (166.5–167.4)
184.8 (185.9–187.0)
109.1 (109.9 –110.6)
137.1 (138.0 –139.0)
United States
171.9 (172.4 –172.9)
192.0 (192.5–193.1)
176.3 (176.7–177.2)
221.5 (222.0 –222.6)
Brazild
10.1 (10.3–10.5)
11.3 (11.5–11.7)
3.4 (3.5–3.6)
4.3 (4.4 – 4.5)
31.7 (32.5–33.2)
35.4 (36.3–37.1)
4.2 (4.5– 4.7)
5.3 (5.6 –5.9)
a
Romania
d
Last available year 2006 Last available year 2005 c Last available year for THA 2006 d Other sources (not OECD data) e Last available year for TKA 2005 f Last available year 1999 g Last available year for TKA 2006 h Estimated from Valdoltra Arthroplasties Register. Doesn’t include revisions a
b
ASIR (95% CI)
13
Epidemiology and Variability of Orthopaedic Procedures Worldwide Legend SMR - Total Hip Arthroplasties 50% and below
Legend Age Standardized Incidence Rate Total Hip Arthroplasties
50% - 100%
1st quintile
100%
2nd quintile
100% - 150%
3rd quintile
150% - 200%
4th quintile
200% and above
5th quintile
Legend Total Knee Arthroplasties SMR 50% and below 50% - 100%
Legend
Age Standardized Incidence Rate Total Knee Arthroplasties
1st quintile 2nd quintile
100% 100% - 150%
3rd quintile
150% and above
5th quintile
4th quintile
Fig. 1 Standard morbidity rate (SMR) and age-standardized incidence rates (ASIR), of total hip and knee arthroplasties worldwide in 2007
much more accentuated, being the ratio H:L = 34.6. Figure 1 presents the geographical distribution of ASIR and SMR of THA and TKA, worldwide in 2007. A closer look at age-standardized incidence rates in Europe is showed in Figs. 2 and 3, respectively for THA and TKA where a cluster of countries with high ASIR seems to emerge. The significance of such spatial clusters were tested with the Local Index of Spatial Autocorrelation – LISA, and confirmed only for THA (Fig. 4), with six countries of central-north Europe aggregating with higher ASIR of THA. For ASIR of TKA there were no significant
spatial clusters. The global index of spatial autocorrelation, Moran Index (I) was moderate for THA (0.26) and null for TKA (0.0043). To develop the regression analysis, a correlation matrix (Table 2) was prepared, regarding the ASIR of THA and TKA, and the selected variables. Statistically significant correlations were observed mainly with economic and macro-economic variables. Using the enter method, significant models emerged for THA (F7.42 = 11.737 p < 0.0005, adjusted R2 = 0.782) and for TKA (F5.16 = 8.702 p < 0.0005, adjusted R2 = 0.647).
14
M.de F.de Pina et al.
Fig. 2 Age-standardized incidence rates of total hip arthroplasties (THA) in Europe (2007)
After removing independent variables that were highly correlated among them, the determinant variables for ASIR of THA were: Growing of the number of THA between 2000 and 2007; number of medical consultations per capita; % of population recognizing the health system as good or very good (p < 0.05); Total Expenditure on Health, related to % of Gross Domestic Product; % of people with 65 and more years old (p < 0.05); Human Development Index; and Gross Domestic Product of 2007 (p < 0.05). The coefficients of the model are presented in Table 3. Regarding the ASIR of TKA, the determinant variables were: % of population perceiving the health system as good or very good; Human Development Index 2007; Gross Domestic Product of 2007; % of people with 65 and more years old; and Total Expenditure on Health , per capita (US$). The coefficients of this model are presented in Table 4.
Discussion In this study we aimed to look for determinants of the incidence of hip and knee arthroplasties worldwide, that could explain the strong geographic inequalities. We wish to understand if the inequalities were related to differences in health risks among countries, for instance, differences in the percentage of elderly people, overweight and obesity, incidence of osteoarthrosis or other health-related variables. A set of several variables covering health, demographic, economic and social aspects were selected to define the model that better explains the variability of arthroplasty procedures worldwide. Contrary to that described in other studies, obesity and incidence of osteoarthritis (measured as number of hospital discharges of patients with coxarthrosis and gonarthrosis) were not determinants of the incidence of THA and TKA.
15
Epidemiology and Variability of Orthopaedic Procedures Worldwide
Fig. 3 Age-standardized incidence rates of total knee arthroplasties (TKA) in Europe (2007)
The disparities between poorer and wealthier, [10] seem to be the better explanation for the high variability among ASIR between countries, being the economic variables those who presented the highest Pearson’s correlation coefficients (Table 2). Quality of health systems measured as the proportion of the population who declared it to be good or very good, as well as the public investments in health systems, were determinants of the incidence of arthroplasties, both for the THA and TKA. However, variables such as Investment on Medical Facilities, number of medical doctors per hundred thousand inhabitants, number of medical consultations per capita and number of hospital beds per thousand inhabitants are not associated with the incidence of arthroplasties. The Moran’s I coefficients for spatial auto-correlation presented moderate values for ASIR of THA, showing a positive spatial auto-correlation, meaning that in nearby countries the incidence of THA tends to have similar
values. However, for the incidence of TKA no spatial autocorrelation was found. The statistically significant local spatial auto-correlation (Fig. 4) showed that one cluster of six countries with higher incidence rates of THA is occurring in central Europe. The highest incidence of such arthroplasties is related to higher Gross Domestic Product and Human Development Index. We believe that the disparities encountered in the geographical distribution of age-standardized incidence rates of THA and TKA are not related to differences in risk, rather reflects differences in health priorities. Arthroplasties help to reduce pain and improve the quality of life of patients. They are not surgeries to “save lifes” and health systems in less rich countries may have other priorities. Nevertheless, future studies need to be done to test this hypothesis. This study has some limitations, mainly related to the availability of data. For countries which are non-members of the OECD, the variety of social, economic and health
16
M.de F.de Pina et al.
Fig. 4 Spatial clusters of age-standardized incidence rates for total hip arthroplasties
variables was much more limited than for the OECD countries. The last available data was not the same for all the variables, and when 2007 data were not available, we used data from 2006 or 2005. If data were from previous years we considered that were not available, except in what respects to number of arthroplasties in Greece (last available data was from 1999).
Our study brings a new contribution to the knowledge of the incidence of arthroplasties worldwide. The determinants for such procedures seemed to be oriented by economic aspects, rather than health needs, with wealthier countries having a better performance. It would be interesting to evaluate the medical protocols for indications for an arthroplasty in the different countries.
Table 2 Correlation matrix between analysed variables, THA and TKA Age standardized incidence rate of THA
Age-standardized incidence rate of TKA
0.292
0.420
Number of coxarthrosis hospital discharges
Pearson correlation Sig. (2-tailed)
0.148
0.041
Number of gonarthrosis hospital discharges
Pearson correlation
0.208
0.476
Sig. (2-tailed)
0.307
0.019
Number of THA procedures
Pearson correlation
0.276
0.459
Sig. (2-tailed)
0.133
0.014
Pearson correlation
0.177
0.463
Sig. (2-tailed)
0.369
0.013
Number of TKA procedures
17
Epidemiology and Variability of Orthopaedic Procedures Worldwide Table 2 (continued) Age standardized incidence rate of THA Growing in the number of THA between 2000 and 2007 Growing in the number of TKA between 2000 and 2007 Medical doctors per 100,000 inhabitants
Pearson correlation Sig. (2-tailed) Pearson correlation
− 0.447
Age-standardized incidence rate of TKA 0.144
0.025
0.502
− 0.365
0.009
Sig. (2-tailed)
0.087
0.968
Pearson correlation
0.284
− 0.087
0.159
0.694
− 0.315
− 0.240
Sig. (2-tailed)
0.133
0.282
Number of hospital beds per 1,000 inhabitants
Pearson correlation
0.043
0.102
Sig. (2-tailed)
0.840
0.651
Perception of health system as good or very good (%)
Pearson correlation
0.663
0.515
Sig. (2-tailed)
0.000
0.014
Public Current Expenditure on Health per capita (US$)
Pearson correlation
0.841
0.524
Sig. (2-tailed)
0.000
0.009
% of Public Expenditure on Health compared to Total Expenditure on Health
Pearson correlation
0.495
0.055
Sig. (2-tailed)
0.010
0.804
Investment on Medical Facilities (% of the total current expenditure on health)
Pearson correlation
0.029
0.170
Sig. (2-tailed)
0.889
0.448
Total Expenditure on Health, % of Gross Domestic Product
Pearson correlation
0.535
0.616
Sig. (2-tailed)
0.003
0.001
Total Expenditure on Health per capita (US$)
Pearson correlation
0.746
0.783
Sig. (2-tailed)
0.000
0.000
− 0.098
0.021
0.634
0.926
Number of medical consultations per capita
Sig. (2-tailed) Pearson correlation
Percentage of overweight and obesity
Sig. (2-tailed)
Age-Standardized Prevalence of diabetes (%)
− 0.265
0.063
Sig. (2-tailed)
0.173
0.767
% of population with 65 and above years
Pearson correlation
0.434
0.344
% of women, among population with 65 and more years
Pearson correlation
Life expectancy GINI Index Human Development Index Gross Domestic Product
Pearson correlation Pearson correlation
0.015
0.073
− 0.177
− 0.133
Sig. (2-tailed)
0.340
0.500
Pearson correlation
0.654
0.681
Sig. (2-tailed)
0.000
0.000
Sig. (2-tailed)
− 0.477
− 0.289
Sig. (2-tailed)
Pearson correlation
0.007
0.136
Pearson correlation
0.711
0.716
Sig. (2-tailed)
0.000
0.000
Pearson correlation
0.777
0.521
Sig. (2-tailed)
0.000
0.008
18
M.de F.de Pina et al.
Table 3 Coefficients from the regression analysis of THA (R2 = 0.782) Variables
Standardized beta
(Constant) Growing of the number of THA between 2000 and 2007
−.074
t
p-value
−1.183
0.256
−.518
.613
Number of medical consultations per capita
.240
1.376
.191
% of population perceiving the health system as good or very good
.556
2.808
.014
Total Expenditure on Health, related to % of Gross Domestic Product
.107
.765
.457
.397
2.602
.021
Human Development Index 2007
% of people with 65 and more years old
−.172
−.689
.502
Gross Domestic Product of 2007
.542
2.816
.014
t
p-value
Table 4 Coefficients from the regression analysis of TKA (R2 = 0.647) Variables
Standardized beta
(Constant) % of population perceiving the health system as good or very good
.098
−2.209
0.042
.549
.590
Human Development Index 2007
.379
1.765
.097
Gross Domestic Product of 2007
− .363
−1.653
.118
% of people with 65 and more years old
.136
.916
.373
Total Expenditure on Health, per capita (US$)
.704
3.104
.007
References 1. Faulkner A, Kennedy LG et al (1998) Effectiveness of hip prostheses in primary total hip replacement: a critical review of evidence and an economic model. Health Technol Assess 2(6):1–133 2. Kane RL, Saleh KJ et al (2003) Total knee replacement. Evidence Reports/Technology Assessment No. 86, Agency for Healthcare Research and Quality, Rockville, pp 1–8 3. Allepuz A, Serra-Sutton V et al (2008) Hip and knee arthroplasties in Catalonia [Spain] from 1994 to 2005. Gac Sanit 22(6):534–540 4. Ingvarsson T, Hagglund G et al (1999) Incidence of total hip replacement for primary osteoarthrosis in Iceland 1982– 1996. Acta Orthop Scand 70(3):229–233 5. Mehrotra C, Remington PL et al (2005) Trends in total knee replacement surgeries and implications for public health, 1990–2000. Public Health Rep 120(3):278–282 6. Lai Y-S, Wei H-W et al (2008) Incidence of hip replace ment among national health insurance enrollees in Taiwan. J Orthop Surg Res 3:42 7. Liu YE, Hu S et al (2009) The epidemiology and surgical outcomes of patients undergoing primary total hip replacement: an Asian perspective. Singapore Med J 50(1):15–19 8. Liu SS, Valle AGD et al (2009) Trends in mortality, complications, and demographics for primary hip arthroplasty in the United States. Int Orthop 33(3):643–651
9. Memtsoudis SG, Valle AGD et al (2009) Trends in demographics, comorbidity profiles, in-hospital complications and mortality associated with primary knee arthroplasty. J Arthroplasty 24(4):518–527 10. Dixon T, Shaw M et al (2004) Trends in hip and knee joint replacement: socioeconomic inequalities and projections of need. Ann Rheum Dis 63(7):825–830 11. Jain NB, Higgins LD et al (2005) Trends in epidemiology of knee arthroplasty in the United States, 1990–2000. Arthritis Rheum 52(12):3928–3933 12. Kim HA, Kim S et al (2008) The epidemiology of total knee replacement in South Korea: national registry data. Rheu matolology (Oxford) 47(1):88–91 13. Kurtz S, Mowat F et al (2005) Prevalence of primary and revision total hip and knee arthroplasty in the United States from 1990 through 2002. J Bone Joint Surg Am 87(7):1487–1497 14. Ostendorf M, Johnell O et al (2002) The epidemiology of total hip replacement in the Netherlands and Sweden: present status and future needs. Acta Orthop Scand 73(3):282–286 15. Pedersen AB, Johnsen SP et al (2005) Total hip arthroplasty in Denmark: incidence of primary operations and revisions during 1996–2002 and estimated future demands. Acta Orthop 76(2):182–189 16. Wells VM, Hearn TC et al (2002) Changing incidence of primary total hip arthroplasty and total knee arthroplasty for primary osteoarthritis. J Arthroplasty 17(3):267–273
Epidemiology and Variability of Orthopaedic Procedures Worldwide 17. Manek NJ, Hart D et al (2003) The association of body mass index and osteoarthritis of the knee joint: an examination of genetic and environmental influences. Arthritis Rheum 48(4):1024–1029 18. Wendelboe AM, Hegmann KT et al (2003) Relationships between body mass indices and surgical replacements of knee and hip joints. Am J Prev Med 25(4):290–295 19. Cobos R, Latorre A et al (2010) Variability of indication criteria in knee and hip replacement: an observational study. BMC Musculoskelet Disord 11:249 20. Herickhoff PK, Callaghan JJ et al (2010) Primary hip and knee replacement: “are we all operating on the same patients, even at the same institution?”. Iowa Orthop J 30:109–114 21. Kurtz S, Ong K et al (2007) Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 89(4):780–785 22. Khatod M, Inacio M et al (2008) Knee replacement: epidemiology, outcomes, and trends in Southern California: 17,080 replacements from 1995 through 2004. Acta Orthop 79(6):812–819 23. Fevang BT, Lie SA et al (2010) Improved results of primary total hip replacement. Acta Orthop 81(6):649–659 24. Bozic KJ, Kurtz SM et al (2009) The epidemiology of revision total hip arthroplasty in the United States. J Bone Joint Surg Am 91(1):128–133 25. Bozic KJ, Kurtz SM et al (2010) The epidemiology of revision total knee arthroplasty in the United States. Clin Orthop Relat Res 468(1):45–51 26. Stargardt T (2008) Health service costs in Europe: cost and reimbursement of primary hip replacement in nine countries. Health Econ 17(1 Suppl):S9–S20 27. Labek G, Janda W et al (2010) Organisation, data evaluation, interpretation and effect of arthroplasty register data on the outcome in terms of revision rate in total hip arthroplasty. Int Orthop 35(2):157–163
19 28. Serra-Sutton V, Allepuz A et al (2009) Arthroplasty registers: a review of international experiences. Int J Technol Assess Health Care 25(1):63–72 29. Labek G (2009) Handbook for the development and operation of an outcome register for medical devices [Available from: http://www.ear.efort.org/getdoc/932a6677-3167-461580f5-2b3c5df9c824/HandbookRegisterdevelopmentfinal. aspx] 30. (2010) Raport 2010 – Endoprotezarea de Sold in perioada 2003–2009, Registrul National de Endoprotezare [Available from: http://www.rne.ro/rnemedia/download/RNE_Raport_ 2010_Sold.pdf] 31. (2006) National Register of Joint Replacement, Coordination center for departmental medical information systems. Czech Rep. Arthroplasty Register. [Available from: http://www. ksrzis.cz/] 32. Nečas L, Katina S et al (2010) Six years of Slovakian arthroplasty register, 2003–2008 data analysis. 11th EFORT Congress Madrid 2010. [Available from: https://sar.mfn.sk/] 33. Bergen HF, H (2009) Annual Report 2009, Norway, Department of Orthopaedic Surgery, Haukeland University Hospital, Bergen [Available from: http://www.haukeland. no/nrl/] 34. https://www,cia,gov/library/publications/the-world-factbook/fields/2172,html, Retrieved Dec 2010 35. http://hdrstats,undp,org/en/indicators/49806,html, Retrieved Dec 2010 36. National Joint Registry (NJR) (2010) 7th annual report, surgical data to December 2009, National Joint Registry for England and Wales 37. Bailey T, Gatrell A (1995) Interactive spatial data analysis. Longman, London 38. Anselin L (1995) Local indicators of spatial association — LISA. Geogr Anal 27(2):93–115
Part II Bone and Joint Tumours
Cartilage – Forming Bone Tumours Antonie H.M. Taminiau, Judith V.M.G. Bovée, Carla S.P. van Rijswijk, Hans A.J. Gelderblom, and Michiel A.J. van de Sande
Introduction
Benign Cartilage Tumours
Cartilaginous tumours form the second largest group of primary bone tumours. They all share the characteristic of production of chondroid matrix by tumour cells. Cartilage tumours range from completely benign lesions to highly malignant and are sub-divided by location in intramedullary, central and surface or peripheral sites. The WHO classification describes benign cartilaginous bone tumours as being: osteochondroma, enchondroma, periosteal chondroma, chondromyxoid chondroma and chondroblastoma. Multiple chondromatosis is described as the only benign cartilaginous joint lesion [1]. The malignant cartilage bone tumour is called chondrosarcoma. They are known to occur as primary bone tumours (de novo) or as a malignant degeneration of pre-existing benign cartilaginous bone tumour (secondary). Four histological sub-types of chondrosarcoma are described: peripheral-, dedifferentiated-, mesenchymal- and clear-cell chondrosarcoma [1, 2]. The clinical presentation of these tumours differs according to site, location, age group and gender. Clinical features, radiological appreciation, histology and treatment protocols are therefore described separately for both benign and malignant cartilaginous tumours. Epidemiology data are based on the Netherlands Committee of bone tumours database.
Benign cartilage tumours arise primarily in young age groups, with a peak incidence between the second and third decade. Osteochondroma and enchondroma occur most frequently as single lesions, but can sometimes be found in multiple skeletal locations as part of rare hereditary syndromes such as Multiple osteochondroma MHE, Ollier’s disease and Maffucci’s syndrome.
A.H.M. Taminiau () Department of Orthopaedics, Leiden University Medical Centre, Albinusdreef 2, Leiden 2300 RC, The Netherlands e-mail:
[email protected]
Osteochondroma Osteochondroma (cartilaginous exostosis) is the most frequently found benign bone neoplasm (35%), is located in the metaphysis and is thought to derive from the epi-metaphysis in the long bones. This tumour is seldom diagnosed before the age of 10 as the median age diagnosis averages around the second decade of life, They increase in size with time and are often present for a considerable time before causing complaints. Growth will continue during skeletal growth and stops when the epiphyseal growth-plates have fused. Osteochondroma can occur in any site but are most frequent around the knee, wrist, hip, shoulder for the long bones and pelvis, scapula and ribs for the flat bones (Fig. 1). Most osteochondroma will present as single asymptomatic lesions. However size and location may cause functional restrictions, bursitis, tendonitis, pressure on adjacent neurovascular structures and cosmetic discomfort. Fifteen percent of diagnosed osteochondroma exist as part of a syndrome where at least two osteochondroma in one long bone are diagnosed. They often exist as part of an autosomal dominant disorder but can occur de novo (multiple hereditary osteochondroma (MHE)). In these patients addition skeletal deformities especially around the knee, ankle and wrist joints are frequently found (Fig. 2).
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_3, © 2011 EFORT
23
24
A.H.M. Taminiau et al.
Fig. 1 Osteochondroma. (a) Radiograph shows an osteochondroma originating from the thoracic surface of the scapula with a regular outer surface. (b) Axial T2-weighted MR image shows a large neo-bursa
Clinical Presentation Osteochondroma can be palpated as a hard mass consisting of bony stalk fixed to the bone that is covered by a cartilaginous cap. They are seldom painful, but can be covered by a painful bursa. Malignant transformation for single lesions is very rare (1%), but somewhat more frequent 1–5% in multiple osteochondroma [3, 4]. Flat bones such as the pelvis and scapula also have a higher risk for malignant degeneration. Growth of an osteochondroma after skeletal maturity suggests malignant transformation as does pain at rest and can suggest malignancy. Complications that can occur include nerve and vascular compression, pseudo-aneurysm formation, fracture of the stalk and growth disturbances.
MR imaging features of osteochondroma are characteristic with normal-appearing bone marrow and cortex extending into the lesion with a thin overlying cartilage cap demonstrating high signal intensity on T2-weighted MR images. Bone scintigraphy will show increased activity in the presence of both persistent enchondral ossification and malignant change. However, normal background activity over a symptomatic osteochondroma excludes malignancy. In malignant degeneration, the thickness of the cartilage cap on MR imaging is often more than 1 cm. and a typical ring and arcs configuration with this cartilage cap can be appreciated.
Differential Diagnosis Radiology The radiographic appearance of osteochondroma using conventional radiography is usually diagnostic. Osteochondromata are metaphyseal lesions projecting away from the adjacent joint. They can be pedunculated or broad-based (sessile) and vary in size. On radiography, there is continuity of the cortex and bone marrow of the underlying bone into the osteochondroma, which helps to differentiate them from other surface bone lesions.
The diagnosis of an osteochondroma is generally not difficult but should be differentiated from surface lesions such as parosteal osteosarcoma.
Histology (Macro- and Micro-) Osteochondroma consists of a bony stalk, usually consisting of mature lamellar bone, covered by a cartilaginous cap (Fig. 3). The interface between bone and cartilage strongly
25
Cartilage – Forming Bone Tumours
a
b
c
Fig. 2 Osteochondromatosis. (a) Radiographs of the pelvis, knee and ankle of a child show multiple osteochondromas in the metaphyses and metadiaphysis of the proximal femur on the left and around the knee and ankle. Due to the osteochondromas normal remodeling of the metaphyses cannot take place giving them a broad plump shape. (b) Radiograph of chest shows
multiple osteochondromas. (c) Additional CT is helpful in anatomically complex regions and in anatomically complex osteochondroma to show the extension of the medullar cavity and the cortex of the underlying bone into the osteochondroma, which is the key feature for the differential diagnosis
resembles the growth-plate, in which resting, proliferating and hypertrophic chondrocytes are recognized, although cells are less organized. When the growth-plates are still open, the cartilage can be cellular. After closure of the growth-plate the cap becomes less cellular. Nuclear atypia and mitoses are absent.
Genetics EXT1 and EXT2 Within the context of the multiple osteochondromas (MO) syndrome (previously known as multiple hereditary exostoses MHE). MO is autosomal dominantly inherited and caused by germline mutations in the EXT1 or -2 genes Stickens [5].
26
a
A.H.M. Taminiau et al.
b
Fig. 3 Histology (macro and micro). (a) Osteochondroma transversal section showing trabeculated bone covered by a cartilaginous cap of limited size and a thin layer of periostium. (b) The
outer line of perichondrium is covering the cartilaginous cap. The chondrocytes are organized into cords and show a variable amount of enchondral ossification
Treatment
The enchondromata make up about 10–25% of all benign bone tumours. The are found in all age groups ranging from 5 to 80, most frequently in the third to fifth decade of life. The ratio m ale-to-female is equal (1/1). Enchondromas are especially common in small tubular bones of the hand (50%) and feet. Other sites where enchondromata occur more frequently are long tubular bones such as the femur, humerus and tibia. Enchondromata occur in ribs but are (rarely) seen in other flat bones such as the pelvis.
En bloc resection of osteochondroma including the pseudocapsule is curative and results in very low recurrence rate. Reconstruction using either bone or osteosynthesis is seldom necessary as the structural integrity of long bones is not changed much with resection. Asymptomatic, osteochondroma without discomfort can be left untreated. Indications for surgery are functional discomfort, severely cosmetic disabling lesions, joint or bone deformity and a growing lesion after growth has stopped. In young children care should be taken not to damage the growthplate. Lesions left untreated can be followed-up using plain radiographs after skeletal maturity and some years thereafter. MO/MHE patients should be screened using either total body MRI or PET- or bone scintigraphy, for undiagnosed osteochondromata located around the pelvis, scapula, ribs and vertebral column. Patients should seek medical consultation if the osteochondroma starts growing again or becomes painful at rest. Adequately resected cases do not need long-term follow-up.
Clinical Presentation Clinically enchondromata of the hand can show some swelling or present with some pain due to small fractures (usually non-displaced) and can be appreciated on standard radiographs. In long bones enchondromata are mostly asymptomatic and are often a coincidental finding on plain radiographs. Malignant transformation of enchondroma towards chondrosarcoma (usually low-grade) in the long bones including the metacarpal bones infrequently occurs in less than 1–5% of cases. In the phalanges malignant transformation is very exceptional.
Enchondroma Enchondroma is a benign tumour of hyaline cartilage originating from the medulla of bone. The majority of lesions are solitary but they can occur multiply as a manifestation of a congenital syndrome (M. Ollier, Maffucci).
Radiology The typical radiographic appearance is a lytic lesion arising in the medulla with a geographic pattern of bone destruction.
27
Cartilage – Forming Bone Tumours
a
b
Fig. 4 Enchondroma. (a) Antero-posterior radiograph shows dense calcifications in the proximal femur. The lesion is smaller than 5 cm and there is no associated osteolysis or endosteal scalloping of the cortex. The patient is asymptomatic in this region
and additional MR imaging is not required. (b) Hypocellular tumor with abundant hyaline cartilage matrix. No or sparse double nuclei: enchondroma
In the hands and feet, bony expansion with cortical thinning is common and they may be purely lytic. In larger bones, the lesion arises usually centrally within the meta(-dia)physis such that endosteal resorption is only seen in larger lesions. If the lesions are larger than 5 cm. the possibility of a low-grade chondrosarcoma is questioned. Identification of classic popcorn-like mineralization indicates the cartilaginous matrix (Fig. 4a).
and grade I chondrosarcoma is extremely difficult both at radiology [6] and histology [7] resulting in high interobserver variability [8]. The presence of entrapment of pre-existing host bone and mucomyxoid matrix changes favour the diagnosis of chondrosarcoma [8]. With the new treatment options using curettage and adjuvant phenol or cryosurgery [9] for enchondroma as well as grade I central chondrosarcoma the distinction is not always essential for clinical decision-making.
Differential Diagnosis Treatment The differential diagnosis includes bone infarction, chondromyxoid fibroma, giant cell tumor, aneurysmal bone cyst, solitary bone cyst or fibrous dysplasia. The major diagnostic problem is the differentiation between enchondroma and low-grade chondrosarcoma. In sites other than the phalanges, malignant transformation of enchondroma to chondrosarcoma is not uncommon.
Histology Histologically the enchondroma has an abundant hyaline cartilaginous matrix, sometimes calcified. The lesions are lowly cellular, unless located in the phalanges or in the context of enchondromatosis where increased cellularity is tolerated (Fig. 4b). The distinction between enchondroma
Enchondroma of the phalanges are usually left alone unless repeated fractures or impairment of function or daily activities requires treatment. In these cases simple curettage and if needed bone grafting is the treatment of choice. The recurrence rate is rather low (1–5%). Asymptomatic enchondroma of the long bones are to be followed-up by plain radiography every other year. Symptomatic enchondroma should be carefully investigated to exclude progression towards malignancy using gadolinium enhanced (dynamic) MRI.
Multiple Enchondromatosis Multiple enchondromatosis (Ollier’s disease) is an uncommon developmental bone abnormality characterized by
28
A.H.M. Taminiau et al.
Fig. 5 Multiple enchondromatosis. Radiographs of the hands show multiple lytic lesions in the phalanges and metacarpals with osteolysis or endosteal scalloping of the cortex. Some calcifications can be noticed
multiple enchondromata located in multiple bones without hereditary or familial tendency (unlike hereditary multiple osteochondromata). Malignant degeneration to a central low-grade chondrosarcoma occurs in about 20% of cases [10]. Multiple enchondromata in combination with hemangioma (soft tissue and rarely visceral) is called Mafucci’s syndrome. Mafucci’s syndrome is also a developmental non-hereditary disorder, however, with higher malignant potential than enchondromatosis alone [11, 12] (Fig. 5).
Juxtacortical (Periosteal) Chondroma Although very uncommon, periosteal chondroma is a benign cartilaginous tumor arising at the periosteal surface of bone. All age groups can be affected and the male-female ratio is equal. This lesion can arise in both long and small tubular bones usually measuring fewer than 6 cm. If it is situated close to a joint it is called juxta-articular chondroma. The mass is palpable and can be painful.
Differential Diagnosis The major differential diagnoses are periosteal osteosarcoma, periosteal chondrosarcoma and giant cell tumour of the tendon sheath (especially when situated in the hand and feet).
Histology In periosteal chondroma increased cellularity as compared to enchondroma is seen; distinction from periosteal chondrosarcoma is made based on size (>5 cm) and cortical invasion in cases of periosteal chondrosarcoma.
Treatment Marginal or wide resection of the lesion en-bloc is mandatory resulting in a low recurrence rate (<5%). If intralesional resection is performed the recurrence rate does increase significantly.
Radiology
Chondromyxoid Fibroma The lesion causes a soft tissue mass with cortical pressure erosion. Calcifications can be appreciated in the soft tissue mass in about 50% of cases. Although cortical erosion is frequently apparent, cortical destruction should raise a high suspicion for malignancy (Fig. 6).
Chondromyxoid fibroma (CMF) is a very rare benign cartilaginous lesion. They most frequently occur in the second and third decades, more frequent in males and have a predilection for the long bones of the lower limb, especially
29
Cartilage – Forming Bone Tumours
a
d
b
c
e
Fig. 6 Juxtacortical chondroma. (a, b) Radiographs show a juxtacortical lesion in the distal femur metaphysis. (c) Axial MR images [proton density]. (d) T2-weighted with fat suppression.
(e) T1-weighted after intravenous contrast injection show the characteristics of a chondroid lesion separated by an intact cortex from the underlying medullar cavity
the proximal tibia and distal femur. Although they are usually small, CMF can become tall as well. Clinical presentation shows mild pain present for some time (years) and swelling when small bones (hand, feet) are involved.
u ncommon. MR imaging will reveal a typical lobulated pattern suggestive of a cartilage tumour (Fig. 7a–c).
Radiology The radiographic appearances are those of a well-defined oval or round lytic lesion with often lobulated margins, and usually eccentrically located in the metaphysis. They may cause mild cortical expansion. Matrix calcification is
Differential Diagnosis The differential diagnosis includes aneurysmal bone cyst and non-ossifying fibroma.
Histology CMF is a lobular tumour in which spindle-shaped or stellate cells are seen embedded in an abundant myxoid matrix.
30
A.H.M. Taminiau et al.
Towards the periphery of the lobules the cellularity increases and the cells are more rounded. Multinucleated giant cells are present. Nuclear atypia and (non-atypical) mitoses can be found which may lead to an erroneous diagnosis of high-grade chondrosarcoma (Fig. 7d).
Treatment Treatment is by excision or curettage with local adjuvant (cryotherapy or phenol) and bone grafting or cementation depending on the site and extension of the lesion
a
The recurrence rate is relative high at about 15%, intensive follow-up is therefore necessary.
Chondroblastoma Chondroblastoma is a rare benign cartilaginous tumour found in skeletally-immature patients (10–25 years), located almost exclusively in the epiphysis, and less frequently the apophysis. If however, there is partial closure of the epiphyseal plate, the lesion may extend into the metaphysis.
b
c
Fig. 7 Chondromyxoidfibroma. (a) Radiographs show a welldemarcated eccentric lytic lesion in the metatarsal bone without matrix calcification. (b) Coronal T2-weighted MR images after intravenous contrast injection with fat-suppression show enhancement of the lesion and extensive surrounding bone marrow edema. The differential diagnosis is enchondroma. (c) Radiograph
d
of the pelvis shows a lobulated well-demarcated lytic lesion with ridges. The differential diagnosis is aneurysmal bone cyst and fibrous dysplasia. (d) Usually lobulated pattern of stellate and spindle shaped cells in a myxoid background. Osteoclast-like giant cells are present
31
Cartilage – Forming Bone Tumours
The most common site of involvement is the proximal humerus, followed by the proximal femur, distal femur and proximal tibia. They can also occur in the pelvis, calcaneus, patella, mid- and hind-foot and in an older age group (40–50) and involvement of the skull has been reported. At clinical presentation localised mild pain (existing sometimes for a long period of time) is a frequent complaint. A minority of patients has some joint effusion. In 15–30% of the chondroblastomas a secondary aneurysmal bone cyst can be found.
Radiology On radiography the lesion is predominantly lytic with geographic and sclerotic margins. About 50% of cases demonstrate some amount of chondroid matrix. Most chondroblastomata elicit a thick periosteal reaction along the metaphysis. MR imaging shows a lobulated lesion with iso-intense signal intensity compared to muscle on T1-weighted MR images and intermediate to high heterogeneous signal intensity on T2-weighted MR images. MR also shows oedema in the adjacent bone marrow and soft tissue (Fig. 8a–c).
Differential Diagnosis The differential diagnosis in the adolescent includes giantcell tumour extending into the epiphysis, articular lesions with large cysts e.g. PVNS and clear-cell chondrosarcoma. In children it includes eosinophilic granuloma and epiphyseal osteomyelitis. Needle biopsy is therefore recommended.
Histology Chondroblastoma consists of cells with a uniform round polygonal nucleus, sometimes with a nuclear groove, and obvious cytoplasm (Fig. 8d). Osteoclast-type giant cells are always present, and therefore giant-cell tumour of bone is the most important histological differential diagnosis. Cartilaginous foci, sometimes with “chicken-wire”-like calcifications are characteristic of chondroblastoma. Mitotic figures are rare. Moreover, the cells in chondroblastoma are most often S100 positive, while giant-cell tumour is negative for S100.
Treatment Treatment of choice is curettage and bone grafting, often combined with local adjuvant (phenol or cryosurgery) with
a reasonable risk for local recurrence (14–18% within 2 years). This is probably due to the combination of a difficult epiphysial location and the initial attempt to save the joint or physis. In rare cases metastases to the lung have been reported, which are usually non-progressive and often treated successfully by metastasectomy.
Synovial Chondromatosis Synovial chondromatosis is a synovial metaplasia in which (osteo)cartilaginous nodules arise in synovial joints. The nodules may grow and variably become loose or re-attach to the synovium. This process, osteochondromatosis, can occur in any site covered with synovium (joint capsule, tendon sheet, bursa and periarticular tissue). When the nodules mature they will be released into the synovial space (joint). Patients are often males (>40 years old) presenting with complaints including: pain, decreased function and sometimes locking. Knee, elbow, hip and wrist are most the frequently involved joints. There are three stages of disease progression: stage one: the early phase, no nodules, only synovitis, stage two: the transition phase: active synovitis with nodules and stage three: the late phase with nodules present even without active synovitis.
Radiology Multiple round bodies of similar size and variable in mineralization in the joint can be seen on conventional radiographs. The round bodies can appear trabeculated. MR imaging is usually not required to make the diagnosis but can be diagnostic if the round bodies lack mineralization (Fig. 9).
Differential Diagnosis The main differential diagnosis is loose bodies seen in osteoarthritis.
Histology Histologically multiple rounded cartilaginous nodules are seen, covered by a thin layer of synovial lining. Increased cellularity, nuclear atypia and an occasional mitosis are accepted; malignant synovial chondromatosis is extremely rare.
32
A.H.M. Taminiau et al.
a
b
c
d
Fig. 8 Chondroblastoma. (a) Anteroposterior radiograph shows a lytic oval lesion in the proximal epiphyseal humerus with a geographic pattern of destruction. (b) Coronal T2-weighted MR images shows bone marrow edema surrounding the lesion. (c) Axial T1-weighted MR image after intravenous contrast injection
Treatment Treatment of choice is total open synovectomy, with still a recurrence rate of 10–20%. Post-operative stiffness of the involved joint has shifted primary treatment to more conservative measures involving a partial synovectomy and removal of loose bodies using arthroscopy. Total synovectomy using arthroscopy is hard to achieve and has increased the recurrence rate up to 30–40%.
shows peripheral septonodular enhancement. (d) Histology on HE: pale cytoplasma and octagonal shape due to abutment against other tumor cells. Homogenous, oval and grooved nuclei, clumps of giant cells are found adjacent to areas of spindle cell stroma
Malignant Cartilage Tumours Malignant Chondrosarcoma Malignant cartilaginous tumours are the second largest group of primary bone tumours [13]. Most of them occur as de novo malignancies in previously normal bone. A small sub-set appears secondary to pre-existing benign
33
Cartilage – Forming Bone Tumours Fig. 9 Synovial chondromatosis. (a, b) Antero-posterior and lateral radiograph op the hip show multiple intraarticular chondral fragments
a
c artilage tumours (enchondroma, osteochondroma). Highest prevalence is found between the fourth and sixth decade, with an equal male-to-female ratio. Approximately 90% of chondrosarcoma are described as conventional type. Chondrosarcoma are sub-divided into central and peripheral sub-types, based upon radiological presentation. Central chondrosarcoma compose about 75% of all chondrosarcoma, the majority are low-grade (grade I). They arise centrally in the metaphysial region of long bones, but can also develop in flat bones such as pelvis, rib and scapula. A minority (up to 15%) of conventional chondrosarcomata develop from the surface of bone as a result of malignant transformation within the cartilage cap of a pre-existent osteochondroma and is therefore called a secondary or peripheral chondrosarcoma. Rarer sub-types include mesenchymal chondrosarcoma and clear-cell chondrosarcoma. Conventional chondrosarcoma can “dedifferentiate” into a very high-grade tumours with dismal prognosis, so-called de-differentiated chondrosarcoma [14]. Most chondrosarcoma are solitary, but they can occur as multiple lesions in patients with syndromes such as multiple osteochondroma and enchondromatosis (M. Ollier, Maffuccis disease) [11, 12, 15].
Clinical Presentation Central low-grade chondrosarcomata are frequently asymp tomatic and are coincidentally found on radiographs. High grade cartilage tumours almost always are symptomatic.
b
Pain at the site of a cartilaginous lesion may be an indicator of malignancy. The differentiation between enchondroma and grade I chondrosarcoma can be difficult. Peripheral chondrosarcomata often present with a painless mass, but can be symptomatic because of size and related functional loss. Malignant cartilage tumours in the phalanges of the hand and feet are extremely rare, but in the other long bones central cartilaginous lesions should be considered low-grade chondrosarcoma till proven otherwise. Treatment is therefore mandatory for all grades of chondrosarcoma. Preferred surgical management is related to grade, type and site of the lesion. Chemo- and radiotherapy do not have a significant role in the treatment of these patients. Adequate medical imaging by conventional X-ray and (enhanced) MRI is essential in the work-up and is indicative for optimal treatment options.
Radiology On conventional radiography, the distinction between enchondroma and central grade I chondrosarcoma cannot be made reliably. The localization in the axial skeleton and size greater than 5 cm are the only reliable predictors for malignancy. Low-grade central chondrosarcoma can be geographic in appearance and may show mild cortical expansion and/or endosteal scalloping. The presence of chondroid matrix is variable, ranging from pure lytic lesions, to few or dense calcifications. They have no associated soft tissue mass. On MR imaging low grade
34
A.H.M. Taminiau et al.
c hondrosarcoma will show a lobulated endomedullary mass with high signal intensity on T2-weighted MR images and typical peripheral and septonodular (guirlande-like) enhancement on MR images after intravenous contrast injection corresponding to the fibrovascular septation between lobules of hyaline cartilage (Fig. 10). Although dynamic contrast-enhanced MR-imaging shows increased sensitivity, suggesting that chondrosarcoma tend to show more rapid enhancement than enchondroma, an absolute distinction between enchondroma and low-
grade chondrosarcoma cannot be made on radiological grounds alone [8] (Figs. 11 and 12). The differential diagnosis of osteochondroma versus low-grade peripheral chondrosarcoma can be reliably made measuring the thickness and enhancement characteristics on MR of the cartilaginous cap (Fig. 13). Aggressive chondrosarcoma sub-types, such as mesenchymal and dedifferentiated chondrosarcoma often contain less extensive areas of matrix mineralization that suggest a chondroid neoplasm. They may demonstrate intra-osseous
a
b
c
d
e
f
Fig. 10 Chondrosarcoma low grade. (a) Radiograph shows chondroid calcifications in the distal meta-diaphyseal femur. Minimal scalloping of the anterior cortex adjacent to the lesion. (b, c) Work-up with MR imaging suggesting a cartilaginous tumor. Sagittal T1- and T2-weighted MR image after intravenous contrast injection. (d–g) Consecutive dynamic contrast-enhanced
g
MR images at the same position with a temporal resolution of 3 s. Arrival of the bolus contrast in the popliteal artery is demonstrated on images. Lesional enhancement is demonstrated within 6 s after arterial enhancement suggestive for malignancy. Histology revealed conventional chondrosarcoma grade I. (h, i) Radiographs after curettage, phenolisation and boneplasty
35
Cartilage – Forming Bone Tumours
h
lesion. Matrix mineralization is not as frequently apparent in clear-cell chondrosarcoma as compared with conventional chondrosarcoma. Clear-cell chondrosarcoma is often mistaken for a chondroblastoma because it is non-aggressive in appearance and arises in the epiphysis. However, clearcell chondrosarcoma occurs in older patients than those with osteoblastoma. CT is only recommended in the pelvis and other flat bones where it may be difficult to discern the pattern of bone destruction and the presence of matrix mineralization. MRI is used to delineate the extent of the intra-osseous and soft- tissue involvement. Core needle biopsy is the preferred method when grading is debated due to imaging studies [6].
i
Histology Fig. 10 (continued)
lytic areas and show aggressive patterns of bone destruction with a “moth-eaten” to permeate pattern in combination with large soft tissue masses. Evidence of a large non-mineralized soft tissue mass associated with a lesion with radiological features indicative of a chondrosarcoma should raise the level of suspicion for dedifferentiation. Clear-cell chondrosarcoma typically has a predilection for the proximal end of the femur and humerus, often with epiphyseal involvement. Radiographs reveal a predominantly lytic
a
Fig. 11 Central chondrosarcoma (macro and micro). (a) Central chondrosarcoma macroscopy with medullary channel filled with cartilaginous tumor replacing the bone marrow. Scalloping is clearly visible on specimen. (b) No clear scalloping in microscopic slide. (c, d) Abundant hyaline matrix with a moderate cellularity and more than sparse double nuclei in c and some entrapment of bone in d. Central chondrosarcoma grade I,
Central and peripheral chondrosarcoma are histologically similar, and for both three different grades are discerned, which is at present the best predictor of clinical behaviour [16], but subject to interobserver variability [8]. Grade I chondrosarcomata are lowly cellular, with an abundant hyaline cartilage matrix and do not metastasize. Grade II chondrosarcoma have increased cellularity, frequent double-nucleated cells and mitotic figures can be found (Figs. 11 and 12). By contrast, grade III chondrosarcomata are highly cellular, with a mucomyxoid matrix, nuclear atypia and mitoses, with metastases developing in 70% of
b
36
c
A.H.M. Taminiau et al.
d
Fig. 11 (continued)
the patients [16]. Up to 13% of recurrent chondrosarcomata exhibit a higher grade of malignancy than the original neoplasm [16, 17] suggesting chondrosarcoma may biologically progress.
Genetics At the genetic level a multi-step genetic model is assumed, with increased genetic instability with increasing histological grade [4]. Central and peripheral chondrosarcoma differ at the genetic level [18]. EXT mutations, disturbing hedgehog diffusion in the growth-plate as a result of a lack of heparan sulphate, underlie the development of osteochondroma [4]. Additional as yet unidentified genetic hits are expected to cause malignant transformation of osteochondroma towards secondary peripheral chondrosarcoma. Peripheral chondrosarcoma is aneuploid, and a subset of tumours demonstrates near-haploidy [18] with polyploidization upon progression towards higher grade. This is however not specific for peripheral chondrosarcoma and is also seen in central chondrosarcoma [19]. Different signalling pathways have been studied and Indian hedgehog, PTHLH/ Bcl-2, Cox-2 and oestrogen signalling have been suggested as therapeutic targets for peripheral chondrosarcoma [4, 20]. Systematic genetic studies in the rare chondrosarcoma sub-types are lacking.
Dedifferentiated Chondrosarcoma Dedifferentiated chondrosarcoma is a highly anaplastic sarcoma next to a (usually low-grade) malignant cartilageforming tumour, with a remarkably sharp junction between the two components (Fig. 14). The rare genetic reports on dedifferentiated chondrosarcoma demonstrate that both components share identical genetic aberrations [18, 21] with additional genetic changes in the anaplastic component [18, 21–23] indicating a common precursor cell with early diversion of the two components [14, 24]. No targets for therapy have been reported for dedifferentiated chondrosarcoma so far.
Mesenchymal Chondrosarcoma Mesenchymal chondrosarcoma is a highly malignant lesion that can occur in bone and soft tissue of relatively young patients and is characterized by scattered areas of differentiated cartilage intermingled with undifferentiated small round-cells with typical staghorn-like vessels [25]. Sixty-one percent of the tumours demonstrate p53 over-expression, but no mutations were found. Recent evidence suggests that mesenchymal chondrosarcoma may be chemotherapy- sensitive, and may be considered for adjuvant or neo-adjuvant therapy [26, 27] (Fig. 15).
Rare Chondrosarcoma Subtypes Clear-cell Chondrosarcoma In addition to conventional chondrosarcoma, several rare sub-types of chondrosarcoma are discerned, together constituting 10–15% of all chondrosarcoma.
Clear-cell chondrosarcoma is a tumour of low-grade malignancy characterized by tumour cells with clear, empty
37
Cartilage – Forming Bone Tumours
a
b
c
d
Fig. 12 Chondrosarcoma grade II. (a) Radiograph of the shoulder in a 24-years-old female shows a well-defined lobulated lytic lesion in the acromion. (b) Axial T1-weighted MR image after intravenous contrast injection shows typical peripheral (guirlandelike) enhancement consistent with a chondroid tumor. (c) CT
guided percutaneous biopsy in prone position was performed because the bony expansion, the cortical disruption in combination with the age of the patient suspected a higher grade of malignancy. (d) Abundant hyaline matrix with moderate cellularity, some spindle shaped nuclei and more than sparse double nuclei
cytoplasm depositing both hyaline cartilage as well as scattered mineralizing osteoid (Figs. 16 and 17). Metastases are rare but may occur up to 24 years after initial diagnosis, thus long-term follow-up is mandatory [28]. A cytogenetic study on four cases suggested that extra copies of chromosome 20 and loss or re-arrangements of 9p may be recurrent.
Treatment Although, for all grades of non-metastatic chondrosarcoma en-block resection offers the best recurrence-free survival, surgical management is related to grade, type and site. In low-grade central chondrosarcoma of the long bones extensive intralesional curettage, followed by local adjuvant
38
A.H.M. Taminiau et al.
treatment has become the treatment of choice [29] (Fig. 10i, j). Local adjuvant in lesions confined to the bone includes phenolization, cryosurgery (liquid nitrogen) and the use of
a
b
d
e
Fig. 13 Peripheral chondrosarcoma. (a, b) Radiographs of the proximal femur with the characteristics of an osteochondroma originating from the minor trochanter region. (c) MR imaging showing the thickness of the cartilage cap (>1 cm) and a typical ring and arcs configuration with this cartilage cap can be
polymethylmethacrylate (PMMA) [30]. Promising longterm clinical results and satisfactory local control have been published using all of the above [9, 31]. Application of phenol (85%) as adjuvant treatment with additional
c
a ppreciated. (d) Macroscopy of the resected specimen reveals the thickness of the cartilaginous cap important for the differential diagnosis of osteochondroma versus low grade peripheral chondrosarcoma. (e) Postoperative radiograph after resection and reconstruction with inlay allograft
39
Cartilage – Forming Bone Tumours
washing with ethanol (96%) has been proven effective causing sufficient tissue necrosis in chondrosarcomaderived cell lines. The use of phenol as adjuvant following intralesional curettage in low-grade chondrosarcoma of
a
long bones is safe and has a 7% recurrence rate after an average follow-up of 6 years. Additional osteosynthesis is not necessary due to its use, therefore MRI-scans can be readily used for follow-up [31, 32].
b
d
Fig. 14 Dedifferentiated chondrosarcoma. (a) Later radiograph showing chondroid calcifications and some lytic changes in the meta-epiphysis of distal femur. (b) MRI revealing cortex invasion and soft tissue extension suggesting malignant dedifferentiation of previous cartilage tumor. (c) Macroscopy of distal
c
e
femur resection specimen of a dedifferentiated chondrosarcoma. Permeative cortex destruction anterior and posterior with soft tissue extension. (d, e) Low grade cartilaginous component abrupt entrapped in a high grade sarcoma with atypical high cellularity suggesting dedifferentiated chondrosarcoma
40
a
A.H.M. Taminiau et al.
b
c
repair resulting in normal bone stock after 2 years. The presumed infectious risk of allograft use is negligible. In some cases of low-grade chondrosarcoma intralesional excision is not a reliable treatment, because of the large size, site (pelvis) and extension (intra-articular). In these cases wide resection is preferred. In peripheral chondrosarcoma complete surgical removal of the osteochondroma including the cartilage cap with the pseudo-capsule is the preferred treatment with excellent long-term clinical and local results. For intermediate (grade II) and high grade (grade III) chondrosarcoma en-bloc resection is considered the preferred surgical treatment (Fig. 18). High grade chondrosarcoma (including clear-cell chondrosarcoma) and all chondrosarcomata of the axial skeleton should be surgically resected with wide margins with a intermediate risk of local recurrence. There is a very high risk of local recurrence following resection of dedifferentiated chondrosarcoma, particularly in the presence of a pathological fracture. The presence of a pathologic fracture must be considered as a sign of a higher grade of malignancy. If wide margins cannot be reliably achieved with limb salvage then amputation must be considered. Chondrosarcomata in the skull and spine are often not resectable with sufficient margins. In these occasions proton-beam radiotherapy could be considered following de-bulking (Flowcharts 1 and 2).
Local Recurrences Fig. 15 Mesenchymal chondrosarcoma. (a) Radiograph of tumor in proximal tibia producing calcification and with cortex destruction and soft tissue extension. (b) MRI showing extra-osseous extension anterior and posterior suggesting high grade malignancy. (c) Biphasic tumor pattern composed of small round cells with lack of differentiation and islands of hyaline cartilage. The small blue round cells can simulate Ewing sarcoma
Cryosurgery uses very low temperatures to induce tissue necrosis with the intent of ablation by freezing, holding of freeze, thawing and repetition of this cycles. The local extent of treatment with cryosurgery is at least 7–12 mm beyond the surgical margin. Side-effects of cryosurgery are (temporarily) nerve damage, fractures and infections. The use of polymethylmethacrylate (PMMA) is based on the hypothesis that it will kill the residual tumour cells following curettage by thermal heating of the bone cavity. The maximum peripheral extent of a thermal lesion induced by PMMA varies from 2 to 5 mm in cancellous bone. An advantage of using PMMA is the possibility of early weight-bearing. The use allograft bone chips does not give the possibility for early weight bearing, but does result in a biological
Local recurrences can occur 10 years post-operatively and demand long term follow-up. The clinical course cannot always be predicted by histological grade alone. Outcome of low-grade chondrosarcoma of long bones and metacarpals are good. Location of the lesion, especially in the pelvis and skull, is an important risk factor for local recurrence and dedifferentiation. Distant metastasis in low-grade chondrosarcoma average at 2–5%. Survival rates after 5 years are described between 85% and 90%.
Prognosis Grade I tumours do not have 100% survival, mainly due to problematic local recurrence or progression into high-grade upon occurrence. The histological grading is subject to variability in interpretation, with grade II and III chondrosarcoma often grouped together even though there is a wide spectrum of outcome [8]. The most important predictors in chondrosarcoma for poor survival are histological grade and age above 50.
41
Cartilage – Forming Bone Tumours
a
c
b
d
Fig. 16 Clearcell chondrosarcoma. (a) Radiograph shows a welldemarcated lytic lesion, slightly lobulated, in the proximal epiphysis of the femur. The lesion is surrounded by a thin sclerotic rim. No matrix calcification is present. (b, c) Coronal T1 and T2- weighted MR images show the lobulated lesion. (d) Axial T2-weighted MR image with fat suppression shows surrounding
bone marrow edema. The differential diagnosis is chondroblastoma or avascular necrosis. Because of the classical location in the epiphyses of long bones the radiographic pattern of clear cell chondrosarcoma is mostly indistinguishable from chondroblastoma. Patient age is considered as an important discriminator between both entities
In case of a pathological fracture of the long bones, wide excision with adequate reconstruction is preferable to reduce the risk for local recurrence. The prognosis in dedifferentiated chondrosarcoma is very poor, despite adequate wide surgical resection
and adjuvant therapy. Inoperable, locally-advanced and m etastatic high grade chondrosarcoma are insensitive to c onventional adjuvant treatment such as radio- and c hemotherapy, reducing life expectancy to minimal [33].
42
A.H.M. Taminiau et al.
a
b
Fig. 17 Clearcell chondrosarcoma (microscopy). (a, b) Cells with round large centrally located nuclei with clear cytoplasm and distinct cytoplasmic membranes. Multinucleated giant cell may be present
Adjuvant Treatment Recent evidence suggests that mesenchymal chondrosarcomata may be chemotherapy-sensitive, and may be considered for adjuvant or neo-adjuvant chemotherapy. Uncertainty remains on the use of chemotherapy for de-differentiated chondrosarcoma, however with poorer outcome they can be treated with a chemotherapy-protocol comparable to the treatment of osteosarcoma [24]. Chondrosarcoma are considered relatively radiotherapy insensitive. In cases where wide resection is not possible or would cause unacceptable morbidity, doses over 60 Gy are needed for maximal local control. As high-dose radiotherapy is not always feasible, new techniques, such proton-beam radiotherapy or charged particles are used. Particle therapy
with protons has the advantage of a minimal exit dose after energy deposition in the target volume, and hence better sparing of critical structures close to the tumour. Proton RT has been found beneficial in cartilage tumours of the skull base and axial skeleton. Local control rates of 85–100% with mixed photon-proton or proton-only protocols (doses up to 79 CGE) are reported by several authors, with limited severe late effects (<10% RTOG Grade 3 toxicity) [33]. Chemotherapy is possibly effective in mesenchymal chondrosarcoma, and of uncertain value in dedifferentiated chondrosarcoma. Chemotherapy should preferably be used is clinical trials to define its definite role in chondrosarcoma. Several new drug targets have been identified and phase II studies are currently ongoing. There is an urgent need for new standard systemic treatment options for the minority of patients with unresectable or metastatic disease.
Local recurrence in central chondrosarcoma Chondrosarcoma
Long bones Central
Flat bones
Peripheral
Grade I involving: Borderline, grade I
Dedifferentiation, grade II, III
All grades
Intralesional
En-bloc resection
En-bloc resection
Intralesional
Curettage + local adjuvant therapy + bone grafting
Wide margin
Wide margin including pseudocapsule
Curettage + local adjuvant therapy + bone grafting
Bone
Soft tissue
Grade II, III, dedifferentiation
All grades
En-bloc resection
Flowchart 1 Flowchart of the surgical management of primary central and peripheral chondrosarcoma [33]
Wide margin
Flowchart 2 Flowchart of the surgical management of local recurrence in central chondrosarcoma [33]
43
Cartilage – Forming Bone Tumours
a
b
c
d
Fig. 18 Chondrosarcoma grade III. (a–c) Axial and coronal CT and T1-weighted MR images after intravenous contrast injection show a large lobulated mass in the acetabulum. The mass contains numerous stippled calcifications and there is expansion of the medial surface with neo-cortex formation. The size of the
References 1. Fletcher CDM, Unni KK, Mertens F (2002) World Health Organization classification of tumours: pathology and genetics of tumours of soft tissue and bone. IARC Press, Lyon 2. van der Eijken JW, van der Heul RO, Mulder JD et al (1987) Wegwijzer tumoren van het skelet. Commissie voor Beentumoren, Leiden 3. Ahmad S, Shen FH, Bleyer WA (1978) Methotrexateinduced renal failure and ineffectiveness of peritoneal dialysis. Arch Intern Med 138:1146–1147 4. Bovee JV, Hogendoorn PC, Wunder JS, Alman BA (2010) Cartilage tumours and bone development: molecular
tumor and the expansion suggested a higher grade of malignancy than conventional chondrosarcoma grade I. (d) Anteroposterior radiograph after tumor resection and reconstruction with LUMiC prosthesis
p athology and possible therapeutic targets. Nat Rev Cancer 10:481–488 5. Stickens D, Brown D, Evans GA (2000) EXT genes are differentially expressed in bone and cartilage during mouse embryogenesis. Dev Dyn 218:452–464 6. Geirnaerdt MJ, Hogendoorn PC, Bloem JL, Taminiau AH, van der Woude HJ (2000) Cartilaginous tumors: fast contrast-enhanced MR imaging. Radiology 214:539–546 7. Mirra JM, Gold R, Downs J, Eckardt JJ (1985) A new histologic approach to the differentiation of enchondroma and chondrosarcoma of the bones. A clinicopathologic analysis of 51 cases. Clin Orthop Relat Res Dec(201):214–237 8. Eefting D, Schrage YM, Geirnaerdt MJ et al (2009) Assessment of interobserver variability and histologic parameters to
44 improve reliability in classification and grading of central cartilaginous tumors. Am J Surg Pathol 33:50–57 9. Veth R, Schreuder B, van Beem H, Pruszczynski M, de Rooy J (2005) Cryosurgery in aggressive, benign, and lowgrade malignant bone tumours. Lancet Oncol 6:25–34 10. Pansuriya TC, Kroon HM, Bovee JV (2010) Enchondro matosis: insights on the different subtypes. Int J Clin Exp Pathol 3:557–569 11. Maffucci A (1881) Di un caso encondroma ed angioma multiplo. Mov medico chirurgico 3:399–412, 565–575 12. Ollier M (1900) Dyschondroplasie. Lyon Med 93:23–25 13. Bertoni F, Bacchini P (1998) Classification of bone tumors. Eur J Radiol 27(Suppl 1):S74–S76 14. Grimer RJ, Gosheger G, Taminiau A et al (2007) Dedifferentiated chondrosarcoma: prognostic factors and outcome from a European group. Eur J Cancer 43:2060–2065 15. Rozeman LB, Hameetman L, van Wezel T et al (2005) cDNA expression profiling of chondrosarcomas: Ollier disease resembles solitary tumours and alteration in genes coding for components of energy metabolism occurs with increasing grade. J Pathol 207:61–71 16. Evans HL, Ayala AG, Romsdahl MM (1977) Prognostic factors in chondrosarcoma of bone: a clinicopathologic analysis with emphasis on histologic grading. Cancer 40:818–831 17. Bjornsson J, McLeod RA, Unni KK, Ilstrup DM, Pritchard DJ (1998) Primary chondrosarcoma of long bones and limb girdles. Cancer 83:2105–2119 18. Bovee JV, Cleton-Jansen AM, Rosenberg C, Taminiau AH, Cornelisse CJ, Hogendoorn PC (1999) Molecular genetic characterization of both components of a dedifferentiated chondrosarcoma, with implications for its histogenesis. J Pathol 189:454–462 19. Hallor KH, Staaf J, Bovee JV et al (2009) Genomic profiling of chondrosarcoma: chromosomal patterns in central and peripheral tumors. Clin Cancer Res 15:2685–2694 20. Hopyan S, Gokgoz N, Poon R et al (2002) A mutant PTH/ PTHrP type I receptor in enchondromatosis. Nat Genet 30:306–310 21. Ropke M, Boltze C, Neumann HW, Roessner A, SchneiderStock R (2003) Genetic and epigenetic alterations in tumor progression in a dedifferentiated chondrosarcoma. Pathol Res Pract 199:437–444
A.H.M. Taminiau et al. 22. Coughlan B, Feliz A, Ishida T, Czerniak B, Dorfman HD (1995) p53 expression and DNA ploidy of cartilage lesions. Hum Pathol 26:620–624 23. Grote HJ, Schneider-Stock R, Neumann W, Roessner A (2000) Mutation of p53 with loss of heterozygosity in the osteosarcomatous component of a dedifferentiated chondrosarcoma. Virchows Arch 436:494–497 24. Dickey ID, Rose PS, Fuchs B et al (2004) Dedifferentiated chondrosarcoma: the role of chemotherapy with updated outcomes. J Bone Joint Surg Am 86-A:2412–2418 25. Nakashima Y, Unni KK, Shives TC, Swee RG, Dahlin DC (1986) Mesenchymal chondrosarcoma of bone and soft tissue. A review of 111 cases. Cancer 57:2444–2453 26. Cesari M, Bertoni F, Bacchini P, Mercuri M, Palmerini E, Ferrari S (2007) Mesenchymal chondrosarcoma. An analysis of patients treated at a single institution. Tumori 93:423–427 27. Dantonello TM, Int-Veen C, Leuschner I et al (2008) Mesenchymal chondrosarcoma of soft tissues and bone in children, adolescents, and young adults: experiences of the CWS and COSS study groups. Cancer 112:2424–2431 28. Donati D, Yin JQ, Colangeli M et al (2008) Clear cell chondrosarcoma of bone: long time follow-up of 18 cases. Arch Orthop Trauma Surg 128:137–142 29. Leerapun T, Hugate RR, Inwards CY, Scully SP, Sim FH (2007) Surgical management of conventional grade I chondrosarcoma of long bones. Clin Orthop Relat Res 463: 166–172 30. Persson BM, Wouters HW (1976) Curettage and acrylic cementation in surgery of giant cell tumors of bone. Clin Orthop Relat Res Oct(210):125–133 31. Verdegaal SH, Corver WE, Hogendoorn PC, Taminiau AH (2008) The cytotoxic effect of phenol and ethanol on the chondrosarcoma-derived cell line OUMS-27: an in vitro experiment. J Bone Joint Surg Br 90:1528–1532 32. Lack W, Lang S, Brand G (1994) Necrotizing effect of phenol on normal tissues and on tumors. A study on postoperative and cadaver specimens. Acta Orthop Scand 65: 351–354 33. Gelderblom H, Hogendoorn PC, Dijkstra SD et al (2008) The clinical approach towards chondrosarcoma. Oncologist 13:320–329
Part III Paediatrics
The Current State of Treatment for Clubfoot in Europe Rüdiger Krauspe, Kristina Weimann-Stahlschmidt, and B. Westhoff
Introduction Clubfoot is one of the most frequent congenital skeletal deformities. Clubfoot is classified in two groups: postu ral or structural. It is characterized as a multi-dimensional complex deformity consisting of inversion at the subtalar joint, adduction at the talo-navicular joint and equinus at the ankle joint to different degrees. It can be acquired or congenital and is diagnosed clinically. The Ponseti method has changed the management of idiopathic, and increasingly also secondary, clubfoot in young children from a typically surgical to a primarily conservative approach. The rate and extent of surgical treatment is substantially reduced and the functional outcome improved.
Epidemiology The incidence of congenital clubfoot in central Europe and North America is known to be 1–2/1,000 newborns in the Caucasian genome. In the Chinese genome it is rare (0.5 –1 per 1,000), in the Polynesic genome it is six times more frequent than in central Europe. The male-to-female ratio is 3:1, and in 30–50%. bilateral involvement is present.
Aetiology The cause of idiopathic clubfoot is as yet still unknown. A multi-factorial cause of this deformity is assumed which affects the development and differentiation of soft
R. Krauspe () Orthopaedic Surgery, Heinrich-Heine-Universität Düsseldorf, Moorenstraße 5, 40225 Düsseldorf, Germany e-mail:
[email protected]
tissues, bone and joint surfaces. Intrinsic [1, 2] and extrinsic causes [2–5] can be proposed (Table 1).
Aetiology of the Congenital Clubfoot The congenital clubfoot has to be distinguished from the secondary clubfoot: The secondary clubfoot results from neurological or systemic disease (Table 2).
Aetiology of the Secondary Clubfoot
Pathophysiology of the Clubfoot Clubfoot is characterized by malrotation of the subtalar joint: in relation to the talus; navicular, os calcis and cuboid are displaced medio-plantarwards [6, 7] (Fig. 1).
Bone Talus: All relationships of the talus to the surrounding bones are abnormal. The talus is fixed in equinus. The talar body being in external rotation and extruded antero-laterally. It is uncovered and can be palpated easily. The neck of the talus is plantarflexed and deviated medially. According to Virchow, Scheel and McKay [8] clubfoot is characterized by sub-talar malrotation caused by thickening of the joint capsule and ligaments and a hyperactive tibialis posterior muscle. Navicular : The navicular is subluxed medially over the talar head. Os calcis: It is medially rotated as an equinus and adduction deformity exist. Cuboid: It is subluxed medially over the calcaneal head. Tibia: According to the equinus the talus in relation to the tibia is dislocated anteriorly and rotated medially.
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_4, © 2011 EFORT
47
48
R. Krauspe et al.
Table 1 Aetiology of congenital clubfoot External influence
Internal influence
• Environmental aspects (nutrition, virus infection, temperature)
• Genetic
• Smoking during pregnancy
• Alteration of collagen
• Amniocentesis
• Bony deformity of the talus
• Oligohydramnios • Arrest of foetal development
• Neuromuscular imbalance • Accessory M. soleus • Vascular Genesis
Table 2 Aetiology secondary clubfoot Neuromuscular Overactivity of supinator and flexor disease: muscles: cerebral palsy, cerebrospinal trauma, apoplexy, multiple sclerosis, spastic spinal paralysis Impairment of pronators: poliomyelitis, arthrogryposis, spina bifida, muscular dystrophy, hereditary motor and sensory neuropathy, spinal muscular atrophy, tethered cord Post-traumatic
Compartment syndrome, osseous trauma, injury of tendons, impairment of peroneal nerve, contracture after burn injury.
Iatrogenic:
Over-correction after operation of pes planovalgus, injury of pronators and nerve lesions respectively.
Psychogenic:
Causalgia syndrome (so called “hysterical clubfoot”)
Metabolic disorder:
Rachitis, neurofibromatosis
Associated syndromes
Mobius syndrome, Freeman-Sheldon syndrome, Larsen syndrome, Ehlers Danlos syndrome, strangulation mark syndrome
Deformity:
Fig. 1 Severe congenital bilateral clubfoot of a newborn
Tibia aplasia, Tibia hemimelia
Metatarsalia: The forefoot in relation to the hindfoot is adducted and pronated with cavus and a dropped and hypotrophic first metatarsal. The fifth metatarsal is hypertrophic and lengthened.
Muscles and Soft Tissues Atrophy of the peroneal group can be found in congenital clubfoot. The number of fibres is normal [9], but the fibres
are smaller in size. The calf is of smaller size and this persists also after successful correction of the foot. Tendon sheaths, joint capsules and ligaments of the medial part are shortened and thickened and on the lateral part of the foot over-lengthened. Contractures, especially of the talonavicular and calcaneaeo-cuboidal but also of the posterior ankle capsule and sub-talar capsule occur. Thickening of the tendon sheath and shortening especially of the tibialis posterior – the so called clubfoot muscle – are frequent findings. The deltoid, the long and short plantar, spring and bifurcate ligaments are contracted and
49
The Current State of Treatment for Clubfoot in Europe
fibrotic. Shortening of the long plantar ligament, the deltoid and retinaculum contribute to the cavus.
Diagnosis Presentation Congenital Clubfoot is diagnosed immediately after birth by inspection and palpation Inspection of the clubfoot with characteristic clinical features (Fig. 2):
• Inversion
at the subtalar joint (talar neck is easily palpable) • Equinus and varus position of the ankle joint (empty heel) • Adduction of the forefoot • Pronation of the forefoot in relation to the ankle • Talipes cavus with plantarflexion (pronation) of the first metatarsal • Skin fold medio-plantar and/or closed as sign of a structural clubfoot
Palpation
• The talar neck is easy to palpate. • The position of the medial malleolus, talus and navicular are to be noted. With increasing deformity the navicular is subluxed medially, the medial malleulos is difficult to palpate and is often in contact with the navicular. The normal navicular–malleolar interval is reduced.
The ability to correct the mal-position by manipulation helps to classify the severity of the clubfoot. Several classifications for clubfeet are described. The classification of Dimeglio et al. [10] and Pirani et al. [11] are frequently used. They allow for a standardized documentation and help to assess progress of the results of treatment.
Clubfoot Scoring Dimeglio Scoring System Dimeglio et al. [10] introduced a very detailed score to evaluate clubfoot. Because of its reliability and simple use it is used in many studies [12–18]. A scale of 0–20 was established evaluating four crucial parameters and their reducibility:
• The heel may be small and “empty” as a result of the equi- • Equinus deviation in the sagittal plane (Fig. 3) nus. The Achilles tendon can be palpated as it is shortened. • Varus deviation in the frontal plane (Fig. 4)
Fig. 2 Secondary clubfoot in the right foot in a newborn with strangulation mark on the right lower leg. Supination and adduction of the forefoot can easily be seen. Equinus in contrast to the
healthy left side, cavus, forefoot adduction and pronation of the forefoot in relation to the hindfoot can also be seen easily
50
R. Krauspe et al.
F.B.
: 4 points : 3 points : 2 points : 1 point
−20°
: 4 points : 3 points : 2 points : 1 point
0° F.B
90°
20°
45°
45°
90° 20°
−20°
0°
Fig. 3 Sagittal plane: evaluation of equinus
Fig. 5 Horizontal plane: evaluation of de-rotation of the calcaneopedal block
90°
0° 20°
45°
−20°
45° 20°
F.B.
0°
: 4 points : 3 points : 2 points : 1 point
−20°
90°
Fig. 4 Sagittal plane: evaluation of varus
• De-rotation of calcaneo-pedal block in the horizontal plane (Fig. 5) • Adduction of the forefoot relative to the hind-foot in the horizontal plane (Fig. 6)
Reducibility of these four parameters must be assessed in terms of angles:
• 90°– 45°: 4 points • 45°–20°: 3 points
F.B : 4 points : 3 points : 2 points : 1 point
Fig. 6 Horizontal plane: evaluation of the forefoot relative to the hind-foot (From Diméglio et al. [19])
• 20°–0° : 2 points • 0 – minus 20°: 1 points • < minus 20°: 0 points The sum of these parameters constitute a total on a 16 points scale.
51
The Current State of Treatment for Clubfoot in Europe Table 3 Classification of clubfoot according to severity Classification
Frequency
Score
Type
I: Soft-soft
20%
<5
Benign, “postural”
II: Soft-stiff
33%
5–10
Moderate clubfoot
III: Stiff-soft
35%
10 –15
Severe clubfoot
IV: Stiff-stiff
12%
15–20
Very severe
The true clubfoot must be differentiated from “postural” deformity (“soft-soft” according to Dimeglio classification) that can be easily corrected with passive stretching and usually does not require active treatment
Four additional characteristics, indicating negative prog nosis and another four points have to be analyzed:
reliability [11, 20]. This scoring system is based on six clinical signs of contracture. Each contracture is scored according to one of the principles:
• 0: no abnormality • 0.5 : moderate abnormality • 1: severe abnormality The foot is divided into three sections: Hind-foot:
• Severity of the posterior crease • Emptiness of the heel • Rigidity of the equinus Mid-foot:
• Curvature of the lateral border of the foot • Posterior crease is marked: 1 point • Severity of the medial crease • Medio-tarsal crease is marked: 1 point • Position of the lateral part of the talar head • Plantar retraction or cavus exists: 1 point According to this score, one foot can achieve a hind-foot • Hypotrophy, contracture or weakness of triceps surae: score and a mid-foot score each between 0 and 3 (0 = nor1 point
Accordingly 0–20 points can be achieved: Severity of clubfoot can be divided into four groups (Table 3).
Pirani Scoring System Pirani et al. in 1999 introduced a classification for idiopathic clubfoot which proved to have very good inter-observer-
mal, 6 = severe deformity) and a total score between 0 and 6. (0 = normal, 6 severe deformity) This scoring system is simple and reliable. The Coleman block test shows the flexibility of hindfoot varus: The lateral border of the foot is placed on a wooden block 0.5–1 cm thick: hind-foot varus is flexible if it is completely corrected. If it corrects to a normal position the varus is due to forefoot pronation (Fig. 7).
Fig. 7 Coleman block test: as described above, the Coleman block-test in this patient shows a rigid hind-foot varus
52
R. Krauspe et al.
Medical Imaging Radiographs Indications for radiographic examination of clubfoot may be surgical planning or clinical monitoring [21, 22]. X-ray should only be performed in babies older than 3 months because of the still cartilagineous structures , which ossifies later than in normal feet. The adjustment technique according to Simons [23] has proved to be useful: The foot is placed in the position of maximum correction obtained. The child is placed in a sitting position with hips and knees flexed at 90°. The feet are placed as close to a plantigrade position and to as much external rotation as possible. In older children and adolescents it can be performed in the standing position. The antero-posterior view is obtained with the x-ray tube directed towards the head of the talus, at an angle of 30° to the vertical. For the lateral view the x-ray tube is directed at 90° to the cassette (Figs. 8–10).
d
b a
Ultrasound There are several methods described to assess clubfoot [24–26]. Yet no method has achieved wide acceptance.
Fig. 8 Anterior-posterior view of a 2 year-old girl with bilateral idiopathic clubfoot weight-bearing. The navicular is not yet calcified. Reduction of the talo-calcaneal angle is characteristic (norm, a–b 30°–50°). Severe forefoot adduction is demonstrated by increase of the tarso-metatarsal I angle (norm. a–d: plus-minus 10°)
Computertomography (CT)/Magnet Resonance Imaging (MRI) CT and MRI normally are not necessary. In complex clubfeet, especially in those persisting for a long time or in secondary clubfeet, the CT can be helpful in pre-operative planning. Three-dimensional reconstruction should be taken into consideration. MRI and CT can be useful in assessing treatment outcome in long-term studies [27–30]. a
Pedobarography
b c
As a dynamic examination pedobarography can be helpful for clinical monitoring [31, 32]. Trobisch and Neidel [33] evaluated the correlation of pedobarography and clinical results in 33 clubfeet with an average age at examination of 64 months (range 47–105 months). They showed that pedobarography can be a useful tool to analyze outcome after clubfoot treatment although up to now only few results are of clinical significance.
d
Fig. 9 Lateral view of a 2 year-old girl with bilateral clubfoot, standing position. Typical reduction of the talocalcaneal angle (a–b norm. 30°–50°) and the reduced calcaneal pitch (b–c norm. 15°–30°). In relation to the tibia the fibula seems to be dislocated posteriorly. The talo-metatarsal I angle (a–d norm. plus-minus 10°) is increased
53
The Current State of Treatment for Clubfoot in Europe
d
a b
a
b
d
Fig. 10 X-ray according to Simons of a 14-month old baby with severe unilateral idiopathic clubfoot: characteristic reduction of the talo-calcaneal angle (a–b) in the antero-posterior and lateral views. Increased forefoot adduction in the antero-posterior view. As a result of the varus, the talo-metatarsal I angle (a–d) in the lateral view is increased
De-rotation of the sub-talar complex beneath the talar head is the basic principle of conservative therapy. All components of clubfoot deformity, except for ankle equinus, are corrected simultaneously. The manipulation consist of abduction of the foot beneath the talar head. Locating the talar head is very important to gain correction. The talar head – easily palpable in clubfoot – is the fulcrum for correction. Most clubfeet are corrected in approximately 6 weeks by weekly manipulations followed by plaster cast application. A long-leg-cast should be applied with hip and knee flexed 90°. After complete correction with 60°–70° of foot abduction and 20°dorsiflexion, a brace is necessary to maintain the foot in this position. This abduction brace should be worn for 24 hours for the first 3 months after the cast is removed. After that, the brace should be applied at night and for 2–4 hours in the middle of the day. If dorsiflexion remains limited after 6 weeks percutaneous achilles tenotomy is indicated. Plaster-of-Paris is recommended because it can be more precisely moulded than fibreglass. Before cast application, the foot is manipulated. Do not touch the heel to allow the calcaneus to rotate with the foot. According to a study by Zionts et al. [34] the rate of extensive surgery to treat idiopathic clubfoot has decreased substantially. They analyzed the data from the Centre for Disease Control and Prevention and the nationwide In-patient Sample in the United States. The number of surgical releases in patients less than 12 months-old decreased from 1641 in 1996 to 230 in 2006. This trend is likely due to the increased use of the Ponseti method. A growing body of evidence has shown this method to be a viable treatment option for clubfoot.
Therapy Manipulation and Cast Correction in Detail Conservative Therapy For successfull treatment of clubfoot the understanding of the complex deformity of the clubfoot and functional (patho-) anatomy of the foot are mandatory. In congenital clubfoot, treatment should be started as soon as possible after birth.
Ponseti Technique for Therapy of Clubfoot Over the last decades the Ponseti management has become established as the most effective and economical treatment of clubfeet.
The first step of correction is to reduce the cavus deformity by positioning the forefoot in proper alignment with the hindfoot. Cavus is due to the pronation of the forefoot in relation to the hindfoot. In newborns it is usually corrected with supination of the forefoot in 1–2 casts. Sufficient alignment of the forefoot with the hindfoot is important for effective abduction of the foot to correct the adductus and cavus. Manipulation: Locate the talar head (Fig. 11). Stabilizing the talus provides a pivot point around which the foot is abducted. Elevate the first ray of the forefoot to achieve a normal longitudinal arch of the foot (supination). Do not pronate!!!
54
The Calcaneus is never touched Casting: The aim of the first cast is to correct the cavus and adductus. The foot remains in marked equinus (Fig. 12). In the next 2–4 casts the adductus and varus are fully corrected. • Correction is achieved by abducting the foot in supination while counterpressure is applied over the lateral aspect of the talar head (Fig. 13). Navicular, cuboid and the rest of the midfoot and forefoot de-rotate in relationship to the talar head. The anterior part of the calcaneus also follows the de-rotation – subtalar de-rotation – and
Fig. 11 Location of the talar head
R. Krauspe et al.
varus deformity is corrected. This can only be achieved when the calcaneus is not fixed > do not touch the calcaneus! 70° of abduction in relation to the frontal plane of the tibia should be achieved. When the sub-talar joints are adequately de-rotated, 0°–5° of dorsiflexion are possible. Do not apply forceful dorsiflexion to the foot [35] > rocker bottom-deformity Tenotomy When cavus, adductus and varus are fully corrected, but ankle dorsiflexion remains less than 10° above neutral, tenotomy of the achilles tendon is indicated. In 85% a percutaneous tenotomy is necessary. According to Ponseti, tenotomy is performed under local anesthesia in the out-patient clinic. The achilles tendon has to be dissected completely under aseptic conditions. In our hands this procedure is safely performed in the operating room. After correction of equinus by tenotomy, the cast is applied in 60°–70° abduction and 15° dorsiflexion. This cast is left in place for 3 weeks. After 3 weeks the tendon is healed and the foot appears to be over-corrected into abduction but it is only full abduction. Recurrent discussion is held over gap healing after Achilles tenotomy. Several studies prove complete healing of the tendon after 6 –12 weeks [36, 37]. Maranho et al. [38] analyzed 37 tenotomies in 26 patients with a mean follow up of 1 year after section. Ultrasonographic scanning was performed after section to measure the stump separation at 3 weeks, 6 month and 1 year post-tenotomy to assess the reparative process. They concluded that there is a fast
Fig. 12 The first cast is applied in marked equinus correcting the cavus and adductus by elevating and supinating the first metatarsal in relation to the hindfoot
55
The Current State of Treatment for Clubfoot in Europe
Fig. 14 Foot-abduction-brace by MD Orthopaedics
Fig. 13 Cast in increasing abduction until 70° abduction is achieved
reparative process that re-establishes continuity between the stumps. The reparative tissue evolved to tendon tissue with a normal ultrasonographic appearance except for mild thickening, suggesting a predominantly intrinsic repair mechanism. In the Ponseti management tenotomy is performed with the child awake in an outpatient clinic. However tenotomy under general anesthesia offers potential advantages of better pain control and the ability to perform the procedure in a more controlled manner. A study by Parada et al. [39] showed that percutaneous tenotomy under general anesthesia in patients less than 1 year old can be performed safely. The Ponseti protocol now calls for a brace to maintain the completely corrected foot in 70° abduction and dorsiflexion to prevent relapse. Calcaneus and forefoot are held in the achieved abducted position and medial soft tissue remains stretched. The brace (Fig. 14) consists of high-top straight-last
shoes attached to a bar. In unilateral cases, the brace is set at 60°–70° of external rotation on the clubfoot side and 30°– 40° of external rotation on the normal side. The bar should be as long as the heels of the shoes are at shoulders’ width. To hold the feet in dorsiflexion, the bar should be bent 10° with the concavity to the child. For the first 3 months after cast removal the abduction brace should be worn for 24 h a day. In neuromuscular and syndrome-associated clubfeet the Ponseti method can be used [40]. Although non-idiopathic clubfeet require more casts and have a higher rate of failure, recurrence, and additional procedures correction can be achieved and maintained in most patients. Manipulation and percutaneous tenotomy according to Ponseti achieve an excellent result (Figs. 15–17). Never theless relapses may occur in more than 80% of the fullycorrected clubfeet when the bracing programme is not diligently followed or due to poor compliance. Treatment of recurrent clubfeet according to the Ponseti protocol According to Ponseti et al. [41] relapses occur when the child shows loss of foot abduction and/or of dorsiflexion and recurrence of adductus and cavus. Supination of the forefoot is another sign of loss of correction. At the first sign of relapse, casting according to the original Ponseti casting method should be applied again. This should also be performed in children of walking age. When the deformity is corrected, the bracing programme is started again. Tenotomy can be necessary again when equinus recurs. There are several reports discussing the results and clinical experience of the Ponseti method in older age groups with a history of failed manipulation [14, 28, 42–44]. Their
56
R. Krauspe et al.
Fig. 15 Fourteen months-old child with bilateral clubfoot treated with the Ponseti method: Equinus, cavus and forefoot adduction are diminished; only a small scar after tenotomy can be found
Fig. 16 Ten year follow-up of a child treated with the Ponseti method
ultimate overall results are satisfactory, reducing the need for extensive surgery. In children more than 2.5 years-old a dynamic supination deformity may occur. These patients benefit from an anterior tibialis tendon transfer to the lateral cuneiform [45]. Only in persistent or recurring relapse extensive operative treatment, based on the severity and type of deformity and age of the patient, is indicated. Soft tissue release and bony procedures may become necessary. Extensive surgery for relapsed clubfoot has a high rate of poor long-term outcome. Ponseti et al. [41] described a
modified treatment of the Ponseti method to treat relapsed clubfeet by casting and tendon transfer in 75 complex idiopathic clubfeet. They concluded that these feet were successfully corrected without the need for extensive surgery. Park et al. [46] reviewed 48 recurrent or selective clubfeet initially treated by the Ponseti method. Selective “a la carte” soft-tissue release was performed at an average of 2.3 years and pre-operative Pirani score of 2.8 points. At the mean follow-up of 3.6 years (2–5.3) the Pirani score improved to 1.1 points but six out of 13 patients needed further surgery.
The Current State of Treatment for Clubfoot in Europe
57
Fig. 17 Nineteen months-old child with bilateral neurogenic clubfoot treated with Ponseti method
Transfer of the anterior tibial tendon can restore balance and may provide improvement of forefoot adduction. A review of Farsetti, Ippolito et al. [45] revealed that transfer of the anterior tibial tendon to the third cuneiform corrected 16 relapsed clubfeet treated with the Ponseti method. Outcome was evaluated clinically and by plain radiographs and CT scan. Transfer of the anterior tibial tendon to the third cuneiform underneath the extensor retinaculum corrects and stabilizes relapsing clubfeet by restoring their normal function of foot dorsiflexion/eversion. As a consequence, the cuneiforms and the cuboid are shifted more laterally than normal, as shown by x-ray and CT scan. However a study by Lampasi et al. [47] reported 11 out of 38 relapsed clubfeet at an average of 24.8 years followup after anterior tendon transfer which were regarded as failures. They concluded that transfer of the anterior tibial tendon has a considerably high complication rate,
including failure of transfer, over-correction and weakening of dorsiflexion. In their opinion the procedure should be reserved for those limited cases in which muscle imbalance is a causative or contributing factor. Complications of the Ponseti method: Burghardt et al. [48] reported a case of pseudoaneurysm after Ponseti percutaneous Achilles tenotomy in a 8 weeksold baby. Diagnosis was confirmed by colour ultrasonography. The large pseudoaneurysm together with incomplete correction made 4 weeks additional Ponseti casting necessary. Emphasis on applying pressure over the pseudoaneurysm was made. Repeat ultrasonogram after 4 weeks casting showed a completely resolved pseudoaneurysm. A possible effect of the foot abduction brace in the Ponseti method, on the femoral anteversion and tibial torsion, is often discussed. In the study of Boehm and Sinclair [49] these changes were assessed by ultrasound and clinical
58
examination in 20 children. There were no pathological changes of femoral anteversion or tibial torsion found.
Classical Method of Correction of Idiopathic Clubfoot According to Mc Kay [8] and Krauspe/Parsch [9, 50] this method of manipulation also starts immediately after birth. Long leg casts are changed every 3–7 days with sequential correction of the sub-talar complex by de-rotation. In contrast to the Ponseti management, the calcaneus is fixed. Force has to be applied anterior to the ankle joint on the medial side of the calcaneal cuboid joint and posterior to the ankle on the lateral side of calcaneal tuberosity towards the tibial malleolus. A normal appearance of the foot in the first 6 weeks is the aim of this procedure. In contrast to the Ponseti method, there is no overcorrection and only the mid- and forefoot, but not the calcaneus, do de-rotate under the talar head. Correction needs to be obtained by the use of splints and consistent physiotherapy. After 6 weeks of casting the limit of this treatment method becomes obvious and 85% [8] need extensive surgery. Until surgery at the age of 4–8 months further manipulation and reduction as well as physiotherapy should be performed. The residual deformity after conservative care needs to be addressed by surgery. It should be performed when the foot has an approximate length of 8 cm. Surgical intervention performed as soft tissue release “a la carte” depending on the residual deformity may be necessary. The aim of this clubfoot regime is to provide a painfree, plantigrade and functioning foot when the child starts walking. Long-term studies [51] show that surgical correction of clubfoot is can be successful in providing a functional plantigrade foot. Nevertheless limitations include foot pain, limited range of motion, especially in ankle dorsiflexion and weakness of dorsiflexion. Our treatment of primary choice is the Ponseti method.
French Functional Physiotherapy Method Souchet and Diméglio [10, 52] developed a functional method for non-surgical treatment of the newborn clubfoot. It consists of daily manipulation of the newborn clubfoot by a special passive motion machine, stimulation of the muscles around the foot, and temporary immobilization of the foot with non-elastic adhesive strapping. The special passive motion machine is adapted to the size and type of clubfoot. The machine is first adapted in the horizontal
R. Krauspe et al.
plane, addressing equinus. Once derotation of the calcaneo- pedal block has been achieved correction of the sagittal plane starts. This functional method implies a well-performed, organized and supervised Orthopaedic treatment. Success of the method relies on an early start of treatment in the first 3 months of life and strict compliance with the guidelines of the functional method. Dimeglio [10] showed that this treatment makes surgery more elective, less severe and less extensive especially in grade II and III clubfeet according to the Dimeglio classification. The prospective study of Steinmann et al. [53] compared the outcome in the treatment of 267 idiopathic clubfoot by the Ponseti method and the French functional (physiotherapy) method with an average follow up of 43 years. They concluded, there was a trend showing improved results with the use of the Ponseti method especially of the equinus but the difference was not significant. Additionally, the Ponseti method was the parents’ first choice twice as often as the French method. According to these findings, S. Richards and M. Dempsey [30] evaluated MRI of 6 infants with congenital clubfeet treated with the French method and compared those to a similar study, assessing clubfeet treated by the Ponseti method [54]. Chondro-osseous abnormalities improved in clinical and MRI examination by both methods but in the French functional method equinus persisted more often and in greater degree.
Botulinum Type A Toxin There is little data in the literature regarding therapy of clubfoot with botulinum toxin [55, 56]. Injection of botulinum toxin does provide transient partial paralysis of the respective muscle. According to Alvarez et al. 2009, in a 5 year follow-up study with 65 congenital clubfeet treated with the Ponseti method, this effect can be used to support reduction and casting as well as to avoid the need of tenotomy in 48%. It has to be noted, this kind of drug-application is still off-label use.
Operative Therapy Operative treatment of clubfoot is indicated when clinically and radiologically proven residual deformities exist after conservative treatment of at least 12 weeks. Therefore non-operative treatment should be attempted before surgery [46, 51, 57–59].
59
The Current State of Treatment for Clubfoot in Europe
Fig. 18 Peritalar arthrolysis in a 7 month old child with bilateral clubfoot. The child is in supine position, the feet overhanging the end of the table. Pre-operatively the medial crease is marked in both feet. Post-operatively right foot is temporarily fixed with
a Kirschner wire. After Cincinnati approach and peritalar release the medial crease is diminished and a good plantigrade position is achieved (Fig. 20)
Surgical principles of the Peritalar Arthrolysis: Simons and McKay [8] as well as Krauspe and Parsch [50] described the surgical technique of the so-called “peritalar release” in detail (Fig. 18). Based on the pathological findings, sub-talar de-rotation “a la carte” by the Cincinnati approach is performed. The peritalar release allows a complete medial, posterior and lateral exposure of hind- and mid-foot as well as a correction of any deformity of the sub-talar, talo-navicular, and calcaneo-cuboid joints. Contracture of the joint capsules and ligaments of the sub-talar joint complex and if necessary of the ankle joint are dissected and lengthened. Restoration of the joint axis is obtained by correcting the position of the talar head to the sub-talar joints (Fig. 18). In older clubfeet bony deformity must be taken into account and the best possible position, especially of the articulating bearing areas must be achieved. Additional fixation with Kirschner wires of the talonavicular and the sub-talar joints is recommended.
Cast and dressing changes are performed on the second or third postoperative day in sedation and after 2 weeks. With change of the cast, a mould is made for the socalled “Copenhagen night splint” (Fig. 19). Removal of the suture after 2 weeks and of K-wires after 4–6 weeks, depending on the age of the child, is performed. 6–8 weeks after surgery. The cast is then removed and the Copenhagen night splint is applied.
Post-operative Care Post-operatively a padded long-leg plaster cast is applied with 90° knee flexion and the foot in 15°external rotation. Splitting the cast and keeping the foot elevated reduces post-operative complications due to swelling.
Therapy of Relapsed, Persistent Clubfoot or Clubfoot in Older Patients …clubfoot has a stubborn tendency to relapse… (Ignatio Ponseti)
The management of rigid clubfoot presenting after failure of surgical treatment is difficult and the results of further surgical intervention are sub-optimal. The options of treatment include extensive soft tissue procedures, osteotomies and triple arthrodesis. These procedures are associated with a shortened stiff foot. The External fixator can be a useful option because of its ability to correct by gradual distraction in a 3-dimensional manner as described by Franke et al. [60], Wallander et al.[61] and others [60, 62, 63]. Prem et al. [64] reviewed 14 patients with rigid clubfeet treated with the Ilizarov external fixator with a minimum follow-up of
60
5 years. They confirmed that the good short-term results reported previously are maintained at least 5 years after operation. In contrast, J. Freedman et al. [65] reviewed 21 resistant clubfeet treated with the Ilizarov external fixation after an average of 6.64 years. Mean age at correction was 5.7 years. They concluded that only the talo-first
R. Krauspe et al.
metatarsal angle displayed a clinically and statisticallysignificant correction. The rest of the deformity results in poor outcome associated with residual or recurrent deformity. As salvage procedure triple arthrodesis is recommended (Figs. 21–23).
Fig. 19 The Copenhagen night splint keeps the knee joint in 90° flexion and, a hinge permits dorsiflexion at the ankle. The foot is fixed in 15° external rotation, the hindfoot is stabilized by a padded strap
Fig. 20 Eight year follow-up of a boy treated by classical casting and peritalar artholysis with reposition of the tibio-talar, talo- calcaneal, talo-navicular and calcaneo-cuboid joints, lengthening
of the ligaments at the age of 8 months. The scar is aesthetically pleasing and the boy is able to actively dorsiflex and plantarflex with slight overactivity of the supinators
The Current State of Treatment for Clubfoot in Europe
Fig. 21 Bilateral neurogenic clubfoot in a 29 years-old woman
Conclusions The Ponseti method has proven to be a very good and safe treatment in idiopathic clubfeet and secondary clubfeet in young patients. Relapse may again be treated by serial casting, splinting and soft tissue release if necessary. The French functional method is another option for conservative treatment in idiopathic clubfoot, but beginning early in the first 3 months is
61
Fig. 22 Three year follow-up after triple arthrodesis. Plantigrade, pain-free feet are achieved. The patient is able to wear normal shoes
mandatory. The effort in organisation, time and costs are more complex and equinus is often not fully corrected. Extensive surgery should only be performed after conservative treatment has failed to correct the deformity because stiffness and a poor functional outcome may occur. In relapsed clubfoot or older patients peritalar release or the use of the external fixator is the method of choice to achieve a plantigrade foot. As a salvage procedure triple arthrodesis may be a reasonable option in adults.
62
Fig. 23 X-ray ap. and lateral of the same patient 3 years after triple arthrodesis with full consolidation of the osteotomies
References 1. Ester AR, Tyerman G, Wise CA, Blanton SH, Hecht JT (2007) Apoptotic gene analysis in idiopathic talipes equinovarus (clubfoot). Clin Orthop Relat Res 462:32–37 2. Shyy W, Wang K, Sheffield VC, Morcuende JA (2010) Eva luation of embryonic and perinatal myosin gene mutations and the etiology of congenital idiopathic clubfoot. J Pediatr Orthop 30(3):231–234 3. Carney BT, Coburn TR (2005) Coburn, demographics of idiopathic clubfoot: is there a seasonal variation? J Pediatr Orthop 25(3):351–352
R. Krauspe et al. 4. Honein M, Paulozzi L, Moore C (2000) Family history, maternal smoking and clubfoot: an indication of gene- environment interaction. Am J Epidemiol 152:658–665 5. Mattison DR (2010) Environmental exposures and development. Curr Opin Pediatr 22(2):208–218 6. Carroll N, McMurty R, Leete S (1978) The pathoanatomy of congenital clubfoot. Orthop Clin North Am 9(1):225–232 7. Westhoff B, Weimann-Stahlschmidt K, Krauspe R (2008) Der kongenitale Klumpfuß. Orthop Unfallchir up2date 3(2):95–114 8. McKay D (1982) New concept of and approach to clubfoot treatment: section I – principles and morbid anatomy. J Pediatr Orthop 2:347–356 9. Krauspe R, Westhoff B, Wild A (2006) Der Klumpfuß. Thieme-Verlag, Leipzig 10. Diméglio A, Bonnet F, Mazeau P, De Rosa V (1996) Orthopaedic treatment and passive motion Achine: consequences for the surgical treatment of clubfoot. J Pediatr Orthop 5(3):173–180 11. Pirani S, Hodges D, Sekeramayi F (2008) A reliable and valid method of assessing the amount of deformity In the congenital clubfoot deformity. J Bone Joint Surg Br 2008(90):53 12. Boehm S, Limpaphayom N, Alaee F, Sinclair MF, Dobbs MB (2008) Early results of the Ponseti method for the treatment of clubfoot in distal arthrogryposis. J Bone Joint Surg Am 90(7):1501–1507 13. Bor N, Yusef A (2008) The Ponseti treatment for clubfeet: the Afula experience with minimum of 5 years follow-up. J Bone Joint Surg Br 2008:512 14. Bor N, Herzenberg JE, Frick SL (2006) Ponseti management of clubfoot in older infants. Clin Orthop Relat Res 444:224–228 15. Chesney D, Barker S, Maffulli N (2007) Subjective and objective outcome in congenital clubfoot; a comparative study of 2004 children. BMC Musculoskelet Disord 8:53 16. Cosma D, Vasilescu D, Vasilescu D, Valeanu M (2007) Comparative results of the conservative treatment in clubfoot by two different protocols. J Pediatr Orthop B 16(5):317–321 17. Dobbs MB, Nunley R, Schoenecker PL (2006) Long-term follow-up of patients with clubfeet treated with extensive soft- tissue release. J Bone Joint Surg Am 88(11):2537 18. Eberhardt O, Schelling K, Parsch K, Wirth T (2006) Die behandlung des kongenitalen klumpfußes mit der Ponsetimethode. Z Orthop 144:497–501 19. Diméglio A (1995) Classification of clubfoot. J Pediatr Orthop B 4:129–136 20. Wainwright AM, Auld T, Benson MK, Thealogis TN (2002) The classification of congenital talipes equinovarus. J Bone Joint Surg Br 84-B:1020–1024 21. de Gheldere A, Docquier PL (2008) Analytical radiography of clubfoot after tenotomy. J Pediatr Orthop 28(6):691–694 22. Radler C, Manner HM, Suda R, Burghardt R, Herzenberg JE, Ganger R, Grill F (2007) Radiographic evaluation of idiopathic clubfeet undergoing Ponseti treatment. J Bone Joint Surg Am 89(6):1177–1183 23. Simons GW (1977) Analytical radiography of clubfeet. J Bone Joint Surg Br 59-B:485–489
The Current State of Treatment for Clubfoot in Europe 24. Maranho DA, Nogueira-Barbosa MH, Simão MN, Volpon JB (2009) Ultrasonographic evaluation of Achilles tendon repair after percutaneous sectioning for the correction of congenital clubfoot residual equinus. J Pediatr Orthop 29(7):804–810 25. Desai S, Aroojis A, Mehta R (2008) Ultrasound evaluation of clubfoot correction during Ponseti treatment: a preliminary report. J Pediatr Orthop 28(1):53–59 26. Suda R, Suda AJ, Grill F (2006) Sonographic classification of idiopathic clubfoot according to severity. J Pediatr Orthop B 15(2):134–140 27. Cahuzac J, Baunin C, Luu S et al (1999) Assessment of hindfoot deformity by three-dimensional MRI in infant club foot. J Bone Joint Surg 81B:97 28. Herzenberg JE, Carroll NC, Christofersen MR, Lee EH, White S, Munroe R (1988) Clubfoot analysis with threedimensional computer modeling. J Pediatr Orthop 8:257–262 29. Pirani S, Zesnik L, Hodges D (2001) Magnetic resonance imaging study of the congenital clubfoot treated with the Ponseti method. J Pediatr Orthop 21:719–726 30. Richards B, Stephens MD, Molly MD (2007) Dempsey, magnetic resonance imaging of the congenital clubfoot treated with the French functional (physical therapy) method. J Pediatr Orthop 27(2):214–219 31. Jacks D, Alvarez C, Black A, DeVera M (2009) Pedobarographic profiles of children with clubfeet. J Bone Joint Surg Br 2:235 32. Jeans KA, Karol LA (2010) Plantar pressures following Ponseti and French physiotherapy methods for clubfoot. J Pediatr Orthop 30(1):82–89 33. Trobisch P, Neidel J (2009) Comparison of clinical and pedobarographic measures in clubfeet treated with posteromedial soft-tissue release. Curr Orthop Pract 20(2):170–174 34. Zionts LE, Zhao G, Hitchcock K, Maewal J, Ebramzadeh E (2010) Has the rate of extensive surgery to treat idiopathic clubfoot declined in the United States? J Bone Joint Surg Am 92(4):882–889 35. Koureas G, Rampal V, Mascard E, Seringe R, Wicart P (2008) The incidence and treatment of rocker bottom deformity as a complication of the conservative treatment of idiopathic congenital clubfoot. J Bone Joint Surg Br 90-B(1):57–60 36. Mangat KS, Kanwar R, Johnson K, Korah G, Prem H (2010) Ultrasonographic phases in gap healing following Ponsetitype Achilles tenotomy. J Bone Joint Surg Am 92(6): 1462–1467 37. Saini R, Dhillon MS, Tripathy SK, Goyal T, Sudesh P, Gill SS, Gulati A (2010) Regeneration of the Achilles tendon after percutaneous tenotomy in infants: a clinical and MRI study. J Pediatr Orthop B 19(4):344–347 38. Maranho DA, Nogueira-Barbosa MH, Simão MN, Volpon JB (2009) Ultrasonographic evaluation of Achilles tendon repair after percutaneous sectioning for the correction of congenital clubfoot residual equinus. J Pediatr Orthop 29:804–810 39. Parada SA, Baird GO, Auffant RA, Tompkins BJ, Caskey PM (2009) Safety of percutaneous tendoachilles tenotomy performed under general anesthesia on infants with idiopathic clubfoot. J Pediatr Orthop 29(8):916–919
63 40. Janicki JA, Narayanan UG, Harvey B, Roy A, Ramseier LE, Wright JG (2009) Treatment of neuromuscular and syndrome-associated (nonidiopathic) clubfeet using the Ponseti method. J Pediatr Orthop 29(4):393–397 41. Ponseti IV, Zhivkov M, Davis N, Sinclair M, Dobbs MB, Morcuende JA (2006) Treatment of the complex idiopathic clubfoot. Clin Orthop Relat Res 451:171–176 42. Hegazy M, Nasef NM, Abdel-Ghani H (2009) Results of treatment of idiopathic clubfoot in older infants using the Ponseti method: a preliminary report. J Pediatr Orthop B 18(2):76–78 43. Khan SA, Kumar A (2010) Ponseti’s manipulation in neglected clubfoot in children more than 7 years of age: a prospective evaluation of 25 feet with long-term follow-up. J Pediatr Orthop B 19(5):385 44. Nagaraju KD, Vidyadhara S, Shetty AP, Venkatadass K, Rajasekaran S (2008) Use of Ponseti’s technique in recurrent clubfeet following Kite’s method of correction. J Pediatr Orthop B 17(4):189–193 45. Farsetti P, Caterini R, Mancini F, Potenza V, Ippolito E (2006) Anterior tibial tendon transfer in relapsing congenital clubfoot: long-term follow-up study of two series treated with a different protocol. J Pediatr Orthop 26(1):83–90 46. Park SS, Kim SW, Jung BS, Lee HS, Kim JS (2009) Selective soft-tissue release for recurrent or residual deformity after conservative treatment of idiopathic clubfoot. J Bone Joint Surg Br 91(11):1526–1530 47. Lampasi M, Bettuzzi C, Palmonari M, Donzelli O (2010) Transfer of the tendon of tibialis anterior in relapsed congenital clubfoot: long-term results in 38 feet. J Bone Joint Surg Br 92-B(2):277–283 48. Burghardt RD, Herzenberg JE, Ranade A (2008) Pseudo aneurysm after Ponseti percutaneous Achilles tenotomy: a case report. J Pediatr Orthop 28(3):366–369 49. Boehm S, Sinclair M (2007) Foot abduction brace in the Ponseti method for idiopathic clubfoot deformity: torsional deformities and compliance. J Pediatr Orthop 27(6):712–716 50. Krauspe R, Parsch K (1995) Die peritalare Arthrolyse zur Klumpfußkorrektur über den sogenannten CincinnatiZugang. Oper Orthop Traumatol 7:125–140 51. Graf A, Hassani S, Krzak J, Long J, Caudill A, Flanagan A, Eastwood D, Kuo KN, Harris G, Smith P (2010) Long-term outcome evaluation in young adults following clubfoot surgical release. J Pediatr Orthop 30(4):379–385 52. Souchet P, Bensahel H, Themar-Noel C et al (2004) Functional treatment of clubfoot :a new series of 350 idiopathic clubfeet with long-term follow up. J Pediatr Orthop B 13:189–196 53. Steinman S, Richards BS, Faulks S, Kaipus K (2009) A comparison of two nonoperative methods of idiopathic clubfoot correction: the Ponseti method and the French functional (physiotherapy) method. J Bone Joint Surg Am 91(Suppl 2, Part(2)):299–312 54. Pirani S, Carlson W (2006) Factors affecting compliance in the treatment of congenital clubfoot: the Uganda clubfoot project: 202 [abstract]. J Investig Med 54(1):S114–S115 55. Alvarez CM, De Vera MA, Chhina H, Williams L, Durlacher K, Kaga S (2009) The use of botulinum type A toxin in the treatment of idiopathic clubfoot: 5-year follow-up. J Pediatr Orthop 29(6):570–575
64 56. Alvarez CM, Tredwell SJ, Keenan SP, Beauchamp RD, Choit RL, Sawatzky BJ, De Vera MA (2005) Treatment of idiopathic clubfoot utilizing botulinum A toxin: a new method and its short-term outcomes. J Pediatr Orthop 25(2):229–235 57. Goyal R, Gujral S, Paton RW (2006) Long-term follow-up of patients with clubfeet treated with extensive soft-tissue release. J Bone Joint Surg Am 88-A(11):2536 58. Valentine KM, Uglow MG, Clarke NMP (2009) A comparison of Ponseti versus surgical treatment in congenital talipes equinovarus. J Bone Joint Surg Br 2:215 59. Vitale MG, Choe JC, Vitale MA, Lee FY, Hyman JE, Roye DP Jr (2005) Patient-based outcomes following clubfoot surgery: a 16-year follow-up study. J Pediatr Orthop 25(4):533–538 60. Franke J, Grill F, Hein G et al (1990) Correction of the clubfoot relapse using Ilizarov’s apparatus in children 8–15 years old. Arch Orthop Trauma Surg 110(1):33–37
R. Krauspe et al. 61. Wallander H, Aurell Y, Hansson G (2007) No association between residual forefoot adduction and the position of the navicular in clubfeet treated by posterior release. J Pediatr Orthop 27(1):60–66 62. Lapidus L, Odessky J, Shitrit R, Copeliovich L (2008) Design of Ilizarov fixator permitting simultaneous and independent clubfoot correction. J Bone Joint Surg Br 2008: 513 63. Mahendra A, Jain UK, Shah K, Khanna M (2008) Contro lled, differential, distraction in resistant, relapsed and neglected clubfeet. J Bone Joint Surg Br 2008:495 64. Prem H, Zenios M, Farrell R, Day JB (2007) Soft tissue Ilizarov correction of congenital talipes equinovarus – 5 to 10 years postsurgery. J Pediatr Orthop 27(2):220–224 65. Freedman Jason A, Watts Hugh, Norman Y (2006) The Ilizarov method for the treatment or resistant clubfoot: is it an effective solution. J Pediatr Orthop 26(4): 432–437
Part IV Polytrauma: Pelvis
Management of Pelvic Fractures Peter V. Giannoudis and Frangiskos Xypnitos
When to Treat?
Introduction Injuries to the pelvic ring are relatively uncommon, with a prevalence of 20–37/100,000 of the general population [1], whereas, in the polytrauma patients, their prevalence rises to 20% [2]. Overall, pelvic fractures account for approximately 3% of all skeletal fractures [3]. These injuries range from low energy stable fractures to high energy unstable patterns. Enormous injury forces are required to cause an unstable pelvic injury, particularly in young patients. The magnitude of this force is associated with substantial soft tissue injuries. Furthermore, this force is usually applied to other parts of the body causing injuries to other organ systems. A pelvic fracture therefore reflects only a portion of the destructive energy sustained by the patient. It is important to understand that this injury is usually seen in the spectrum of polytrauma and must be considered as a potentially lethal injury. Indeed despite the improvements made in prevention of injury, pre-hospital care, the widespread use of the ATLS protocol and the advances made in intensive care medicine, the mortality rate following pelvic fractures remains high in the region of 15% [4]. Mortality usually is due to both the pelvic fracture hemorrhage and the associated injuries to the central nervous system and to the chest. The management of pelvic fractures can be divided into the acute and the delayed reconstruction phase (Fig. 1). Hypotensive patients with these fractures present a major challenge in diagnosis and treatment of the bleeding source. A thorough knowledge of the anatomical structures contributing to pelvic stability and the source of the bleeding is essential for the assessment and treatment of these injuries.
P.V. Giannoudis () Academic Department of Trauma and Orthopaedics, Leeds General Infirmary, Clarendon Wing, Level A, Great George Street, LS1 3EX Leeds, UK e-mail:
[email protected]
Acute Treatment
Delayed Reconstruction
(Haemorrhage control) Within hours Clinical Outcome
Within 3-7 days Functional Outcome
Final Outcome
Fig. 1 Phases and timing of pelvic ring reconstruction
Pelvic Anatomy and Stability The pelvis is a ring structure formed by the two innominate bones and the sacrum. The stability of the ring derives from the surrounding supporting ligaments. Anteriorly, the ring is stabilised by the fibrocartilaginous pubic symphyseal ligaments (Fig. 2). Posteriorly, the ring is maintained by the posterior sacro-iliac ligaments, the posterior interosseous ligaments, the anterior sacro-iliac ligaments and the sacrotuberous ligaments (Fig. 3). The pelvic floor with its muscular layer also acts as a stabiliser of the pelvic ring. A stable pelvis is defined as one that can withstand normal physiological forces without undergoing abnormal deformation. The stability of the pelvic ring depends not only on the bone structures but also on the strong ligamentous structures holding firmly together the three bones of the pelvis. A pelvic disruption can involve an osseous injury, ligamentous injury or a combination of both. Tile illustrated the concept of stability and integrity of the ligamentous structures in a study where the pelvic ligaments were sequentially sectioned: (a) division of the pubic symphysis alone led to a pubic diastasis of 2.5 cm or less, (b) division of the anterior SI ligaments (sacrospinous and
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_5, © 2011 EFORT
67
68
P.V. Giannoudis and F. Xypnitos 15
d issections, all specimens with radiographic evidence of isolated anterior injuries had posterior injuries as well [7].
14
1
13 2 12 11
3 4
10
5
9
6
8
7
Fig. 2 Anterior stabilising ligaments of the pelvic ring. 1 Prom ontory, 2 Sacrum, 3 Sacrotuberous ligament, 4 Anterior inferior iliac spine, 5 Coccyx, 6 Obturator membrane, 7 Pubic symphysis, 8 Pubic tubercle, 9 Ischial spine, 10 Sacrospinous ligament, 11 Inguinal ligament, 12 Anterior superior iliac spine, 13 Anterior sacroiliac ligaments, 14 Iliolumbar ligament, 15 Anterior longitudinal ligament
14
1
13 12
2
11 10 9
3 4 5
6
8 7
Sources of Bleeding Arterial bleeding (iliac vessels and their branches to the inferior abdominal viscera and pelvic organs) is a major contributor to hemorrhagic shock in pelvic fractures (Fig. 4). Other sources of bleeding include the low pressure venous plexus and fractured cancellous bone surfaces. The retroperitoneum can contain up to 4 L of blood and bleeding will continue until intra-vascular pressure is overcome and tamponade has occurred.
Classification of Pelvic Ring Injuries Several classification systems have been developed over the years based on fracture location, pelvic stability, injury mechanism and direction of injury force applied [8]. Tile modified the Pennal system, based on the concept of pelvic stability, to create an alphanumeric system involving three groups: A (stable), B (rotationally unstable but vertically stable), and C (rotationally and vertically unstable) [9] (Table 1). Tile’s classification of pelvic ring fractures relates directly to the type of treatment indicated and the prognosis of the injury. The Young and Burgess classification [10] system is an expansion of the original classification developed by Pennal and Sutherland [11] where the fractures were classified
1 2
Fig. 3 Posterior stabilising ligaments of the pelvic ring. 1 Iliac crest, 2 Interosseous sacroiliac ligaments, 3 Greater sciatic foramen, 4 Lesser sciatic foramen, 5 Obturator membrane, 6 Coccyx, 7 Ischial tuberosity, 8 Sacrotuberous ligament, 9 Ischial spine, 10 Sacrospinous ligament, 11 Posterior sacroiliac ligaments, 12 Ilium, 13 Iliolumbar ligament, 14 L4 spinous process
3 13 12
11
sacrotuberous) led to a rotationally unstable but vertically stable hemipelvis, and (c) division of the posterior SI ligaments led to a completely unstable hemipelvis being able to displace both vertically and posteriorly [5]. Because the pelvis is a true ring structure, if the ring is broken in one area then there must be a fracture or a dislocation at another portion of the ring. This concept has been supported by the studies of Gertbein et al. who used bone scans to show that all anterior fractures that appeared nondisplaced on plain films featured a posterior ring injury as well [6]. Bucholz also reported that during cadaveric
4 5
10
9
Fig. 4 Major vessels in close proximity to the pelvis
6
7
8
69
Management of Pelvic Fractures
based on the direction of three possible injury forces: Anterior posterior compression (APC), lateral compression (LC) and vertical shear (VS). Young and Burgess developed
subsets on the LC and APC injuries to quantify the forces applied. They also added a forth injury force category of combined mechanical injury. Their classification system (Fig. 5, Table 2) helps with the detection of the posterior ring injury, predicts local and distant associated injuries, resuscitation needs and expected mortality rates. [10] AP III injuries require the most blood replacement, followed by VS patterns, followed by CM (combined injury patterns) followed by LC III injuries.
Table 1 Tile classification of pelvic fractures Type A
Stable
A1
Fractures not involving the pelvic ring; avulsion injuries of the iliac spines or the ischial tuberosity and isolated fractures of the iliac wing
A2
Minimal displacement of the pelvic ring
Type B
Vertically stable – Rotationally unstable
B1
“Open-book” injury
B2
LC injury – internal rotation instability – ipsilateral fractures
B3
LC injury – bilateral rotational instability
Type C
Vertically and rotationally unstable
C1
Unilateral injury
C2
Bilateral injuries with one hemipelvis vertically stable and the other unstable
C3
Bilateral fractures that are both vertically and rotationally unstable
Initial Assessment A high index of suspicion is mandatory for the diagnosis of pelvic fractures especially when the patient is uncooperative or unconscious due to the injuries sustained. It can be useful to speak to the ambulance attendants and get a history of the accident and to know if the patient developed signs of shock at any time. Inspection of the soft tissues both anteriorly and posteriorly is essential. The Destot sign (superficial hematoma above the inguinal ligament or in the scrotum or thigh) may indicate a pelvic fracture, as may a limb-length discrepancy or an obvious rotational deformity of the pelvis or
LC lateral compression
a
b
e
c
f
d
g
Fig. 5 Young-Burgess classification system. (a) AP I. (b) AP II. (c) AP III. (d) LC I. (e) LC II. (f) LC III. (g) VS
70
P.V. Giannoudis and F. Xypnitos
Table 2 Young and Burgess classification system Lateral compression (LC)
Direct or indirect lateral applied force – internal rotation of the ipsilateral hemipelvis. The anterior SI, SS and ST ligaments are shortened. Oblique fractures of the pubic rami.
LC – I
Sacral impaction on the side of impact – Transverse fractures of the pubic rami – Injury pattern stable
LC – II
Anterior transverse pelvic fracture with internal rotation of the hemipelvis towards the midline – Crescent iliac wing fracture – Injury pattern vertically, but not rotationally stable
LC – III
LC – I or LC – II with contralateral “open-book” (APC) injury (“windswept pelvis”) – SI, ST, SS ligamentous disruption
Antero-posterior compression (APC)
Force impact in the sagittal plane – direct or indirect. External rotation injuries – Pubic symphysis diastasis – Longitudinal pubic rami fractures.
APC – I
<2.5 cm symphyseal diastasis – Pubic rami vertical fracture (one or both) – Intact posterior elements – Injury pattern rotationally and vertically stable
APC – II
³2.5 cm symphyseal diastasis – Disruption of anterior SI, ST, SS and symphyseal ligaments – “Open book” injury – Injury pattern rotationally unstable and vertically stable
APC – III
Complete disruption of pubic symphysis (or anterior vertical fracture pattern) – Disruption of ST, SS, SI ligaments – Injury pattern vertically and rotationally unstable
Vertical Shear (VS)
Massive axial loading – Complete disruption of pubic symphysis or a vertical fracture pattern of one or both pubic rami – Disruption of ST, SS, SI ligaments – Vertical displacement of the hemiplevis – Occasionally, posterior injury presents as vertical transsacral or transiliac fractures
Combined Mechanical (CM)
Combination of injuries, usually VS + LC – Crush mechanism
SI (Sacroiliac), SS (Sacrospinous), ST (Sacrotuberous) ligaments
lower extremity. The patient should be kept immobilized for protection of the spine and the suspected pelvic fracture. Bimanual compression and distraction of the iliac wings as well as log rolling is not currently recommended. Such an approach could potentially dislodge an already formed clot (tamponade effect) and lead to recurrence of haemodynamic instability. Radiographs are the main diagnostic tool in the assessment of patients for suspected pelvic fractures. A single antero-posterior x-ray of the pelvis usually provides sufficient information to identify the pattern of injury and presence of fractures (Fig. 6). Lately the availability of whole body CT in the clinical setting has become a fast and efficient way to make an early diagnosis of the overall injuries sustained and initiate specific targeted treatment to the areas of clinical concern. In the haemodynamically stable patient further information of the nature of injury can be obtained by obtaining inlet and outlet radiographs as described by Pennal et al. [12]. Computer-assisted tomographic scans (CT) provide further detailed cross-sectional information and are very useful for identification of injuries of the sacro-iliac complex and occult fractures. Furthermore, three-dimensional reconstructions of the pelvis from computed tomography scans have also proven valuable in the non-emergency evaluation of these injuries. Rectal and vaginal examination should be performed to determine whether the pelvic injury communicates with
Fig. 6 AP radiograph illustrating a vertical shear pelvic fracture
these structures. A urological examination looking for scrotal hematoma, high riding prostate or blood at the urethral meatus in a man should raise the suspicion of a urethral injury. A distended abdomen can indicate an intraabdominal injury or may indicate the development of a retroperitoneal haematoma. Neurologic examination of the lower extremities is essential and should be clearly documented. Also importantly the distal pulses of the lower extremities must be carefully examined and documented.
71
Management of Pelvic Fractures
Haemodynamic Instability During the acute phase, the goal of treatment of high-energy pelvic ring disruptions is the prevention of early death from haemorrhage. The management therefore of the internal blood loss is of paramount importance initially. The source of bleeding must be identified and addressed accordingly. Sources of bleeding could be in one of the following anatomical areas: pelvic cavity, chest cavity, abdominal cavity, external sources (i.e., lacerations to skull or to lower extremity with profuse venous or arterial bleeding) (Fig. 7). Be aware that not infrequently, especially in the polytrauma setting, all of the above anatomical areas could be contributing to the development of haemorrhagic shock. The first step in the treatment of haemorrhagic shock includes the administration of intravenous crystalloid fluids and whole blood. When replacement of fluid and blood does not stabilise the patient’s vital signs, additional steps must be taken. Assuming that there is no solid organ injury and the primary source of bleeding is the pelvis, the legs should be tied together. Provisional stabilisation of the pelvic ring with a pelvic binder, a specially designed circumferential sheet, an external fixator or a C-clamp can restore a distorted pelvic ring to a stable configuration and thereby aid in tamponade of bleeding with persistent haemorrhage. The C-clamp consisting of two pins is applied on the posterior ilium in the region of the SI joints to provide compression and stability at the posterior aspect of the ring, at the point where usually the greatest bleeding occurs. Both the external fixator and the C-clamp may be applied in the trauma room or in theatre depending on the status of the patient (Fig. 8a, b).
In cases where haemodynamic instability continues despite the above measures then angiography could be considered as the next treatment option. Embolization avoids the retroperitoneal contamination that is associated with operative ligation of bleeding vessels and it preserves whatever tamponade effect is present in the retroperitoneal space. However, it should be stated that arterial bleeding amenable to embolisation is present in only about 10% of the cases [13]. Operative control of hemorrhage is generally not recommended and is reserved for bleeding from large vessels. The technique of retroperitoneal packing has been successfully used in some institutions where tamponades are applied in the pre-vesicular and pre-sacral spaces in an attempt to control bleeding [14] (Fig. 9). The packing is changed or removed 48 hours after injury. Other techniques for consideration include the ligation of hypogastric artery and the temporary occlusion of the aorta. A summary of the tools to be used for the control of acute pelvic bleeding is shown in Fig. 10.
Mechanical Instability The restoration of the mechanical stability of the pelvic ring by means of surgical reconstruction is performed when the physiological state of the patient has been stabilized. Usually, this phase of treatment takes place between the third and the seventh days following injury. However, this time window can be influenced by many unexpected parameters [15]. The indications for open reduction and internal fixation include pure ligamentous injury posteriorly, opening of the
Critical decision making: Source of bleeding Is the blood? In the chest?
Fig. 7 Sources of bleeding
In the belly?
On the floor?
Around pelvis?
72
a
P.V. Giannoudis and F. Xypnitos
b
Fig. 8 (a) Application of external fixator for an “open book” pelvic fracture. (b) Application of c-clamp for disruption of the right SI joint
Fig. 9 Laparotomy with pelvic packing performed
pubic symphysis anteriorly of more than 2 cm, associated acetabular fractures and vertically unstable fracture patterns. The surgical approaches most often utilised for reconstruction of the pelvic ring include the Phannestial approach, the Stoppa approach, the ilio-inguinal and the posterior approach to the SI joint [16]. Application of a plate anteriorly at the site of a symphyseal disruption after a laparotomy can simplify treatment. The current techniques for posterior internal fixation include percutaneous ilio-sacral screw insertion under fluoroscopic control, anterior or posterior fixation of the sacro-iliac joint with a plate or screws, use of cobra plates and binding of the injured hemipelvis to the opposite ilium with sacral bars [17–19]. Lately, in an attempt to overcome the morbidity of extensile surgical approaches percutaneous pelvic fixation has been receiving increasing attention [20].
These procedures may result in a shorter surgical time and reduce exposure-related hazards. The soft-tissue disruption with the potential for devascularisation or denervation and consequent necrosis is virtually eliminated. This technique does not decompress the pelvic haematoma, so early surgical stabilization is possible without the risk of additional haemorrhage. Percutaneous fixation is recommended when a number of essential criteria are met and only after an accurate reduc tion is achieved, thus avoiding residual displacement which can endanger the adjacent neural and vascular structures which is associated with compromised outcomes and function [20]. This technique is indicated in patients with significant soft tissue injuries that would complicate or prevent open treatment techniques providing a stable skeletal frame in anterior and posterior pelvic injuries. It can also be applied in patients with severe open fractures, faecal or environmental contamination, extensive closed de-gloving injuries, and abrasions or lacerations. Recently, lateral compression type I pelvic injuries have had a lot of attention in the scientific community with the argument being whether they are mechanically stable lesions (Fig. 11a–c). In the authors’ institution, during the past 5 years, for these types of lesions mechanical stability of the pelvic ring has been assessed in the operating theatre under general anaesthesia. Instability was defined as displacement >2 cm of the anterior or posterior elements. Out of 210 patients admitted with pelvic fractures, 40 had sustained fractures of LC1 type pattern. There were 23 female 17 male and with a mean age of 33.5 (range 18–68). The mean ISS was 10 (range 9–19). 23 patients (group 1) were found to have more than 2 cm. rotational displacement
73
Management of Pelvic Fractures Fig. 10 Shock management for pelvic ring injuries
Acute treatment: Shock management Pelvic bleeding
Endorotation legs (open book) Pelvic sling
fracture bleeding
C-clamp (complete unstable) External fixation (open book)
fracture bleeding
85–90%
ORIF anterior ring Peripelvic packing
venous bleeding
Selective embolisation
arterial bleeding
Surgical haemostasis 10 –15%
during EUA and were stabilised with SI screws posteriorly and a combination of retropubic screws, external fixator or plating anteriorly. 17 patients (group 2) exhibited minimal displacement less than 5 mm and were not stabilised. Rotational instability >2 cm was characterised by complete fracture of the sacrum posteriorly. Stabilisation of the pelvic ring in group I was associated with a significant reduction of the visual analogue score (VAS) for pain within 72 hours from surgery, early ambulation and discharge from the hospital.
Open Pelvic Fractures Open pelvic fractures are a special entity as they involve a direct communication between the pelvic fracture site and a vaginal, rectal, perineal or skin laceration. Contamination with perineal and enteric organisms is relative common in this type of injury increasing mortality and morbidity. Early diagnosis and aggressive management is essential and a thorough examination must be done so that no such injuries are missed. A diverting colostomy with pre-sacral drainage, rectal repair and distal rectosigmoid wash-out are the mainstays of treatment [21] (Fig. 12a, b).
Complications Genito-Urinary Injuries Injuries to the urinary tract are a recognised complication of serious pelvic trauma either from the initial impact or due to secondary injury from fracture fragments. The incidence of lower urinary tract injuries has been reported to be from 10% to 25% after pelvic fractures [22]. The spectrum of genitourinary trauma ranges from rare disruptions to kidney and ureter to more common injuries of urethra and bladder. The fixed membranous part of the urethra is more often the site of major injury in male patients. Bladder tears can be either intraperitoneal or extraperitoneal. Intraperitoneal bladder tears require immediate repair due to the potential contamination of the peritoneal cavity. Bladder neck injuries require immediate repair to prevent loss of sphincter function and incontinence. Urethral tears are usually managed with early re-alignment and delayed repair. A recent study revealed that early or late repair of the urethra is associated with the same rate of strictures but late repair has a lower incidence of impotence. A retrograde urethrogram-cystogram and an intr avenous pyelogram have become essential components of a genito-urinary work–up after pelvic fractures.
74
Fig. 11 (a) AP pelvic radiograph with lateral compression type I lesion. (b) CT scan revealing injury to the right sacral ala. (c) Fluoroscopic images illustrating stabilisation of pelvic
P.V. Giannoudis and F. Xypnitos
f racture with right SI and retropubic screw insertion following examination under anaesthesia and marked rotational instability
Fig. 12 (a) Open pelvic fracture. (b) Open pelvic fracture stabilised with external fixator. Diverting colostomy has been performed
75
Management of Pelvic Fractures
Neurological Injuries The incidence of neurological injuries following pelvic fractures ranges from 3% to 10%. The prevalence of neurological injury increases with the severity of the trauma and with the posterior involvement of the pelvic ring. In particular sacral fractures and disruptions of the SI joint have been noted to have a high prevalence of neurological injury, including avulsion of the lumbar nerve roots, superior gluteal nerve, involvement of the obturator nerve, sciatic nerve and cauda equina syndrome. Knowledge of the anatomy of the lumbosacral plexus and the major nerves in the pelvis, their dermatological and muscular innervations is mandatory for the correct diagnosis of the neurological lesion. Electromyography and nerve conduction studies are helpful in the diagnosis and evaluation of neurological deficits. Exploration has not been advocated for nerve root avulsions. However, if there is compression of the cauda equina, operative decompression is indicated. The long-term prognosis of neurological injury following pelvic fractures is variable and in general terms depends on the degree and level of the nerve-root involvement.
DVT-PE Pelvic fractures have a high incidence of proximal vein thrombosis leading to a high rate of pulmonary embolism (PE) (2–10%) [23]. However, in the acute phase where hemodynamic instability may be present anticoagulation treatment is not indicated. The argument exists that in this group of patients coagulation mechanisms are disrupted. Prophylaxis is usually initiated when the patient is stable and low molecular weight heparin is prescribed. Other means of anticoagulation include the use of mechanical pumps, and placement of vena cava filters for prevention of PE in the face of known or likely thrombosis in high risk patients.
Post-operative Treatment The goal of pelvic stabilisation is the early mobilisation of the patient to prevent respiratory failure, joint stiffness and DVT [24]. The post-operative mobilisation protocol is assigned by the surgeon and depends on the fracture stability and the existence or not of other associated injuries [25, 26]. Usually after a period of 8–10 weeks of partial weightbearing the patient is allowed to progress to full weightbearing. During the rehabilitation phase regular radiographic follow-up is essential to monitor the possibility of loss or reduction or metalwork failure.
References 1. Gansslen A, Pohlemann T, Paul C, Lobenhoffer P, Tscherne H (1996) Epidemiology of pelvic ring injuries. Injury 27 (Suppl 1):S-A13–S-A20 2. Matewski D, Szymkowiak E, Bilinski P (2008) Analysis of management of patients with multiple injuries of the locomotor system. Int Orthop 32(6):753–758 3. Jerrard DA (1993) Pelvic fractures. Emerg Med Clin North Am 11:147–163 4. Giannoudis PV, Grotz MR, Tzioupis C, Dinopoulos H, Wells GE, Bouamra O, Lecky F (2007) Prevalence of pelvic fractures, associated injuries, and mortality: the United Kingdom perspective. J Trauma 63(4):875–883 5. Tile M (1988) Pelvic ring fractures: should they be fixed? J Bone Joint Surg 70(B):1–12 6. Gertzbein SD, Chenoweth DR (1977) Occult injuries of the pelvic ring. Clin Orthop Relat Res 128:202–207 7. Bucholz RW (1981) The pathologic anatomy of Malgaigne fracture dislocations of the pelvis. J Bone Joint Surg Am 63:400–404 8. Fallinger MS, McGanity PLJ (1992) Current concepts review: unstable fractures of the pelvic ring. J Bone Joint Surg Am 74(A):781–791 9. Tile M (1996) Acute pelvic fractures: I. Causation and classification. J Am Acad Orthop Surg 4(3):143–151 10. Young JW, Burgess AR, Brumback RJ, Poka A (1986) Pelvic fractures: value of plain radiography in early assessment and management. Radiology 160(2):445–451 11. Pennal GF, Sutherland GO (1961) Fractures of the pelvis. American Academy of Orthopedic Surgeons, Park Ridge 12. Pennal GF, Tile M, Waddell JP, Garside H (1980) Pelvic disruption: assessment and classification. Clin Orthop Relat Res 151:12–21 13. Buckle R, Browner BD, Morandi M (1995) Emergency reduction for pelvic ring disruptions and control of associated hemorrhage using the pelvic stabiliser. Tech Orthop 9:258–266 14. Papakostidis C, Giannoudis PV (2009) Pelvic ring injuries with haemodynamic instability: efficacy of pelvic packing, a systematic review. Injury 40(Suppl 4): S53–S61 15. Katsoulis E, Giannoudis PV (2006) Impact of timing of pelvic fixation on functional outcome. Injury 37(12): 1133–1142 16. Jimenez ML, Vrahas MS (1997) Surgical approaches to the acetabulum. Orthop Clin North Am 28:419–446 17. Routt MLC, Kregor PJ, Simonian PT, Mayo KA (1995) Early results of percataneous iliosacral screws placed with the patient in the supine position. J Orthop Trauma 9: 207–214 18. Reynolds JH, Attum B, Acland RJ, Giannoudis P, Roberts CS (2008) Anterior versus posterior pin placement of pelvic C-clamp in relationship to anatomical structures: a cadaver study. Injury 39(8):865–868 19. Giannoudis PV, Papadokostakis G, Alpantaki K, Kontakis G, Chalidis B (2008) Is the lateral sacral fluoroscopic view
76 essential for accurate percutaneous sacroiliac screw insertion? An experimental study. Injury 39(8):875–880 20. Giannoudis PV, Tzioupis CC, Pape HC, Roberts CS (2007) Percutaneous fixation of the pelvic ring: an update. J Bone Joint Surg Br 89(2):145–154 21. Grotz MR, Allami MK, Harwood P, Pape HC, Krettek C, Giannoudis PV (2005) Open pelvic fractures: epidemiology, current concepts of management and outcome. Injury 36(1):1–13 22. Harwood PJ, Grotz M, Eardley I, Giannoudis PV (2005) Erectile dysfunction after fracture of the pelvis. J Bone Joint Surg Br 87(3):281–290
P.V. Giannoudis and F. Xypnitos 23. Montgomery KD, Geerts WH, Potter HG et al (1996) Thromboembolic complications with pelvic trauma. Clin Orthop Relat Res 329:68–87 24. Giannoudis PV, Bircher M, Pohlemann T (2007) Advances in pelvic and acetabular surgery. Injury 38(4): 395–396 25. Papakostidis C, Kanakaris NK, Kontakis G, Giannoudis PV (2009) Pelvic ring disruptions: treatment modalities and analysis of outcomes. Int Orthop 33(2):329–338 26. Madhu TS, Raman R, Giannoudis PV (2007) Long-term outcome in patients with combined spinal and pelvic fractures. Injury 38(5):598–606
Part V Shoulder, Elbow, Arm and Forearm
The Reverse Shoulder Prosthesis Carlos Torrens
Is There a Need for a Reverse Solution? In 1983 Neer described the clinical and pathological findings developed in some patients with long-standing massive tears of the rotator cuff in what he called “the cuff-tear arthropathy”. He believed that this condition may have to be considered as a distinct pathological entity which is extremely difficult to treat. Neer managed to treat these patients by total shoulder replacement with limited results [1]. Since then, several different constrained designs have been tried to improve function but all of them presented with early loosening of the components and no functional improvement at all [2]. Franklin in 1988 found that the superior migration of the humeral head associated with rotator cuff insufficiency was closely correlated with the degree of glenoid loosening. Superior displacement of the humeral head conditioned superior tipping of the glenoid component in what was called the “rocking horse” phenomena that lead to glenoid loosening [3]. Hemiarthroplasty was then considered to be the “gold standard” treatment of such patients. Since then many papers have been published detailing the results obtained with prosthetic replacement of the humeral head. Most of them conclude that good pain relief can be expected but with limited improvement in motion. Pre-operatively patients average 70º of forward elevation and after hemi-arthroplasty is performed there is an improvement to 90º or 120º. But if a deeper analysis of the results is done it can be appreciated that pre-operatively patients ranged from 30º to 150º of forward elevation, revealing that there were two distinct populations to be considered; on the one hand patients
C. Torrens Hospital del Mar de Barcelona, Passeig Maritim 25-29, 08003 Barcelona, Spain e-mail:
[email protected]
p resenting with a “pseudo-paralytic” arm (30º) and on the other hand patients who even though presenting with a massive cuff tear still have a reasonably good range of motion (150º). If we then look at the results it can still clearly be seen that hemi-arthroplasty has little effect on motion and that these two populations persist after treatment with ranges of motion from 15º to 160º, meaning that those “pseudo-paralytic” arm patients remain the same after treatment, and those with good range of motion slightly improve their motion [4, 5]. It can be said that some patients with massive cuff tears present as “pseudoparalytic” and have no improvement in motion after anatomical shoulder replacement is performed. This clinical situation is also present any time that biomechanically the rotator cuff does not work as in fracture sequela or when tuberosities ,after hemi-arthroplasties implanted for fracture, fail to heal. Even though these patients may not complain of pain their function is extremely limited and cannot be properly managed with an anatomical prosthesis. Orthopedic surgeons always try to mimic anatomy when replacing hips and knees thinking that this is the better way to obtain the best outcome, but things are different in cuffarthropathy shoulders and restoring anatomy just turns on pain relief. However, there is a need to change the shoulder anatomy and biomechanics if the aim is to restore motion.
What Does It Mean – A “Reverse System”? The relative incongruence in volume and radius of curvature of the humeral head and the glenoid provides a wide range of motion for the shoulder but requires the integrity of the “cuff-system” to hold the humeral head against the glenoid to allow the deltoid to raise the arm. Massive cuff tears fail to maintain the humeral head in contact with the glenoid and progressive upper migration of the humeral head is to be expected, limiting the function of the deltoid. The main purpose of the reversed system is to optimize the
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_6, © 2011 EFORT
79
80
C. Torrens
deltoid muscle to restore function. There have been previous constrained designs to improve biomechanics but all of them failed to restore function and components rapidly loosened too. In the 80’s Grammont introduced several changes in the constrained design that completely changed the outcome. Grammond’s design was semi-constrained, placing the ball in the glenoid and the socket in the humeral side. The glenoid was increased in size to obtain more stability and mobility. He also eliminated the neck of the glenosphere placing the centre of rotation directly in the articular surface of the glenoid. By medialization of the centre of rotation more deltoid fibers are recruited and also the shear forces are decreased. Bringing the humerus to an inferior location also favours deltoid action. Humeral components also have a non-anatomic inclination of 155º that diminishes the covering of the glenosphere, pushing the humerus inferiorly and giving more range of motion because the space between the prostheses and the acromion is increased [6–8]. Fig. 1 Shoulder radiograph in an elderly female patient with a cuff-tear arthropathy
When Is a Reverse System Indicated and What Results Can Be Expected? Paul Grammont initially designed the reverse system to treat cuff arthropathy patients, but recently the indication has been extended to all the clinical situations where the cuff is non-functional (massive cuff tears without arthritis, acute fractures, rheumatoid arthritis, fracture sequelae, primary arthritis in static posterior subluxation, revision surgery and tumour surgery).
Patients complaining of cuff-arthropathy normally have a preserved glenoid because of proximal humeral migration, but four different morphologies of glenoid erosion have been described with different clinical and long-term prognoses: glenoid without erosion (E0), concentric erosion (E1), superior erosion (E2) and superior erosion extended to the inferior pole (E3). E2 and E3 morphologies are associated with worse results [9–15]. Tha pre-operative status of the teres minor also has an influence on outcome. Patients with extended atrophy of the teres minor give worse clinical results [16].
Cuff-Arthropathy Cuff-arthropathy represents the main indication for a reverse system (Fig. 1). Its indication is limited to those patients with massive cuff tears and arthritic changes that clinically present a “pseudo-paralytic” arm. Patients having relative good function and needing only impro vement of pain can be properly managed with a hemi- arthroplasty. Patients treated with a reverse system can expect pain relief in the vast majority of cases (96%) and, as measured by the Constant Score patients with a preoperative level of pain of 3 improve to 13 after reverse shoulder implantation. Forward elevation and abduction also improves (from 73º pre-op to 138º post-op) while lateral rotation remains unchanged and internal rotation even can get worse.
Massive Cuff Tears Without Arthritis and “Pseudo-Paralytic” Arm In elderly patients with massive irreparable cuff tears and clinically a “pseudo-paralytic” arm a reverse system can be indicated even in the absence of arthritic changes (Fig. 2). In this selected population cuff repair and tendinous transfers are not indicated and reverse prostheses can offer the same result as in cuff-arthropathy patients with pain impro vement (from 3.8 pre-op. to 12 post-op. in the Constant score) as well as functional improvement (from 16.9 pre-op. to 28.4 post-op.) [11, 12, 15]. These patients with severe massive cuff tears but without arthritic changes present with a relatively-preserved anatomy that facilitates surgery.
The Reverse Shoulder Prosthesis
Fig. 2 Shoulder radiograph in a patient with a massive irreparable cuff-tear without associated arthrithic changes
81
Fig. 3 Shoulder radiograph in an osteoporotic elderly female patient with a 4-part fracture
affecting shoulder may still have some cuff this is usually very thin and is to be considered non-functional. Some of Given the unpredictable results obtained with hemi- these patients are too young to receive a reverse prosthearthroplasties for the treatment of complex acute proximal ses but good functional results have been reported in humeral fractures, the reverse solution has been considered elderly people with rheumatoid arthritis, but despite in especially elderly patients (Figs. 3 and 4). Bufquin in a these encouraging results rheumatoid patients develop series of 43 patients obtained a mean forward elevation of more complications than patients with other conditions. 97º, with a mean lateral rotation of 30º and a mean Constant Ritmeister, in a series of eight patients, reported two Score of 44. Even though a large series of complications loosenings of the glenoid component, one infection and was also described in 12 of these patients including glenoid three non-unions of acromial osteotomy used to approach fracture, neurological injuries, acromion fracture and dislo- the joint. Walch also reported that rheumatoid patients cation. Radiological examination shows 53% of tuberosity are the ones with largest number of complications migration, peri-prosthethic calcification in 90% of the [14, 18]. patients and 25% of scapular notch development. The good thing is that the results obtained are independent healing of the tuberosity. Probably over time, the reversed solution Fracture Sequelae will be more and more considered for the treatment of acute complex proximal humeral fractures, and new improved When non-union or mal-union of the tuberosities are presdesigns specific for fractures are imminent to obtain better ent following a fracture, the reverse system is indicated. results and decrease complications [17]. Limited results are to be expected but usually better than those obtained in similar situations with hemi- or total shoulder arthroplasties. Walch has reported a significant Rheumatoid Arthritis improve in pain (from 3.2 to 12.2 in the Constant Score) and function (from 10 to 20.6 in the Constant Score) The shoulder is a common site of rheumatoid affection although a higher number of complications also can be and even though some patients with rheumatoid arthritis present [14].
Acute Fractures
82
C. Torrens
described in a series of 19 patients an improvement of forward elevation from 49.7º to 76.1º with the use of the reverse system but with an incidence of complications of 32% [19, 20].
Tumour Surgery When surgery for tumours means the resection of all the proximal humerus with all muscular insertions, the reverse system can be considered to avoid proximal migration associated with the hemi-arthroplasty implanted in such situations. There are few studies but even though the functional outcome is reasonable, these procedures are associated with a high number of complications [21].
Do the Results Last?
Fig. 4 Radiograph of a reverse prostheses implanted for a 4-part fracture
Primary Arthritis in Static Posterior Subluxation Primary arthritis can be properly managed either with a hemi- or a total shoulder arthroplasty and never constitutes an indication for the reverse system except when a static posterior subluxation is present. In such a situation treatment with a hemi- or a total replacement arthroplasty does not address the posterior subluxation. Walch has reported good results with the reverse prosthesis for this specific condition and patients gain a significant improve in pain (from 3 pre-op. to 12 post-op. in the Constant Score) and function (from 12 pre-op. to 28 post-op. in the Constant Score) [14].
Revision Surgery When hemi- or total arthroplasty fails and there is a large cuff defect, the reverse system can be considered. Even though significant pain relief can be expected, improvements in function are more limited. This surgery is also associated with a high number of complications. Levy
Glenoid loosening has always been the major concern in any total shoulder replacement, so when the reverse system came into the market most surgeons were aware of possible complications on the glenoid side. Mid-term follow-up has shown, as far as glenoid loosening is concerned, survival rates of 84% at 120 months for the glenoid component and of 91% at 120 months when prostheses replacement was the end-point. However there are significant differences according to aetiology. Patients with massive rotator cuff tear showed a survival rate of 95% compared with 77% for shoulders that had another aetiology. Functional outcomes do not seem to score as well over time, and the overall survival rate with a Constant score <30 as the end-point was 88% at 72 months and 58% at 120 months with no significant difference according to aetiology. The same has been observed concerning pain with an overall survival rate with a pain score <10 as the end-point of 81% at 72 months and 61% at 120 months with no significant differences according to aetiology [14].
What Is Still Unsolved with the Reverse System? The reverse system has been proved to consistently restore forward elevation in “pseudo-paralytic” arms because of massive cuff tears but still has some weak points and carries a number of complications. It has already been said that although component loosening does not appear to be a major concern, indications for reverse prostheses have been extended to any clinical
83
The Reverse Shoulder Prosthesis
situation where the cuff system will not work and these other indications carry an increased risk of component loosening. Infection has been reported to be as high as 4% but with a significant relation to aetiology. Patients who have had prior surgeries, revision surgery and tumour surgery are more likely to develop an infection than those with primary surgery for massive cuff tears [14, 15]. Dislocation remains the most common complication related to the reverse system and can be as high as 7.5%. Once again aetiology seems to have a strong influence on that. The vast majority of cases of dislocation are secondary to bad tensioning of the components. It has also been shown that delto-pectoral approach is more likely to produce dislocation than the antero-superior approach. Biomechanically bigger glenospheres also provide more range of motion and more stability. Early dislocations can be properly treated with closed reduction followed by a 30º of abduction immobilization for 4–6 weeks. In chronic dislocations open reduction is imperative and there is commonly a need to restore tension by adding lengtheners to the humeral component [14, 15]. Near 50% of the reverse prostheses develop what has been called “scapular notch”, that is to say, inferior and sometimes posterior glenoid erosion that can be progressive (Fig. 5). Several factors seem to facilitate scapular notch development, such as medialization of the centre of rotation, increasing the angle between the prostheses and scapular neck and the positioning of the glenoid component.
There are several different morphologies of the glenoid that also favour scapular notch development. Polyethylene disease also seems to contribute to scapular notch development. It has been classified as stage 0 when is not present, 1 when its extension does not reach the inferior screw, 2 when it reaches the inferior screw and 3 when it extends beyond the inferior screw. It has not been proved that the development of a notch has any relationship to glenoid loosening yet and it seems to be self-limiting over time. Some authors have found a correlation with scapular notch development and worst functional results while others do not. It is still not clear why some patients develop scapular notch and others do not and if this scapular notch has any influence on clinical outcome [9, 13, 14, 16, 22, 23]. Although the reverse system has been proven to be effective restoring forward elevation, it has no effect on external rotation. Medialization of the centre of rotation results in decreasing the number of deltoid fibres that can help in external rotation leaving the teres minor as the sole muscle to favour external rotation and also the limited lateral offset limits external rotation because of posterior bone impingement. To provide some external rotation a latissimus dorsi transfer has been postulated in association with the reverse prostheses in cases where the teres minor is torn. It is not clear if there is a significant improvement of external rotation by transferring the latissimus dorsi but significant better outcomes measured with the Constant score have been reported. It seems that there is a better arm control at 90º of abduction and that this improvement provides a more useful arm [24, 25].
Author’s Rationale for the Use of the Reverse System
Fig. 5 Shoulder radiograph with a scapular-notch following a reversed prostheses
The reverse system has proven to be useful in patients with massive irreparable cuff tears that clinically present with a “pseudo-paralytic” arm. In such patients reliable pain relief can be expected and restoration of forward elevation and abduction can also be achieved. Patients have to be aware that no improve in lateral rotation will be present and that internal rotation can even get worse, so if patients wish to restore lateral rotation, latissimus dorsi transfer may be considered. Few complications are to be expected in this selected population. As has been said before, functional outcome as measured with the Constant score may decrease over time so the best candidates for the reverse system are elderly people with massive irreparable cuff tears. In acute complex fractures where replacement is to be considered, pro. and cons. have to be discussed with the patient knowing that hemi-arthroplasty result is
84
unpredictable and that the reverse system carries a higher complication rate but with a more predictable result as far as forward elevation and pain relief is concerned as it is independent of healing of the tuberosities. Rheumatoid arthritis can be properly managed with the reverse system because the pathology resembles cuff deficiency. Care must be taken to select elderly patients because benefits may decrease over time and higher a number of complications. Revision and tumour surgery have to be carefully considered when no other option is available. Limited outcomes are to be expected with a high number of complications, especially dislocation and infection.
References 1. Neer CS, Craig EV, Fukuda H (1983) Cuff-tear arthropathy. J Bone Joint Surg Am 65A:1232–1244 2. Post M, Hskell SS, Jablon M (1980) Total shoulder replacement with a constrained prosthesis. J Bone Joint Surg Am 62A:327–335 3. Franklin J, Barrett W, Jackins S et al (1988) Glenoid loosening in total shoulder arthroplasty association with rotator cuff deficiency. J Arthroplasty 3:39–46 4. Sanchez-Sotelo J, Cofield RH, Rowland CM (2001) Shoulder hemiarthroplasty for glenohumeral arthritis associated with severe rotator cuff deficiency. J Bone Joint Surg Am 83A:1814–1822 5. Williams GR Jr, Rockwood CA Jr (1996) Hemiarthroplasty in rotator cuff-deficient shoulders. J Shoulder Elbow Surg 5:362–367 6. Baulot E, Chabernaud D, Grammont PM (1995) Résultats de la prothèse inversée de grammont pour des omarthroses associées à de grandes destructions de la coiffe à propos de 16 cas. Acta Orthop Belga 61:112–119 7. Boileau P, Watkinson DJ, Hatzidakis AM et al (2005) Grammont reverse prosthesis: design, rationale, and biomechanics. J Shoulder Elbow Surg 14:147S–161S 8. De Wilde L, Audenaert E, Barbaix E et al (2002) Con sequences of deltoid muscle elongation on deltoid muscle performance: a computerised study. Clin Biomech 17: 499–505 9. Boileau P, Watkinson D, Hatzidakis AM et al (2006) The Grammont reverse shoulder prosthesis: results in cuff tear arthritis, fracture sequelae, and revision arthroplasty. J Shoulder Elbow Surg 15:527–540 10. Frankle M, Siegal S, Pupello D et al (2005) The reverse shoulder prosthesis for glenohumeral arthritis associated with severe rotator cuff deficiency. J Bone Joint Surg Am 87A:1697–1705 11. Guery J, Favard L, Sirveaux F et al (2006) Reverse total shoulder arthroplasty. J Bone Joint Surg Am 88A:1742–1747
C. Torrens 12. Matsen FA, Boileau P, Walch G et al (2006) The reverse total shoulder arthroplasty. J Bone Joint Surg Am 88A:660–667 13. Sirvaux F, Favard L, Oudet D et al (2004) Grammont inverted total shoulder arthroplasty in the treatment of glenohumeral osteoarthritis with massive rupture of the cuff. J Bone Joint Surg Br 86B:188–195 14. Wall B, Nové-Josserant L, O’Connor DP et al (2007) Reverse total shoulder arthroplasty: a review of results according to etiology. J Bone Joint Surg Am 89A:1476–1485 15. Werner CML, Steinmann PA, Gilbart M et al (2005) Treatment of painful pseudoparesis due to irreparable rotator cuff dysfunction with the delta III reverse-ball-andsocked total shoulder prosthesis. J Bone Joint Surg Am 87A:1476–1486 16. Simovitch RW, Helmy N, Zumstein MA et al (2007) Impact of fatty infiltration of the teres minor muscle on the outcome of reverse total shoulder arthroplasty. J Bone Joint Surg Am 89A:934–939 17. Bufquin T, Hersan A, Hubert L et al (2007) Reverse shoulder arthroplasty for the treatment of three- and four part fractures of the proximal humerus in the elderly. J Bone Joint Surg Br 89B:516–520 18. Rittmeister M, Kerschbaumer F (2001) Grammont reverse total shoulder arthroplasty in patients with rheumatoid arthritis and nonreconstructible rotator cuff lesions. J Shoulder Elbow Surg 10:17–22 19. Levy J, Frankle M, Mighell M et al (2007) The use of the reverse shoulder prosthesis for the treatment of failed hemiarthroplasty for proximal humeral fracture. J Bone Joint Surg Am 89A:292–300 20. Levy JC, Virani N, Pupello D et al (2007) Use of the reverse shoulder prosthesis for the treatment of failed hemiarthroplasty in patients with glenohumeral arthritis and rotator cuff deficiency. J Bone Joint Surg Br 89B:189–195 21. De Wilde L, Plasschaert FS, Audenaert EA et al (2005) Functional recovery after a reverse prosthesis for reconstruction of the proximal humerus in tumor surgery. Clin Orthop Relat Res 430:156–162 22. Simovitch RW, Zumstein M, Lohri EL et al (2007) Predictors of scapular notching in patients managed with the delta III reverse total shoulder replacement. J Bone Joint Surg Am 89A:588–600 23. Torrens C, Corrales M, Gonzalez G et al (2008) Cadaveric and three-dimensional computed tomography study of the morphology of the scapula with reference to reversed shoulder prosthesis. J Orthop Surg 3:49 24. Boileau P, Chuinard C, Roussanne Y et al (2007) Modified latissimus dorsi and teres major transfer through a single delto-pectoral approach for external rotation deficit of the shoulder: as an isolated procedure or with a reverse arthroplasty. J Shoulder Elbow Surg 16: 671–682 25. Gerber C, Pennington SD, Lingenfelter EJ et al (2007) Reverse delta-III total shoulder replacement combined with latissimus dorsi transfer. J Bone Joint Surg Am 89A: 940–947
Part VI Spine (incl. Trauma)
Spine Injury: Polytrauma Management Benny Dahl
Introduction
Definitions
Patients with polytrauma often have injuries of the spine as a component in their injury pattern. Hence, the spinal injury is contributing to the systemic inflammatory response seen in trauma patients. The primary injury is often referred to as the first hit, and in general trauma treatment subsequent secondary events/hits, such as delayed surgical procedures, can cause further inflammatory response resulting in acute respiratory distress syndrome (ARDS) or multiple organ dysfunction and multiple organ failure (MOD/MOF). A prolonged primary surgical procedure can also have this effect, and this has led to the concept of initial damage control surgery awaiting definitive treatment until the inf lammatory response has settled [1, 2]. The development in Orthopaedic implants, surgical techniques and understanding of trauma-related pathophysiology through the last two decades has improved the treatment of patients with polytrauma [3, 4]. In spite of this development, there are still major challenges to be met, if further progress is to be obtained. One area is the treatment of polytrauma patients with spinal injuries. In these patients, the clinical question is rarely related to the choice between conservative or surgical treatment, but rather the type and timing of surgical intervention. This review will focus on the current status in these areas regarding surgical treatment of polytrauma patients with spine injury.
Polytrauma
B. Dahl Spine Section, Department of Orthopaedic Surgery, Rigshospitalet and University of Copenhagen, Copenhagen, Denmark e-mail:
[email protected]
Although most surgeons to a certain extend, have the same understanding of the term polytrauma, there is no established, evidence-based definition [5]. In some countries the term polytrauma refers to a situation where simultaneous life-threatening injuries are present in the same patient. In other countries polytrauma is designated to patients with a certain injury severity score and finally the two terms are used interchangeably [6]. In most trauma literature the Injury Severity Score (ISS) that is based on the Abbreviated Injury Scale (AIS) score is still widely used, even though there is general agreement that this method has major drawbacks; one being the lack of physiological parameters. This fact, combined with the fact that some injuries are often diagnosed some time after the patient’s arrival to the hospital, makes ISS less useful for clinical, prospective use; e.g., triage. In trauma research, though, it is valuable a methods for describing populations of trauma patients. Using the ISS a score >15 is considered severe injury and in some studies equals polytrauma, especially when more than one body region is involved. Even though this discussion can be regarded as being somewhat semantic in nature, it is imperative that international consensus is established to be able to compare the research results in this patient population. In patients with spine injury, the spine trauma itself can result in an ISS > 15, not necessarily reflecting injuries to other organ systems. Butcher et al. have therefore proposed that the term Polytrauma is defined at an injury to at least two body regions with AIS ³ 3 and with the presence of Systemic Inflammatory Response Syndrome (SIRS) at least 1 day during the first 72 hours after injury [5].
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_7, © 2011 EFORT
87
88
Spine Injury Versus Spine Fracture The terms spine injury and spine fracture should be used synonymously, underlining the importance of assessing both bony and non-bony structures related to the spine in the evaluation of the polytrauma patient. None of the terms, however, should be used if no structural pathology has been demonstrated after diagnostic imaging.
Neurologic Injury The extend of neurologic injury is assessed using the Frankel classification which is the basis of the ASIA imp airment scale [7]. The ASIA impairment scale assesses the injury of the spinal cord and the cauda equina, and most authors do not regard isolated root deficits as a true spinal cord injury, consequently classifying such cases as ASIA/ Frankel E [8].
Epidemiology of Spinal Injury in Polytrauma In a recent study, the experience with spinal fracture pati ents from a Level 1 trauma centre over a period of 5 years was reported [9]. The majority of fractures were located at the thoracolumbar region from T11 to L2, and 54% of all patients had their fracture in association with another injury. This risk depended on the level of the spine fracture so that a patient with a cervical fracture had a 65% risk of an associated injury, whereas this risk was around 50% for patients with thoracic and lumbar fractures. Approximately 20% of the patients had an additional spine fracture, and the likelihood of multisegmental fractures correlated with the number of associated injuries. E.g., a patient with a multilevel spinal fracture had more than 95% risk of an associated injury. These findings correspond to a previous study and underline the importance of always having the suspicion of one or more spinal injuries in a patient with polytrauma [10].
Fracture Classification It can roughly be estimated that about 60% of spinal fractures are located in thoracolumbar region from T11 to L2, with the remaining 40% being equally distributed at the other levels; C1-C1, C3-C7, T1-T10, and L3-L5 [9]. The principle of evaluating the three columns of the spine, proposed by Denis almost 30 years ago is still valid;
B. Dahl
also in patients with polytrauma [11]. The original scoring system, however, did not take into account the trauma mechanism, and most studies find that the reproducibility between surgeons is higher when the choice of category has direct consequences for the choice of treatment [12]. Therefore, regarding fractures in the thoracolumbar region, the Thoracolumbar Injury Classification and Seve rity Score (TLICS) has been developed and validated in a number of studies [13, 14]. The principle behind TLICS is to evaluate trauma mechanism/morphology, integrity of the Posterior Ligamentous Complex (PLC) and neurologic involvement (Figs. 1–4). In a panel review by spine surgeons from a number of international trauma centres, the process of deciding the treatment strategy in a patient with a thoracolumbar fracture relies on an assessment of injury morphology, neurologic status and the integrity of the posterior ligament complex [8]. For the cervical spine a Subaxial Cervical Injury Classification System (SLIC) has recently been developed, also aiming at guiding the surgeon in his/her choice of treatment [15].
Timing of Surgery The optimal timing of surgical treatment in patients with spinal fractures has been investigated in a number of studies. A number of weaknesses characterise the comparison of these studies. First of all, different time limits have been used to differentiate between early and late surgery; in most cases 24 and 72 hours, but in come studies several days. Secondly, the number of patients with neurologic injury varies, and in some studies is not described in detail. Finally, the studies to a variable degree, includes thoracolumbar fractures alone or fractures at other spinal levels. Most studies have divided the timing of surgery into less than 72 hours (early) and more than 72 hours (late) [16]. The reason for this distinction is somewhat unclear, but could be based on experimental studies showing benefit of early surgical treatment in an animal model of acute spinal injury [17]. These findings, however, have not been consistently found in clinical studies [18, 19]. One clinical randomised study in patients with cervical fractures did not demonstrate any benefit of surgery within 72 hours after the injury [20]. In the polytrauma patient with spine fracture and no neurological deficits, the question of spine surgery within 24 hours can be a special problem, since prioritising treatment of the additional injuries will be necessary. A number
89
Spine Injury: Polytrauma Management Fig. 1 (a) Flexion – distraction injury through bone. (b) Distraction extension injury (With permission from Dr. Vaccaro [8])
a
b
Fig. 2 (a) Flexion distraction translational injury. (b) Translational injury (With permission from Dr. Vaccaro [8])
of factors may delay the surgical treatment of patients with spinal injury such as transportation from accident site to hospital, but also the lack of relevant resources at the primary hospital can play at role [10, 20, 21]. In a recent review assessing studies on surgical timing in spinal injury, it was concluded that there is low evidence that surgical stabilization of thoracic fractures within 3 days after the trauma reduces morbidity, primarily length of stay in the ICU and length of hospital stay. In patients with lumbar fractures only length of hospital stay was reduced, and no effect was found for thoracolumbar fractures or
mortality [22]. In that review, the authors did not differ between patients with and without neurologic injury. Also, no specific considerations regarding polytrauma or monotrauma were made, but judged by the Injury Severity Scores (ISS) of the studies included in the review; polytrauma had been a part of the patient population. In a major study including data from the National Trauma Data Bank, the data on more than 16,000 patients with surgically-treated spinal fractures were evaluated [23]. Fifty-nine percent of the procedures were performed within 3 days after injury defined by the authors as early
90
B. Dahl
Fig. 3 (a) Burst injury without disruption of the posterior ligamentous complex. (b) Nerve root injury in the setting of an L5 burst injury (With permission from Dr. Vaccaro [8])
Fig. 4 (a) Flexion distraction injury partially through soft tissue. (b) Burst fracture with disruption of the posterior ligamentous complex (With permission from Dr. Vaccaro [8])
fixation. The database allowed comparison between a small group of patients receiving early surgery and a group operated more than 72 hours after the injury (n = 374 and 497 respectively). It was concluded that the majority of patients undergo surgical fixation within 3 days after injury with fewer complications and less utilisation of resources. The weaknesses of studies focusing on timing of surgical treatment of spinal fractures, was underlined in a major recent review including 11 studies over the last 20 years, comparing early and late surgery [24]. Although the evidence is weak, there is strong recommendation that early surgery should be performed in polytrauma patients. Since most studies have demonstrated that the morbidity of a
posterior instrumented fusion is low, the benefits of early stabilisation in these patients regarding days in the ICU and pulmonary complications outweigh the risk of complications. In patients with compression of neural elements posterior decompression is also advocated. In patients with cervical fractures and neurologic compromise, the primary purpose of early surgical stabilisation is aimed at decompressing the spinal cord. There is, however, weak evidence and conflicting results supporting a strategy of early surgery and recent publications have underlined the need for development of evidence-based treatment algorithms in patients with cervical fractures [15, 20, 25–27].
91
Spine Injury: Polytrauma Management
Choice of Surgical Solution Surgical decision-making in patients with polytrauma and spine fractures depends on the general condition of the patient, the type of associated injuries and the characteristics of the spine fracture itself. In the thoracic and thoracolumbar fracture these variables include injury morphology (i.e. compression, translation, rotation, or distraction), neurologic status, and injury to the posterior ligamentous complex [8]. The majority of these fractures can be treated with early stabilisation with posterior pedicle screw instrumentation and decompression in case of neurological compromise. With the development of spinal implants, surgical and anesthesiological techniques, this procedure has a low morbidity with a complication rate of 1–2% [28]. Recently, an increasing number of studies have described similar results with percutaneous stabilisation [29]. Early posterior stabilisation increases the possibility of mobilising the patient, thereby reducing the risk of pulmonary complications. Also, supplemental radiological examinations can be performed allowing the surgeon to evaluate the need for additional anterior decompression and stabilisation at a later time. In patients with cervical fractures the same biomechanical considerations as in thoracolumbar fractures are necessary. Hence, the evaluation of fracture morphology and neurologic status has to be made [30]. Most compression and distraction fractures can be treated with a single anterior approach, discectomy or corpectomy and instrumented fusion, whereas fractures with an element of translation or rotation in most cases have to be stabilized with a posterior or combined approach. In most cases MRI is necessary to assess disc fragments in the spinal canal.
Conclusion The evaluation of the spinal column is routine in the assessment of any polytrauma patient, once the initial evaluation of airway, breathing, and circulation (A,B,C) has taken place. A neurologic evaluation is mandatory. In the early phase, the purpose of the evaluation is to identify unstable, or potentially unstable spinal fractures and provide initial stabilisation. The primary imaging procedures consist primarily of plain radiographs and CT. MRI is used to assess intervertebral discs, ligaments, and hematoma in the spinal canal or intramedullary. If cervical injuries are diagnosed a stiff collar or traction is often the primary treatment. In cases with neurological compromise
and malalignment of the spine, primary reduction is necessary. It is recommended that early stabilisation, within 24–72 hours, be done in polytrauma patients with spinal fractures, taking the general condition of the patient as well as associated injuries into consideration. The majority of thoracic and thoracolumbar fractures can be stabilised with posterior instrumentation, planning additional procedures at a later time. In these cases percutaneous instrumentation may prove to be a better solution than open surgery.
References 1. Giannoudis PV (2003) Surgical priorities in damage control in polytrauma. J Bone Joint Surg Br 85(4):478–483 2. Keel M, Trentz O (2005) Pathophysiology of polytrauma. Injury 36(6):691–709 3. Probst C et al (2009) 30 years of polytrauma care: an analysis of the change in strategies and results of 4849 cases treated at a single institution. Injury 40(1):77–83 4. Pape HC et al (2005) Timing of fixation of major fractures in blunt polytrauma: role of conventional indicators in clinical decision making. J Orthop Trauma 19(8):551–562 5. Butcher N, Balogh ZJ (2009) The definition of polytrauma: the need for international consensus. Injury 40(4):S12–S22 6. Sikand M et al (2005) The financial cost of treating polytrauma: implications for tertiary referral centres in the United Kingdom. Injury 36(6):733–737 7. Waring WP 3rd et al (2010) _ 2009 review and revisions of the international standards for the neurological classification of spinal cord injury. J Spinal Cord Med 33(4):346–352 8. Vaccaro AR et al (2006) Surgical decision making for unstable thoracolumbar spine injuries: results of a consensus panel review by the Spine Trauma Study Group. J Spinal Disord Tech 19(1):1–10 9. Leucht P et al (2009) Epidemiology of traumatic spine fractures. Injury 40(2):166–172 10. Harris MB, Sethi RK (2006) The initial assessment and management of the multiple-trauma patient with an associated spine injury. Spine (Phila Pa 1976) 31(11):S9–S15, discussion S36 11. Denis F (1983) The three column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine 8(8):817–831 12. Oner FC et al (2010) Therapeutic decision making in tho racolumbar spine trauma. Spine (Phila Pa 1976) 35(21): S235–S244 13. Vaccaro AR et al (2005) A new classification of thoracolumbar injuries: the importance of injury morphology, the integrity of the posterior ligamentous complex, and neurologic status. Spine (Phila Pa 1976) 30(20):2325–2333 14. Rihn JA et al (2008) A review of the TLICS system: a novel, user-friendly thoracolumbar trauma classification system. Acta Orthop 79(4):461–466
92 15. Patel AA et al (2010) Classification and surgical decision making in acute subaxial cervical spine trauma. Spine (Phila Pa 1976) 35(21):S228–S234 16. Croce MA et al (2001) Does optimal timing for spine fracture fixation exist? Ann Surg 233(6):851–858 17. Dimar JR 2nd et al (1999) The influence of spinal canal narrowing and timing of decompression on neurologic recovery after spinal cord contusion in a rat model. Spine (Phila Pa 1976) 24(16):1623–1633 18. Tator CH, Fehlings MG (1991) Review of the secondary injury theory of acute spinal cord trauma with emphasis on vascular mechanisms. J Neurosurg 75(1):15–26 19. Fehlings MG, Sekhon LH, Tator C (2001) The role and timing of decompression in acute spinal cord injury: what do we know? What should we do? Spine (Phila Pa 1976) 26(24):S101–S110 20. Vaccaro AR et al (1997) Neurologic outcome of early versus late surgery for cervical spinal cord injury. Spine (Phila Pa 1976) 22(22):2609–2613 21. Levi AD et al (2006) Neurologic deterioration secondary to unrecognized spinal instability following trauma – a multicenter study. Spine (Phila Pa 1976) 31(4):451–458 22. Bellabarba C et al (2010) Does early fracture fixation of thoracolumbar spine fractures decrease morbidity or mortality? Spine (Phila Pa 1976) 35(9):S138–S145 23. Kerwin AJ et al (2008) Best practice determination of timing of spinal fracture fixation as defined by analysis of
B. Dahl the National Trauma Data Bank. J Trauma 65(4):824–830, discussion 830-1 24. Dimar JR et al (2010) Early versus late stabilization of the spine in the polytrauma patient. Spine (Phila Pa 1976) 35(21):S187–S192 25. Mirza SK et al (1999) Early versus delayed surgery for acute cervical spinal cord injury. Clin Orthop Relat Res 359:104–114 26. Grauer JN et al (2009) The timing and influence of MRI on the management of patients with cervical facet dislocations remains highly variable: a survey of members of the Spine Trauma Study Group. J Spinal Disord Tech 22(2): 96–99 27. Kerwin AJ et al (2005) The effect of early spine fixation on non-neurologic outcome. J Trauma 58(1):15–21 28. Verlaan JJ et al (2004) Surgical treatment of traumatic fractures of the thoracic and lumbar spine: a systematic review of the literature on techniques, complications, and outcome. Spine (Phila Pa 1976) 29(7):803–814 29. Ni WF et al (2010) Percutaneous pedicle screw fixation for neurologic intact thoracolumbar burst fractures. J Spinal Disord Tech 23(8):530–537 30. Dvorak MF et al (2007) The surgical approach to subaxial cervical spine injuries: an evidence-based algorithm based on the SLIC classification system. Spine (Phila Pa 1976) 32(23):2620–2629
Surgical Management of Tuberculosis of the Spine Ahmet Alanay and Deniz Olgun
Introduction Tuberculosis of the spine is one of the most ancient diseases known to mankind, with reports of it dating back 5,000 years [1]. Despite the advances in the previous centuries, tuberculosis remains an important public health problem with close to 10 million new reported cases in 2008 [2]. First characterized by Pott in the late eighteenth century as ‘Pott’s distemper of the spine’, it still represents one-third of spinal infections today. Owing to the advent of effective public health measures, anti-tuberculous drugs and, although controversial, the Bacille-Calmette–Guerin vaccine, the incidence of tuberculosis has been declining steadily in the latter half of the twentieth century. However, with the emergence of first the AIDS and then the diabetes epidemics, tuberculosis is back on the rise even in developed countries. Today, patients with co-morbidities make up the bulk of cases, while antibiotic therapy remains the mainstay of treatment. Although spinal tuberculosis remains an uncommon diagnosis, it must yet be kept in mind in patients with spinal complaints whose aetiology is not readily apparent.
Aetiology Tuberculosis is caused by the pathogen Mycobacterium tuberculosis. It is transmitted mainly through inhalation or ingestion of the bacterium. Less than 10% of tuberculosis
A. Alanay Department of Orthopaedics and Traumatology, Istanbul Bilim University Faculty of Medicine, Istanbul, Turkey e-mail:
[email protected]
patients have musculo-skeletal involvement, yet 50% of these have involvement of the spine [3–6]. Neurologic deficit at the time of presentation is also common, reported to be between 10% and 60% [7]. Extrapulmonary tuberculosis seems to be increasing worldwide [8, 9]. With the increasing number of immune-compromised patients due to AIDS, auto-immune disease, cancer therapy and organ transplantation, the incidence of diseases caused by atypical mycobacteria has also increased. Atypical mycobacteria and fungi constitute a small percentage of the causes of spinal infection, but their clinical and radiologic appearances resemble those of Mycobacterium tuberculous spondylitis. The most common atypical mycobacterium isolated from vertebral osteomyelitis in one series was found to be mycobacterium avium-intracellular complex [10]. Although anti-tuberculosis drugs provide an effective weapon in the treatment of tuberculous spondylitis, the emergence of multi-drug resistant strains has caused a setback. Tuberculosis treatment is started with four so-called first-line drugs: isoniazid, rifampicin, pyrazinamide and ethambutol. Multi-drug-resistant tuberculosis is defined as that resistant to isoniazid and rifampicin [11–13]. The term “extensively drug- resistant tuberculosis” has been coined by the US CDC and the WHO to describe tuberculosis resistant to at least isoniazid and rifampicin and several second-line drugs [14, 15]. Multi-drug resistant spondylitis has been reported [16, 17]. Multi-drug and extensively drug-resistant tuberculosis represent failures of the aforementioned public health measures to control the disease and emphasizes the necessity of a proper drug regimen of appropriate duration and complete patient compliance. The incidence of these problems have been on the rise as well [18]. Before the discovery of anti-tuberculous drugs and modern surgical techniques, bed-rest and conservative immobilization were the mainstays of treatment of tuberculous spondylitis. This led to an extensive knowledge regarding the natural history of the disease [19, 20]. Untreated tuberculosis of the spine has three stages. The first is the stage of
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_8, © 2011 EFORT
93
94 Fig. 1 (a) A-P and lateral x-ray of a 42 year old male patient suffering back pain and neurological symptoms. Patient had a pathologic compression fracture of T9 vertebrae due to tuberculosis. (b) Sagittal MRI views demonstrate the abscess at T9 vertebral body and epidural compression due to abscess. (c) A-P and lateral x-rays. After anterior debridement, reconstruction with allograft and instrumentation was performed
A. Alanay and D. Olgun
a
b
95
Surgical Management of Tuberculosis of the Spine Fig. 1 (continued)
c
onset, lasting from 1 month to 1 year, the second the stage of destruction which can go on for up to 3 years and the last stage, the stage of repair and ankylosis. Abscess formation and destruction are seen in the second stage, which a third of the patients do not survive, while in the third stage, the joint or spine heals with bony ankylosis or fusion. Nonunion is associated with recurrences and super-infections with pyogenic bacteria, generally, an unfavorable outcome. Historic treatments of tuberculosis included bed rest, heliotherapy and sometimes plaster immobilization in order to pre-empt spinal deformity. Despite these measures, kyphosis still was a problem, many times accompanied by paraplegia as described by Pott [20, 21]. Spinal tuberculosis most commonly affects the thoracic or thoraco-lumbar spine, although cervical and lumbosacral involvement has been reported [22]. Spinal involvement has been classified by Mehta et al. according to anterior and posterior column involvement into four groups: anterior involvement only, anterior and posterior involvement, anterior or global with thoracotomy presenting grave risk, and posterior involvement only [23]. The most common is anterior involvement with destruction of the disc space and loss of anterior stability, making posterior laminectomy a greater destabilizing factor, should it be chosen as the method for treatment.
Pathophysiology Tuberculosis reaches the spine either through direct extension through the lungs or haematologic dissemination from a pulmonary or genitor-urinary source. Direct extension is rare, whereas the haematologic form of dissemination is far more common. The infection can appear in three distinct patterns: peri-discal, central and anterior [24], the most common of which is peri-discal involvement. The disease begins in the vertebral end-plate adjacent to the disc, extends anteriorly underneath the anterior longitudinal ligament and in this way multiple levels are infected while the intervertebral discs are spared. This presents a contrast to pyogenic spinal osteomyelitis where the disc is involved. Central involvement can lead to deformity. Anterior involvement can lead to spinal abscesses that span many levels. Primary posterior involvement is rare. As in spinal trauma, stability of the spine is lost if two or more columns are affected [25, 26], but this definition is not as clear-cut as in the case of acute fracture. Inflammation and destruction in tuberculosis co-exist with repair and fibrosis. Yet, the occurrence of a pathologic fracture or global disease affecting posterior elements as well may lead to the loss of stability [27]. With the loss of the support of the anterior column, acute
96
kyphosis develops. Once the disease reaches the healing stage, bony ankylosis is complete and the kyphosis is rigid. Once the pathogen is safely ensconced in living tissue, the inflammatory response of the immune system causes pus and debris to accumulate, forming abscesses and fluid collection. In contrast to pyogenic infection where proteolytic enzymes cause most of the destruction, in
A. Alanay and D. Olgun
tuberculosis the delayed -type hypersensitivity reaction of the body itself is the culprit [28]. Bone resorption follows. This may take place anterior to the anterior longitudinal ligament, extending downward to the psoas sheath and causing the well-defined psoas abscesses of Pott’s disease. It may also end up in the spinal canal, causing compression of the spinal cord. The neural structures may also be
a
b
Fig. 2 (a) A-P and lateral x-ray of a 60 year old male who had tuberculosis at T9 and T10 vertebrae. (b) Sagittal MRI scans demonstrating the abscess at T9-T10 vertebral bodies and the disc space. There is also an epidural abscess. (c) Figures demonstrating
the technical steps of decompression, fusion and instrumentation via a single posterior approach. (d) Follow-up A-P and lateral x-rays
97
Surgical Management of Tuberculosis of the Spine Fig. 2 (continued)
c
d
affected directly by tubercle formation, leading to neurologic deficit and even paraplegia. Causes of neurologic deficit include direct involvement of neural structures with the disease, compression by abscess and fluid formation, vascular compromise and compression by bony debris left over from the destructive process.
Diagnosis The presentation of tuberculosis of the spine can be variable. It depends on the extent of the disease, the nutritional status of the patient and the time that has elapsed since the
98
onset of disease. Back pain is a common presenting symptom. The pain is less severe than in pyogenic infection [24], follows an indolent course, often waxing at night and increases as instability progresses. Pott’s paraplegia, the gravest complication of the disease, is a presenting symptom in nearly 10% of the patients [29, 30]. Constitutional symptoms are also common such as fatigue, malaise, lowgrade fever, weight loss and the anaemia of chronic disease. Acute phase reactants such as white blood cell count, sedimentation rate and C-reactive protein may be elevated, but normal values do not rule out the disease. The patient may or may not have a history of pulmonary tuberculosis. Immunosuppression is a risk factor for the development of tuberculous spondylitis. In underdeveloped countries, patients may present with obvious deformity, sinus tract formation and even neurological deficit and paraplegia. Elderly patients are more likely to present with neurologic deficit. Late-onset paraplegia is defined as new-onset neurologic deficit after the first spinal infection has healed. It can occur many years after soft-tissue and bony healing have been completed. The reasons for late-onset paraplegia are numerous, some of which are re-activation, development of anterior bony ridges and subsequent cord compromise, chronic instability, increase in kyphotic deformity and rarely, degenerative changes in segments adjacent to those that have healed with significant deformity [31]. Radiographs in early disease are most commonly normal. Osteoporosis is the first sign that can be noted in x-ray studies, with loss of definition at the end-plates and only slight narrowing of the disc space [32]. These changes progress to loss of vertebral body height. Disk space is preserved until the disease progresses. Fusiform soft-tissue swelling in the thoracic region and the darkening of the psoas shadow in the lumbar region are other radiological changes that have been previously defined. The destruction of the anterior portions of multiple levels with sparing of the posterior elements will lead to a progressively worsening kyphotic deformity [28]. This kyphosis will progress until the last stage of the disease if it is left untreated. Sinus tract formation can occur during this process, and lead to pyogenic super-infection, which will in turn increase bony destruction and worsen deformity. Plain radiographs usually do not usually indicate the extent of the disease. Further imaging, preferably with MRI, is always necessary. CT scanning shows bony destruction and can be used for pre-operative planning of complex deformity or, more commonly, as a guide for needle biopsy in order to achieve tissue diagnosis. Bone scanning can be performed
A. Alanay and D. Olgun
but cannot differentiate tuberculous spondylitis from other causes of infectious disease and although it can be helpful in some cases, its use is limited. MRI remains the most helpful imaging modality in the diagnosis of tuberculosis, showing abscess formation, epidural involvement and involvement of the spinal cord as well as bony destruction. MRI is the modality of choice in vertebral osteomyelitis of any kind as it has very high sensitivity and specificity [33]. Also, MRI is non-invasive and has unequalled resolution for soft, especially neural tissues. MRI is the only modality to distinguish spondylitis of different aetiology [7, 34–37]. The earliest finding is end-plate oedema, which appears as a decreased T1-weighted signal and increased T2-weighted signal. Short-tau inversion recovery images are usually superior to other modalities as they allow the suppression of the bright fat signal of the bone marrow [38]. If the disc space is found to be preserved, the diagnosis of tuberculosis will become more than likely, as it is a pathognomonic finding of this disease. This relative sparing of the disc space is what differentiates it most from pyogenic infection. The infection progresses into the retropharyngeal soft tissue or sub-ligamentously to involve further spinal levels and the paraspinal areas. Abscesses show rim-enhancement with the addition of Gadoliniumcontaining contrast material, and therefore, cases with suspicion for spinal infection should always be examined with contrast unless otherwise contra-indicated [32]. This abscess wall in tuberculous is thick, and calcifications, though not always present, are also characteristic of the disease. Tuberculosis is known to mimic other conditions of the spine. One of these is metastatic disease, which can be differentiated from tuberculosis of the spine by the absence of paraspinal and other abscesses. Fungal spondylitis and spondylitis caused by atypical mycobacteria are far more difficult to differentiate by imaging findings alone and require tissue diagnosis. Radiographic changes may progress with the initiation of medical treatment for more than a year and should not be mistaken for the failure of treatment [39]. Gibbus formation (sharp kyphosis at affected levels), due to anterior column destruction is, seen in late untreated disease and conservatively-treated disease. This deformity may progress despite skeletal maturity and lead to late paraplegia. However, the increase in deformity is not the only cause of late paraplegia in healed disease. Other causes are compression of the spinal cord by bony bridges, calcified caseous material, fibrosis and disease re-activation [28].
99
Surgical Management of Tuberculosis of the Spine
Laboratory diagnosis is difficult. Purified protein derivative (PPD) or tuberculin skin testing has lost importance in the passing years. It is especially non-specific in areas where tuberculosis is endemic, BCG vaccination is routine, or the population is frequently exposed to subclinical disease [28]. A new blood test measuring interferon-gamma response after in vitro stimulation of the patient’s T-cells with tuberculosis antigens is being developed and could replace the less specific tuberculin skin testing and provide a tool for the detection of latent tuberculosis. There are also studies attempting to increase the specificity of the tuberculin skin test [40]. Sputum smears for acid-fast bacilli are one of the primary methods of laboratory diagnosis but are negative in patients without pulmonary tuberculosis and in a significant portion of patients with it. Although mycobacterial culture is quite sensitive, it requires direct tissue sampling in the case of spinal tuberculosis and is slow to yield results. Newer liquid culture systems such as BACTEC have reduced this delay to days rather than weeks with conventional methods and have been found to be more sensitive as well [41]. Diagnostic tests using nucleic acid amplification techniques and polymerase-chain reaction methods have been developed and show high specificity for tuberculosis, yet their cost and requirement for high-technology laboratory facilities coupled with their modest sensitivity have precluded widespread use [40]. Direct visualization of the granulomatous reaction and the presence of intracellular pathogens(acid-fast bacilli) under direct microscopy are the gold standard methods of diagnosis.
Indications for Surgery Today, the mainstay of treatment for tuberculous spondylitis is medical. Shortened time to disease onset and diagnosis have allowed tuberculous spondylitis to be caught before the development of complex spinal deformity. However, medical therapy alone has been shown to increase healing with kyphosis and deformity in many cases. The addition of bed rest and/or cast or brace immobilization was found to be ineffective in the development of kyphotic deformity in the British Medical Research Council Working Party on Tuberculosis of the Spine reports [42–45]. Multi-drug regimens (three or more drugs) of at least 6 months duration showing good healing responses, and advancement in minimally-invasive techniques to evacuate
huge abscesses led to re-definition of surgical indications. These are:
• Lesions not healing after 6 months of anti-tuberculosis
therapy • Lesions developing after 6 months of anti-tuberculosis therapy • Gross instability of the spine • New-onset neurologic deficit or worsening of prior neurologic deficit while under anti-tuberculosis therapy • Unacceptable or impending deformity
Pre-operative Planning Once the diagnosis of tuberculosis of the spine has been established, the patient should be started on anti-tuberculous therapy as soon as possible, preferably under the supervision of an infectious diseases specialist. Drug regimens based on isoniazid and rifampicin for at least 6 months have shown good results [46]. According to the recommendations issued by the United States Centers for Disease Control, a fourdrug regimen should be used to treat Pott’s disease. Rifampin and isoniazid should be administered during the therapy and another first-line drug chosen for the first 2 months along with one second-line drug. The duration of therapy should be at least 6 months, but as studies concerning special circumstances such as neurologic deficit and the involvement of multiple vertebral levels are scanty, some specialists still recommend therapy to last for 9–12 months. In the case of suspicion of multi-drug or extensively drug-resistant tuberculosis, proper consultations should be obtained.
Surgical Technique Many approaches to tuberculous spondylitis have been described. Before the advent of effective anti-mycobacterial therapy, surgery carried the quite large risk of sinus tract formation, leading to pyogenic infection and death of the patient. For this reason, indirect operations were favoured in order to increase stability and decrease recurrence, leading to the description of posterior fusion techniques. After effective anti-tuberculous therapy was shown to heal sinus tracts and ulcers, surgical therapy could directly deal with the problem at hand. Many techniques were described, most of them including radical resection of diseased tissues and massive reconstruction using structural grafts or cages.
100
Non-instrumented Posterior Fusion Posterior fusion was the preferred method of treatment in many centres before anterior spinal surgery was found to be safe and effective. The rationale behind posterior fusion is the achievement of a stable spinal segment in order to hasten healing and decrease the progress of kyphotic deformity. However, results of this technique were disappointing. Kyphotic deformity increased despite posterior fusion and prolonged immobilization, pseudarthrosis was common and healing was not found to be more rapid in several published series [47, 48]. Today, non-instrumented posterior fusion has been abandoned in the treatment of tuberculous spondylitis.
Anterior Radical Resection and Bone Grafting The “Hong Kong operation” was described by the British Medical Research Council Working Party on Tuberculosis of the Spine. It is the radical removal of all affected tissue until healthy, bleeding bone is encountered and subsequent reconstruction with bone graft with or without internal fixation, a modification of the original technique of Hodgson [49–51]. The reports on the Hong Kong operation, which does not employ instrumentation, are favourable in the long-term with very little loss of correction of kyphosis. However, there is a need for external bracing at least for 3–6 months until bony healing and incorporation of the graft material. On the other hand, it may be difficult to preserve the sagittal plane correction when more than one vertebral level has to be resected and either anterior or posterior instrumentation should be added when reconstruction spans more than one vertebral body. Debridement of all the necrotic and diseased segments and reconstruction of the anterior defect is still the key surgical principle for the treatment of tuberculosis. However, surgeons nowadays prefer to do either anterior or posterior instrumentation in addition to the Hong-Kong procedure to increase stability, preserve the correction in sagittal plane and to obviate the need for external braces.
Debridement (Anterior or Posterior) and Instrumentation The study by Oga et al. reporting the lack of glycocalyx capsule formation by tuberculosis bacilli has been a revolutionary step in the surgical treatment of tuberculosis
A. Alanay and D. Olgun
s pondylitis [52]. Many studies in the recent years have shown successful use of implants either anteriorly or posteriorly after debridement of necrotic tissues with no recurrence and exacerbation of the infection [50–53]. Both anterior and posterior instrumentation have been used in tuberculous spondylitis with success. Many combinations of the aforementioned approaches exist and should be chosen according to the patient’s special features, the resources available and Surgeon preference. Staged operations beginning with anterior debridement and continuing with posterior instrumentation, anterior debridement, posterior instrumentation and subsequent anterior instrumentation, and simultaneous anterior and posterior debridement and instrumentation have been defined and used with success [52, 54]. The thoracic vertebral column can be approached by an anterior trans-thoracic or posterior extra-pleural method. While the trans-thoracic method is straightforward, it may be associated with pulmonary complications post- operatively. The pulmonary condition of the patient before the operation should be carefully assessed and the risks weighed. The posterior extra-pleural method requires more surgical finesse, but may prevent further deterioration in patients with pulmonary co-morbidity. It also may be indicated in severe osteoporotic patients where anterior instrumentation may be unsafe and can be an alternative for combined anterior debridement and posterior instrumentation surgery. The postero-lateral approach as used for posterior vertebral column resection provides adequate exposure and allows the insertion of cages and other anterior struts. This is performed by a posterior approach. The upper and lower end levels are instrumented using pedicle screws. Once this is performed, one rod is inserted in order to prevent accidental movements. On the other side, costo-transversectomy is performed on as many levels as necessary. Nerve roots and intercostal veins are visualized, tied and then cut. Using a periosteal elevator, the exposure is extended to cover the entire circumference of the vertebral body. Once the anterior column is visualized, debridement is commenced. Debridement should remove all necrotic tissue, pus and loose bone fragments, but viable bone is not resected. Tissue sampling should be performed, with mycobacterial cultures and specimens for pathological study. After debridement and decompression anterior structural bone graft is placed and a rod is placed on the costo-transversectomy side and pedicle screws are compressed to increase the stability of the anterior graft. Authors have reported good results in tuberculosis as well with this technique [55]. Good results were achieved with the use of the posterior approach alone [53].
101
Surgical Management of Tuberculosis of the Spine
As deformity is often a result of tuberculous spondylitis, the necessity for instrumentation should be carefully evaluated.
Late Deformity With recent advances in surgical implants and techniques, the contemporary approach to severe kyphotic deformity includes instrumentation and spinal osteotomy, which can be done in an anterior-posterior-anterior fashion, simultaneous anterior surgery or posterior vertebral column resection (PVCR). These procedures, although challenging and prone to severe complications, have been used successfully for the treatment of late deformity [56–59]. Following spondylectomy, the resulting bone defect is filled with autograft or titanium mesh cages. Pedicle screw instrumentation and vertebral osteotomy are effective in the treatment of most forms of kyphosis, although spondylectomy is more appropriate for sharp, angular kyphosis as occurs following tuberculosis [59]. Previous studies have found that instrumentation of the spine afflicted with tuberculosis is safe [52] and that titanium mesh cages can safely be used in pyogenic infection as well. Fusion rates with any approach are acceptable and deformity correction is best with spondylectomy and pedicle screw instrumentation.
Minimally-Invasive Techniques Video-assisted thoracoscopic techniques have been described in the treatment of tuberculosis of the spine. They are especially appropriate for the procurement of tissue material for biopsy and culture, and mid-thoracic disease affecting few levels which is unrelated to pulmonary tuberculosis [60, 61]. Complications of tuberculosis of the spine, such as discrete abscesses and collections, can be successfully treated by percutaneous drainage placed under ultrasound or CT guidance [62].
Post-operative Care and Rehabilitation The post-operative care for a tuberculosis patient is no different than for any other spine patient, except for the obvious need for anti-tuberculous therapy. Anterior transthoracic
approaches are involved with a high degree of pulmonary compromise and may necessitate intensive care and prolonged intubation. Once the patient’s general condition permits, the patient may be mobilized according to the rigidity achieved by the instrumentation. Routine immobilization is not required with posterior pedicle screw fixation. Orthoses can be used for 6–12 months in those in whom a spondylectomy has been performed. The physical therapy regimen should follow the standards for spine patients.
Complications The complications that can be encountered depend on the extent of the disease, previous neurological deficit, the surgical approach selected, the type of graft used and the presence or absence of instrumentation. In patients with neglected disease, deformity is common. The spine usually heals in a kyphotic position. Kyphotic deformity in excess of 60° is associated with late paraplegia even in healed disease. Pain and cosmetic problems can also be seen. Kyphosis may increase with age despite fusion. Recurrence and re-activation of the disease if not treated properly with anti-tuberculous medication is also possible. Pyogenic infection may supervene in a spine already de-stabilized by tuberculosis and open to the exterior by sinus tracts. Complications of surgery include pulmonary complications especially for the anterior approach. Vessel injury and epidural bleeding can also be encountered during debridement due to the ossification and fibrosis of the tissues. Pedicle screw instrumentation is a safe and effective technique for the treatment of spinal disorders. Complications related to the use of pedicle screws can be related to the mal-positioning and faulty technique. Pull-out in osteoporotic bone has been reported and can be avoided in most cases with careful pre-operative planning. Neurological injury during the placement of pedicle screws is rare but catastrophic. The use of motor and sensory-evoked potential monitoring has been revolutionary in the safety of deformity surgery. Patients presenting with neurological deficit at the time of diagnosis usually have a favourable outcome with decompression and medical therapy. With better supportive care, intensive-care facilities and the better nutritional status of the patients, post-operative mortality has decreased. Miliary tuberculosis following surgery is rare with concomitant medical therapy. Non-union and mal-union are uncommon. Fusion rates in surgery for the tuberculosis of the spine have been
102
favourable even in historical reports where instrumentation was not available. Loss of correction is also a minor concern.
Summary Tuberculosis of the spine is an ancient disease that as a large public health problem has inspired research, the development of many surgical techniques and new drugs. While poor living conditions nurture the disease in developing countries, the falling incidence in developed countries following the discovery of effective anti-tuberculosis drugs has been pre-empted by the appearance of modern epidemics leading to overt or functional immuno-compromise. Starting in the pulmonary system, the disease spreads to the vertebral column via the haematologic route and causes significant disability and deformity, and may lead to neurological deficit. Several characteristic radiographic changes point to tuberculosis of the spine, the most notable of which is the early sparing of disk space. MRI is the best imaging modality in the diagnosis of tuberculous spondylitis. Diagnosis often requires tissue biopsy which can be done with minimally-invasive techniques, under CT guidance or during surgery. Treatment of tuberculosis of the spine is with antituberculosis drugs, but drug resistance is becoming a problem. According to the results of a series of studies by the British Medical Research Council on Tuberculosis of the Spine, multi-drug therapy combined with surgical intervention leads to best results. Combined with the advances in surgical technique, anaesthetic procedures and implant technology, the preferred treatment today is debridement, instrumentation and fusion of the spine. While good results are being obtained in patients with tuberculosis of the spine, further support of the public health measures are required in order to obtain eradication.
References 1. Nerlich AG et al (1997) Molecular evidence for tuberculosis in an ancient Egyptian mummy. Lancet 350(9088):1404 2. World Health Organization (2009) Global tuberculosis control: a short update to the 2009 report, World Health Organization, Geneva 3. Davidson PT, Horowitz I (1970) Skeletal tuberculosis. A review with patient presentations and discussion. Am J Med 48(1):77–84 4. Gropper GR, Acker JD, Robertson JH (1982) Computed tomography in Pott’s disease. Neurosurgery 10(4):506–508
A. Alanay and D. Olgun 5. Martini M, Ouahes M (1988) Bone and joint tuberculosis: a review of 652 cases. Orthopedics 11(6):861–866 6. Tuli SM (1975) Tuberculosis of the spine. New Delhi Springfield, Va.: Published for the National Library of Medicine, U.S. Dept. of Health, Education, and Welfare available from the U.S. Dept. of Commerce, National Technical Information Service. xviii, 163 p. 7. Boachie-Adjei O, Squillante RG (1996) Tuberculosis of the spine. Orthop Clin North Am 27(1):95–103 8. Kruijshaar ME, Abubakar I (2009) Increase in extrapulmonary tuberculosis in England and Wales 1999-2006. Thorax 64(12):1090–1095 9. Peto HM et al (2009) Epidemiology of extrapulmonary tuberculosis in the United States, 1993-2006. Clin Infect Dis 49(9):1350–1357 10. Petitjean G et al (2004) Vertebral osteomyelitis caused by non-tuberculous mycobacteria. Clin Microbiol Infect 10(11):951–953 11. Pablos-Mendez A et al (1998) Global surveillance for antituberculosis-drug resistance, 1994-1997. World Health Organization-International Union against Tuberculosis and Lung Disease Working Group on Anti-Tuberculosis Drug Resistance Surveillance. N Engl J Med 338(23):1641–1649 12. Espinal MA et al (2001) Global trends in resistance to antituberculosis drugs. World Health Organization-International Union against Tuberculosis and Lung Disease Working Group on Anti-Tuberculosis Drug Resistance Surveillance. N Engl J Med 344(17):1294–1303 13. Aziz MA et al (2006) Epidemiology of antituberculosis drug resistance (The Global Project on Anti-tuberculosis Drug Resistance Surveillance): an updated analysis. Lancet 368(9553):2142–2154 14. Holtz TH, Cegielski JP (2007) Origin of the term XDR-TB. Eur Respir J 30(2):396 15. Holtz TH (2007) XDR-TB in South Africa: revised definition. PLoS Med 4(4):e161 16. Cherifi S, Guillaume MP, Peretz A (2000) Multidrugresistant tuberculosis spondylitis. Acta Clin Belg 55(1): 34–36 17. Pawar UM et al (2009) Multidrug-resistant tuberculosis of the spine-is it the beginning of the end? A study of twentyfive culture proven multidrug-resistant tuberculosis spine patients. Spine 34(22):E806–E810 18. Migliori GB et al (2009) Multidrug-resistant and extensively drug-resistant tuberculosis in the west Europe and United States: epidemiology, surveillance, and control. Clin Chest Med 30(4):637–665; vii 19. Bailey HL et al (1972) Tuberculosis of the spine in children. Operative findings and results in one hundred consecutive patients treated by removal of the lesion and anterior grafting. J Bone Joint Surg Am 54(8):1633–1657 20. Tuli SM (2007) Tuberculosis of the spine: a historical review. Clin Orthop Relat Res 460:29–38 21. Bick EM (1976) Classics of orthopaedics. Series from clinical orthopaedics and related research 1, Lippincott, Philadelphia, xviii, 541 p 22. Hoffman EB, Crosier JH, Cremin BJ (1993) Imaging in children with spinal tuberculosis. A comparison of radiography,
Surgical Management of Tuberculosis of the Spine computed tomography and magnetic resonance imaging. J Bone Joint Surg Br 75(2):233–239 23. Mehta JS, Bhojraj SY (2001) Tuberculosis of the thoracic spine. A classification based on the selection of surgical strategies. J Bone Joint Surg Br 83(6):859–863 24. Tay BK, Deckey J, Hu SS (2002) Spinal infections. J Am Acad Orthop Surg 10(3):188–197 25. Denis F (1984) Spinal instability as defined by the threecolumn spine concept in acute spinal trauma. Clin Orthop Relat Res 189:65–76 26. Jain AK, Sinha S (2005) Evaluation of systems of grading of neurological deficit in tuberculosis of spine. Spinal Cord 43(6):375–380 27. Jain AK, Dhammi IK (2007) Tuberculosis of the spine: a review. Clin Orthop Relat Res 460:39–49 28. Luk KD (1999) Tuberculosis of the spine in the new millennium. Eur Spine J 8(5):338–345 29. Hodgson AR, Skinsnes OK, Leong CY (1967) The pathogenesis of Pott’s paraplegia. J Bone Joint Surg Am 49(6):1147–1156 30. Hodgson AR, Yau A (1967) Pott’s paraplegia: a classification based upon the living pathology. Paraplegia 5(1): 1–16 31. Luk KD, Krishna M (1996) Spinal stenosis above a healed tuberculous kyphosis. A case report. Spine 21(9):1098–1101 32. Joseffer SS, Cooper PR (2005) Modern imaging of spinal tuberculosis. J Neurosurg Spine 2(2):145–150 33. Modic MT et al (1985) Vertebral osteomyelitis: assessment using MR. Radiology 157(1):157–166 34. Jain R, Sawhney S, Berry M (1993) Computed tomography of vertebral tuberculosis: patterns of bone destruction. Clin Radiol 47(3):196–199 35. Kim NH, Lee HM, Suh JS (1994) Magnetic resonance imaging for the diagnosis of tuberculous spondylitis. Spine 19(21):2451–2455 36. Naim Ur-R et al (1997) Neural arch tuberculosis: radiological features and their correlation with surgical findings. Br J Neurosurg 11(1):32–38 37. Nussbaum ES et al (1995) Spinal tuberculosis: a diagnostic and management challenge. J Neurosurg 83(2):243–247 38. Stabler A, Reiser MF (2001) Imaging of spinal infection. Radiol Clin North Am 39(1):115–135 39. Boxer DI et al (1992) Radiological features during and following treatment of spinal tuberculosis. Br J Radiol 65(774):476–479 40. Pai M, O’Brien R (2008) New diagnostics for latent and active tuberculosis: state of the art and future prospects. Semin Respir Crit Care Med 29(5):560–568 41. Cruciani M et al (2004) Meta-analysis of BACTEC MGIT 960 and BACTEC 460 TB, with or without solid media, for detection of mycobacteria. J Clin Microbiol 42(5): 2321–2325 42. (1973) A controlled trial of ambulant out-patient treatment and in-patient rest in bed in the management of tuberculosis of the spine in young Korean patients on standard chemotherapy a study in Masan, Korea. First report of the Medical Research Council Working Party on Tuberculosis of the Spine. J Bone Joint Surg Br 55(4):678–697
103 43. (1973) A controlled trial of plaster-of-paris jackets in the management of ambulant outpatient treatment of tuberculosis of the spine in children on standard chemotherapy. A study in Pusan, Korea. Second report of the Medical Research Council Working Party on Tuberculosis of the Spine. Tubercle 54(4):261–282 44. (1976) A five-year assessment of controlled trials of inpatient and out-patient treatment and of plaster-of-Paris jackets for tuberculosis of the spine in children on standard chemotherapy. Studies in Masan and Pusan, Korea. Fifth report of the Medical Research Council Working Party on tuberculosis of the spine. J Bone Joint Surg Br 58-B(4): 399–411 45. (1985) A 10-year assessment of controlled trials of inpatient and outpatient treatment and of plaster-of-Paris jackets for tuberculosis of the spine in children on standard chemotherapy. Studies in Masan and Pusan, Korea. Ninth report of the Medical Research Council Working Party on Tuberculosis of the Spine. J Bone Joint Surg Br 67(1):103–110 46. (1998) A 15-year assessment of controlled trials of the management of tuberculosis of the spine in Korea and Hong Kong. Thirteenth Report of the Medical Research Council Working Party on Tuberculosis of the Spine. J Bone Joint Surg Br 80(3):456–462 47. Aksoy M et al (1995) Retrospective evaluation of treatment methods in tuberculous spondylitis. Hacettepe J Orthop Surg 5:207–209 48. Upadhyay SS et al (1994) The effect of age on the change in deformity after radical resection and anterior arthrodesis for tuberculosis of the spine. J Bone Joint Surg Am 76(5): 701–708 49. Cavusoglu H et al (2008) A long-term follow-up study of anterior tibial allografting and instrumentation in the management of thoracolumbar tuberculous spondylitis. J Neurosurg Spine 8(1):30–38 50. Benli IT et al (2000) The results of anterior radical debridement and anterior instrumentation in Pott’s disease and comparison with other surgical techniques. Kobe J Med Sci 46(1–2):39–68 51. Benli IT et al (2003) Anterior radical debridement and anterior instrumentation in tuberculosis spondylitis. Eur Spine J 12(2):224–234 52. Oga M et al (1993) Evaluation of the risk of instrumentation as a foreign body in spinal tuberculosis. Clinical and biologic study. Spine 18(13):1890–1894 53. Guzey FK et al (2005) Thoracic and lumbar tuberculous spondylitis treated by posterior debridement, graft placement, and instrumentation: a retrospective analysis in 19 cases. J Neurosurg Spine 3(6):450–458 54. Moon MS et al (1995) Posterior instrumentation and anterior interbody fusion for tuberculous kyphosis of dorsal and lumbar spines. Spine 20(17):1910–1916 55. Sundararaj GD et al (2009) Extended posterior circumferential approach to thoracic and thoracolumbar spine. Oper Orthop Traumatol 21(3):323–334 56. Thomasen E (1985) Vertebral osteotomy for correction of kyphosis in ankylosing spondylitis. Clin Orthop Relat Res 194:142–152
104 57. Berven SH et al (2001) Management of fixed sagittal plane deformity: results of the transpedicular wedge resection osteotomy. Spine 26(18):2036–2043 58. Wang Y et al (2009) Posterior-only multilevel modified vertebral column resection for extremely severe Pott’s kyphotic deformity. Eur Spine J 18(10):1436–1441 59. Macagno AE, O’Brien MF (2006) Thoracic and thoracolumbar kyphosis in adults. Spine 31(19 Suppl): S161–S170
A. Alanay and D. Olgun 60. Huang TJ et al (2000) Video-assisted thoracoscopic surgery in managing tuberculous spondylitis. Clin Orthop Relat Res 379:143–153 61. Kapoor SK et al (2005) Video-assisted thoracoscopic decompression of tubercular spondylitis: clinical evaluation. Spine 30(20):E605–E610 62. Pieri S et al (2009) Percutaneous management of complications of tuberculous spondylodiscitis: short- to mediumterm results. Radiol Med 114(6):984–995
Part VII Hand and Wrist
Scaphoid Fractures Joseph J. Dias
Applied Anatomy Anatomy (Shape, Blood Supply) The word “Scaphoid” is derived from the Greek word “skaphe” meaning a boat. This bone has an oblique orientation and bridges the distal and proximal carpal rows on the radial side. The scaphoid is unique among the carpal bones as it has six surfaces, four of which are articular facets. The surface area of the scaphoid [1] is 1482 (SD 212) mm2. Forty-two percent of this surface is articular covered with cartilage (662 (SD 95) mm2) and 58% is non-articulating (860 (SD137) mm2). The proximal surface of the scaphoid is convex and articulates with the scaphoid fossa of the radius forming the radio-scaphoid joint. The ulnar facet of the scaphoid is semilunar and articulates with the lunate at the scapholunate [SL] joint. Distally, the ulnar portion of the scaphoid is concave and articulates with the radial portion of the head of the capitate ([SC] joint). Finally, the most distal aspect of the scaphoid is convex and is divided by a sagittal smooth ridge separating the articulation between the trapezium laterally, and trapezoid medially (scapho-trapeziotrapezoid [STT] joint) [2]. The blood supply of the scaphoid is precarious and arises mainly from branches of the radial artery. A dorsal ridge serves as the attachment for the dorsal joint capsule and perforating branches from the radial artery supply 75% of the intra-osseous blood. Through the retrograde flow, the dorsal branches also supply the proximal pole. Only 67% of cadaver scaphoids from the 297 examined by Obletz and Halbstein had multiple arterial foramina throughout their
J.J. Dias Department of Orthopaedic Surgery, University Hospitals of Leicester, The Glenfield Hospital, Groby Road, Leicester, LE3 9QP, UK e-mail:
[email protected]
length. Of the remaining, 13% had foramina predominantly in the distal third, and 20% had most of the arterial foramina in the waist area of the bone with no more than a single foramen near the proximal third. One third of scaphoid fractures occurring in the proximal third may therefore be left without adequate blood supply and the prevalence of osteonecrosis in fractures here can be 35%. Fractures in the proximal third can be expected to take longer to heal and may have higher rates of non-union [3]. Gleberman and Menon found that branches of the radial artery, entering through the dorsal ridge supplied 70–80% of the interosseous circulation and the proximal pole. In the distal tuberosity, 20–30% of the bone receives its blood supply from the palmar branches of the radial artery [4]. In a recent study, Oehmke et al. using India ink instead of Ward’s blue latex solution and decalcification and staining instead of Spalteholz technique, have demonstrated that the palmar and dorsal side of the scaphoid had sufficient blood supply and even the proximal third of the scaphoid was supplied by multiple branches of the palmar carpal artery. They felt that inadequate vascularity was unlikely to be the cause of scaphoid non-union [1].
Ligamentous Attachments Although the scaphoid is mainly intra-articular and covered with cartilage, there are important sites for attachment of ligaments. Along the ulnar aspect of the proximal pole, the scapho-lunate interosseous ligament joins the scaphoid to the lunate [5]. When a scaphoid fractures, the proximal fragment extends with the attached lunate, and the distal fragment flexes, creating a “humpback” deformity [6]. Attaching directly on the palmar scapho-lunate interosseous ligament is the radio-scapho-lunate ligament, which acts as a neurovascular conduit and not a true ligament. Even more radial is the radio-scapho-capitate ligament, which has substantial insertions on the waist of the scaphoid. At the distal articulation of the scaphoid is the v-shaped scapho-trapezial ligament [7]. The scapho-capitate
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_9, © 2011 EFORT
107
108
J.J. Dias
ligament is almost confluent with the fibro-osseous tunnel of the flexor carpi radialis tendon, which runs directly palmar to the distal pole of the scaphoid as it heads toward the trapezium [8].
Radiological Anatomy (Fig. 1) Compson et al. [9] studied 50 dry Caucasian scaphoids and described the complex three-dimensional anatomy of the scaphoid. The shape of scaphoid can be simplified by look-
ing at the body and the tubercle as separate entities, with the body being bean-shaped and the tubercle offset from the distal end by about 45°. This explains the twisted appearance of the bone. The dorsal non-articular surface is long and thin and in the neutral position of the wrist, lies transverse to the long axis of the limb and parallel to the distal articular surface of the radius. It has two prominent features, a dorsal sulcus and a dorsal ridge. They further outlined these salient anatomical features using radiopaque markers, setting the bones in wax blocks and obtaining radiographs of the blocks in the same axis as
RADIAL
VOLAR
TUB TUB ST
LA LA
DA
DORSAL SULCUS DORSAL RIDGE
R-S
R-S DISTAL
TUB PROXIMAL S-T-T
S-C LA DA
Fig. 1 Four views of the right scaphoid demonstrate the complex shape. The fracture line depicts the pattern of waist fracture (With permission from Elsevier, from Compson et al. [9])
DORSAL SULCUS DORSAL RIDGE R-S
DORSAL
DA
S-L
ULNAR
109
Scaphoid Fractures
six “standard” scaphoid views. The pictures obtained were compared with clinical radiographs to define which films most clearly and consistently showed the anatomical landmarks. They concluded that, to assess scaphoid anatomy from plain radiographs, five views are needed. These are: ulnar deviated PA; ulnar deviated PA with 20° angulation to the elbow (or alternatively Ziter [10] or Stecher views [11]); lateral 45° PA (semi-pronated) oblique; and 45°AP (semi-supinated) oblique. The true anatomical waist is best seen on the ulnar deviated PA with 20° angulation to the elbow [12]. The 45° AP (semi-supinated) oblique is the best view to assess the flexion deformity found in some non-unions and essential for the peri-operative assessment of waist fractures.
Aetiology and Classification Mechanism of Injury Extreme dorsiflexion of the wrist with compressive force to the radial side of the palm causes the common middlethird scaphoid fractures. Weber and Chao created scaphoid fractures in cadavers with the wrist in 95–100° of dorsiflexion. The force was magnified four times at the radioscaphoid joint when the proximal pole was locked in the scaphoid fossa of the radius and the distal pole was forced dorsally. The palmar ligaments transmit tensile loads when the wrist is in marked dorsiflexion while the dorsal ligaments are lax [13]. This occurs during a fall on an outstretched hand. Falls backward, with a hand directed anteriorly, are most likely to force such extreme dorsiflexion [14]. Kozin examined several cadaveric studies using varying injury mechanisms and loading conditions. He concluded that the mechanism of injury does not predict a scaphoid fracture [15]. Avulsion fractures of the distal scaphoid occur in children and young adults and comprise nearly 2% of all hand and wrist fractures. The avulsion occurs following dorsiflexion-ulnar deviation stresses [14]. A few scaphoid fractures result from forced dorsiflexion by other means, such as ball sports where the ball strikes the palm of the hand, or with the force of the palm against a steering wheel in a motor vehicle accident. In the past, this “crank-handle kickback” was a frequent cause of scaphoid fracture. A direct blow to the scaphoid can also result in fracture. The great forces involved produced a high incidence of displaced, oblique, or unstable fractures [16]. Less common mechanisms of injury may involve forced palmar-flexion of the wrist [17] and axial loading of the
wrist with the hand clenched into a fist [18]. The history should also include the time of onset of pain and swelling after an injury.
Classification Three common classifications used for scaphoid fracture include the Russe, the Mayo, and the Herbert classification. The Russe classification is based on the inclination of the fracture line, and may predict the healing. Fractures may be horizontal oblique, transverse, or vertical-oblique [19]. The vertical-oblique type accounts for only 5% of fractures. This fracture pattern has the most shear forces across the fracture site, thus making it the unstable type. Horizontal oblique types have the most compressive forces across the fracture site, whereas transverse fractures have a combination of compressive and shear forces. The Mayo Clinic classification divides scaphoid fractures into the proximal (30% of fractures), middle (65% of fractures) and distal (5% of fractures) thirds. Within the distal third, the classification distinguishes between the distal articular surface and the distal tubercle. The location of the fracture influences both the union rate and time taken to heal. The rate of union in proximal, middle and distal third scaphoid fractures is 64%, 80% and 100% respectively [20]. The Herbert classification is difficult to understand and is supposed to be based on stability. The type A Herbert classification fracture is a stable acute fracture, and a type B is an unstable acute fracture. However stable fractures include fractures of the tubercle (A1) and an incomplete fracture of the waist (A2). These fractures can be treated non-operatively. By this definition all waist fractures of the scaphoid are unstable. All other types of fractures “may require surgical treatment”. Type B (acute unstable fractures) includes sub-types B1 (oblique fractures of the distal third); B2 (displaced or mobile fractures of the waist); type B3 (proximal pole fractures); type B4 (fracture-dislocations); and B5 (comminuted fractures). Type C fractures are those that demonstrate delayed union after >6 weeks of plaster immobilization, and type D fractures are established non-unions, either fibrous (D1) or sclerotic (D2) [21]. Compson et al. devised a radiological classification system by mapping the radiological fracture lines on transparent methylmethacrylate models of scaphoid bone. They matched the anatomical landmarks from 50 dry adult Caucasian scaphoids onto models by looking at standard radiological views. From the scaphoids where the fracture lines could be accurately defined they found three main fracture patterns: (1) the “surgical waist”, (2) the dorsal
110
sulcus, (3) the proximal pole. The fracture line in the dorsal sulcus was at 45° to the surgical waist and so was in the long axis of the bone. This group was further sub-divided into three variants: fracture line passing proximally, distally or on both sides of the apex. The butterfly fragment in the third sub-group showed displacement and comminution. He believed that the fractures of the distal third or the distal pole are due to radiological overlap of bone from the collapse of the sulcal fracture. The sulcal fracture was thought to be much less stable than a surgical waist fracture leading to the “hump-back” deformity [22]. Prosser et al. expanded the classification of distal pole fractures [23]. Type I were fractures of the tuberosity; type II, a distal intra-articular fracture; and type III, an osteochondral fracture. Some series have demonstrated limited prognostic value and poor inter- and intra-observer reliability of scaphoid fracture classification schemes [24]; nevertheless, these classifications are in common use in publications, and many believe they are helpful in determining treatment options and providing a prognosis.
Diagnosis Clinical Tests The diagnosis of a scaphoid fracture can sometimes be difficult, as patients may have normal radiographs early in their clinical course. Most patients demonstrate tenderness over the anatomic snuffbox or over the distal scaphoid tubercle, pain with longitudinal compression of the thumb, and limited range of motion and pain at the end arc of motion, especially with flexion and radial deviation [25, 26]. However, not all patients have pain over the scaphoid even with a fracture seen on radiographs. Overall, sensitivity is quite high for the clinical examination, although specificity is only 74–80% [13, 14]. An injured wrist with a fractured bone will have a bloody effusion that is palpable or visible and may be detected on ultrasound scanning but an effusion predicts a fracture with a sensitivity of only 50% and a specificity of 91%. Munk et al. concluded that ultrasound examination is unreliable for the diagnosis of acute scaphoid fractures. If the injury is very recent (<4 hours), the effusion might not have developed to a detectable level. Similarly, if the injury is several days old (>4 days), the effusion might have resolved [27]. Dias suggested that swelling in the anatomical snuff box, best seen by abducting the thumb to produce a concavity and comparing the depth of the concavity on the both sides, may only be seem after an interval. There is usually more than 20% reduction in grip strength measured with a Jamar dynamometer.
J.J. Dias
The “classic” sign of a scaphoid fracture is, of course, tenderness in the anatomical snuffbox. This finding alone is not sufficient to diagnose fracture of the scaphoid or even an occult fracture of the scaphoid. Snuffbox tenderness has a sensitivity of 90%, but its specificity is 40% for a scaphoid fracture. But tenderness at the scaphoid tubercle supports a diagnosis of scaphoid fracture with sensitivity of 87% and specificity of 57%. The scaphoid tubercle is located at the intersection of the distal wrist crease and the tendon of flexor carpi radialis. With radial deviation of the wrist, the scaphoid flexes producing a prominent bump on the radial side of the volar wrist. Palpating the scaphoid tubercle applies direct pressure on the scaphoid bone and will stress the fracture, especially in the radially-deviated wrist, whereas snuffbox palpation is less direct [28]. The absence of snuffbox and scaphoid tubercle tenderness virtually excludes a diagnosis of scaphoid fracture. Most wrist injuries resulting in joint effusion will produce snuffbox tenderness. A traumatized wrist with an effusion might be diffusely tender, and careful identification of the point of maximum tenderness is essential. Tenderness in the snuffbox alone may be found with fracture of the trapezium or radial styloid, as well as with de Quervain’s disease or osteoarthritis of the first carpo-metacarpal joint [29]. Chen has described the “Scaphoid Compression Test”, which is intended to discriminate scaphoid fracture from other causes of snuffbox tenderness. The Scaphoid Compression Test is performed by grasping the thumb of the affected limb in one hand while stabilizing the forearm in the other hand. In his series of 52 traumatized wrists with snuffbox tenderness, he reports very high sensitivity and specificity of this test for scaphoid fracture. Several authors describe tests for scaphoid fracture involving forced deviation of the wrist. These tests have poor specificity for scaphoid fracture. Powell et al. noted that in patients with scaphoid fractures pronation of the affected wrist followed by ulnar deviation produces pain in the anatomic snuff box that this is not present in patients without scaphoid fractures. The test gave a 52% positive predictive value and a 100% negative predictive value [30]. Others [31] have not found such high specificity. Davis et al. suggests the assessment should start on the ulnar side of the wrist and that one should assess the region of particular concern last. Tenderness over the scaphoid bone is worse than that over the scaphoid on the other wrist when a similar amount of pressure is applied. But these authors agreed that none of the special clinical tests or manoeuvres for diagnosis were specific and useful in the assessment of 12 clinical features for scaphoid fracture in 52 patients; 23 of these had a radiologically confirmed scaphoid facture. The signs were tested within a few days of injury and again 2 weeks later.
111
Scaphoid Fractures
Two hundred and fifteen consecutive patients with suspected scaphoid fractures were examined in a prospective study on two separate occasions to evaluate tenderness in the anatomical snuff box (ASB), tenderness over the scaphoid tubercle (ST), pain on longitudinal compression of the thumb (LC) and the range of thumb movement. At the initial examination ASB, ST and LC were all 100% sensitive for detecting scaphoid fracture with specificities of 9%, 30% and 48% respectively. These clinical signs used in combination, within the first 24 h following injury, produced 100% sensitivity and an improvement in the specificity to 74%. The authors suggested that these clinical signs are inadequate indicators of scaphoid fracture when used alone and should be combined to achieve a more accurate clinical diagnosis [26].
Various authors differ in their management of the injured wrist, when symptoms have failed to resolve 2 weeks later, and a scaphoid fracture is still suspected. The possibilities are to: (a) Advise the patient of the possibility of a fracture and 10% or so rate of non-union. (b) Advise the patient to restrict activity or to use a removable splint when using the hand to limit wrist, and therefore scaphoid, movement. (c) Give the patient the option of being treated in a cast with a 0.5% chance of subsequently identifying a fracture, and (d) Review the patient after 2 or 3 weeks; if at this interval, clinical signs still suggest a fracture (swelling, tenderness in the anatomical snuff box and more than 20% restriction of grip strength) further radiographs or other imaging are obtained.
Investigations “Clinical” Scaphoid fracture (diagnosis not certain): (Fig. 2) Initial radiographs detect a fracture when present in 70–90% cases [16] but it can be occult in approximately 16% of cases [32]. It can be up to 6 weeks before the fracture becomes evident on the plain films [31].
It is common to immobilize the wrist when the clinical suspicion is high but radiographs are normal. These patients are recalled at 10–14 days for further clinical and radiographic assessment until a definite diagnosis is established or symptoms resolve. As Barton stated “we overtreat a lot of patients to avoid undertreating a few” [33]. The time delay allows for the bony resorption adjacent to the fracture site, making the fracture visible [34]. But Dias et al. in their study of assessment of radiographs of scaphoid fractures by 20 different observers found that the errors in diagnosis made on the 2- to 3-week radiographs were comparable to those made on the initial films and that reliability did not improve when both sets were viewed together. The seniority and experience of the observer did not improve the ability to interpret radiographs correctly. It appears, therefore, that radiographs are of little value in the early management of suspected scaphoid injuries, and their management should depend upon careful clinical examination or different imaging [35]. Standard radiographs for a scaphoid fracture include four views of the wrist in most units:
• Postero-anterior (PA) with ulnar deviation; • Lateral; • Semi-pronated oblique; and • Semi-supinated oblique
Fig. 2 Bones scan showing a “hot spot” highly suggestive of scaphoid fracture
To improve diagnostic accuracy, it has been suggested that three [36], four [19] or even 16 radiographic views [37] were needed. Leslie and Dickson [16] reported that, in a series of 222 scaphoid fractures, 98% were visible on x-ray film at first examination. The remaining 2% became visible after 2 weeks. Compson [22] has suggested that to define all fractures, a series of scaphoid views should
112
J.J. Dias
include 45° PA oblique (semi-pronated) and 45° AP obli que (semi-supinated) views and one showing the long axis of the bone, such as an ulnar-deviated PA view with 20° angulation of the beam towards the elbow. He showed that surgical waist fractures are best seen on true lateral film and an ulnar deviated PA view with 20° angulation to the elbow. Dorsal sulcal fractures are best seen on a 45° PA (semi-pronated) oblique view on which the fracture line runs from the dorsal apex of the ridge adjacent to the lunate, and passes obliquely. Proximal pole fractures are specifically seen on a 45° AP (semi-supinated) oblique film, which will show the fracture crossing the radioscaphoid joint. Terry and Ramin in 1975 38suggested that wrist radiographs should also be evaluated for soft tissue signs of fracture. In particular, displacement of the scaphoid fat stripe (SFS), a radiolucent stripe adjacent to the radial side of the scaphoid as visualized on the PA film, should be sought. Radial convexity or obliteration of the SFS is considered diagnostic of fracture and sensitivity for this test was reported to be in the range of 95% [38] but other authors have found the soft tissue radiographic signs were unreliable [39].
Ultrasound Hauger et al. [40] described the use of high-spatial-resolution sonography in the diagnosis of occult fractures of the waist of the scaphoid. He studied 54 patients with clinically suspected scaphoid fracture and normal findings on initial radiographs, including scaphoid views with sonograms, which were compared with the results of radiography repeated 10–14 days after the initial trauma. Using cortical disruption as a diagnostic criterion, they found the sensitivity, specificity, and accuracy of high-resolution sonography in identifying scaphoid fracture to be 100%, 98%, and 98%, respectively. Fusetti et al. [41] suggested that three positive criteria of cortical interruption, radio-carpal effusion, and scapho-trapezium-trapezoid effusion could be interpreted as being highly indicative of fracture. But ultrasonography is highly user- dependent.
Bone Scan (Fig. 3) Leaving the injured wrist in a plaster cast for 2 weeks when a scaphoid fracture is suspected may, result in unnecessary
Clinical Scaphoid 10 to 16% of tender snuff box with normal xray
Nonoperative treatment
Investigate
2−6 weeks in plaster
SCAPHOID FAM COST £16
Risks
Cost £50−90
Can pick up other injuries
Overtreating
MRI scan
Specificity of Clinical exam 74 − 80%
95 −100% sensitivity & specificity
11−19%
Plan management
Loss of work time 2−6 weeks
Radiation
Cost
Cost
Cost per day $ 44.37
CT scan 95% specific
Limited CT (£57)
Plan other management Poor Sensitivity 16 − 21% Over treatment
Bone scan
Poor specificity False positive 25%
Other injuries Positive predictive (66%)
Fig. 3 Flow diagram of options and outcomes after clinical scaphoid fracture
113
Scaphoid Fractures
immobilization with an impact on return to work, need repeat radiographs and clinical examinations, and may require splint or cast changes [42]. An alternative is to investigate early with a bone scan, a CT or MRI scan. A Bone scan is a useful investigation for suspected cases of scaphoid fracture and it is sensitive but not specific [43]. Scintigraphy showed localized increased uptake in the scaphoid bone in only 19 of 84 patients with a “clinical” scaphoid fracture but only seven of these demonstrated a scaphoid fracture subsequently on repeat radiographs (five cases) or a CT scan (two cases),The rest were all normal . In another study by Murphy, only 8 of 54 patients (14.8%) had a scaphoid fracture. In six of these, the diagnosis was made on technetium bone scans, as the radiographs remained negative on re-assessment. This study confirmed the low incidence of scaphoid fractures in patients with a diagnosis of a “clinical scaphoid fracture”.
MRI Scan A magnetic resonance imaging (MRI) is superior to repeat radiograph for detecting an occult scaphoid fracture [44]. MRI scan has become the “gold standard” for early diagnosis of scaphoid fractures especially in the “clinical scaphoid”. MRI has better intra-observer agreement and fewer false positives and allows diagnosis of ligamentous injury or carpal mal-alignment. 195 patients attending emergency department with suspected scaphoid fracture and normal scaphoid series plain films over a 25-month period had a wrist MRI that allowed change in management in 90% [45]. The sensitivity and specificity of MRI for occult scaphoid fractures was better than that of bone scintigraphy [46]. MRI has sensitivity of 95–100% and with specificity of nearly 100% [47]. Brooks et al. [42] analysed the cost effectiveness of a magnetic resonance imaging scan (MRI) done within 5 days of injury compared to the usual management of occult scaphoid fracture in a randomized controlled trial involving 28 patients. The group having an MRI had shorter immobilization, decreased use of health care resources, but increased the cost of treatment. The cost differences bet ween standard follow-up and MRI was small in another study as 75% of patients with clinical evidence of a scaphoid fracture would be immobilized [48]. In acute fractures, Dynamic MRI scans after bolus administration of gadolinium estimates blood flow through bone although it doesn’t measure perfusion [49]. This technique assesses bone marrow vascularity and by implication, scaphoid vascularity [50]. There is no clear correlation between sclerosis seen on plain radiographs and blood flow assessed by dynamic MRI [51].
Magnetic resonance imaging can predict the vascularity of the ununited scaphoid with an accuracy of 100% compared to only 80% for surgical inspection and 40% for conventional radiography [52]. The signal is patchy and variable on T1- and T2-weighted sequences in both necrotic and viable bone [53]. Un-enhanced MR imaging cannot determine the degree of ischemia. Gadolinium contrast-enhanced MR imaging quantifies the extent of necrosis of the proximal fragment and has a good correlation with surgical and histologic findings and the subsequent healing of the nonunion [54].
CT Scan CT defines of the location, pattern, and displacement of the fracture. Sanders [55] identified the true longitudinal axis of the scaphoid. Sagittal-plane images are obtained by placing the patient prone in the scanner with the hand over the head, in full pronation and neutral flexion [56]. A reproducible image can be obtained with attention to the alignment of the scanning plane to the longitudinal axis of the scaphoid on the scout image, and verified with the “target sign”. The sagittal plane of the scaphoid is similar to that defined by the axis of the thumb metacarpal when fully abducted from the plane of the hand [56]. This allows assessment of bony architecture. Displacement is associated with failure of scaphoid fracture union [20, 57]. The following criteria define displacement in scaphoid fractures: (1) gapping of the fracture fragments of ³1 mm; (2) translation of the fracture fragments of ³1 mm; (3) angulation at the fracture site and (4) dorsal tilting of the lunate of >15° on a true lateral radiograph with the third metacarpal parallel to the radius. In a recent study [58], computed tomography scans of the scaphoid helped improve the intra- and inter-observer reliability of measurement of fracture displacement and angulation at the fracture site was the most reliable indicator of displacement. Bain et al. [59] described the lateral intra-scaphoid angle, the dorsal cortical angle, and the height-to-length ratio for measuring the “Hump-Back” deformity. The intra- and inter-observer reliability of the measurement of intra-scaphoid angle was poor; the dorsal cortical angle was moderate to excellent, and the height-to-length ratio was excellent. The measurement of the height-to-length ratio, dorsal cortical angle and lateral intra-scaphoid angle from sagittal MR slices of acute scaphoid fractures [60] and CT scans of normal scaphoids [61] is inaccurate and prone to inter- and intra-observer variability. However the height-to-length ratio is considered the most reproducible of the three measures of mal-union [59, 61], though not by all [60].
114
J.J. Dias
In a series of 50 patients, Bhat et al. [60] identified translation and gapping of the fracture fragments in the sagittal plane on the MRI scans and found that the three fractures that failed to heal were displaced. There were too few displaced fractures for their analysis to be reliable. Filan and Herbert [62] noted poor correlation between radiographic signs of displacement and operative findings. Amadio et al. [6] described the use of the trispiral tomography for the study of displacement. He found that patients with greater than 45° of lateral intra-scaphoid angulation present at the time of union had a satisfactory clinical outcome in 27% and post-traumatic arthritis in 54%.
Treatment of Scaphoid Fractures Cast Immobilization (Fig. 4) Casting for 2–3 months will predictably heal 90–95% of scaphoid waist fractures [39, 63], but possible patient inconvenience and work restrictions when in the cast have prompted some surgeons to advocate internal fixation with a screw [64, 65]. Level 1b prospective randomised studies have shown that short arm casts with the thumb left free provide adequate immobilisation of scaphoid fractures. Clay et al. [66] randomly allocated 392 scaphoid waist fractures to below-elbow casts with or without immobilization of the thumb to the interphalangeal joint. The outcomes were stratified according to fracture pattern and indicate a nonunion rate of 10% for transverse fractures regardless of the type of cast. Such a cast allows the use of the hand and the elbow is not immobilized. The position of the wrist does not affect healing. Hambidge et al. [67] randomized 121 waist and distal scaphoid fractures to immobilization in either slight flexion or slight extension using a below-elbow plaster cast with the thumb free and found an equal rate of failure of union, wrist flexion, grip strength, and pain, but patients immobilized in flexion had more trouble regaining extension.
Undisplaced Fractures There are at present six clinical trails comparing casting with surgery in acute scaphoid fractures. The rate of bony union for both methods is greater than 90%. The best approach to a patient with undisplaced acute scaphoid fracture is to consider patient’s personal circumstances and to discuss the risks and benefits of both non-operative management and surgery (Table 1). Yin et al. [63] performed a systematic review comparing casting and surgery for undisplaced scaphoid fractures. They found no difference in union rate or time to return to
Fig. 4 Below-elbow plaster cast used for treatment of scaphoid fracture
work, and any surgical benefits were transient (p > 0.05). Minor complications were higher in the group managed operatively. Dias et al. reported the greatest number of probable failures (10 of 44 with a cast vs 0 of 44 for operative treatment, p < 0.001). They defined non-union as absence of radiographic signs of healing at 12 weeks and a gap on CT scan at 16 weeks; however, one such non-union healed without additional treatment and 4 of 10 patients did not have visible fracture line or evidence of mobility at the time of surgery and their fractures could represent partial union [74] (Fig. 5). Davis et al. [75] performed a cost-utility analysis of open reduction and internal fixation (ORIF) versus cast immobilization for acute non-displaced mid-waist scaphoid fractures in a long-arm plaster which immobilized the elbow at a right angle and put the arm and hand in a position which would make use almost impossible. They concluded that time off work would be about 0.17 years (8.8 weeks) for surgery and 0.33 years (17.2 weeks) for casting. When only considering direct costs incurred by Medicare reimbursement, casting was less costly than ORIF ($605 vs $1,747). Arora et al. [76] allocated patients to cast or surgical
No of patients
53
25
88
60
62
52
Study
Adolfsson et al. [68]
Bond et al. [69]
Dias et al. [70]
McQueen et al. [71]
Saéden et al. [72]
Vinnars et al. [73]
Undisplaced
Acute
Scaphoid waist (displaced and undisplaced)
Undisplaced
Undisplaced waist
Undisplaced
Fracture type
Herbert screw versus plaster cast
Operated patients had scaphotrapezial arthritis 10 years after injury
Many operated patients had scaphotrapezial arthritis 12 years after injury
Operative group regained grip and pinch strength and range of motion more quickly
Colles’ cast thumb free, cannulated Acutrak screw Below elbow cast with thumb included versus operative group
Union rate 95% in plaster group, 100% in operative group
No differences between the groups in respect of function, radiological healing of the fracture No significant difference in symptoms, motion, grip strength or union
No differences In union rate, grip strength, range of motion at 2-year follow-up No differences after 12 weeks, 30% complication rate in operated group (minor scar-related complications) No difference in union rate, strength, or range of motion at 1 year
No difference in rate of union, final motion, or grip strength
Significantly better motion in operated group initially Time to return to work shorter after screw fixation
Comments
Differences
8 weeks in below elbow plaster cast, thumb free versus Herbert screw fixation
Long arm cast versus percutaneous cannulated Acutrak screw
Below-elbow plaster cast with thumb immobilized versus Acutrak screw
Treatment groups
Table 1 Review of randomized controlled trials: cast versus surgical fixation for acute scaphoid fractures
Scaphoid Fractures 115
116
J.J. Dias
Fig. 5 CT scan of scaphoid showing partial union
fixation and found a significant difference (p = −0.05) in immobilization time between the surgical and cast group (11 days vs 76 days) and a significantly shorter time (p = −0.05) off work (8 days vs 55 days). However, there were higher direct costs of surgery. The authors reported a 19% complication rate (4 of 21) with surgery, including one nonunion, one superficial wound infection, and two with complex regional pain syndrome. Unfortunately, the costs of complications were not considered.
Displaced Fractures Displaced fractures risk non-union. They can heal with conservative treatment but may mal-unite. Early fixation of acute scaphoid fractures offers no clear benefit over “aggressive conservative management” [70], and mild mal-union is well tolerated. A review of 49 scaphoid waist fractures treated conservatively for up to 13 weeks showed all 40 undisplaced fractures united, and six of nine displaced fractures also united, presumably with mal-union as no attempt was made to reduce the fractures. [60] This study found that the only three of the 49 (6%) fractures might have benefited from operative treatment to achieve union. Displaced fractures could benefit from re-alignment of the fracture fragments followed by stable internal fixation. In most cases, palmar exposure of the scaphoid limits injury to the blood supply of the scaphoid [77]. It is easier to address very proximal fractures through a dorsal exposure. The headless screw has been a popular device but using the alignment jig appropriately is technically difficult
and has been abandoned. The use of the jig could damage the scapho-trapezial joint [78]. Cannulated screw fixation, which does not require the use of a jig, has become popular for open as well as percutaneous fixations. Kirschner wires inserted into the each scaphoid fragment as joysticks allow manipulation of the fragments to achieve reduction. Additional Kirschner wires can be inserted to stabilize the fragments in a reduced position while the wire intended to guide the screw is placed. In patients with fracture comminution, particularly with compromise of the palmar cortex, primary bone grafting may be considered but is very uncommon. The position of the screw within the scaphoid may influence healing as the time to healing was shorter when the screw was placed in the central third. Central screw placement was achieved more consistently with cannulated screws than with Herbert screws [79]. The rate of success in obtaining a satisfactory reduction with the use of either closed or limited open techniques has not been evaluated. The study of the degree of mal-union is difficult and the correlation between mal-union and clinical symptoms has also not been established.
Return to Activity Most studies report improved grip strength and range of motion in those fixed compared with that for those treated in different casts between 8 and 16 weeks but no difference after that time or at final assessment. Prolonged aboveelbow cast immobilization is not well-tolerated, especially by younger patients who want to return to work and sports
117
Scaphoid Fractures
as soon as possible. However, most are happier with a below-elbow cast with the thumb left free. For athletes, treatment programs have been modified such that standard fibreglass casts are exchanged for soft or padded casts on game days. The soft casts are required to minimize the potential for injury to other athletes. This has not caused healing problems provided treatment is not delayed [80]. An alternative approach is operative fixation of minimally- or non-displaced fractures which allowed 12 athletes to return to sports such as basketball, baseball, and archery within a mean of 6 weeks. Clinical and radiographic union was evident in 11 subjects at an average follow-up of 3 years. The healing rate was comparable with other treatment methods like non-operative treatment in a plaster cast.
Fractures of the Proximal Pole of the Scaphoid Ten to fifteen percent of all scaphoid fractures involve the proximal pole [15]. Union rate after open reduction and internal fixation is around 66% [81]. The primary concerns are the small size and avascularity of the proximal fragment. The proximal pole has the most tenuous blood supply and this could explain the higher rates of non-union [4]. There is very little data on the behaviour of acute fractures of the proximal pole. However, prolonged immobilization and a higher rate of non-union have been reported in a few small series [66]. As a result many favour operative fixation of all fractures of the proximal third of the scaphoid. The choice of fixation depends on the size of the proximal fragment. If the fragment is large, a headless compression screw can be used. The type of screw is not as important as the starting point, which should be proximal and dorsal. It is critical to obtain good, preferably central, purchase on the proximal fragment, and ideally the screw should be placed orthogonal to the plane of the fracture. If the fragment is too small to accept such a screw, then Kirschner wires can be used to hold the fracture reduced, and sometimes trans-articular fixation is required.
Operative Technique Proximal pole fractures are approached dorsally and care taken not to disturb any cartilage healing. Bone graft may be required which can be harvested from the distal radius and one to two screws used to stabilize the fracture. Distal pole fractures are sometimes so thin and subchondral that they cannot be fixed using screws and multiple Kirschner wires may be needed.
Considerations 1. Patient’s Occupation 2. Time since Injury. (If more than 4 weeks without plaster, Consider ORIF) 3. Whether Definite fracture. (If not sure, consider CT/MRI scan) 4. Site of fracture. (a) Proximal (Consider ORIF) (b) Waist Fracture (c) Distal fracture 5. Displacement: More can 1 mm gap, step, translation, or DISI deformity, consider CT scan and/ or Fixation 6. Discuss Options with patient regarding time in plaster 7. Adequate follow-up and review 8. Possible CT scan at 8–12 weeks to confirm union
Consenting Process More than 90% will heal with either conservative or operative treatment. Risk of Non-union is present with both methods. Non-union should be stabilised to avoid osteoarthritis in future.
Pre-operative Preparation and Planning
Conservative Management
Important decisions before surgical fixations are the surgical approach to depending on site of fracture, extent of the exposure, ligament preservation or repair, graft requirement, possibly from iliac crest, and the use of a burr to shape the graft. Smokers are offered advice and help to curtail smoking as it impairs healing. Patient should be examined also for generalised ligament laxity.
1. 6 –8 weeks in plaster 2. Stiffness possible so allow function early in plaster 3. 7 –10% risk of Non-union 4. Time off contact sports 3–4 months 5. No difference in function at 6 months with either treatment
118
If fracture is displaced markedly leading to DISI deformity, consideration should be given to operative fixation as risk of Mal-union/Non-union.
Operative Management [70] 1. 5 –7% risk of Non-union 2. Scar complications (15%) 3. CRPS (3%) 4. Nerve injury (Palmar cutaneous Branch) (2%) 5. Infection (<1%) 6. Need for metalwork removal 7. Time off contact sports 3–4 months 8. No difference in function at 6 months with either treatment.
J.J. Dias
Dorsal Approach This approach allows visualization of the proximal portion of the scaphoid with the wrist in flexion. This is the preferred open approach to proximal pole fracture and is centred over Lister’s tubercle. A transverse or oblique skin incision is made and the extensor retinaculum is transversely incised in order to retract the tendons of the second and third dorsal compartments. The septum between the second and third compartments may need division. Care must be taken not to detach the capsule attachment to the dorsal ridge, as the main blood supply to the scaphoid runs within it. The wrist capsule is incised transversely and extended ulnarwards distally as needed, without injuring the deeper scapho-lunate ligament. A guide-wire is placed for a cannulated screw system or a mini-Herbert screw can be introduced freehand.
Observations Open Techniques
Five observations are made at surgery:
Palmar Approach
1. The presence of a gap at the fracture site. 2. Any mobility at the fracture site is noted as none, slight or marked with any sclerosis of the opposing fracture surfaces noted. 3. The vascularity of the fragments is noted as good bleeding from each surface, sparse bleeding or no bleeding. 4. The radio-scaphoid joint is assessed for the degree and extent of any arthritis. 5. Finally, scapho-lunate joint laxity is assessed by displacing the scaphoid proximal pole relative to the lunate.
This approach yields excellent visualization with less risk of injury to the main blood supply, which is on the dorsal side. This exposure is required if a screw is to be used retrograde. A longitudinal incision is made just radial to the flexor carpi radialis tendon, which is retracted to the ulnar side. Distally, the incision is carried over the tubercle of the scaphoid, forming a hockey-stick shaped incision. A longitudinal incision then is made in the volar wrist capsule, with partial division of the radioscaphocapitate ligament. The capsule over the scapho-trapezial joint is incised horizontally and the non-articular portion of the proximal part of the trapezium may be resected in order to gain central access to the distal part of the scaphoid in stiff individuals. The fracture is reduced using Kirschner wire “joysticks” if needed. Kirschner wires may also be employed to stabilize the fracture. A reduction/alignment guide can be used for a cannulated screw. A guide wire is passed under fluoroscopic guidance starting at the distal end for a cannulated screw system. This exposure allows inspection of the entire palmar surface of the scaphoid. The disadvantages are the risk of injury to the volar radio-carpal ligaments; the inability to assess the dorsal scapho-lunate ligament and the potential for scarring, which could limit wrist extension. There is also the risk to the palmar cutaneous branch of the median nerve lying on the ulnar surface of the flexor carpi radialis sheath.
Percutaneous Technique This approach is reserved for undisplaced fractures or those that can be anatomically reduced by closed or arthroscopically-assisted means. The patient is placed supine with the arm on a hand-table. With the palmar percutaneous approach, the distal aspect of the scaphoid is used as the entry point for fixation. The guide wire may need to be passed through the trapezium to obtain a central position in the scaphoid. A Mini-fluoroscopy unit is used to check position. In the dorsal approach, the proximal pole of the scaphoid is used as the entry point. A large-bore needle through the skin of the dorsum of the wrist guides the wire into the proximal pole of the scaphoid with the wrist held in flexion and ulnar deviation.
119
Scaphoid Fractures
Arthroscopy may be used in conjunction to see the fracture site but is seldom needed. If the fracture needs reduction, this is done with fluoroscopic guidance or arthroscopic visualisation using Kirschner wires as joysticks to manipulate the fractured ends of the scaphoid. If the fracture is not reducible by closed means, then an open approach is recommended. The arthroscopy allows assessment of other associated intraarticular injuries such as ligament disruption. A guide-wire is introduced percutaneously along the central axis of the scaphoid and a cannulated screw is used for fixation. The screw must be placed as central as possible as this provides most compression across the fracture. Various fixation methods can be used with the percutaneous technique including headless compression screws, HerbertWhipple screw, the Acutrak screw or other cannulated systems. The implant has to be advanced below the level of the cartilage on both ends of the scaphoid to prevent the development of radio-scaphoid or scapho-trapezial arthritis. The screw length is often about 4 mm shorter than the measured guide wire. Often the small incisions can be approximated using steristrips or one or two nylon sutures.
Post-operative Care and Rehabilitation This has to be individualised to the patient and the “personality” of the fracture. Hands are usually rested in a bulky bandage or a Futuro splint if patient compliance is assured. We use a below-elbow plaster cast for 6–8 weeks and then re-assess the need for a Futuro splint if we are uncertain of patient compliance. The patient is advised against contact sport for 2–3 months and repeatedly counselled about the risk of re-fracture. Assessment of union is by repeated clinical observations and serial radiographs and we believe union is a process rather than a single event at a specific time in the natural history of healing. Time to union is a flawed observation as it depends on when the observer conducts the radiographs or scans.
Complications Malunion Lindström and Nyström [82] assessed 229 acute scaphoid fractures which had united with non-operative treatment and found that 11% experienced persistent symptoms, including pain at rest (3%), restricted range of motion
(6%), pain with wrist motion (10%) and weakness of grip (11%). Twelve cases developed radiological changes indicating post-traumatic osteoarthritis between the radial styloid process and the scaphoid, though in all but one case this was only Grade 1 with slight decrease in joint space. Amadio et al. [6] has reported that scaphoid fracture malunion with foreshortening and flexion resulting in the humpback deformity is associated with impairment of function and the development of post-traumatic osteoarthritis. However he studied 45 of 105 patients and 26 of these had fracture non-unions that needed open surgery and bone grafting. Mal-union was assessed on trispiral tomography of the scaphoid, measuring the lateral intra-scaphoid angle and the functional outcome was measured with a modification of the Green and O’Brien’s scale. The lateral intra-scaphoid angle has subsequently been shown to be an unreliable assessment of scaphoid anatomy [59], and Green and O’Brien’s functional grading is not validated for scaphoid fractures [83]. The report may have been further confounded by the 20 of the 27 non-unions in the mal-union group, but only 6 of 19 without deformity in the control group. Treated non-union is likely to have a poorer outcome than an acute fracture if both united with the same degree of mal-union. While this study suggests the possibility of symptoms due to mal-union it does not provide evidence for this. Another study of 26 non-unions of scaphoid fractures treated successfully with Russe bone grafting found no association between mal-union and functional impairment 11 years after treatment, but a small but significant association between mal-union and reduced grip strength, but no association with the flexion – extension arc [84]. The impact of mal-union after an acute fracture is uncertain, as adhesions may have confounded and masked any specific effect of mal-union.
Avascularity Increased radio-opacity of the proximal fragment is thought to represent avascular necrosis of the scaphoid. Russe [19] observed increased density in the proximal fragments in about 30% of acute scaphoid fractures and felt that this was a transient phenomenon caused by damage to the nutrient blood vessels. The appearance of increased density may be due to surrounding osteoporosis or due to new bone deposition on the dead trabeculae within the proximal fragment. There is poor agreement between observers on whether there was sclerosis at or near the fracture and on whether the proximal part of the scaphoid was avascular in radiographs taken 12 weeks after a scaphoid fracture [39].
120
J.J. Dias
common in patients with a scaphoid non-union to lose wrist extension due to carpal collapse and palmar capsular contracture [87]. There may also be limited radial and ulnar deviation and decreased grip strength. The diagnosis is usually made on plain radiographs. Displacement of fracture fragments, a clear gap, cyst formation, and sclerosis all suggest non-union. These features may take several months to appear. Dias et al. [39] found that even experienced observers (both surgeons and radiologists) were unable to consistently agree on radiologic scaphoid union on radiographs taken 12 weeks after injury. Computed tomography scans, particularly thin slices through the longitudinal axis of the scaphoid, will confirm a non-union. Magnetic resonance imaging, though poor in assessing bony architecture, can demonstrate the vascular status of fracture fragments, particularly any suspected avascular necrosis of the proximal pole. Bone scans are not specific, but they may objectively highlight the area of nonunion. None of these methods demonstrate movement at the fracture site. Scaphoid non-union has been classified into four main groups. Type D1 is fibrous union and D2 is pseudoarthrosis at the non-union site. Type D3 a non-union with a fixed dorsal intercalated segment instability (DISI) deformity and a sclerotic surface of any pseudoarthrosis. Type D4 nonunion has avascular necrosis with collapse of the proximal pole [88].
This appearance is a poor predictor of vascularity of scaphoid seen at operation. However, increased density may also be produced by proximal pole rotation. Vascularity can also be assessed at the time of surgery and by taking a biopsy. Biopsy can be misleading because of the patchy pattern of avascular necrosis. Biopsy specimens are likely to contain both viable and dead osteocytes and cannot be used to accurately predict the histologic status of the entire fragment; therefore, serial sections of the entire proximal pole would be necessary to prove complete avascular necrosis [85]. Radionuclide bone scanning is sensitive and can reveal early avascular necrosis but is inaccurate in acute fractures and is not quantifiable. It has a specificity of only 18% in one series because areas of minor damage or synovitis may give a positive result [86].
Non-union (Fig. 6) The diagnosis of scaphoid non-union can often be delayed and many patients cannot recall an injury to the hand or wrist and many have taken part in contact sports when younger. When an acute fracture is being treated a persistent gap suggests a non-union. Lucency around a screw also suggests failure of union. Patients may have dorsal and radial swelling and tenderness. The scaphoid tuberosity may also be tender with pain felt at the extreme of dorsiflexion. It is
Delayed diagnosis
–
Fibrous union / Stable Types of nonunion
–
Rigid fixation –
Fibrous union / Unstable
Rigid fixation /? Graft –
Fixation + Graft
Non union + Avascularity
–
Non/ Vascularised Graft + Fixation
Non union with OA changes
–
Salvage, PIN+ AIN denervation, Limited / Complete fusion
NONUNION SCAPHOID 95 − 100% risk of osteoarthritis, ?Symptoms
Risk factors
Management options
Also for further nonunion – after vascularised graft
–
91% union rate in absence vs 48% in absence of AVN –
Operative –
Delays OA
Avascularity Displacement
Benefits
Grip strength Function/ROM
Instability of fracture
20% risk of further nonunion
Proximal location
Infection (<1%)
Smoking Union rate 81% Nonsmoker vs 46% smokers
–
Risks
–
CRPS (3%) Removal of metal Scar complication (15%) Nerve injury (2%)
Fig. 6 Flow diagram for risk factors, types and management of non-union of scaphoid fractures
121
Scaphoid Fractures
Slade and Geissler [65] classified scaphoid non-unions by using the treatment classification system. They recommended that acute scaphoid fractures presenting in a delayed fashion, later than 1 month, should be treated with rigid fixation (Grade I). Fibrous unions appear solidly healed, but insufficient remodeling has occurred to resist the stresses of bending and torque (Grade II). These often require only rigid fixation to heal. Correctly-aligned scaphoid non-unions with minimal fracture sclerosis suggest micro-motion and early resorption at the fracture site (Grade III). With gaps less than 1 mm in a stable fracture, healing can proceed across this gap. These non-unions require rigid fixation to achieve union. A CT scan should be used to confirm that the non-union front represents only a minimal sclerotic line (less than 1 mm) and that the two parts of the scaphoid are in correct alignment. Scaphoid non-unions with cystic changes at the fracture represent extensive re-absorption and non-viable tissue at the fracture site (Grade IV). These non-unions present with sclerotic zones between 1 and 5 mm. These non-unions require at minimum debridement, bone grafting and rigid fixation. MR imaging should be considered if bone viability of these fracture fragments is of question. A CT scan is required to define the extent of local destruction and confirms correct structural alignment. Scaphoid non-unions with pseudoarthrosis and/or deformity require structural bone grafting for mechanical support (Grade V) and percutaneous bone grafting is questionable. Scaphoid non-unions with necrosis require vascularised bone graft and non-unions with SNAC deformity (Grade VI) are candidates for scaphoid excision and carpal reconstruction. Mack and Lichtman [89] classified scaphoid nonunions based on the amount of displacement. The nondisplaced stable non-union without degenerative changes (Mack-Lichtman type I) may be treated with bone grafting with or without hardware. Type II non-unions, which are unstable owing to fragment displacement, require restoration of normal carpal stability to prevent the downward spiral from instability to collapse and arthritis. Scaphoid non-union with accompanying mild arthritis are classified as Mack-Lichtman type III. Initial findings of radio-carpal arthritis include beaked changes to the radial styloid and narrowing of the joint space between the radius and scaphoid. Treatment includes addressing the non-union as well as the arthritis and open reduction and internal fixation with bone graft is required with or without radial styloidectomy. Mack-Lichtman types IV and V non-unions are those associated with mid-carpal arthritis, without and with radio-lunate arthritis, respectively. They require partial or complete wrist arthrodesis for optimal treatment.
The result of non-union is a distinctive pattern of osteoarthritis [90] affecting firstly the joint between the radius and the distal fragment of the scaphoid, which is the rationale for radial styloidectomy as a palliative procedure. Three-dimensional CT can detect the earliest stages of this process [91]. Next affected is the mid-carpal joint between the capitate and the proximal scaphoid and later between the capitate and the lunate. The joints between the radius and the proximal fragment of the scaphoid and between the radius and the lunate are seldom affected. Fisk [92] described the goal of treatment of scaphoid non-union with wedge graft was to restore normal scaphoid anatomy and to re-establish the normal tension in the palmar radio-carpal ligaments. Fernandez [93] modified the original Fisk procedure with use of a pre-operative plan based on radiographs of the opposite wrist, the use of a palmar approach, the resection of the non-union site and insertion of an iliac graft and the use of internal fixation. Zaidemberg et al. [94] described vascular bone grafting for scaphoid non-union with a distal radius graft taken from between first and second extensor compartment based on a 1,2-intercompartmental supra-retinacular artery. Various other vascularised grafts have been described which include those based on the distal radius, index metacarpal, scaphoid tubercle and pisiform. Free iliac crest and medial femoral condyle vascular-pedicled bone grafting has also been reported.
Bone Grafting (Fig. 7) Non-vascularised bone grafting is probably sufficient for most waist fracture non-unions and those with preserved vascularity of the proximal pole. If standard bone grafting fails, future surgery is likely also to be unsuccessful [95]. The benefits of vascularised bone grafting for scaphoid non-union include preservation of the blood supply, primary bone healing, and maintenance of structural integrity. Therefore, vascularised bone grafting can be considered. In a recent study [96], the outcomes and complications of vascularised bone grafting for scaphoid non-union were described in a series of 52 non-unions in 51 patients. The 1,2-intercompartmental supra-retinacular artery (1,2ICRSA) was used as a reverse-flow vascularised bone graft for scaphoid non-union. Overall, 72% of the scaphoid fractures achieved union with vascularised bone grafting (36/50), and healing occurred at an average of 16 weeks (range 8–40 weeks). Factors adversely affecting the union rate included female sex (union rate: 30% vs. 82% in males), tobacco use (union rate: 81% in non-smokers vs. 46% in smokers), and proximal pole avascularity [48% union rate in the presence of avascular necrosis (AVN) vs. 91% in the
122
J.J. Dias
Outcome
Fig. 7 A clinical photograph showing wedge graft placed in the gap after shaping with burr
absence of AVN]. Simple K-wire fixation resulted in a 53% union rate, whereas screw fixation resulted in an 88% union rate. Carpal collapse with formation of a hump-back deformity was present in 50% of the failures versus 11% of patients who went on to union. Fracture-dislocation did not affect the union rate, with waist fractures achieving 70% union and proximal poles achieving 72% union. Prior surgery resulted in a healing rate of 64% compared to 73% in those with no previous operations. Outcomes of 34 patients who had undergone vascularised bone grafting for a chronic scaphoid non-union were reviewed. In 18 cases the fracture involved the proximal and in 16 cases the middle third of the scaphoid. In 26 patients the proximal scaphoid fragment was deemed avascular. Sixteen patients had previously undergone scaphoid fixation and non-vascularised bone grafting. At a follow-up of 1–3 years (mean 1.6 years), 15 of the 34 scaphoid nonunions had united. Injury to the dominant hand and duration of the non-union significantly increased the risk of failure. Persistent non-union was more common in proximal third fractures and in the presence of an avascular proximal pole but these findings did not reach statistical significance [97].
Dias, Brenkel and Finlay found 20% patients had some pain and tenderness 1.7–2.6 years after healed scaphoid fracture but grip strength and wrist movement were nearly normal. They felt the persistent symptoms were attributable to damage to the articular cartilage at the time of the injury [98]. Lindström and Nyström also found osteoarthritis in 5% of wrists with fracture of the scaphoid which had healed normally and suggested that it would probably progress [82]. Duppe et al. [99] found marked radio-carpal osteoarthritis developed in only 2% of patients who had a healed fracture; it was far more common in the group that had a non-union, in which the prevalence was 5 of 9 patients. Manifest osteoarthritis also seemed to be associated with pain or weakness: it had developed in only 6% of the patients who did not have any symptoms at re-examination, compared with 3 of the 7 who had symptoms. In [72] Saéden et al. reviewed patients with CT scans 10–12 years after treatment in a cast or a Herbert screw. Seven of sixteen patients treated conservatively had osteoarthritis in the radio-carpal joint and seven in the scapho-trapezial joint. The higher incidence of osteoarthritis reported in this series was probably because the CT scans revealed it. The effect of mal-union of scaphoid fractures on the clinical outcome at 1 year in 42 consecutive patients with united scaphoid waist fractures, which had been treated non-operatively, has been studied. They underwent longitudinal CT scans to confirm union and assess mal-union at 12–18 weeks after injury. No significant relationships between any of the outcome measures (range of motion, grip strength and PEM and DASH scores) and any of the three measures of malunion (height-to-length ratio, the dorsal cortical angle and the lateral intra-scaphoid angle) were identified [58]. Jiranek et al. [84] compared 13 patients with mal-union, defined as a lateral intra-scaphoid angle of >45°, and 13 with acceptable union. Both followed Russe procedures for non-union. There was no difference in symptoms or function and 12 patients with mal-union returned to high level of function despite deformity. Many are uncertain about osteotomies for correction of mal-union. Lynch and Linscheid [100] reviewed five corrective osteotomies after 1.5–19 years and showed that despite improvement in grip strength, it did not prevent osteoarthritis. Nakamura, Imaeda and Miura [101] showed improvement in grip strength and movement after union of corrective osteotomy but their criterion for mal-union was a DISI deformity that suggests a severe mal-union. Other authors remain cautious about osteotomising a bone, which remains difficult to heal in first place.
123
Scaphoid Fractures
Long-term follow-up after untreated scaphoid nonunion has shown radiological osteoarthritis but many patients remain symptom-free. Mack et al. [89] noticed three roentgenographic patterns after 5–53 years follow-up of scaphoid non-union: sclerosis, cyst formation, or resorptive changes confined to the scaphoid bone (Group I), radio-scaphoid arthritis (Group II), and generalized arthritis of the wrist (Group III). They found a high correlation between fracture displacement with carpal instability and the severity of degenerative changes. Based on the high probability of arthritis, they recommended that all displaced un-united scaphoid fractures be reduced and grafted, regardless of symptoms, before degenerative changes occur. Asymptomatic patients with an undisplaced, stable non-union should be advised of the possibility of late degenerative changes. Inoue and Sakuma [102] found osteoarthritis in 100% of patients with symptomatic nonunion after 10 years but symptoms did not correlate with the severity of arthritis or the duration of non-union. Lindström and Nyström [103] reviewed 33 patients with untreated fracture after 12 years, all patients had radiological evidence of osteoarthritis but five remained symptom free. Further follow-up after 17 years showed two had developed pain, weakness and stiffness and one wrist was slightly swollen but not painful. Two patients had died. Results of surgery for non-union of scaphoid are unpredictable. There is a discrepancy between clinical and radiological outcome after non-union surgery. Patients in whom definite union has been achieved after bone grafting may have persistent pain. Inoue, Shionoya and Kuwahata [104] reviewed 215 patients in whom non-union was treated with a bone graft and Herbert screw. Thirty patients had mild osteoarthritis before the operation and they had worse results both symptomatically and radiologically. Filan and Herbert [62] followed up 304 patients for 6–34.2 months (163 were treated for non-union) and found 40% had radiocarpal osteoarthritis before the operation and 49% afterwards, though for severe osteoarthritis the increase was only from 3% to 7%. They concluded that the progress of osteoarthritis was reduced by successful internal fixation, but could not establish by how much in the long-term. The case for a prophylactic operation of an asymptomatic patient with scaphoid non-union to prevent osteoarthritis is weak. The operation will necessarily involve inconvenience and a period off work. It may result in reduced movement at the wrist. One in five will fail to achieve union and some of the remaining four will still develop osteoarthritis, but perhaps later than they would have done without operation. There might be a better case for operating to prevent the onset of pain but union does not guarantee the absence of pain [105]. On the basis of the current literature we still rec-
ommend internal fixation and bone grafting for most non-union but discuss the pros. and cons. with each patient based on the type of non-union and its site. Despite the best efforts in diagnosis and treatment, a scaphoid non-union may fail to unite. With collapse of the carpus and painful arthritis, a salvage procedure is likely to be necessary. These salvage procedures include conservative management with observation, occasional injections or the use of a splint, radial styloidectomy with partial scaphoid excision, and/or posterior and anterior interosseous neurectomy. More complex procedures include limited intercarpal fusion, proximal row carpectomy if the capitate surface is preserved, scaphoid excision and four-corner fusion, and, lastly, a total wrist fusion.
Summary We treat most of the patients with scaphoid fracture in below-elbow plaster with the thumb free for 6–8 weeks. We consider surgery for displaced scaphoid fractures causing DISI deformity, proximal pole fractures, fractures associated with perilunate injuries, open fractures, and fractures in multiply-injured patients. Other decision-making factors are whether there is a great potential for morbidity from prolonged immobilization, the occupation of the patient, and a clear failure of healing after non-operative treatment of the fracture.
Fracture of the Hamate Fractures of Hamate are rare accounting for 2–4% of carpal fractures. They are usually caused by a direct blow or indirect injury during gripping of an object such as a baseball bat, golf club, or tennis racquet. Fractures of the hamate can involve the hook, the body, and various articular surfaces. Pain is elicited at the heel of the hand with firm grasp and with pressure against the bony prominence just radial and slightly distal to the pisiform. A carpal tunnel view (Fig. 8) may show the fracture, but this is usually better seen on a CT scan (Fig. 9). The patient’s hands can be placed together in the praying position so both wrists can be viewed. Bilateral abnormality suggests congenital variation of the hamate. Fractures of the hook can be treated with casting for 6–4 months. These unite in approximately 50% of cases. Non-union can be treated with excision and rarely needs open reduction and fixation. Fractures of the body are treated in a cast, unless displacement is significant. Articular
124
Fig. 8 Carpal tunnel view showing hamate and Trapezium
J.J. Dias
with peri-lunate fracture/dislocation. It usually is a result of axial load of the middle metacarpal and may not unite in rare cases. Scapho-capitate syndrome (Fig. 10) is associated with scaphoid waist fracture and a proximal capitate fracture, with or without distal radius fracture. It presents with spontaneous reduction of a trans-scaphoid, trans- capitate perilunar dislocation in which the proximal pole of the capitate rotates on partial reduction after initial displacement. Occasionally the proximal pole may be flipped so that the articular surface abuts the fracture line. This injury needs open reduction and internal fixation of scaphoid and capitate using a headless screw or trans-articular wires. Vander Grend et al. [108] studied intra-osseous capitate vascularity by in vitro arterial injection studies to correlate this with the clinical problem of avascular necrosis. Palmar vessels were found to contribute the majority of the blood supply to the capitate. The proximal pole received its blood supply exclusively in a retrograde fashion across the capitate waist analogous to the proximal scaphoid. Aseptic necrosis without collapse of the proximal pole was successfully managed with curettage and bone grafting in three patients. The remaining two patients, with collapse and peri-capitate degenerative changes, were managed with intercarpal fusion.
Trapezium and Trapezoid Fractures Fig. 9 CT scan showing non-united fracture of hook of hamate (With permission from Elsevier, from De Schrijver and De Smet [106])
fractures needs open reduction and internal fixation if displacement is 1 mm or more and especially if there is a step. A stress fracture may develop in the hook of the hamate with some repetitive activities, such as golf. Initial diagnosis can be difficult. Transient ulnar nerve motor palsy can be caused by an undiagnosed stress fracture of the hook of the hamate. In most instances, unless the diagnosis is delayed, union is likely after immobilisation, but excision of the fragment may be necessary for nonunion, persistent pain, or ulnar nerve palsy.
Capitate Fractures Capitate fractures are usually associated with dorsal carpometacarpal dislocation and occur especially at the dorsal ridge of capitate. It is rarely isolated but more commonly presents as a “scapho-capitate syndrome” when associated
Fractures of the trapezium and trapezoid are rare (1–5% of wrist fractures) and may be comminuted when seen in conjunction with radial fracture-dislocations and other carpal bone fractures. These fractures usually can be seen radiographically on the carpal tunnel view of the wrist and on a CT scan. Fractures occur through the body or the trapezial ridge and may occur with a dislocation of the trapeziometacarpal joint. Palmer classified trapezial ridge fractures into two types: Type I is a fracture of the base of the ridge which may heal when treated by immobilization in plaster; type II is an avulsion at the tip of the ridge which usually fails to heal when immobilized. Displaced trapezial fractures require open reduction. Fractures of the body can be exposed through a J-shaped incision along the radial side of the thumb metacarpal, curving medially at the wrist flexion crease. Un-united fragments of the trapezial ridge can be excised using the proximal limb of the J-incision or through a longitudinal incision in the thenar crease. Care should be taken to avoid injury to the palmar cutaneous branch of the median nerve. The trapezoid is fractured least often of all the carpal bones and usually is injured at the time of other carpo-metacarpal
125
Scaphoid Fractures
a
b
Fig. 10 (a) Postero-anterior and (b) oblique radiographs following injury show the scapho-capitate syndrome (With permission from Elsevier, from Sawant and Miller [107])
injuries, especially injuries to the index metacarpal. Displaced fractures can require reduction and fixation.
Fractures of the Lunate Fractures of the lunate are rare and negative ulnar variance may be a risk factor for chronic repetitive trauma syndromes. Fractures occur with forces acting along the longitudinal axis of the limb, such as falls on the hand and punching. The dorsal pole may fracture with hyperextension of the wrist. They can be difficult to detect on plain radiography and CT scans may be required to diagnose the fracture. Blood supply to the lunate is from the palmar radial carpal arch and blood vessels branch within the bone. Injuries of the lunate may damage the circulation, leading to osteonecrosis. Gelberman et al. [109] described three patterns of extraosseous and intra-osseous vascularity of the lunate in 35 fresh cadaver limbs. The specimens were injected with latex, debrided by a non-dissection technique, and cleared by a modified Spalteholtz method. The extra-osseous vascularity was profuse through two to three dorsal and three to four palmar vessels feeding dorsal and palmar capsular plexi.
The intra-osseous vascularity formed one of three (cross, y-shaped or single vessel) consistent patterns with anastomoses of dorsal and palmar vessels in each specimen. The vascular patterns support a theory of compression fracture from repeated trauma as one cause of Kienböck’s disease. The lunates which are believed to be most at risk for osteonecrosis are those with a single vessel or one surface exposed to the blood supply, representing about 20% of lunates. Fractures of the lunate may be non-displaced; displaced with large fragments; avulsed, especially the dorsal pole; or comminuted. Non-displaced and non-displaced comminuted fractures can be treated with cast immobilization. Fractures with more than 1 mm. offset and avulsion fractures may require reduction. Internal fixation technique includes Kirschner wires, small-cannulated screws, and suture anchors.
Triquetrum Fractures Triquetrum fractures are usually caused by a direct blow. Three types are recognized: Avulsion fractures of dorsal radio-triquetral or ulno-triquetral ligament, dorsal impaction
126
J.J. Dias
Fig. 11 Radiograph showing chip avulsion fracture of Triquetrum
fractures, and fractures of the body. They require careful clinical examination and PA, lateral and oblique radiographic views of wrist or CT scans. They can be minimallydisplaced, when it should be treated in a short arm cast for 4 – 6 weeks. It may show more than 1 mm displacement and diastases in excess of 2 mm require open or percutaneous reduction with fixation. They can be approached via a dorsal or ulnar incision through the fifth extensor compartment. Chip or avulsion fractures (Fig. 11) are common and are caused by forced hyper-flexion. They occur due to avulsion of attachment of radio-carpal ligaments and are treated symptomatically with immobilisation.
Pisiform Fracture These fractures are rare (1–3%) and are caused by a direct blow or rarely due to flexor carpi ulnaris avulsion. Pisiform fractures are treated symptomatically or with excision if unsuccessful. Carrol and Coyle [110] treated 42 patients by excision of the pisiform. Ulnar neuropathy was noted in association with fractures and subluxations or dislocations of the pisiform. The abductor and flexor digiti minimi and the palmar carpal ligament with their common fibrous origin were
the most common compressing structures on the ulnar nerve. Excision of the pisiform provided complete relief of localized hypothenar pain with no loss of wrist motion or strength. Neurolysis produced full sensory and motor recovery.
References 1. Oehmke MJ, Podranski T, Klaus R et al (2009) The blood supply of the scaphoid bone. J Hand Surg Eur 34:351–357 2. Patterson RM, Moritomo H, Yamaguchi H et al (2003) Scaphoid anatomy and mechanics: update and review. Oper Tech Orthop 13:2–10 3. Obletz BE, Halbstein BM (1938) Non-union of fractures of the carpal navicular. J Bone Joint Surg Am 20:424–428 4. Gelberman RH, Menon J (1980) The vascularity of the scaphoid bone. J Hand Surg Am 5:508–513 5. Berger RA (1996) The gross and histologic anatomy of the scapholunate interosseous ligament. J Hand Surg Am 21:170–178 6. Amadio PC, Berquist TH, Smith DK et al (1989) Scaphoid malunion. J Hand Surg Am 14:679–687 7. Bettinger PC, Linscheid RL, Berger RA, Cooney WP III, An KN (1999) An anatomic study of the stabilizing ligaments of the trapezium and trapeziometacarpal joint. J Hand Surg Am 24:786–798
Scaphoid Fractures 8. Berger RA (2001) The anatomy of the scaphoid. Hand Clin 17:525–532 9. Compson JP, Waterman JK, Heatley FW (1994) The radiological anatomy of the scaphoid. Part 1: osteology. J Hand Surg Br 19:183–187 10. Ziter FM Jr (1973) A modified view of the carpal navicular. Radiology 108:706–707 11. Stechers WR (1937) Roentgenography of the carpal navicular bone. Am J Roentgenol Radium Ther Nucl Med 37:704–705 12. Compson JP, Waterman JK, Heatley FW (1997) The radiological anatomy of the scaphoid. Part 2: radiology. J Hand Surg Br 22:8–15 13. Weber ER, Chao EY (1978) An experimental approach to the mechanism of scaphoid waist fractures. J Hand Surg Am 3:142–148 14. Cockshott WP (1980) Distal avulsion fractures of the scaphoid. Br J Radiol 53:1037–1040 15. Kozin SH (2001) Incidence, mechanism, and natural history of scaphoid fractures. Hand Clin 17:515–524 16. Leslie IJ, Dickson RA (1981) The fractured carpal scaphoid natural history and factors influencing outcome. J Bone Joint Surg Br 63:225–230 17. Shestak K, Ruby LK (1983) An unusual fracture of the scaphoid. J Hand Surg Am 8:925–928 18. Horii E, Nakamura R, Watanabe K, Tsunoda K (1994) Scaphoid fracture as a “puncher’s fracture”. J Orthop Trauma 8:107–110 19. Russe O (1960) Fracture of the carpal navicular. Diagnosis, nonoperative treatment and operative treatment. J Bone Joint Surg Am 42A:759–768 20. Cooney WP, Dobyns JH, Linscheid RL (1980) Fractures of the scaphoid: a rational approach to management. Clin Orthop Relat Res 149:90–97 21. Herbert TJ, Fisher WE (1984) Management of the fractured scaphoid using a new bone screw. J Bone Joint Surg Br 66:114–123 22. Compson JP (1998) The anatomy of acute scaphoid fractures: a three-dimensional analysis of patterns. J Bone Joint Surg Br 80:218–224 23. Prosser AJ, Brenkel IJ, Irvine GB (1988) Articular fractures of the distal scaphoid. J Hand Surg Br 13:87–91 24. Desai VV, Davis TR, Barton NJ (1999) The prognostic value and reproducibility of the radiological features of the fractured scaphoid. J Hand Surg Br 24:586–590 25. Grover R (1996) Clinical assessment of scaphoid injuries and the detection of fractures. J Hand Surg Br 21(3):341–343 26. Parvizi J, Wayman J, Kelly P, Moran CG (1998) Combining the clinical signs improves diagnosis of scaphoid fractures. A prospective study with follow-up. J Hand Surg Br 23:324–327 27. Munk B, Bolvig L, Kroner K et al (2000) Ultrasound for diagnosis of scaphoid fractures. J Hand Surg Br 25:369–371 28. Freeland P (1989) Scaphoid tubercle tenderness: a better indicator of scaphoid fractures? Arch Emerg Med 6:46–50 29. Chen SC (1989) The scaphoid compression test. J Hand Surg Br 14:323–325 30. Powell JM, Lloyd GJ, Rintoul RF (1988) New clinical test for fracture of the scaphoid. Can J Surg 31:237–238
127 31. Waizenegger M, Barton NJ, Davis TR, Wastie ML (1994) Clinical signs in scaphoid fractures. J Hand Surg Br 19:743–747 32. Hunter JC, Escobedo EM, Wilson AJ et al (1997) MR imaging of clinically suspected scaphoid fractures. AJR Am J Roentgenol 168:1287–1293 33. Barton NJ (1992) Twenty questions about scaphoid fractures. J Hand Surg Br 17:289–310 34. Watson-Jones R (1952) Fractures and joint injuries, 4th edn. Williams & Wilkins, Baltimore, pp 86–89 35. Dias JJ, Thompson J, Barton NJ, Gregg PJ (1990) Suspected scaphoid fractures the value of radiographs. J Bone Joint Surg Br 72:98–101 36. Watson-Jones R (1943) Fractures and joint injuries, 3rd edn. E & S Livingstone, Edinburgh 37. Graziani A (1940) L’esame radiologico del carpo. Radiol Med (Torino) 27:382–392 38. Terry DW Jr, Ramin JE (1975) The navicular fat stripe: a useful roentgen feature for evaluating wrist trauma. Am J Roentgenol Radium Ther Nucl Med 124:25–28 39. Dias JJ, Taylor M, Thompson J, Brenkel IJ, Gregg PJ (1988) Radiographic signs of union of scaphoid fractures an analysis of inter-observer agreement and reproducibility. J Bone Joint Surg Br 70:299–301 40. Hauger O, Bonnefoy O, Moinard M, Bersani D, Diard F (2002) Occult fractures of the waist of the scaphoid: early diagnosis by high-spatial-resolution sonography. AJR Am J Roentgenol 178:1239–1245 41. Fusetti C, Poletti PA, Pradel PH et al (2005) Diagnosis of occult scaphoid fracture with high-spatial-resolution sonography: a prospective blind study. J Trauma 59(3): 677–681 42. Brooks S, Cicuttini FM, Lim S et al (2005) Cost effectiveness of adding magnetic resonance imaging to the usual management of suspected scaphoid fractures. Br J Sports Med 39:75–79 43. Nielsen PT, Hedeboe J, Thommesen P (1983) Bone scintigraphy in the evaluation of fracture of the carpal scaphoid bone. Acta Orthop Scand 54:303–306 44. Kukla C, Gaebler C, Breitenseher MJ, Trattnig S, Vecsei V (1997) Occult fractures of the scaphoid. The diagnostic usefulness and indirect economic repercussions of radiography versus magnetic resonance scanning. J Hand Surg Br 22:810–813 45. Brydie A, Raby N (2003) Early MRI in the management of clinical scaphoid fracture. Br J Radiol 76:296–300 46. Fowler C, Sullivan B, Williams LA et al (1998) A comparison of bone scintigraphy and MRI in the early diagnosis of the occult scaphoid waist fracture. Skeletal Radiol 27:683–687 47. Thorpe AP, Murray AD, Smith FW, Ferguson J (1996) Clinically suspected scaphoid fracture: a comparison of magnetic resonance imaging and bone scintigraphy. Br J Radiol 69:109–113 48. Dorsay TA, Major NM, Helms CA (2001) Cost-effectiveness of immediate MR imaging versus traditional follow-up for revealing radiographically occult scaphoid fractures. AJR Am J Roentgenol 177:1257–1263
128 49. Cova M, Kang YS, Tsukamoto H et al (1991) Bone marrow perfusion evaluated with gadolinium-enhanced dynamic fast MR imaging in a dog model. Radiology 179:535–539 50. Munk PL, Lee MJ, Janzen DL et al (1998) Gadoliniumenhanced dynamic MRI of the fractured carpal scaphoid: preliminary results. Australas Radiol 42:10–15 51. Downing ND, Oni JA, Davis TR et al (2002) The relationship between proximal pole blood flow and the subjective assessment of increased density of the proximal pole in acute scaphoid fractures. J Hand Surg Am 27:402–408 52. Perlik PC, Guilford WB (1991) Magnetic resonance imaging to assess vascularity of scaphoid nonunions. J Hand Surg Am 16:479–484 53. Cerezal L, Abascal F, Canga A et al (2000) Usefulness of gadolinium-enhanced MR imaging in the evaluation of the vascularity of scaphoid nonunions. AJR Am J Roentgenol 174:141–149 54. Munk PL, Lee MJ (2000) Gadolinium-enhanced MR imaging of scaphoid nonunions. AJR Am J Roentgenol 175:1184–1185 55. Sanders WE (1988) Evaluation of the humpback scaphoid by computed tomography in the longitudinal axial plane of the scaphoid. J Hand Surg Am 13:182–187 56. Bain GI, Bennett JD, Richards RS, Slethaug GP, Roth JH (1995) Longitudinal computed tomography of the scaphoid: a new technique. Skeletal Radiol 24:271–273 57. Eddeland A, Eiken O, Hellgren E, Ohlsson NM (1975) Fractures of the scaphoid. Scand J Plast Reconstr Surg 9:234–239 58. Forward DP, Singh HP, Dawson S, Davis TR (2009) The clinical outcome of scaphoid fracture malunion at 1 year. J Hand Surg Eur 34:40–46 59. Bain GI, Bennett JD, MacDermid JC et al (1998) Measurement of the scaphoid humpback deformity using longitudinal computed tomography: intra- and interobserver variability using various measurement techniques. J Hand Surg Am 23:76–81 60. Bhat M, McCarthy M, Davis TR, Oni JA, Dawson S (2004) MRI and plain radiography in the assessment of displaced fractures of the waist of the carpal scaphoid. J Bone Joint Surg Br 86(5):705–713 61. Ring D, Patterson JD, Levitz S, Wang C, Jupiter JB (2005) Both scanning plane and observer affect measurements of scaphoid deformity. J Hand Surg Am 30:696–701 62. Filan SL, Herbert TJ (1996) Herbert screw fixation of scaphoid fractures. J Bone Joint Surg Br 78:519–529 63. Yin ZG, Zhang JB, Kan SL, Wang P (2007) Treatment of acute scaphoid fractures: systematic review and meta- analysis. Clin Orthop Relat Res 460:142–151 64. Rettig ME, Kozin SH, Cooney WP (2001) Open reduction and internal fixation of acute displaced scaphoid waist fractures. J Hand Surg 26:271–276 65. Slade JF III, Geissler WB, Gutow AP, Merrell GA (2003) Percutaneous internal fixation of selected scaphoid nonunions with an arthroscopically assisted dorsal approach. J Bone Joint Surg Am 85-A(Suppl 4):20–32 66. Clay NR, Dias JJ, Costigan PS, Gregg PJ, Barton NJ (1991) Need the thumb be immobilised in scaphoid fractures? A randomised prospective trial. J Bone Joint Surg Br 73:828–832 67. Hambidge JE, Desai VV, Schranz PJ et al (1999) Acute fractures of the scaphoid. Treatment by cast immobilisation with the wrist in flexion or extension? [see comment]. J Bone Joint Surg Br 81:91–92
J.J. Dias 68. Adolfsson L, Lindau T, Arner M (2001) Acutrak screw fixation versus cast immobilisation for undisplaced scaphoid waist fractures. J Hand Surg Br 26:192–195 69. Bond CD, Shin AY, McBride MT, Dao KD (2001) Percutaneous screw fixation or cast immobilization for nondisplaced scaphoid fractures. J Bone Joint Surg Am 83:483–488 70. Dias JJ, Wildin CJ, Bhowal B, Thompson JR (2005) Should acute scaphoid fractures be fixed? A randomized controlled trial. J Bone Joint Surg Am 87:2160–2168 71. McQueen MM, Gelbke MK, Wakefield A, Will EM, Gaebler C (2008) Percutaneous screw fixation versus conservative treatment for fractures of the waist of the scaphoid: a prospective randomised study. J Bone Joint Surg Br 90:66–71 72. Saeden B, Tornkvist H, Ponzer S, Hoglund M (2001) Fracture of the carpal scaphoid. A prospective, randomised 12-year follow-up comparing operative and conservative treatment. J Bone Joint Surg Br 83:230–234 73. Vinnars B, Pietreanu M, Bodestedt A, Ekenstam F, Gerdin B (2008) Nonoperative compared with operative treatment of acute scaphoid fractures. A randomized clinical trial. J Bone Joint Surg Am 90:1176–1185 74. Singh HP, Forward D, Davis TRC et al (2005) Partial union of acute scaphoid fractures. J Hand Surg Br 30:440–445 75. Davis EN, Chung KC, Kotsis SV, Lau FH, Vijan S (2006) A cost/utility analysis of open reduction and internal fixation versus cast immobilization for acute nondisplaced mid-waist scaphoid fractures. Plast Reconstr Surg 117:1223–1235 76. Arora R, Gschwentner M, Krappinger D et al (2007) Fixation of nondisplaced scaphoid fractures: making treatment cost effective. Prospective controlled trial. Arch Orthop Trauma Surg 127:39–46 77. Gelberman RH, Wolock BS, Siegel DB (1989) Fractures and non-unions of the carpal scaphoid. J Bone Joint Surg Am 71:1560–1565 78. Adams BD, Blair WF, Reagan DS, Grundberg AB (1988) Technical factors related to Herbert screw fixation. J Hand Surg Am 13:893–899 79. Trumble TE, Clarke T, Kreder HJ (1996) Non-union of the scaphoid. Treatment with cannulated screws compared with treatment with Herbert screws. J Bone Joint Surg Am 78:1829–1837 80. Riester JN, Baker BE, Mosher JF, Lowe D (1985) A review of scaphoid fracture healing in competitive athletes. Am J Sports Med 13:159–161 81. Cooney WP, Linscheid RL, Dobyns JH, Wood MB (1988) Scaphoid nonunion: role of anterior interpositional bone grafts. J Hand Surg 13:635–650 82. Lindström G, Nyström A (1990) Incidence of post-traumatic arthrosis after primary healing of scaphoid fractures: a clinical and radiological study. J Hand Surg Br 15:11–13 83. Dias JJ (2001) Definition of union after acute fracture and surgery for fracture nonunion of the scaphoid. J Hand Surg Br 26:321–325 84. Jiranek WA, Ruby LK, Millender LB, Bankoff MS, Newberg AH (1992) Long-term results after Russe bone-grafting: the effect of malunion of the scaphoid. J Bone Joint Surg Am 74:1217–1228 85. Urban MA, Green DP, Aufdemorte TB (1993) The patchy configuration of scaphoid avascular necrosis. J Hand Surg Am 18:669–674 86. Reinus WR, Conway WF, Totty WG et al (1986) Carpal avascular necrosis: MR imaging. Radiology 160:689–693
Scaphoid Fractures 87. Herbert TJ (1990) The fractured scaphoid. Quality Medical Publishing, St. Louis, p 31 88. Herbert TJ, Filan SL (1999) Proximal scaphoid nonunionosteosynthesis. Handchir Mikrochir Plast Chir 31:169–173 89. Mack GR, Bosse MJ, Gelberman RH, Yu E (1984) The natural history of scaphoid non-union. J Bone Joint Surg Am 66:504–509 90. Vender MI, Watson HK, Wiener BD, Black DM (1987) Degenerative change in symptomatic scaphoid nonunion. J Hand Surg Am 12:514–519 91. Hidaka Y, Nakamura R (1998) Progressive patterns of degenerative arthritis in scaphoid nonunion demonstrated by three-dimensional computed tomography. J Hand Surg Br 23:765–770 92. Fisk GR (1980) An overview of injuries of the wrist. Clin Orthop Relat Res 149:137–144 93. Fernandez DL (1984) A technique for anterior wedgeshaped grafts for scaphoid nonunions with carpal instability. J Hand Surg Am 9:733–737 94. Zaidemberg C, Siebert JW, Angrigiani C (1991) A new vascularized bone graft for scaphoid nonunion. J Hand Surg 16:474–478 95. Steinmann SP, Bishop AT, Berger RA (2002) Use of the 1, 2 intercompartmental supraretinacular artery as a vascularized pedicle bone graft for difficult scaphoid nonunion. J Hand Surg Am 27:391–401 96. Chang MA, Bishop AT, Moran SL, Shin AY (2006) The outcomes and complications of 1, 2-intercompartmental supraretinacular artery pedicled vascularized bone grafting of scaphoid nonunions. J Hand Surg Am 31:387–396 97. Kapoor AK, Thompson NW, Rafiq I et al (2008) Vascularised bone grafting in the management of scaphoid non-union – a review of 34 cases. J Hand Surg Eur 33:628–631 98. Dias JJ, Brenkel IJ, Finlay DB (1989) Patterns of union in fractures of the waist of the scaphoid. J Bone Joint Surg Br 71:307–310
129 99. Duppe H, Johnell O, Lundborg G, Karlsson M, RedlundJohnell I (1994) Long-term results of fracture of the scaphoid. A follow-up study of more than thirty years. J Bone Joint Surg Am 76:249–252 100. Lynch NM, Linscheid RL (1997) Corrective osteotomy for scaphoid malunion: technique and long-term follow-up evaluation. J Hand Surg Am 22:35–43 101. Nakamura P, Imaeda T, Miura T (1991) Scaphoid malunion. J Bone Joint Surg Br 73:134–137 102. Inoue G, Sakuma M (1996) The natural history of scaphoid non-union. Radiographical and clinical analysis in 102 cases. Arch Orthop Trauma Surg 115:1–4 103. Lindström G, Nyström A (1992) Natural history of scaphoid non-union, with special reference to asymptomatic cases. J Hand Surg Br 17:697–700 104. Inoue G, Shionoya K, Kuwahata Y (1997) Herbert screw fixation for scaphoid nonunions. An analysis of factors influencing outcome. Clin Orthop Relat Res 343: 99–106 105. Barton NJ (1997) Experience with scaphoid grafting. J Hand Surg Br 22:153–160 106. De Schrijver F, De Smet L (2001) Fracture of the hook of the hamate, often misdiagnosed as “wrist sprain”. J Emerg Med 20(1):47–51 107. Sawant M, Miller J (2000) Scaphocapitate syndrome in an adolescent. J Hand Surg Am 25(6):1096–1099 108. Vander GR, Dell PC, Glowczewskie F, Leslie B, Ruby LK (1984) Intraosseous blood supply of the capitate and its correlation with aseptic necrosis. J Hand Surg Am 9:677–683 109. Gelberman RH, Bauman TD, Menon J, Akeson WH (1980) The vascularity of the lunate bone and Kienbock’s disease. J Hand Surg Am 5:272–278 110. Carroll RE, Coyle MP Jr (1985) Dysfunction of the pisotriquetral joint: treatment by excision of the pisiform. J Hand Surg Am 10:703–707
Part VIII Hip
Bearing Surfaces Theofilos Karachalios and George Karydakis
Introduction Total Hip Arthroplasty (THA) is an effective surgical intervention for the end stages of hip joint diseases [1]. Wear debris production, which is primarily generated from the articular – bearing surface of the artificial joint, is the major factor limiting the survival of joint implants [2]. Wear debris induced osteolysis is a biological process causing a subtle progression of bone tissue destruction. Osteolysis, from the clinical point of view, is a major challenge, since signs and symptoms may not be clinically apparent until the late stages of failure (Fig. 1) [2]. The first trials of Sir John Charnley, pioneer of THA, showed early failure because of wear-induced implant loosening. After laboratory tests and clinical trials, he established the combination of a small (22 mm) metallic head with a cup of ultra high molecular weight polyethylene (PE), as the low friction arthroplasty principle. Despite the early success of Charnley’s THA, different combinations of metallic on PE and ceramic on PE bearing couplings used in THA carried on producing wear debris and thus osteolysis [3]. For this reason, a number of alternative bearing surfaces have been developed, some of them new and some following old concepts [2, 3] (Fig. 2). The current options a surgeon has as an alternative are metallic couplings, ceramic couplings and a combination of ceramic or ceramised heads with enhanced polyethylenes. Despite the fact that a great amount of basic science data exists
T. Karachalios () Associate Professor in Orthopaedics, Orthopaedic Department, Faculty of Medicine, School of Health Sciences, University of Thessalia, University General Hospital of Larissa, Mezourlo region,, Larissa 41110, Hellenic Republic, Greece e-mail:
[email protected]
Fig. 1 Extensive osteolysis around an asymptomatic THA (15 years follow-up)
concerning in-vitro tribology of these bearing surfaces, long-term clinical data is lacking, new biological problems have appeared and the cost effectiveness of their use has not been determined.
Metal-on-PE This combination is the original bearing coupling introduced by Charnley who utilized a metallic head of 22 mm combined with a cup of UHMWPE (PE). PE is produced
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_10, © 2011 EFORT
133
134 Fig. 2 Comparative in vitro wear rates of different couplings (From Santavirta [3])
T. Karachalios and G. Karydakis 0,2 0,2
0,15 0,1 0,1
0,05 0,002
0,001
0 METAL-PE
CERAMIC-PE
by the polymerization of ethylene (CH2 = CH2), with a molecular weight of 28. Today, PE, after numerous improvements, has a molecular weight of 5–6,000,000 and about 200,000 monomers. The wear of PE depends on molecular weight and less wear is related to larger molecular weight. PE in orthopaedic implants appears in two phases-crystal and amorphous. The crystal phase consists of molecular chains wound in layers of 10–50 nm. and length of 10–50 mm and gives, PE hardness, reduced plasticity and chemical endurance to biological fluids. It occurs in a percentage of 35–55%. The amorphous phase consists of molecular chains, the ends of the chains of the crystal phase and smaller molecules. Equilibrium between these two phases gives the strength, plasticity and hardness necessary for orthopaedic implants. The biphasic structure of PE is responsible for the material’s physical characteristics. When under load, the crystal areas are mechanically more stable ,whereas the amorphous areas can be twisted more easily. Crystallinity and cross-linking (described below), provide a more stable and less ductile structure. A PE with a higher percentage of amorphous phases is more ductile and weaker than a material with a higher percentage of the crystal phase. In the last decade it has been discovered that after exposure to ionized radiation (which is a method of sterilization), free radicals are produced. The latter, under special circumstances, link the molecules of the polymer in between and this improves its resistance to wear (cross linking) [4]. On the other hand, free radicals have the potential to link with oxygen molecules (oxidation) whether during sterilization or later during storage or even after
METAL-METAL
CERAMIC-CERAMIC
implantation, resulting in disruption of polymer chains and, finally, reduction of durability. For this reason, the method of sterilization with ionizing radiation in the air has been abandoned and is now accomplished in an environment free of oxygen; usually nitrogen is used. Also, packaging is performed in a vacuum. Another consideration is the danger of oxidation after material implantation, by the oxygen diluted in the tissues. To avoid this, during the cross-linking process the majority of free radicals has to be linked with polymer molecules. This can be achieved by heating PE above the melting point after the application of radiation and then practically all free radicals are linked with polymer molecules (extensive cross-linking). Finally, this process results in a PE which has a large percentage of cross-linked bonds and nearly zero free radicals (highly cross-linked PE). On clinical grounds one has to realize that not all cross-linked, commercially available PEs are the same; their main characteristics are shown in Table 1. After 1998 the application of acetabular cups from highly cross-linked PE was used in clinical practice (Fig. 3). Early radiological studies have demonstrated a reduction of penetration of the femoral head into the cross-linked PE in comparison with a conventional one [5, 6]. Furthermore, retrieval analysis of conventional and highly cross-linked PEs, has shown significant reduction of wear in the latter [7]. Wear particles generated by conventional PE are mainly spherical in shape, with a size from 0.1 to 1 mm (mean 0.5 mm). Some particles agglomerate to form fibrils, with a width of 0.3–1 mm and length of 10–25 mm, but fibrils of more than 100 mm length have been reported [8]. Wear particles have also been detected in acetabular screw holes in
Company
Smith and nephew
Zimmer
Centerpulse
DePuy/J & J
SHO
Biomet
Trade name
Reflection XLPE
Longevity
Durasul
Marathon
Crossfire
ArCom
None
7.5 Mrad (room temp)
Gamma
5 Mrad (room temp)
Gamma
9.5 Mrad (~125°C)
E-beam
10 Mrad (~40°C)
E-beam
10 Mrad(room temp)
Gamma
X-link protocol
None
Below melt
Melt
Melt
Melt
Melt
Annealing step
Table 1 Characteristics of different commercially available cross-linked PEs
Gamma (N2) 3.3 Mrad
Gamma (N2) 2.5 Mrad
Gas (VHP gas plasma)
Gas (EtO)
Gas (VHP gas plasma)
Gas (VHP or EtO)
Sterilization method
40
90
86
100
89
98
Volumetric wear reduction (%)
45%
94%
87%
No correlation possible
94%
98%
Wear reduction vs. standard poly
Bearing Surfaces 135
136
T. Karachalios and G. Karydakis
Metal-on-Metal
Fig. 3 Ceramic-on-cross-linked PE THA
modular implants. These particles are slightly bigger than the former and this difference in size may play a role in biological activity, as particles that are phagocyted are biologically more active. Wear particles from highly cross-linked PE are smaller than those from conventional ones, usually in nanometers (nm.) [9] and they do not form fibrils. There is a debate about whether they can cause inflammatory reactions that can result in osteolysis. Some authors consider that the smaller particles (0.24 vs. 0.45 mm) cause minor bone resorption in vitro [10]. Furthermore Illgen et al. [11] showed that wear particles of highly cross-linked PE have the same biological activity as the particles of conventional PE. On the other hand, other studies [12] have shown that macrophages are stimulated mainly by highly cross-linked PE wear particles and they release more inflammatory mediators than conventional PEs. Particle morphology may also affect biological reaction. Yang et al. [13] showed that longitudinal particles are more biologically active than spherical ones and this gives an advantage to highly cross-linked PEs. It is clear when cross-linked PEs are compared to conventional polyethylene that some mechanical properties such as the fatigue strength of the material as well as the elongation to failure are diminished. There have been some acetabular liners that have broken and this has been studied. It is suggested that the use of thin and small liners, the use of liners with rim elevation and the mal-positioning of the component should be avoided.
The interplay of materials, macrogeometry (diameter and clearance), microgeometry (surface topography) and lubrication influence the wear of metal-on-metal bearing coupling in THAs to a far greater degree than metal-on-PE. There is a variety of metals used as bearing surfaces, but the preferred alloys are those which consist of cobalt (Co) and chromium (Cr), due to their hardness. Chromium provides corrosion resistance, while the manufacturing process produces carbon-rich compounds of Cr, Co and molybdenum (Mo). These carbides are firmly adherent to the surrounding material and are much harder than it, and relatively brittle. The dispersion of these carbides is of great importance in wear resistance [14]. The macrogeometry is determined by the relative diameter of the ball and socket and the clearance (the size of the gap between the surfaces at the equator) of the bearing couple. The contact area can be increased by increasing the diameter of the bearing surfaces and/or by decreasing the clearance. Contact stresses are a function of material properties and are inversely proportional to contact area. The contact area is the main factor influencing lubrication, as clearance (the size of the gap between the two bearing surfaces) plays an important role in the amount and type of lubrication: smaller clearance encourages fluid film lubrication which generates the lowest friction. However, too small a clearance leads to equatorial contact, high friction and high torque which cause loosening of the implant and early failure. On the other hand, large clearance leads to a reduced contact area, loss of effective lubrication and rapid wear [15]. Industrial production sets the lower limit of clearance at 20 mm. Another important factor relating to the above is where the contact occurs. For bearing couples of equivalent diameters, equatorial contact is associated with higher frictional torques, comparable to the same contact area in a more polar location. Therefore, relatively polar contact is preferred [16, 17]. The type of lubrication is an important variable which influences friction and wear. To maintain low friction between the articulating surfaces, an ideal film thicknessto-surface roughness ratio (lambda – l ratio) is required. This can be achieved by controlling the microgeometry of the contacting surfaces and the elastic properties of the materials [3]. Full film lubrication completely separates the surfaces of a bearing couple, while mixed film lubrication partially separates the surfaces. The latter is the operative mechanism in most metal-on-metal bearing surfaces and the fluid film thickness is dependent on the properties of the
Bearing Surfaces
fluid but can be influenced by the properties of the bearing materials (macrogeometry and microgeometry). After implanting a THA, the actual contact area is at the tips of the asperities of the bearing surfaces and a lubricant film can influence wear significantly. As wear proceeds, the contact area at the tips increases and this “run-in” procedure can produce a more favorable microgeometry for lubricant film to separate the surfaces and reduce wear rate. Fluid film lubrication is enhanced by using as large a femoral head as is practically possible and a clearance as small as possible [18–20]. Wear rates have been measured in laboratory studies as well as in retrieval studies. In the former, Medley et al. [20] described wear volumes ranging from 0.09 to 61 mm3 per million cycles and linear wear rates ranging from 1.3 to 100 mm per million cycles. In most studies, the wear rate decreased substantially after the first 0.1–0.5 million cycles [21]. Other studies have shown similar results. In the retrieval studies, it has been confirmed that the wear rate is much smaller than that typically seen with polyethylene: the worst case estimate of combined femoral and acetabular linear wear was 4.2 mm/year, about 25 times less. It has also been confirmed that the larger the head, the smaller the wear rate (about two times) (Fig. 4) [22, 23]. Recently, it has been recognized that metal-on-metal articulations are sensitive to cup mal-positioning with specific designs showing excessive wear when the acetabular component is placed in an open position [24].
Biological Considerations Metal wear particles are nanometers in linear dimension, which is substantially smaller than PE wear particles [25, 26]. The size of metal particles ranges from 0.01 to 5 mm as
Fig. 4 Metal-on-metal resurfacing arthroplasty
137
reported by electron microscopy studies and most of them are smaller than 50 nm [25]. The number of particles produced per year, calculated by considering the volumetric wear rate and the size of the particles, has been estimated to be 6.7 × 1012–2.5 × 1014, which is 13–500 times the number of PE particles produced per year by a typical metal – polyethylene joint [26]. This large aggregate surface area of metal wear particles may have both local and systemic effects. The local tissue reaction around metal-on-metal prostheses, indicated by the number of histiocytes, is about one grade lower than that around metal-on-PE prostheses [25, 26]. This can be explained because metal particles are smaller than polyethylene particles and histiocytes are able to store a larger number of metal particles and therefore the total number of histiocytes required to store the metal particles is lower. Also the small metal particles enter the histiocytes by pinocytosis instead of phagocytosis, which may alter the cellular response to the particles. Furthermore, Co-Cr particles have greater potential for cytotoxicity than polyethylene particles and the cell may be incapable of the same inflammatory response. Dissolution of metal particles results in elevation of the cobalt and chromium ion concentrations in erythrocytes, serum and urine [27]. In vitro studies have shown that there is a dose-related response to metal particles: low to moderate concentrations stimulate the release of cytokines that can lead to osteolysis. At higher concentrations, Co-Cr particles have been found to be cytotoxic leading to cell death [28]. The incidence of osteolysis associated with metal-on-metal bearings appears to be comparatively low [29]. Co and Cr wear particles have been shown to induce carcinoma in animal models [30, 31]; consequently, there is a concern about the same effect on human tissues. For implant site tumors, there is ambiguous clinical evidence. A study reports 19 cases of periprosthetic tumours; the majority of them were malignant fibrous histiocytoma [19]. A more recent study reports four cases of soft tissue sarcomas surrounding orthopaedic implants [32]. For remote site malignancies, epidemiologic studies have shown an elevated risk of lymphoma and leukemia associated with older metal-onmetal total hip arthroplasties [33]. Other studies did not suggest increased risk for lymphoma or leukemia, for patients operated on after 1973 [34]. The data of these studies are limited because of the small number of patients who underwent metal-on-metal THA. Furthermore, the majority of patients in these reports have less than 10 years of follow-up. Due to the fact the tumour development latent period is more than 20 years, longer follow-up studies of large patient groups is needed to better assess the risk of cancer with any implant system [35]. Currently, the development of pseudo-tumors around metal-on-metal resurfacing hip arthroplasties is under thorough investigation [36].
138
Ceramic-on-Ceramic Ceramic bearing couplings have demonstrated the lowest in vivo wear rates to date of any bearing combination [37]. Generally, the same principles of friction and lubrication reported for metal-on-metal apply to ceramic-on-ceramic bearings. Additionally, they have two important properties: they are hydrophilic, resulting in a uniformly distributed synovial fluid over the whole bearing surface area and they have greater hardness than metal and can be polished to a much lower surface roughness. Here the l ratio is higher, resulting in a reduced coefficient of friction and it is likely to achieve true fluid-film lubrication. The greater hardness of ceramic materials, however, results in a lack of ductility and this represents the major disadvantage. Because the material is not ductile it will not deform, but when it reaches a critical threshold it will fracture. Today’s ceramics (biolox forte) are of high quality with decreased grain size, inclusions, and grain boundaries. They present far greater burst strength than pre-1995 ceramics and today are mated with implants that have excellent fixation records, high taper tolerances, and designs that minimize the risk of ceramic impingement. A major problem is the necessity for a proper implant placement. Hips with a lateral opening of less than 30° or greater than 55° and high neck/shaft angle (>140° are at risk of neck-socket impingement and high wear as a result of stress concentration in the very stiff ceramic material [38]. This would affect the performance of the ceramic bearing. Another problem that surgeons should take into consideration is the liner chipping during insertion, intra-operatively. Generally, ceramic materials have better biocompatibility than metal alloys. A retrieval study [39] reports a double size range of ceramic wear debris: the smallest were 5–90 nm (mean 24 nm) and the largest were 50–3,200 nm (mean 430 nm). The authors suggested that these two types of wear debris are generated by two different wear mechanisms. The small group is generated under normal articulating conditions and the large one under microseparation conditions. The latter situation occurs during the swing phase of gate, when the femoral head and the acetabular liner can separate up to 2 mm. After load application at heel strike, the femoral head moves vertically to relocate in the cup and this may result in changes in the wear performance of the bearing surfaces (stripe wear) [16, 40]. Wear debris from ceramic-on-ceramic bearings may not be as bio-inert as initially assumed, because osteolysis has been described in some patients with COC bearing [41].
T. Karachalios and G. Karydakis
It seems there is less inflammatory reaction compared to metal on metal-and-metal on PE bearings. Ceramic-on-ceramic bearings have been used for over 30 years with excellent survival and low ceramic fracture rates [22, 42]. Many studies with contemporary ceramics [43, 44] have also reported satisfactory results in the midterms.
Oxidised Zirconium (Oxinium) Bearing Surface A zirconium oxide surface (oxinium) is produced by thermally driven oxygen diffusion that transforms the metallic zirconium alloy surface into a durable low friction oxide [45]. This oxidized layer is not a ceramic coating but a transformation of the surface leaving a 5–10 micron coating that is much harder and scratch resistant than the untreated alloy but less so than true alumina ceramic surface. Wear simulator studies have shown lower wear of both smooth and roughened oxinium femoral heads on conventional and crosslinked PE compared to cobalt chrome on PE. One has to understand that this is a new technology with no long clinical record. In a prospective randomized study (Fig. 5) established in our department, oxinium heads (either 28 or 32 mm) have shown very low wear rates when compared to ceramic on conventional or cross-linked PE (Fig. 6). However, retrievals of heavily damaged femoral heads have been reported in patients who suffered recurrent dislocations [46]. Under these circumstances, both the femoral head and acetabular liner should be revised during revision surgery for instability.
Fig. 5 Oxinium-on-cross-linked PE THA
139
Bearing Surfaces 3-D WEAR 1600
1375,35
3-D WEAR (mm3 )
1400 1110,616
1200 1000
754,891
790,9
810,5
265,9
288,11
320,22
324,34
285,33
298,21
300,2
305,8
723
800 600
530,963 420,507
400
240,2
200
1500,9
1420,2
266,345
536,187 310,7 341,946
492,708
0 1 YEAR
2 YEARS 3 YEARS
4 YEARS 5 YEARS
6 YEARS
OXINIUM 32
CER - POLY
OXIN - CROSS 28
CER - CROSS
Fig. 6 Comparative in vivo 3-D wear of different couplings including oxinium-on-cross-linked PE (Polywear software – unpublished personal data)
Ceramic-on-Metal Recent simulator studies have demonstrated superior wear rates for alumina ceramic-on-metal articulations when compared to metal-on-metal articulations [47]. Using 28 mm femoral heads, medical grade alumina biolox forte heads were articulated with acetabular cups manufactured from medical grade high carbon wrought cobalt chrome alloy. These were compared to metal-on-metal articulations utilizing medical grade low carbon cobalt chrome alloy femoral heads on medical grade high carbon cobalt chrome alloy sockets. Metal particles from both surfaces were of nanometer sizes (6–30 nm) and while the ceramic-on-metal articulations produced slightly smaller particles, they were far fewer in number. Clinical studies are currently underway to evaluate the performance of these bearings.
References 1. Harris WH (1995) The problem is osteolysis. Clin Orthop Relat Res 311:46–53 2. Abu-Amer Y, Darwech I, Clohisy JC (2007) Aseptic loosening of total joint replacement: mechanisms underlying osteolysis and potential therapies. Arthritis Res Ther 9(Suppl 1):1–6 3. Santavirta S, Bohler M, Harris WH, Konttinen YT, Lappalainen R, Muratoglou O, Rieker C, Salzer M (2003) Alternative materials to improve total hip replacement tribology. Acta Orthop Scand 74(4):380–388
4. McKellop H, Shen FW, Lu B, Campbell P, Salovey R (2000) Effect of sterilization method and other modifications on the wear resistance of acetabular cups made of ultra-high molecular weight polyethylene. J Bone Joint Surg 82A:1708–1725 5. Nivbrant B, Roerhl S, Hewitt BJ et al (2003) In vivo wear and migration of high cross-linked poly cups: a RSA study. Presented at the 49th annual meeting of the ORS, New Orleans, 2003 6. Martell JM, Verner JJ, Incavo SJ (2003) Clinical performance of a highly cross-linked polyethylene at two years in total hip arthroplasty: a randomized prospective trial. J Arthroplasty 18(7 Suppl 1):55–59 7. Muratoglu OK, Greenbaum E, Bragdon CR, Jasty M, Freiberg AA, Harris WH (2004) Surface analysis of early retrieved acetabular polyethylene liners: a comparison of standard and highly cross-linked polyethylenes. J Arthroplasty 19(1):68–77 8. Shandhag AS, Jacobs JJ, Glant TT, Gilbert JL, Black J, Galante JO (1994) Composition and morphology of wear debris in failed uncemented total hip replacement. J Bone Joint Surg 76B:60–67 9. Ingram JH, Stone M, Fisher J, Ingham E (2004) The influence of molecular weight, crosslinking and counterface roughness on TNF-alpha production by macrophages in response to ultra high molecular weight polyethylene particles. Biomaterials 25(17):3511–3522 10. Green TR, Fisher J, Matthews JB, Stone MH, Ingham E (2000) Effect of size and dose on bone resorption activity of macrophages by in vitro clinically relevant ultra high molecular weight polyethylene particles. J Biomed Mater Res 53:490–497 11. Illgen RL 2nd, Forsythe TM, Pike JW, Laurent MP, Blanchard CR (2008) Highly crosslinked vs. conventional polyethylene particles: an in vitro comparison of biologic activities. J Arthroplasty 23(5):721–731 12. Huddleston JI, Hayata K, Kawashima M et al (2006) Human macrophage response to highly cross-linked UHMWPE debris. Trans Orthop Res Soc 31:700 13. Yang SY, Ren W, Park Y, Sieving A, Hsu S, Nasser S, Wooley PH (2002) Diverse cellular and apoptotic responses to variant shapes of UHMWPE particles in a murine model of inflammation. Biomaterials 23:3535–3543 14. Schmidt M, Weber H, Schon R (1996) Cobalt chromium molybdenum metal combination for modular hip prostheses. Clin Orthop Relat Res 329:35–47 15. Schey JA (1996) Systems view of optimizing metal on metal bearings. Clin Orthop Relat Res 329:115–127 16. Kothari M, Bartel DL, Booker JF (1996) Surface geometry of retrieved McKee-Farrar total hip replacements. Clin Orthop Relat Res 329:141–147 17. Schmalzried TP, Peters PC, Maurer BT, Bragdon CR, Harris WH (1996) Long-duration metal-on-metal total hip arthroplasties with low wear of the articulating surfaces. J Arthroplasty 11:322–331 18. Dowson D, Jin ZM (2006) Metal-on-metal hip joint tribology. J Eng Med 220:107–111 19. Jacobs JJ, Rosenbaum DH, Hay RM, Gitelis S, Black J (1992) Early sarcomatous degeneration near a cementless hip replacement. J Bone Joint Surg 74B:740–744
140 20. Medley JB, Chan FW, Krygier JJ, Bobyn JD (1996) Comparison of alloys and designs in a hip simulator study of metal on metal implants. Clin Orthop Relat Res 329:148–159 21. Isaac GH, Thompson J, Williams S, Fisher J (2006) Metalon-metal bearings surfaces: materials, manufacture, design, optimization, and alternatives. Proc Inst Mech Eng H 220(2):119–133 22. Sedel L, Kerboul L, Christel P, Meunier A, Witvoet J (1990) Alumina on alumina hip replacement: results of survivorship in patients. J Bone Joint Surg Br 72:658–663 23. Willert HG, Buchhorn GH, Göbel D, Köster G, Schaffner S, Schenk R, Semlitsch M (1998) Wear behavior and histopathology of classic cemented metal on metal hip endoprotheses. Clin Orthop Relat Res 356:170–180 24. Grammatopoulos G, Pandit H, Glyn-Jones S, McLardySmith P, Gundle R, Whitwell D, Gill HS, Murray DW (2010) Optimal acetabular orientation for hip resurfacing. J Bone Joint Surg Br 92(8):1072–1078 25. Doorn PF, Campbell PA, Amstutz HC (1996) Metal versus polyethylene wear particles in total hip replacements. A review. Clin Orthop Relat Res 329:S206–S216 26. Doorn PF, Campbell PA, Worrall J, Benya PD, McKellop HA, Amstutz HC (1998) Metal wear particle characterization from metal on metal total hip replacements: transmission electron microscopy study of periprosthetic tissues and isolated particles. J Biomed Mater Res 42:103–111 27. McDonald SJ, McCalden RW, Chess DG, Bourne RB, Rorabeck CH, Cleland D, Leung F (2003) Metal-on-metal versus polyethylene in hip arthroplasty: a randomized clinical trial. Clin Orthop Relat Res 406:282–296 28. Catelas I, Campbell PA, Frausto A, Mills BG, Amstutz HC (2003) Semi-quantitative analysis of cytokines in MM THR tissues and their relationship to metal particles. Biomaterials 24(26):4785–4797 29. Zahiri CA, Schmalzried TP, Ebramzadeh E, Szuszczewicz ES, Salib D, Kim C, Amstutz HC (1999) Lessons learned from loosening of the McKee-Farrar metal-on-metal total hip replacement. J Arthroplasty 14:326–332 30. Freeman MA, Swanson SA, Heath JC (1969) Study of the wear particles produced from cobalt-chromium-molybdenum-manganese total joint replacements prostheses. Ann Rheum Dis 28(Suppl 5):29–32 31. Heath JC, Freeman MA, Swanson SA (1971) Carcinogenic properties of wear particles from prostheses made in cobaltchromium alloy. Lancet 1:564–566 32. Langkamer VG, Case CP, Collins C, Watt I, Dixon J, Kemp AJ, Atkins RM (1997) Tumors around implants. J Arthroplasty 12:812–818 33. Black J (1998) Biomaterials in total hip arthroplasty. In: Rubash H, Callaghan J, Rosenberg A (eds) The adult hip, vol 1. Lippincott-Raven, Philadelphia, pp 46–53
T. Karachalios and G. Karydakis 34. Gillespie WJ, Henry DA, O’Connell DL, Kendrick S, Juszczak E, McInneny K, Derby L (1996) Development of hematopoietic cancers after implantation of total joint replacement. Clin Orthop Relat Res 329:S290–S296 35. Tharani R, Dorey FJ, Schmalzried TP (2001) The risk of cancer following total hip or knee arthroplasty. J Bone Joint Surg Am 83A:774–780 36. Kwon YM, Ostlere SJ, McLardy-Smith P, Athanasou NA, Gill HS, Murray DW (2011) Asymptomatic pseudotumors after metal-on-metal hip resurfacing arthroplasty prevalence and metal ion study. J Arthroplasty 26:28 37. Clarke IC, Good V, Williams P et al (2000) Ultra-low wear rates for rigid-on-rigid bearings in total hip replacements. Proc Inst Mech Eng H 214:331–347 38. Walter A (1992) On the material and the tribology of alumina-alumina couplings for hip joint prostheses. Clin Orthop Relat Res 282:31–46 39. Hatton A, Nevelos JE, Nevelos AA, Banks RE, Fisher J, Ingham E (2002) Alumina-alumina artificial hip joints. Part I: A histological analysis and characterization of wear debris by laser capture microdissection of tissues retrieved at revision. Biomaterials 23:3429–3440 40. Stewart TD, Tipper JL, Insley G, Streicher RM, Ingham E, Fisher J (2003) Severe wear and fracture of zirconia heads against alumina inserts in hip simulator studies with microseparation. J Arthroplasty 18:726–734 41. Yoon TR, Rowe SM, Seon JST, KJ MWJ (1998) Osteolysis in association with a total hip arthroplasty with ceramic bearing surfaces. J Bone Joint Surg 80A:1459–1468 42. Hamadouche M, Boutin P, Daussange J, Bolander ME, Sedel L (2002) Alumina on alumina total hip arthroplasty; a minimum 18.5 year follow up study. J Bone Joint Surg 84A:69–77 43. D’Antonio J, Capello W, Manley M, Naughton M, Sutton K (2005) Alumina ceramic bearings for total hip arthroplasty. Five year results prospective randomized study. Clin Orthop Relat Res 436:164–171 44. Garino JP (2000) Modern ceramic-on-ceramic total hip systems in the United States: early results. Clin Orthop Relat Res 379:41–47 45. Reis MD, Saleahi A, Widding K, Hunter G (2002) Polyethylene wear performance of oxidized zirconium and cobalt-chromium knee components under abrasive conditions. J Bone Joint Surg 84A:129–135 46. Kop AM, Whitewood C, Johnston DJ (2007) Damage of oxinium femoral heads subsequent to hip arthroplasty dislocation. J Arthroplasty 22(5):775–779 47. Firkin S, Fisher J (2001) A novel low wearing differential hardness, ceramic on metal hip joint prosthesis. J Biomech 34(200):1291–1298
Hip Pain in the Young Adult Moritz Tannast, Christoph E. Albers, Simon D. Steppacher, and Klaus A. Siebenrock
Introduction Persistent chronic hip pain in the young adult is a relatively common problem which often represents a frustrating situation both for patients and the treating physician. Many underlying causes are known and should be taken into consideration during the diagnostic algorithm. These include inflammatory, tumour, vascular, muscular, neurological disorders and pathologies of the adjacent joints and body regions. However, based on our experience, mechanical hip problems such as developmental dysplasia of the hip (DDH) and femoro-acetabular impingement (FAI) are the main reasons why young adults present with hip pain. This article focuses on the underlying biomechanical concepts and discusses the diagnosis, potential treatment and outcome when joint-preserving surgery is performed.
General Principles The Normal Hip Joint A normal hip joint is defined by sufficient acetabular coverage, a round and concentric femoral head as well as an impingement-free range of motion. Depending on the activity level, any deviations from this ‘physiological equilibrium’[1] can result in a pathological biomechanical function of the hip which leads sooner or later to degenerative changes of the joint. It should be emphasized that both static and dynamic components can cause the articular
M. Tannast () Department of Orthopaedic Surgery, Inselspital, University of Bern, Freiburgstrasse, 3010 Bern, Switzerland e-mail:
[email protected]
damage. Generally, the static component is the predominant patho-mechanism in DDH [1]. In contrast, the dynamic component plays the dominant role in hips with FAI [2]. In contrast to DDH, the term ‘FAI’ is not a pathological entity per se. It rather describes a patho-mechanical process that is secondary to alterations of the hip morphology.
Pre-operative Clinical Evaluation The pre-operative clinical and radiographic evaluations have been described elsewhere in detail [3]. Clinical examination should comprise an assessment of the range of motion and provocative pain tests. Generally, patients with DDH present with an increased range of motion while patients with FAI have a decreased motion amplitude, particularly in flexion and internal rotation. Combined flexion and internal rotation typically hurts reproducibly in both pathologies. The pain is caused by a stressed, damaged labrum although the underlying patho-mechanism differs between DDH and FAI. Analogously, combined extension and external rotation might hurt in both pathologies as described later.
Pre-operative Radiographical Evaluation The standard radiographic work up for diagnosis and decision-making in joint-preserving hip surgery involves a series of conventional radiographs and a specific MRIarthrogram [3]. Additional imaging with computed tomography or other imaging means might be obtained in addition. The basic set of conventional radiographs includes an antero-posterior radiograph of the pelvis (Fig. 1a) and a cross-table lateral radiograph of the hip (Fig. 1b). These two views provide an imperative overview which shows the main pathological features. For specific questions, a false profile view [4] or a strong lateral pelvis radiograph
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_11, © 2011 EFORT
141
142
M. Tannast et al.
might be obtained in addition. It is of outmost importance that the radiographs are taken according to a standardized technique, particularly in terms of film-focus-distance, centreing of the x-ray beam and patient positioning. Nowadays, electronic tools are available to further quantify the radiographic anatomy of the hip [5, 6].
a
The specific MRI technique [7] is obtained using a flexible surface coil and intra-articular gadolinium contrast agent. Besides the usual axial, coronal oblique, and sagittal oblique sequences, this specific technique also includes a radial proton density-weighted sequence (Fig. 2). These sections lie orthogonal to the acetabular rim and labrum.
b
Fig. 1 (a) The technique for the antero-posterior pelvic radiograph is shown. The film-focus distance should be 1.2 m. The patient is positioned supine with the legs internally rotated to compensate for femoral antetorsion. The x-ray beam should be directed to the mid-point between the symphysis and a line
a
c onnecting the antero-superior iliac spines. (b) The cross-table lateral radiograph is taken with the contralateral hip flexed. The film-focus distance is 1.2 m. The x-ray beam centre is directed to the inguinal fold
b
superior 14
1
2
13
3
su
4
12 posterior
pe
anterior
rio
r
5
11
inf
er
6
10 9
8
ior
7
inferior
Fig. 2 (a) The geometry of the radial sections for the MR-arthrogram is shown. The sequences are oriented radially along the femoral neck axis. (b) A total of 14 sections are obtained
143
Hip Pain in the Young Adult
Hip Dysplasia Patho-mechanism in Hip Dysplasia The classic definition of hip dysplasia is a lateral acetabular deficiency. This is the result of a decreased surface area of the lunate surface [8] and the steeply-orientated acetabular sourcil. The resulting lateral centre-edge angle is typically less than 20° [9]. The articular pressure is concentrated on a small area at the edge of the socket as a result of the static overload (Table 1). The articular stress is increased to many times its normal magnitude (Fig. 3) [1, 9]. An additional subluxation of the joint gradually occurs when the surrounding soft tissues are not able to constrain the femoral head within the deficient acetabulum.
Damage Pattern The static instability of the hip together with the chronic overload leads to a compensatory hypertrophy of the acetabular labrum [10]. As a result, this enlarged and
degenerated labrum develops a partial tear or detaches completely from the acetabular rim. The labrum even may detach with a piece of bone or cartilage (Fig. 4). This leads to an additional femoral head instability. In contrast to FAI [10], ganglia and cysts are more often seen in hips with DDH. According to the most frequent location of acetabular deficiency, the labral alterations are typically seen in the antero-superior acetabular quadrant. During clinical examination, this can lead to painful flexion and internal rotation. In addition, patients with DDH present with painful extension and external rotation where the shear forces on the dysplastic area are most pronounced.
Classification Besides the more classic antero-lateral deficiency, a deficient acetabular coverage can occur in any portion of the acetabulum (Fig. 5a). A lateral acetabular deficiency is defined by a decreased lateral centre-edge angle with a normal version of the acetabulum (Fig. 5b). We define anterior acetabular deficiency with an increased acetabular anteversion while the lateral centre-edge angle is normal (Fig. 5c). Typically,
Table 1 Comparison of developmental dysplasia of the hip and femoro-acetabular impingement Parameter
Hip dysplasia
Femoroacetabular impingement
Prevalence
25–50 per 1,000
10–15 per 100
Gender distribution
f :m = 3.4:1
Cam: m:f = 14:1
Typical age
29 (13–56)
Cam: 32 (21–51)
Pincer: f:m = 3:1 Pincer: 40 (40–57) Types
Anterolateral (71%)
Cam (9%)
Anterior (8%)
Pincer (5%)
Lateral (4%)
Mixed (86%)
Posterior (17%) Predominant patho-mechanism
Static overload of the acetabulum
Dynamic femoro-acetabular conflict
Ganglia/cysts
Frequent
Rare
Natural history
Osteoarthritis if LCE < 16° or acetabular index >15°
Presumably osteoarthritis
Volume increase labrum
Frequent
No
Type of tear
Partial labral tear, sometimes detachment of acetabular cartilage and/or rim
Cam: chondrolabral dissociation
Pain in flexion and internal rotation
Present (due to partial labral tear)
Present in anterior impingement
Pain in extension and external rotation
Often present (due to apprehended joint subluxation)
Present in posterior impingement
Peri-acetabular osteotomy
Surgical hip dislocation, hip arthroscopy, peri-acetabular osteotomy
Pincer: full-thickness labral tear
Location of labral tear Therapy
144
M. Tannast et al.
the posterior acetabular rim is excessive which in turns can lead to impingement symptoms described later. We define a posterior acetabular deficiency with a deficient acetabular posterior wall (Fig. 5d). This is often seen in acetabular retroversion where the anterior acetabulum presents with excessive coverage. It has been shown that one out of six dysplastic hips are retroverted [11]. Based on this definition, it is obvious that both a dysplastic and an impinging component can occur in the same hip.
a
b
Fig. 3 (a) In a normal hip, a more or less equal distribution of the axial load is shown which is reflected in a regular sclerosis of the acetabular sourcil. (b) In hip dysplasia, the resulting articular pressure is concentrated on a small area at the edge of the socket. (c) After reorientation of the acetabulum by peri-acetabular osteotomy, a normal stress distribution is restored
Radiographic Diagnosis As mentioned earlier, the diagnosis is primarily based on conventional radiography. Acetabular dysplasia is defined as a hip with a lateral centre-edge angle (LCE) of less than 20°. An LCE angle between 20° and 25° is considered as borderline dysplasia [9]. Typically, the anterior centre-edge angle (ACE) (which is measured on a false profile view) is decreased analogously [4]. An ACE angle of less than
145
Hip Pain in the Young Adult Fig. 3 (continued)
c
20° is considered as dysplastic; between 20° and 25° as borderline dysplastic, and above 25° as normal [4]. Both definitions were established before the recognition of the importance of the anterior and posterior acetabular coverage. In Table 2, we provide the reference values for dysplastic, normal, and pincer-impingement hips as nowadays used in our department. This evaluation allows a more comprehensive assessment of dysplastic hips.
The Natural History of Hip Dysplasia Hip dysplasia leads to osteoarthritis [12, 13]. Several factors can be identified as very strong negative predictors of the long-term survivorship of the dysplastic hip: subluxation [12] (i.e. a broken Shenton’s line), an LCE angle of less than 16°, and an acetabular index of greater than 15° [13]. In symptomatic patients with these radiographic features, an acetabular re-orientation should be strongly recommended. Asymptomatic patients with the abovementioned radiographic features should be informed accordingly and followed-up on a regular basis. The goal of a surgical intervention is to prevent or delay further degenerative changes of the joint.
Treatment The treatment of DDH consists of a re-orientation of the acetabulum to provide more (but not excessive) acetabular coverage and stability of the hip (Fig. 4c). Peri-acetabular osteotomy (PAO) has become firmly established as the current state of the art for treatment of DDH in the adolescent and adult patient [14]. A modified Smith-Peterson approach is performed followed by four peri-acetabular osteotomies and a controlled fracture to completely mobilize the acetabulum. This provides excessive potential for acetabular reorientation. The posterior column of the pelvis remains intact maintaining stability of the pelvic ring without compromising the dimensions of the birth canal. Since dysplastic hips often present with an oval-shaped femoral head and a decreased femoral head-neck ratio, a routine arthrotomy should be performed to check for a potential femoro-acetabular impingement. A concomitant femoral head-neck osteoplasty is indicated if this problem is present [15]. The most important and most difficult step of the operation is the correct three-dimensional re-orientation of the acetabulum. Persistent deficiency of the acetabulum will lead to persistent dysplasia while overcoverage in any dimension can cause femoro-acetabular impingement [15]. Table 2
146
a
M. Tannast et al.
b
c
d
Fig. 4 (a) The pre-operative antero-posterior radiograph of a patient with hip dysplasia is shown. (b) The acetabular labrum hypertrophies with labral tears and cysts. The aspherical morphology of the head-neck junction is obvious. (c) The same
patient is shown 12 years after correction by peri-acetabular osteotomy. (d) The axial cross-table view shows the restoration of a well-contoured femoral head-neck junction as a result of an additional osteochondroplasty
gives an overview on the target zones for the acetabular re-orientation. Since the lunate surface of the acetabulum is decreased in dysplastic hips [8], it is often impossible to re-orientate the acetabulum in a way that all radiographic factors lie within the normal range. Therefore, as many factors as possible should lie in the normal target zone.
study on the first 75 hips with PAO, Steppacher et al. showed a 87% survivorship of the hip at 10 years and a 61% survivorship at 20 years post-operatively [19]. Since these results represents the follow-up of very first 75 patients that had undergone PAO, an even better survivorship can be expected with the current technique and indications. The following negative predictors are known from the literature: advanced age and pre-operative osteoarthritis, pre-operative joint subluxation, undercorrection, pre-operative low Merle d’Aubigné score, pre-operative pain in deep flexion and internal rotation [17, 19–23].
Results Various reports have proven the long-term potential of PAO as surgical treatment for DDH [16–18]. In their
147
Hip Pain in the Young Adult
Femoro-acetabular Impingement Patho-mechanism in Femoro-acetabular Impingement In contrast to DDH, the leading patho-mechanism is predominantly a dynamic conflict of osseous prominences of the acetabulum and/or femur (Table 1). Two basic types, cam and pincer FAI, are distinguished. FAI only describes the patho-mechanism but can be caused by different underlying pathological entities, such as slipped capital femoral epiphysis, Legg-Calvé-Perthes-disease, etc.
a
b
Fig. 5 An example of an anterolateral (a), lateral (b), anterior (c) and posterior (d) acetabular deficiency is shown. The red line represents the anterior acetabular rim. The blue line shows the posterior acetabular rim. The red area visualizes the anterior acetabular coverage, the blue area the posterior acetabular coverage, respectively
Cam Impingement In cam impingement, the predominant abnormality is the aspherical contour of the antero-superior femoral headneck junction (Fig. 6a). Normally, the antero-superior femoral head-neck junction has a concave, spherical configuration. This is either flattened or convex in hips with cam impingement. The eccentric part then slides into the acetabulum and induces compression and shear forces at the junction between the labrum and the cartilage. The maximum impact force is perpendicular to the joint surface (Fig. 6b). As a result, the labrum is stretched
148 Fig. 5 (continued)
M. Tannast et al.
c
d
and pushed outward while the cartilage is compressed. This leads to a separation between the labrum and the cartilage. The labrum itself with this ‘undersurface lesion’ of its matrix partially remains attached to the acetabular rim. Asphericities of the femoral head-neck junction often are idiopathic but can be secondary to known causes of hip osteoarthritis such as post-traumatic, slipped capital femoral epiphysis, or Legg-CalvéPerthes disease.
Pincer Impingement In pincer impingement, a linear contact occurs between the acetabular rim and the femoral head-neck junction. The maximal impact force is tangential to the joint surface (Fig. 6c, d). The labrum is then compressed between the acetabular rim and the femoral neck and tears completely off the acetabulum. Pincer impingement can be the results of an excessive acetabular rim, a supra-physiological hip
149
Hip Pain in the Young Adult
Table 2 Definition of a dysplastic, normal, and pincer impingement hip. The values given represent reference values for symptomatic hips. Depending on the activity level of the individual patient Excessive pincer impingement Parameter Dysplastic hip Normal Pincer impingement (Protrusio acetabuli) LCE [º]
<22
23 – 33
34 – 39
>40
Acetabular index [º]
>14
3 – 13
(−7) – 2
<−8
Total coverage [%]
<69
70 – 83
84 – 93
>93
Anterior coverage [%]
<14
15 – 26
27 – 32
>33
Posterior coverage [%]
<35
36 – 47
48 – 55
>56
Extrusion index [%]
>27
17 – 27
12 – 16
<11
a
b Fibrocartilaginous separation
Aspherical head
FC
c
Excessive coverage
d Labral tear
FP
Contre-coup lesion
Fig. 6 (a) Cam impingement is caused by an aspherical femoral head. (b) This aspherical portion can lead to shear stresses at the end of the range of motion with subsequent fibrocartilaginous separation. (c) The main cause of pincer impingement is excessive
acetabular coverage. (d) This can lead to a linear contact between the femoral head-neck junction and the acetabular rim. The labrum usually tears entirely off the acetabular rim
150
motion, or a massive asphericity of the femoral head-neck junction (where the aspherical portion cannot even enter the joint). The excessive acetabular rim can occur focal in one area of the joint or circumferential in generally deep hips. Reference values for the different types of overcoverage are provided in Table 2.
Damage Pattern The damage pattern of the articular cartilage differs between cam and pincer-type FAI. In cam impingement, the aspherical portion causes a relatively deep lesion involving approximately one third of the total depth of the acetabulum (Fig. 7a). The lesion is more or less restricted to the antero-superior quadrant of the acetabulum. In pincer impingement, the transmission of force to the cartilage is restricted to a narrow but more circumferential band of the acetabular rim (Fig. 7b). Often, a second cartilage lesion is seen in the postero-inferior portion of the hip joint which reflects a ‘contre coup’ phenomenon.
Classification Cam Impingement Types Cam impingement can be caused by asphericities that typically are located in the antero-superior or lateral portion of the femoral head-neck junction. Antero-superior femoral
a
M. Tannast et al.
head-neck deformities are usually idiopathic due to an extension of the epiphysis [24]. Lateral femoral head-neck deformities (so-called ‘pistol grip’ deformities) are often seen in hips with slipped-capital femoral epiphysis. Other forms of cam impingement are seen in coxa vara and femoral retrotorsion. Although in these two pathologies, the femoral offset might be normal, the femoral head-neck junction is closer to the acetabular rim during normal activities of daily living putting the hip at risk for cam impingement.
Pincer Impingement Types The excessive coverage in pincer hips can be general or focal. General overcoverage is typically seen in hips with coxa profunda or protrusio acetabuli. Focal overcoverage can occur in the anterior or posterior portion of the acetabulum. Anterior overcoverage is seen in patients where the cranial portion of the acetabulum is posteriorly oriented (‘retroverted’). Often this is seen with a deficient posterior wall. Typically, forced flexion and internal rotation (the ‘anterior impingement test’) is positive. Posterior overcoverage is usually seen in patients with an excessive acetabular anteversion and a simultaneous deficient anterior wall. Here, combined forced extension and external rotation (the ‘posterior impingement test’) is reproducibly painful. In contrast to hip dysplasia, this particular motion pattern is painful because of the direct femoro-acetabular impingement and not due to an apprehended subluxation of the hip joint.
b
Fig. 7 The damage pattern in cam (a) and pincer (b) impingement is shown. Explanation in text
151
Hip Pain in the Young Adult
Radiographic Diagnosis Many radiographic parameters have been described to quantify acetabular and femoral patho-morphologies related to FAI. An excellent overview on the correct radiographic technique, interpretation and potential pitfalls is given elsewhere [3]. Table 2 provides an overview on selected radiographic parameters and their reference values. The analyzed features include parameters describing the acetabular depth, coverage, sphericity, and joint congruence. It should be noted that the diagnosis of impingement is based on a positive correlation between symptoms, physical findings on clinical examinations, suggestive conventional radiographs, and signs of chondro-labral degeneration on the radial MR-arthrogram [25]. Asymptomatic patients with incidental findings on conventional radiography exist and should be followed-up closely.
The Natural History of Femoro-acetabular Impingement The natural history of FAI has not yet been described in the literature. In contrast to DDH, radiographic features that will ultimately lead to end-stage osteoarthritis do not exist yet for FAI and are the subject of intensive research. It is suggestive that other factors (e.g. the individual activity level) besides the patho-morphology play an important role in the development and progression of arthritic degeneration.
Treatment Non-operative treatment might be an option if the activity level of a patient can be decreased to relieve symptoms. However, the pathological anatomy can only be addressed surgically. The curative treatment involves the reduction of the excessive coverage and the osteochondroplasty of a potential aspherical femoral head-neck junction to restore an anatomical offset.
Surgical Hip Dislocation Surgical hip dislocation is the gold-standard in treatment of FAI (Fig. 8). This technique allows a complete access to both the acetabular rim and the femoral head and neck. With this technique, the femoral head vascularity is preserved and osteonecrosis avoided. Briefly, after a longitudinal lateral incision in lateral decubitus position, a trochanteric slide osteotomy is performed. The osteotomy must be extracapsular to avoid damage to the blood supply of the femoral
head. The trochanteric fragment is mobilized anteriorly and after a Z-shaped capsulotomy, the femoral head can be dislocated and the joint inspected. After detachment of the acetabular labrum, the excessive acetabular rim can be trimmed with curved chisels. The intact portion of the detached labrum is re-attached afterwards with bone anchors. A potential femoral head asphericity can be quantified with spherical templates. If necessary, an osteochondroplasty is performed to recreate a physiological femoral head-neck offset. This surgical approach also allows to perform additional osteotomies of the femoral head and neck [26] if additional deformities of the proximal femur are present.
Hip Arthroscopy Hip arthroscopy is a valuable alternative to surgical hip dislocation. However, hip arthroscopy should only be performed if the same surgical correction can be achieved compared to the open technique. Arthroscopy is performed in the standard supine or lateral decubitus position, with or without traction, depending on the desired visualization of the central and the peripheral compartments. Hips with pure cam-type deformities in the anterior portion of the femoral head-neck junction are an excellent indication for hip arthroscopy. Arthroscopic treatment of pincer-type deformities is technically demanding because of the limited visualization and access of the acetabular rim. The indications for and techniques of arthroscopic treatment are currently still evolving.
Peri-acetabular Osteotomy The indication for PAO in hips with FAI is excessive acetabular retroversion with a deficient posterior acetabular coverage. In acetabular retroversion, the entire hemipelvis is externally rotated [27]. This produces excessive coverage in the anterior and less-than-normal coverage in the posterior portion of the acetabulum which can be corrected through an anteverting PAO. This ‘reversed’ PAO is performed in the same manner and sequence as described for the treatment of hip dysplasia. The acetabular fragment needs to be flexed and internally rotated to achieve a satisfactory correction. Rather rarely in massive acetabular protrusio, a combined acetabular rim trimming through a surgical hip dislocation and a concomitant PAO to increase the acetabular index might be necessary [28]. The incomplete ischium cut of the PAO can be performed under direct vision of the
152
a
M. Tannast et al.
c
b
d
e
Fig. 8 The antero-posterior pelvic (a) and cross-table lateral radiograph (b) of a young patient with combined Cam-Pincer type of femoro-acetabular impingement is shown. The MR-arthrography shows a labral tear in the antero-superior quadrant of the
a cetabulum (c). The correction was achieved through a surgical hip dislocation with trimming of the acetabular rim (d) and creation of a sufficient femoral head-neck offset (e)
sciatic nerve during surgical hip dislocation. The correct re-orientation involves a lateralization of the entire acetabulum and a rotation towards the midline.
ment of FAI could show an improvement of the Merle d’Aubigné score in 95% of all patients depending on the individual joint alterations at the time of surgery. Good to excellent results were obtained in 91% of all cases. The cumulative 5-year survivorship of the hip was 91%. Despite the very encouraging results, it still has to be proven that the surgical therapy of FAI can delay or even prevent osteoarthritis of the hip.
Results Up to date, only short to mid-term results are available for the surgical treatment of FAI. Very satisfactory results were presented for open surgical hip dislocation [29, 30], hip arthroscopy [31, 32] and PAO [33]. Negative predictors of outcome are advanced degenerative osteoarthritis at the time of surgery as well as over- and undercorrection of both the acetabular and the femoral side. Preliminary results of the study with the longest follow-up with a minimum of 5 years after surgical hip dislocation [34] for treat-
Summary This article gives a short overview on the two main causes of hip pain in the young adult: FAI and DDH. Diagnosis and treatment of acetabular dysplasia have been
Hip Pain in the Young Adult
well-described and long-term results of more than 20 years could prove the superiority of PAO over the natural history of hip dysplasia. The scientific field for diagnosis and treatment of FAI is still young. Mid-term results after open treatment of FAI are encouraging. The future research needs to focus on the development of sophisticated methods to detect and treat early pre-arthritic changes of the articular cartilage.
References 1. Pauwels F (1976) Biomechanics of the normal and diseased hip: theoretical foundation, technique and results of treatment, 2nd edn. Springer, Berlin/New York 2. Ganz R, Parvizi J, Beck M, Leunig M, Notzli H, Siebenrock KA (2003) Femoroacetabular impingement: a cause for osteoarthritis of the hip. Clin Orthop Relat Res 417:112–120 3. Tannast M, Siebenrock KA, Anderson SE (2007) Femoroacetabular impingement: radiographic diagnosis – what the radiologist should know. AJR Am J Roentgenol 6: 1540–1552 4. Lequesne M (2002) The false profile view of the hip: role, interest, economic considerations. Joint Bone Spine 2:109–113 5. Tannast M, Zheng G, Anderegg C, Burckhardt K, Langlotz F, Ganz R, Siebenrock KA (2005) Tilt and rotation correction of acetabular version on pelvic radiographs. Clin Orthop Relat Res 438:182–190 6. Zheng G, Tannast M, Anderegg C, Siebenrock KA, Langlotz F (2007) Hip2Norm: an object-oriented cross-platform program for 3D analysis of hip joint morphology using 2D pelvic radiographs. Comput Methods Programs Biomed 1:36–45 7. Leunig M, Werlen S, Ungersbock A, Ito K, Ganz R (1997) Evaluation of the acetabular labrum by MR arthrography. J Bone Joint Surg Br 2:230–234 8. Hipp JA, Sugano N, Millis MB, Murphy SB (1999) Planning acetabular redirection osteotomies based on joint contact pressures. Clin Orthop Relat Res 364:134–143 9. Wiberg G (1939) The anatomy and roentgenographic appearance of a normal hip joint. Acta Chir Scand 83(58):7–38 10. Leunig M, Podeszwa D, Beck M, Werlen S, Ganz R (2004) Magnetic resonance arthrography of labral disorders in hips with dysplasia and impingement. Clin Orthop Relat Res 418:74–80 11. Li PL, Ganz R (2003) Morphologic features of congenital acetabular dysplasia: one in six is retroverted. Clin Orthop Relat Res 416:245–253 12. Hartofilakidis G, Karachalios T, Stamos KG (2000) Epidemiology, demographics, and natural history of congenital hip disease in adults. Orthopedics 8:823–827 13. Murphy SB, Ganz R, Muller ME (1995) The prognosis in untreated dysplasia of the hip. A study of radiographic factors that predict the outcome. J Bone Joint Surg Am 7:985–989
153 14. Ganz R, Klaue K, Vinh TS, Mast JW (1988) A new periacetabular osteotomy for the treatment of hip dysplasias. Technique and preliminary results. Clin Orthop Relat Res 232:26–36 15. Myers SR, Eijer H, Ganz R (1999) Anterior femoroacetabular impingement after periacetabular osteotomy. Clin Orthop Relat Res 363:93–99 16. Clohisy JC, Barrett SE, Gordon JE, Delgado ED, Schoenecker PL (2005) Periacetabular osteotomy for the treatment of severe acetabular dysplasia. J Bone Joint Surg Am 2:254–259 17. Kralj M, Mavcic B, Antolic V, Iglic A, Kralj-Iglic V (2005) The Bernese periacetabular osteotomy: clinical, radiographic and mechanical 7-15-year follow-up of 26 hips. Acta Orthop 6:833–840 18. Siebenrock KA, Scholl E, Lottenbach M, Ganz R (1999) Bernese periacetabular osteotomy. Clin Orthop Relat Res 363:9–20 19. Steppacher SD, Tannast M, Ganz R, Siebenrock KA (2008) Mean 20-year followup of Bernese periacetabular osteotomy. Clin Orthop Relat Res 7:1633–1644 20. Dagher F, Ghanem I, Abiad R, Haykal G, Kharrat K, Phares A (2003) Bernese periacetabular osteotomy for the treatment of the degenerative dysplasic hip. Rev Chir Orthop Reparatrice Appar Mot 2:125–133 21. Matheney T, Kim YJ, Zurakowski D, Matero C, Millis M (2009) Intermediate to long-term results following the Bernese periacetabular osteotomy and predictors of clinical outcome. J Bone Joint Surg Am 9:2113–2123 22. Sambandam SN, Hull J, Jiranek WA (2009) Factors predicting the failure of Bernese periacetabular osteotomy: a metaregression analysis. Int Orthop 6:1483–1488 23. Troelsen A, Elmengaard B, Soballe K (2009) Medium-term outcome of periacetabular osteotomy and predictors of conversion to total hip replacement. J Bone Joint Surg Am 9:2169–2179 24. Siebenrock KA, Wahab KH, Werlen S, Kalhor M, Leunig M, Ganz R (2004) Abnormal extension of the femoral head epiphysis as a cause of cam impingement. Clin Orthop Relat Res 418:54–60 25. Tannast M, Siebenrock KA (2009) Conventional radiographs to assess femoroacetabular impingement. Instr Course Lect 58:203–212 26. Ganz R, Huff TW, Leunig M (2009) Extended retinacular soft-tissue flap for intra-articular hip surgery: surgical technique, indications, and results of application. Instr Course Lect 58:241–255 27. Jamali AA, Mladenov K, Meyer DC, Martinez A, Beck M, Ganz R, Leunig M (2007) Anteroposterior pelvic radiographs to assess acetabular retroversion: high validity of the “cross-over-sign”. J Orthop Res 6:758–765 28. Leunig M, Huff TW, Ganz R (2009) Femoroacetabular impingement: treatment of the acetabular side. Instr Course Lect 58:223–229 29. Ganz R, Gill TJ, Gautier E, Ganz K, Krugel N, Berlemann U (2001) Surgical dislocation of the adult hip a technique with full access to the femoral head and acetabulum without the risk of avascular necrosis. J Bone Joint Surg Br 8:1119–1124
154 30. Murphy S, Tannast M, Kim YJ, Buly R, Millis MB (2004) Debridement of the adult hip for femoroacetabular impingement: indications and preliminary clinical results. Clin Orthop Relat Res 429:178–181 31. Larson CM, Giveans MR (2008) Arthroscopic management of femoroacetabular impingement: early outcomes measures. Arthroscopy 5:540–546 32. Philippon MJ, Briggs KK, Yen YM, Kuppersmith DA (2009) Outcomes following hip arthroscopy for femoroacetabular
M. Tannast et al. impingement with associated chondrolabral dysfunction: minimum two-year follow-up. J Bone Joint Surg Br 1:16–23 33. Siebenrock KA, Schoeniger R, Ganz R (2003) Anterior femoro-acetabular impingement due to acetabular retroversion. Treatment with periacetabular osteotomy. J Bone Joint Surg Am 2:278–286 34. Tannast M, Siebenrock KA (2010) Open therapy of femoroacetabular impingement. Oper Orthop Traumatol 1:3–16
Bone Loss Around the Acetabular Component Jonathan Howell and Ben Bolland
Introduction Successful reconstruction of the diseased hip requires restoration of normal anatomical parameters and joint biomechanics. This requires that the following conditions are met: (1) the centre of rotation of the hip should be restored to its correct position; (2) the acetabular component should be firmly fixed to the host skeleton; and (3) where it is deficient, bone stock should be restored, and where there is discontinuity, the structural integrity of the pelvis must be re-established. The hip surgeon will encounter a wide spectrum of bone loss around the failed acetabular component and it is highly unlikely that any one surgical technique can be used to treat all scenarios with equal success. The surgeon should therefore be familiar and confident with a range of techniques, so that treatment can be tailored to the individual patient and their pattern of bone loss. In this chapter we will review some of the techniques available for the treatment of bone loss around the acetabular component and we will review how each of these methods has been be used to treat specific patterns of bone loss.
Classification of Bone Loss The aims of a classification system may be several-fold, and may include the following: 1. As an aid to communication that will allow description of a problem (in this case bone loss) to others. 2. As a tool for pre-operative planning and prediction of the likely requirements for equipment and prostheses. J. Howell () Princess Elizabeth Orthopaedic Centre, Barrack Road, EX 2 5DW Exeter, United Kingdom e-mail:
[email protected]
3. As an intra-operative guide to the most suitable method of reconstruction for a particular situation. 4. As a research tool for the description and reporting of defects and reconstruction methods, to allow comparison of the results of different techniques applied to the same problem. There is no one classification system that is perfect in all respects and that fulfills all the possible requirements set out above. The most commonly used classification systems for bone loss around the acetabular component are the AAOS system devised by D’Antonio et al. [1] and the Paprosky classification [2]. The AAOS classification system divides bone defects into segmental and cavitary patterns, as well as combinations of these two, pelvic discontinuity and arthrodesis. This classification system is useful as an intra-operative descriptor of the pattern of bone loss but may be less useful as a pre-operative planning guide. The Paprosky classification system describes three broad categories of bone loss based on four indicators of acetabular bone quality. This classification system was initially devised as a guide to the Orthopaedic surgeon undertaking revision of a cemented acetabular component, to help predict whether or not there was sufficient supportive bone to allow reconstruction with an uncemented acetabular shell. At Exeter we have adapted both of these classification systems to devise a system that describes the location and degree of bone loss that we use as a guide to treatment. We divide the acetabulum into four compartments (three peripheral and one central, shown in Fig. 1) and defects are then divided into five main categories. The classification system is shown in Table 1, with a pictorial description of the patterns of bone loss shown in Fig. 2. A Type 0 defect is one in which there is no significant bone loss. A Type I defect involves cavitary bone loss of any size, which is completely contained by intact acetabular walls. A Type II defect involves minor segmental bone loss, with or without an associated cavitary defect. This type of defect can be further sub-categorised according to the location of the segmental
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_12, © 2011 EFORT
155
156
J. Howell and B. Bolland Superior rim segment
Posterior rim segment
Anterior rim segment
process and symptoms do not generally appear until the cement mantle becomes completely unstable. In contrast, uncemented acetabular components are associated with higher polyethylene wear rates [3, 4] and, in many cases, the emergence of large lytic lesions behind otherwise well fixed components (Fig. 3). Symptoms generally arise when the acetabular component becomes loose or as a result of periprosthetic fracture, but revision may also be indicated to deal with large lytic lesions in patients who are largely asymptomatic.
Local Treatment Methods for Osteolysis Medial segment
Fig. 1 The descriptive divisions of the acetabulum. Red lines show the division of the anterior, superior and posterior rim segments and green lines show division between the central and peripheral segments (Reproduced with permission of Exeter Hip Foundation)
loss. More extensive degrees of bone loss involving two segmental regions are classified as type III defects, whereas the most complex patterns of bone loss, involving three regions of segmental loss, are classified as type IV defects and again these may be further sub-categorised by the location of the segmental loss. A type V defect is one in which there is discontinuity of the anterior and posterior columns and a loss of the structural integrity of the pelvis.
Management Strategies for Dealing with Acetabular Bone Loss Although there are some similarities in the defects created by failing cemented and uncemented acetabular components, there are certainly patterns of osetolysis and modes of component loosening that are specific to each of these two methods of acetabular fixation and thus alternative treatment strategies are required to deal with them. The different patterns of osteolysis seen with cemented and uncemented acetabular components may be a result of differences between these two with respect to the access they provide to the host bone surface for fluid and particles. With a cemented acetabular component the host bone surface is essentially sealed by the presence of the cement mantle. This means that it is rare to see the development of large silent lytic lesions behind such components, and it is more common for these to fail through progressive loosening at the cement-bone interface. Such loosening is usually a slow
The presence of a large osteolytic lesion behind a well-fixed uncemented shell presents the surgeon with a particular dilemma. Removal of the acetabular component in such a case may risk further injury to the surrounding bone, particularly the medial and posterior walls, and may leave the surgeon with a complex reconstruction problem. For this reason, methods of treatment have been advocated that involve retention of the acetabular component, localised debridement of the granulomatous tissue and bone grafting of the lytic cavity. At the same time the polyethylene liner is usually exchanged or replaced in an attempt to switch off the driver of the osteolysis (Fig. 4). However, such local treatment methods are often limited by the access of the surgeon to the lytic cavity and a thorough debridement may not be possible. In addition, damage or wear of the locking mechanism may make simple liner exchange impossible and as a result some surgeons have advocated cementing a new socket into the existing metal shell. The development of new techniques that allow removal of a well-fixed uncemented shell with very little associated bone loss [5] has removed one of the major obstacles to the revision of such components. As a result, localised treatment of lytic lesions may be undertaken less frequently in the future, as surgeons prefer to undertake a more comprehensive revision procedure.
Revision with an Uncemented Shell In some parts of the world, particularly North America, use of an uncemented shell supplemented with screws has become the standard method of acetabular revision. Longterm success with this method of reconstruction depends upon sufficient initial stability of the shell combined with adequate contact with host bone for bone in-growth to occur. Contact of at least 50% of the shell surface with host bone has been advocated as a minimum requirement for the long-term success of this treatment method [6].
157
Bone Loss Around the Acetabular Component Table 1 Exeter classification of acetabular bone defects Type 0: no defect Paprosky Primary-type socket • Focal bone loss only, restricted to previous screw holes or cement holes
AAOS
Reconstructive options
1
–
• Cemented cup • Uncemented cup
2A
Cavitary
• Impaction grafting required with cemented cup, no mesh • Uncemented cup, with reverse-reamed morcellised graft, +/− screws. Rim fit achievable
Type 1: cavitary Contained defect, regardless of size • Rim intact • Floor intact • Columns intact Type 2: minor segmental Cavitary defect + one segmental defect • General socket enlargement • Both columns intact • Sub-classified by the location of the defect
Segmental
2M: Medial wall defect
2C
Central segmental
• Impaction graft requiring medial wall mesh (interior) • Uncemented cup; +/− screws, +/− morcellised graft
2A: Anterior wall defect
–
Anterior peripheral segmental
• Impaction graft requiring anterior wall mesh (interior) • Uncemented cup +/− screws, +/− morcellised graft
2S: Supero-lateral rim defect
2B
Superior peripheral segmental
• Impaction graft requiring superior rim mesh (exterior) • Uncemented cup +/− screws, +/− morcellised graft +/− metal augmentation device
2P: Posterior wall defect
–
Posterior peripheral segmental
• Impaction graft requiring posterior wall mesh (exterior) • Uncemented cup +/− screws, +/− morcellised graft +/− metal augmentation device
Type 3: Major segmental
Paprosky
AAOS
Reconstructive options
Cavitary defect + two segmental defects • Significant socket enlargement • Al least one column intact • Sub-classified by location of defects e.g., 3SM
3A
Combined
• Impaction graft with rim mesh and/or floor mesh • Uncemented cup with screws and morcellised graft +/− metal augmentation device • Structural allograft + cemented/uncemented cup
3B
Central segmental + • Impaction graft with superior, anterior, and medial wall mesh three peripheral • Structural allograft or metal augmentation segmental defects device + uncemented cup • Anti-protrusio cage, morcellised graft and cemented cup
–
Discontinuity
Type 4: complex segmental Cavitary defect + three segmental defects • Significant socket enlargement • One column intact • Sub-classified according to segmental defects e.g., 3ASM Type 5: discontinuity • Both columns discontinuous
Source: Reproduced with permission of Exeter Hip Foundation
• Anti-prostrusio cage, morcellised graft, and cemented cup • ORIF + impaction grafting • Porous metal uncemented shell +/− metal augmentation device +/− ORIF
158
J. Howell and B. Bolland
Type 0
Type 1
Type 2a
Type 2s
Type 2p
Type 2m
Type 3
Type 4
Type 5
Fig. 2 The classification system used by the authors for decision-making in acetabular revision (Reproduced with permission of Exeter Hip Foundation)
159
Bone Loss Around the Acetabular Component
In simple cases in which there is relatively little bone loss it is straightforward for the surgeon to achieve adequate contact between an uncemented shell and host bone. However, in cases with more extensive bone loss, contact with host bone is more limited, particularly where
Fig. 3 Large lytic cavity in the ileum lying adjacent to an uncemented cup (Reproduced with permission of Exeter Hip Foundation)
a
segmental loss has created an aspherical cavity that does not conform to a hemispherical shell. In such cases alternative strategies have been proposed that include the use of extra-large shells, defined as prostheses of diameter greater than 62 mm in women and 66 mm in men [7]. The technique involves wedging the jumbo shell between the ilium superiorly and the ischium and pubis inferiorly, with the posterior column and wall preventing migration of the shell. It necessitates the creation of a hemispherical cavity from an aspherical defect and therefore the surgeon is required to remove additional bone and, as a general rule, the anterior acetabular bone is sacrificed to protect the posterior column. While there have been reports of good results in the medium term with this method of treatment [8–12], it does nothing to address the loss of bone stock and indeed its success requires the removal of further bone. This means that, should revision be required again in the future, there will be even less bone available at the next operation. As a way of protecting bone stock and enhancing prosthesis stability during revision with an uncemented acetabular component, some authors have advocated use of a block allograft to fill segmental defects [13–15]. Such defects are most commonly seen in the superior part of the acetabulum and when block allografts are used the method of reaming differs from that of a jumbo cup technique. In this scenario the surgeon progressively reams at the correct centre of rotation until there is contact of the reamer with the anterior and posterior walls (Fig. 5a). At this point the superior block allograft is inserted and secured (Fig. 5b), before it is then reamed in preparation for insertion of the
b
Fig. 4 Grafting of a lytic cyst behind a well-fixed uncemented acetabular shell: (a) Pre-operative X-ray showing lytic lesion superior and medial to the shell; (b) X-ray taken after localised grafting of the cyst (Reproduced with permission of Exeter Hip Foundation)
160
J. Howell and B. Bolland
a
b
c
d
Fig. 5 The use of block allograft to augment a defect above an uncemented revision shell. (a) The acetabulum is reamed at the correct centre of rotation until contact is made with the anterior and posterior walls. (b) The block allograft is secured initially with K-wires, which are later replaced with screws. (c) The
block allograft is reamed to fit the intended size of acetabular component. (d) The acetabular component lies at the correct centre of rotation, supported superiorly by the block allograft (Reproduced with permission of Exeter Hip Foundation)
hemispherical component (Figs. 5c, d). Although block allografts may help to provide initial stability for the acetabular component, they do not offer any prospect for in-growth and therefore the success of the construct will still depend upon there being adequate contact between the host bone and the shell. The use of bilobed components has been suggested as one method of achieving adequate contact between the host and implant in extensive acetabular defects [16 –18]. The superior part of these components is designed to fill the segmental defect and gain contact with the host bone, whilst the inferior part of the shell lies at the correct centre of rotation. These prostheses may be difficult to fit to the complex geometry of the bone loss surrounding a failed acetabular component and they fill the defect with metal rather than
replenishing bone stock. There are relatively few reports of their use in the literature, with a small number of patients studied in total, but they offer an option for the management of large bone defects. An alternative strategy for achieving adequate contact between host bone and the acetabular shell is the use of a high centre [19]. This method of treatment contravenes one of the pre-requisites for successful long-term function, namely that the acetabular component should be placed at the correct centre of rotation. It risks de- functioning the abductor muscles and, as a result, increasing the joint reaction force, as well as potentially creating a short limb and gait abnormalities. However, it is technically easy to achieve and it offers a quick method of reconstruction for the elderly, infirm patient with low demands, for whom the long-term result may be less important.
161
Bone Loss Around the Acetabular Component
Impaction Grafting and Cemented Revision The use of large amounts of cement in isolation for the treatment of bone stock loss around acetabular components has been reported, but in general the results have been poor [20 –22]. Osteolysis and the loosening of existing implants create cavities in the bone that are characterised by a hard sclerotic surface that provides a poor mechanical environment for the use of cement on its own. However, the technique of acetabular impaction grafting transforms the local environment and provides an ideal surface into which cement may be pressurised, thus achieving reliable fixation for the new prosthesis. In addition, incorporation of the bone graft in the long-term addresses the bone stock loss that is seen with a failed acetabular component.
impactors are used to compress the graft through repeated impaction (Fig. 6e), which prepares a cavity suitable for cementing. As progressively more graft is added and then impacted, so the bone surface becomes firmer until it begins to resemble cortical bone. When the correct centre of rotation has been established and the graft bed feels solid, the impaction process is completed by peripheral packing with large cancellous chips that compress the graft bed even further. The graft bed is then washed and dried and bone cement is prepared and then pressurized into the socket cavity. A flanged acetabular component is inserted and pressure maintained on the prosthesis until the cement is set.
Results Surgical Technique Essential to the technique is the conversion of segmental defects to a contained cavitary defect and the constraint of bone graft through the use of metal meshes, secured firmly to the host bone with screw fixation (Fig. 6a). When a mesh is to be used the surgeon should align it with the intended orientation of the acetabular component (Fig. 6b) and then fix it in place, with the first screw placed in the apex of the mesh. Screws are then placed at the anterior and posterior margins of the mesh and once this initial fixation has been achieved, further screws are inserted at intervals of 1 cm around the periphery (Fig. 6c). It is essential that the surgeon is satisfied with the stability of the mesh before proceeding any further and if a stable mesh cannot be achieved then an alternative method of reconstruction should be used. The host bed should be prepared for receipt of bone graft, and particular attention should be paid to establishing a bleeding host bone surface. This ensures that revascularisation of the graft can take place followed by remodelling and incorporation of the graft in the long-term. In the revision setting, allograft bone is most frequently used and the ideal substrate is fresh frozen allograft harvested from femoral heads removed from patients undergoing total hip replacement. Prior to use, the allograft should be defrosted and all cartilage and cortical bone should be removed. For the best results large cancellous bone chips are made by hand, using large rongeurs (Fig. 6d). In general, commercially available bone mills are to be avoided, as they tend to produce bone slurry composed of small bone chips. The surgeon sequentially adds layers of allograft bone chips and vigorously impacts each layer before adding the next. Hemispherical
Acetabular impaction grafting with cement was first established as a reliable revision technique by the Nijmegen group and they have reported their results in 62 cases, reported at 20–25 years’ follow-up [23]. In this series, 38% of the defects were cavitary and 62% were combined defects. The overall survivorship with end-point revision for aseptic cup loosening was 87% at 20 years. There are now several reports in the literature of successful outcomes from other centres, indicating that the technique is reproducible and reliable. Comba et al. [24] reported the results of 131 patients who underwent revision with acetabular impaction grafting. The series included only 11 cases with severe segmental defects that required reconstruction with a mesh or block allograft. At a mean follow-up of 51 months the survivorship for mechanical loosening was 98%. A series with more extensive bone loss was reported by Garcia-Cimbrelllo [25]. The survivorship of 70 acetabular revisions with Paprosky grade IIIA and IIIB defects at 5–9 years’ follow-up was 98% in this series, with only one cup revised for aseptic loosening. We have reviewed a consecutive series of 339 patients who underwent acetabular impaction grafting with cement at a mean follow-up of 6.1 years (range 4.3–8.4 years) [26]. The patients were classified according to the Paprosky system and the series includes a large number of patients with extensive bone loss (Table 2). At latest follow-up 15 cases (4.4%) had been revised for aseptic loosening of the acetabular component, giving an overall survivorship for aseptic loosening at 6 years of 91.6%. We found that results were worse in cases with extensive bone loss, in which a large rim mesh had been required to close the segmental defect, and these results are borne out by the findings of other authors [27, 28].
162
J. Howell and B. Bolland
a
b
c
d
e
Fig. 6 Acetabular impaction grafting. (a) A trial cup has been placed at the correct centre of rotation, demonstrating the superolateral defect that must be closed prior to graft impaction. (b) A mesh is placed over the defect, aligned with the intended position of the acetabular component, which is judged by holding an impactor in the correct orientation. (c) The mesh is securely fixed
with screws placed at approximately 1 cm intervals around the periphery. (d) Large cancellous bone chips are made by hand. (e) Impaction of the allograft chips with the hemispherical packer, which is hammered into place (Reproduced with permission of Exeter Hip Foundation)
163
Bone Loss Around the Acetabular Component Table 2 Paprosky classification of the cases included in our series of acetabular impaction grafting Grade 1
10
Grade 2A Grade 2B Grade 2C
71 90 44
Grade 3A Grade 3B Pelvic discontinuity Protrusio Dysplasia
55 48 3 13 5
Allograft and Support Rings and Cages In the most challenging cases of extreme acetabular bone loss there is often insufficient host bone to allow successful deployment of the reconstruction methods outlined above, either through lack of host contact with the revision prosthesis or an inability to adequately contain bone graft. These cases include massive combined defects as well as cases with pelvic discontinuity and, in an attempt to restore stability, a support ring or cage may be required, often in conjunction with morcellised or bulk allograft. In our series of 339 acetabular impaction grafting cases [26], a Kerboull-Postell support ring was used on 53 occasions, for many of the more severe defects. In this sub-group, ten of the devices fractured, two were revised and two showed a pattern of continuing painless migration. In the six others the migration of the ring halted and the surrounding graft looked satisfactory and had probably incorporated as it became loaded after the ring failure. Kawanabe et al. [29] reported their results with the same device using morcellised and structural allograft. At 10 years the survivorship for clinical or radiological failure for the morcellised group was only 53%, compared to 82% for the structural group. The results suggest that morcellised graft alone is not sufficient support for this device, which should lie in contact either with host bone, or with a bulk allograft. Several authors have tried alternative, stronger and more rigid acetabular cages to support the morcellised allograft in cases with extensive bone loss. On the one hand these devices are stronger and so may not suffer from the mechanical failure of thinner support rings, but their rigid nature may in turn offload the allograft to such an extent that they cause resorption of the graft. Berry and Muller [30] reported on the results of the Burch-Schneider cage with morcellised allograft in 42 hips at a mean follow-up of 5 years. There were five cases (12%) of aseptic loosening in this series.
In our series of acetabular impaction grafting cases the same device was used in 12 patients with severe bone loss and two of these migrated leading to re-operation. Carroll et al. [31] reviewed 63 cases treated with a variety of support rings in combination with impacted graft; all cases were Paprosky grade III and included four with pelvic dissociation. They reported relatively good results in this difficult-to-treat group, with 84% survival for aseptic loosening at 8.75 years. Similarly, Haddad et al. [32] reported good results using a variety of support rings and morcellised graft, with no aseptic failures and good graft incorporation at mean follow-up of 64 months. Good long-term results of the Birch-Schneider cage with morcellised graft was reported by Wachtl et al. [33] in a series of 38 cases with mean follow-up of 12 years. Only three cages required revision (two for dislocation and one for infection) giving a cumulative survivorship at 21 years with revision of the cage for all causes as 92%. The long-term results of bulk, structural allografts and a cage used to reconstruct massive defects have recently been reported by Regis et al. [34]. In a series of 71 cases treated with this technique, 56 were available for follow-up at mean 11.7 years (12 had died and three were lost to follow-up.) Aseptic loosening of the cage had occurred in five cases and two of these had required revision. Overall the survivorship was 87.5% at 11.7 years in this group of patients with severe acetabular bone loss. The authors stressed the importance of gaining initial stability of the cage in achieving long-term success and advocated supporting the cage on residual host bone and then augmenting the fixation with multiple screws.
Porous Metal Shells and Augmentation Devices A porous material made from tantalum was first developed over 10 years ago [35], since when such materials have been used in the manufacture of a range of shells and augmentation devices that may be used for the treatment of acetabular bone stock loss. They have several potential advantages that include: 1. A high surface coefficient of friction giving enhanced initial stability and scratch-fit with host bone. 2. Improved structural reliability imparted by their resistance to fracture and failure (which may occur with structural allograft over time as a result of revascularisation and remodelling). 3. A favourable environment for the remodelling of morcellised or structural bone graft along with less stress
164
shielding due to their reduced stiffness and a similar Young’s modulus to bone. 4. The open structure, regular pore size and biocompatibility of the material leads to rapid bony in-growth demonstrated in animal models. The manufacturing processes used in the production of these materials allows the creation of a range of shapes and types of prostheses, including hemispherical shells, partial hemispherical augments and prostheses designed to replace large segments of the acetabular walls or columns (Fig. 7). To date the results of these devices have been encouraging. Four main studies have described results for severe acetabular defects with approximately 3 years’ followup [36–39]. Concentrating on IIIA and IIIB defects, 130 cases were treated with porous metal components (Shell
J. Howell and B. Bolland
+/− augment +/− cage) with only two aseptic failures. Common conclusions and main advantages of these prostheses were the immediate structural stability of the construct avoiding the need for cages; an increased surface area contact with host bone provided by the augment resulted in the construct acting as a modular component and a resultant restoration of the hip centre. The encouraging early results of these porous metal implants have led some authors to extend their use to the most challenging acetabular reconstructions. Sporer and Paprosky [40] have reported their early experience with porous metal shells in cases with pelvic discontinuity, in which they restored the acetabular rim where necessary with augmentation devices and then impacted a large shell into the acetabulum, distracting the defect and wedging the shell between remaining host bone and metal augments. Thirteen
a
b
c
d
Fig. 7 Porous metal augment and shell used to revise a combined acetabular defect. (a) Insertion of the trial acetabular component demonstrates the superior acetabular defect. (b) A trial augment device has been inserted to fill the defect above the trial acetabular
component. (c) The augment device is secured to the pelvis with screws. (d) The acetabular porous metal shell is impacted underneath the augment device (Reproduced with permission of Exeter Hip Foundation)
165
Bone Loss Around the Acetabular Component
patients with a grade IIIB defect and pelvic discontinuity were treated in this fashion and reviewed at a mean 2.6 years’ follow-up, at which time one patient had radiographic loosening of the acetabulum but the remaining 12 patients had radiographically stable hips. An alternative method of reconstruction has been described by Hanssen and Lewallen [41] with the so-called “cup-cage” technique. Using this method of reconstruction, morcellised or structural bone graft is impacted into the acetabular defects, onto which a porous metal shell is fixed with multiple screws. To protect the shell prior to bone graft incorporation, a reconstruction cage is inserted into the cup and fixed superiorly with screws and inferiorly with the ischial flange fixed through a slot in the ischium (avoiding the increased rate of screw fracture and sciatic nerve palsy reported with fixing cages to the surface of the ischium [42]). Kosashvili et al. [43] have reported early results in 26 cases of pelvic discontinuity in which this method was utilized. All 26 acetabular components had less than 50% contact with host bone and the mean contact was estimated at 15.8%. At a mean follow-up of 44 months there had been three failures (11.5%), with two undergoing a further revision and the third patient listed for revision. In 23 cases (85%) there had been no change in the acetabular component position at the latest follow-up. Clearly, the number of cases so far reported with these techniques is small and the length of follow-up is relatively short. Further long-term studies will be required to prove that these techniques are effective in the treatment of these complex cases but they provide an exciting avenue of future research.
Summary
›› Acetabular bone stock loss around loose or failed
››
components remains one of the most challenging problems faced by the revision hip surgeon. The aims of treatment should be to restore the centre of rotation to the correct position and to replace lost bone stock whenever possible. In this chapter we have reviewed some of the commonly used techniques for dealing with acetabular bone stock loss, as well as some of the evidence of their results. No one reconstruction technique is applicable to all cases and the revision surgeon must be confident with a range of techniques to match the wide spectrum of defects that are encountered in clinical practice.
References 1. D’Antonio JA et al (1989) Classification and management of acetabular abnormalities in total hip arthroplasty. Clin Orthop Relat Res 243:126–137 2. Paprosky WG, Perona PG, Lawrence JM (1994) Acetabular defect classification and surgical reconstruction in revision arthroplasty. A 6-year follow-up evaluation. J Arthroplasty 9(1):33–44 3. Devane PA et al (1997) Measurement of polyethylene wear in acetabular components inserted with and without cement. A randomized trial. J Bone Joint Surg Am 79(5):682–689 4. McCombe P, Williams SA (2004) A comparison of polyethylene wear rates between cemented and cementless cups. A prospective, randomised trial. J Bone Joint Surg Br 86(3):344–349 5. Mitchell PA et al (2003) Removal of well-fixed, cementless, acetabular components in revision hip arthroplasty. J Bone Joint Surg Br 85(7):949–952 6. Rosenberg AG (2003) Cementless acetabular components: the gold standard for socket revision. J Arthroplasty 18 (3 Suppl 1):118–120 7. Whaley AL, Berry DJ, Harmsen WS (2001) Extra-large uncemented hemispherical acetabular components for revision total hip arthroplasty. J Bone Joint Surg Am 83A(9):1352–1357 8. Patel JV et al (2003) The fate of cementless jumbo cups in revision hip arthroplasty. J Arthroplasty 18(2):129–133 9. Jasty M (1998) Jumbo cups and morsalized graft. Orthop Clin North Am 29(2):249–254 10. Hendricks KJ, Harris WH (2006) Revision of failed acetabular components with use of so-called jumbo noncemented components. A concise follow-up of a previous report. J Bone Joint Surg Am 88(3):559–563 11. Dearborn JT, Harris WH (2000) Acetabular revision arthroplasty using so-called jumbo cementless components: an average 7-year follow-up study. J Arthroplasty 15(1):8–15 12. Obenaus C et al (2003) Extra-large press-fit cups without screws for acetabular revision. J Arthroplasty 18(3):271–277 13. Blackley HR et al (2001) Proximal femoral allografts for reconstruction of bone stock in revision arthroplasty of the hip. A nine to fifteen-year follow-up. J Bone Joint Surg Am 83A(3):346–354 14. Dewal H et al (2003) Use of structural bone graft with cementless acetabular cups in total hip arthroplasty. J Arthroplasty 18(1):23–28 15. Sporer SM et al (2005) The use of structural distal femoral allografts for acetabular reconstruction. Average ten-year follow-up. J Bone Joint Surg Am 87(4):760–765 16. Chen WM et al (2000) Acetabular revision with use of a bilobed component inserted without cement in patients who have acetabular bone-stock deficiency. J Bone Joint Surg Am 82(2):197–206 17. Berry DJ et al (2000) Bilobed oblong porous coated acetabular components in revision total hip arthroplasty. Clin Orthop Relat Res 371:154–160 18. Moskal JT, Higgins ME, Shen J (2008) Type III acetabular defect revision with bilobed components: five-year results. Clin Orthop Relat Res 466(3):691–695
166 19. Dearborn JT, Harris WH (1999) High placement of an acetabular component inserted without cement in a revision total hip arthroplasty. Results after a mean of ten years. J Bone Joint Surg Am 81(4):469–480 20. Amstutz HC et al (1982) Revision of aseptic loose total hip arthroplasties. Clin Orthop Relat Res 170:21–33 21. Kavanagh BF, Ilstrup DM, Fitzgerald RH Jr (1985) Revision total hip arthroplasty. J Bone Joint Surg Am 67(4):517–526 22. Callaghan JJ et al (1985) Results of revision for mechanical failure after cemented total hip replacement, 1979 to 1982. A two to five-year follow-up. J Bone Joint Surg Am 67(7):1074–1085 23. Schreurs BW et al (2009) Acetabular revision with impacted morsellised cancellous bone grafting and a cemented acetabular component: a 20- to 25-year follow-up. J Bone Joint Surg Br 91(9):1148–1153 24. Comba F et al (2006) Acetabular reconstruction with impacted bone allografts and cemented acetabular components: a 2- to 13-year follow-up study of 142 aseptic revisions. J Bone Joint Surg Br 88(7):865–869 25. Garcia-Cimbrelo E, Cordero J (2002) Impacted morcellised allograft and cemented cup in acetabular revision surgery: a five to nine year follow-up study. Hip Int 12(3):281–288 26. Rigby M, Whitehouse S, Timperley AJ (2010) Clinical results: acetabular impaction grafting. In: Ling RSM, Gie GA, Timperley AJ, Hubble MJW, Howell JR, Whitehouse SL (eds) The Exeter hip: 40 years of innovation in total hip arthroplasty. Exeter Hip Publishing, Exeter, pp 339–344 27. van Haaren EH et al (2007) High rate of failure of impaction grafting in large acetabular defects. J Bone Joint Surg Br 89(3):296–300 28. Buttaro MA et al (2008) Acetabular revision with metal mesh, impaction bone grafting, and a cemented cup. Clin Orthop Relat Res 466(10):2482–2490 29. Kawanabe K et al (2007) Revision total hip replacement using the Kerboull acetabular reinforcement device with morsellised or bulk graft: results at a mean follow-up of 8.7 years. J Bone Joint Surg Br 89(1):26–31 30. Berry DJ, Muller ME (1992) Revision arthroplasty using an anti-protrusio cage for massive acetabular bone deficiency. J Bone Joint Surg Br 74(5):711–715
J. Howell and B. Bolland 31. Carroll FA et al (2008) The survival of support rings in complex acetabular revision surgery. J Bone Joint Surg Br 90(5):574–578 32. Haddad FS, Shergill N, Muirhead-Allwood SK (1999) Acetabular reconstruction with morcellized allograft and ring support: a medium-term review. J Arthroplasty 14(7):788–795 33. Wachtl SW et al (2000) The Burch-Schneider antiprotrusio cage in acetabular revision surgery: a mean follow-up of 12 years. J Arthroplasty 15(8):959–963 34. Regis D et al (2008) Long-term results of anti-protrusion cage and massive allografts for the management of periprosthetic acetabular bone loss. J Arthroplasty 23(6):826–832 35. Black J (1994) Biological performance of tantalum. Clin Mater 16(3):167–173 36. Weeden SH, Schmidt RH (2007) The use of tantalum porous metal implants for Paprosky 3A and 3B defects. J Arthroplasty 22(6 Suppl 2):151–155 37. Flecher X, Sporer S, Paprosky W (2008) Management of severe bone loss in acetabular revision using a trabecular metal shell. J Arthroplasty 23(7):949–955 38. Van Kleunen JP et al (2009) Acetabular revisions using trabecular metal cups and augments. J Arthroplasty 24 z(6 Suppl):64–68 39. Siegmeth A et al (2009) Modular tantalum augments for acetabular defects in revision hip arthroplasty. Clin Orthop Relat Res 467(1):199–205 40. Sporer SM, Paprosky WG (2006) Acetabular revision using a trabecular metal acetabular component for severe acetabular bone loss associated with a pelvic discontinuity. J Arthroplasty 21(6 Suppl 2):87–90 41. Hanssen AD, Lewallen DG (2005) Modular acetabular augments: composite void fillers. Orthopedics 28(9):971–972 42. Goodman S et al (2004) Complications of ilioischial reconstruction rings in revision total hip arthroplasty. J Arthroplasty 19(4):436–446 43. Kosashvili Y et al (2009) Acetabular revision using an antiprotrusion (ilio-ischial) cage and trabecular metal acetabular component for severe acetabular bone loss associated with pelvic discontinuity. J Bone Joint Surg Br 91(7): 870–876
Part IX Knee
The Uni-Knee: Indications, and Recent Techniques Sébastien Lustig, Gérard Deschamps, M. Alsaati, C. Fary, and Phillippe Neyret
Introduction UKA’s were first implanted in the late 1960s by Marmor [1] and later in France by Cartier et al. [2]. Multiple studies have been published concerning outcomes of the early UKAs [3, 4]. Compared to total knee replacements, UKAs were shown to have worse outcomes with less reproducible results. We believe that this occurred primarily because the wrong indications were used. Failure rates were due, not just to errors in patient selection, but also problems with design and surgical technique. Recently there has been renewed interest in UKAs. and the results from the Swedish Registry [5] have led to an improved understanding of the causes for failure. This has led to interest in minimally-invasive techniques with lower morbidity and more rapid rehabilitation, which has resulted in improved functional results (“the forgotten knee” ) [6, 7]. Firstly, we explain the basic concepts of the UKA. Secondly, we describe the current indications and techniques.
Concept of the UKA The UKA is implanted where loss of femoro-tibial cartilage from OA has occurred. It acts as a “wedge” joint replacing the lost articular cartilage. The implant compensating for the lost cartilage is constrained in size by the ligamentous envelope that stabilizes the knee. Due to the fixed nature of this envelope the prosthesis cannot
c ompensate for wear or ligamentous instability elsewhere. As a result the appropriately corrected axis by the “wedge” unicompartmental prosthesis is limited to the amount of cartilage loss and a return to the prior pre-pathological axis of the patient. UKA should not change the constitutional tibio-femoral axis of the patient. The residual varus or valgus may be measured on the contra-lateral knee if normal. A UKA is considered perfectly matched if the constitutional alignment of the patient matches. The correct axis is by definition limited to the compensation for wear. The stability of the UKA relies solely only on the ligaments which must be intact.
Indications Indications for the UKA have been well defined since the late 1980s particularly by Kozinn and Scott [8]. Much of the success of a UKA prosthesis depends on the indication. Various aspects are considered, both clinical and radiological, which when combined allow the “ideal” patient to be defined for a UKA.
The Stage of Osteoarthritis (Fig. 1) The first criterion is the location of joint involvement: osteonecrosis of the medial condyle or isolated femorotibial unicompartmental OA (complete or incomplete cartilage loss) but with moderate bone loss (less than 5 mm).
Location and Character of Pain P. Neyret () Pr Neyret’s Orthopaedic Department, Centre A Trillat, University Hospital, 103 Grande Rue de la Croix – Rousse, 69004 Lyon, France e-mail:
[email protected]
It must be localized to the femoro-tibial compartment, be mechanical in nature and relieved by rest. The prime indication for UKA is unicompartmental femoro-tibial OA. Inflammatory arthropathies are a strict contra-indication, even if the cartilage loss appears unicompartmental.
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_13, © 2011 EFORT
169
170
S. Lustig et al.
Fig. 1 Isolated medial femoro-tibial unicompartmental osteoarthritis
An exception may be in the situation of unicompartmental OA associated with mild patellofemoral OA in the elderly where the symptoms are primarily unicompartmental femoro-tibial. An important symptom and sign:
• When asked about the location of pain the patient points to the relevant compartment; and
• Digital pressure in the femoro-tibial space by the exam-
iner reproduces pain and the patient recognizes it as their “usual pain”.
Fig. 2 Good reducibility with varus and valgus stress X-rays
Femoro-tibial Axis Ultimately the aim is to undercorrect the femoro-tibial axis but this has limitations. The prosthesis is a unicondylar wedge that replaces the cartilage and bone wear; it cannot be used to correct a varus or valgus morphotype. The limit of residual varus or valgus after a UKA should not exceed 5° to avoid overloading the compartment prostheses [9, 10]. It is important to take into account the reducibility judged on radiological reduction (varus or valgus stress) (Fig. 2) and the pre-operative deformity in the frontal plane.
171
The Uni-Knee: Indications, and Recent Techniques
The limit is set at a femoro-tibial angle of 170° for medial UKA (varus overall less than 10°) and 194° for lateral UKA (valgus lower overall than 14°). Either way the postoperative axis is often reduced less than 5° and so is undercorrected unless there is extensive soft tissue release or an excessive tensioning of the capsule. Residual physiological laxity of the compartment and therefore the prosthesis are necessary for proper function [9].
The Anterior Cruciate Ligament The integrity of the anterior cruciate ligament is critical. It can be evaluated clinically and radiologically (anterior tibial translation >10 mm; posterior cupula or hooked spines can be considered as a contra-indication [11, 12]).
Pre-operative Mobility It must be normal or nearly normal, with less than 10° flexion contracture and flexion greater than 100°. Limited flexion should be investigated for patellofemoral pain or OA, which moves the indication towards a total knee prosthesis. The presence of a flexion contracture is a relative contraindication if less than 10°. If there is no obvious mechanical cause the patient must be re-assessed until the pathology is understood.
Weight The contact surface of the tibial component is limited and so there should be caution with overweight patients. A moderate BMI is an important concern for us, especially in medial OA. The tolerance of the medial UKA when the patient is overweight is less than lateral UKA as undercorrection and load combine to increase stress on the implant and the medial tibial plateau. Some authors’ indications are up to 125 kg [13] which seems excessive to us; however successful outcomes in obese patients have been reported [14]. The weight of the patient is considered in isolation and is not an absolute contra-indication, but we believe it is reasonable to avoid a UKA in patients weighing over 80 kg.
Age and Activity Level Swienckowski et al. [15] believe that age should not be an absolute limiting factor, and in certain indications (e.g.,
post-traumatic OA) UKA may be indicated for patients under 60 years. Nevertheless, most authors recommend reserving UKA for patients aged over 60, respecting that a joint replacement in general should be delayed in the young until symptoms are significant. But the debate continues between the two views:
• Advocates for a UKA in patients greater than 75 years
old believe that, as they are more sedentary and that revision is required after 10–15 years, the implant will last the life of the patient [16]. • Others who consider the UKA as a temporary solution before TKA will use it for younger patients (60–70 years). They accept that in the future conversion to TKA will be required [17]. Ultimately the level of activity appears to be more important than age. The patient should have a sedentary lifestyle and avoid activities that involving repeated heel impact (jumping, jogging, etc. ...). Fishing, hunting, golf or skiing activities are acceptable for those who previously followed these pursuits prior to the procedure.
Contra-indications Inflammatory arthritis (e.g., rheumatoid arthritis), bi- or tricompartmental OA or any ligamentous injury (e.g., chronic anterior cruciate laxity, medial collateral ligament damage) are absolute contra-indications.
The “Ideal” Patient If we consider all the above indications, the “ideal” patient should have; [18]:
• Pain localized to the site of the radiographic arthritis • Normal range of motion • Normal ligament balance • Passively correctable deformity • Age >65 • BMI <30 • Weight <80 kg • IKDC A, B, C or D • No cupula • Isolated, single compartment arthritis with normal c ontra-lateral and patello-femoral compartments • Deformity <10° with osseous part <5° • Activity which excludes running and jumping
172
Technical Aspects Patient Set-up Supine on a conventional operating table. The knee must be kept flexed to 90° (we use a side wedge and wedgeended table) but allowing intra-operative maximal range of motion. It should be possible to extend and external rotate the knee. A tourniquet is required on the proximal thigh and inflated with the knee in flexion.
Incision Minimally-invasive approaches are part of the philosophy of UKA (Fig. 3), made popular in 1998 and subsequently popularized by literature coverage. An antero-lateral approach is used for a lateral UKA and antero-medial approach for the medial UKA.
Tibial Cutting Coronal Plane The mechanical femoro-tibial axis, the mechanical femoral axis, and the mechanical tibial axis (MTA) enable us to
Fig. 3 Minimally-invasive approach
S. Lustig et al.
analyze the osseous and wear-induced deformities. In some patients, arthritic damages to a tibial plateau may be very severe, making it impossible for the mechanical tibial angle to be measured with certainty. This flaws the interpretation of the tibial wear component (which will be corrected by the UKA) and the extra-articular osseous deformity component (which should not be corrected by the UKA). To determine the wear component, the concept of the proximal tibial epiphyseal axis (PTEA), which represents the extra-articular osseous deformity component, has been introduced by the Lyon School previously[19]. The PTEA is perpendicular to the line connecting the medial and lateral physis. The MTA is determined by a line connecting the tibial spine with the middle of the talar dome. The ∆ (PTEA-MTA) thus represents the angle of the extra-articular osseous deformity. Previous studies performed in our Institute [19] and by Jenny et al. [20] revealed that the angle between a line parallel with the non-affected compartment and the MTA accurately described the same extraarticular osseous deformity without the difficulties of defining the PTEA (Fig. 4). In a tibia with no extra-articular deformity, the extramedullary cutting guide is centred over the mid-point of the mechanical axis in the coronal plane (Fig. 5). In the case of a metaphyseal tibia bowing, the tibial cut should be performed perpendicular to the PTEA. Thus, in the case of a varus or valgus deformity in the tibial metaphysis, the tibial alignment guide will not be parallel to the tibial mechanical axis, but will be pointing more to the lateral or medial malleolus. For lateral UKA the tendency is to accentuate the angle inferiorly and medial which is important to avoid as it may lead to the transverse sliding phenomena. There will always be a necessity for checking for extension in the oblique downward and outward direction for the coronal section to
Fig. 4 Tangent to the tibial plateau, reproducing the coronal plane tibial cut
173
The Uni-Knee: Indications, and Recent Techniques
Fig. 6 Increased tension on the posterior tibial plateau due to failure to correct the pre-pathological tibial slope Fig. 5 Adjustment of the frontal slope by reference to the tibial crest
Femoral Preparation Medial UKA
be exaggerated. An important consideration is to maximize internal rotation of the tibial plateau to minimize risk of impingement between the prosthetic condyles and large osteophytic tibial spines [21].
Sagittal Plane It is essential to determine the change in sagittal plane secondary to articular wear. Failure to correct the pre-pathological sagittal slope of the plateau will lead to hyperflexion with increased tension on the posterior tibial plateau. This results in raising the anterior part of the plateau or the phenomenon of a “rocking” femur (Fig. 6). It is not possible to determine an average slope that would suit all cases. The appropriate slope needs to be determined for each patient. This can be done either by pre-operative radiographic templating, or by using jigs intra-operatively (e.g., pins or blades). Complications from an incorrect slope are rare. We have never observed anterior subluxation associated with the sagittal slope. This may be attributed to the ACL which, by definition, must remain intact [22].
The main debate concerns the choice between UKA “resurfacing” and UKA with distal bone cut. The principle of a “wedge” composed of two elements, one femoral and one tibial, each, of whose point of contact represents the joint line, presents a risk to induce some modification of the joint line. This is dependent on the position of the condylar prosthesis. The main issue is that the condyle may be offset by a tibial overcut resulting in a difference in femoro-tibial spacing with no effect on function, but with risk of decreasing the quality of the tibial fixation [21, 23]. The advantage of re-surfacing is that it allows compensation for joint loss with strong support of the prosthesis on the subchondral bone which has been strengthened by osteoarthritis. Most data on femoral re-surfacing demonstrates high reliability in terms of loosening and survival of femoral implants. Re-surfacing is an excellent choice when there is wear on the weight-bearing area of the femoral condyle. A complication that we have found is that islets of cartilage may persist in the subchondral bone of OA (mostly medial) or in aseptic necrosis of the medial condyle. In these situations it is important to carefully resect all residual cartilage so as to prevent prosthetic loosening.
174
UKA is attractive alternative to TKA for cases with stage 2 tibio-femoral OA which is only visible on a Schuss view (cartilage loss on the posterior condyle) or situations with aseptic necrosis of the medial condyle. Early intervention increases the risk of prosthetic loosening on of the femoral condyle. However delay increases the risk of bone loss on the underlying tibial plateau. An AFTM greater than 180° suggests lateral OA and acts as a guide for the UKA cut. However, the femoral cut raises other issues and other constraints on both the direction of the cut in the frontal plane and the correlation with the posterior cut and its rotation. UKA bone cuts are not forgiving and may be particularly difficult to adjust with a minimally-invasive approach. We avoid using an intramedullary rod as it is a source of significant post-operative bleeding. UKA is in theory designed to minimise blood loss. The prosthesis requires immediate post-operative stable fixation without the risk of movement in hyperflexion. The stability of the prosthesis during full range of motion is important as often the post-operative range of motion is greater than 120°.
S. Lustig et al.
Lateral Compartment Re-surfacing in lateral UKA is particularly suitable if not better than medial UKA. The cartilage loss is frequently secondary to condylar dysplasia (AFM > 92°). In these situations, unlike the medial condyle, the risk of prosthetic loosening is very low. The condylar prosthetic re-surfacing often is correcting dysplasia of the distal condyle. The greatest difficulty is to assess the correct rotation of the femoral condyle. It is important not to be falsely guided by the oblique internal rotation of the distal femoral condyle when the knee is flexed. Aligning with the long axis of the distal condyle will inevitably lead to impingement of the prosthetic condyle anteriorly with large tibial spines in extension regardless of the internal rotation of the tibial plateau. Good fixation of the prosthetic condyle in external rotation is aided by the lateral condyle osteophyte (which should be scrupulously retained). The aim is to find the centre point in extension once the tibial cut has been made or to be guided by instruments that determine the appropriate rotation by ranging the knee from extension to flexion (Fig. 7).
Fig. 7 Specific instrument that determines the appropriate rotation by ranging the knee from extension to flexion
175
The Uni-Knee: Indications, and Recent Techniques
Prosthetic Stability When the trial prosthesis is in place, and prior to cementing the definitve prosthesis, the surgeon should ensure: Perfect stability of the implant components during full range of motion (Fig. 8). It may be possible to detect a “tilt” (tibial insert is anteriorised with flexion), subluxation of the pad, translation, tibial femoral incongruence which are phenomena which reflects tightening in flexion. The tibial tilt may reflect a lack of sagittal tibial slope which can be easily corrected. It may also be due to an insufficient posterior femoral condyle cut causing the posterior condyle of the prosthesis to impinge. In this situation it is often impossible to correct the error once the condyle posts, which anchor the prosthesis, have been made. Increasing the posterior sagittal tibial cut may be an option, but this increases the risk of laxity in extension. It is particularly important to avoid over-tensioning of the ligaments. Over-tensioning should never be treated by surgical release. It is appropriate to decrease the thickness of the PE, or to revise the bone cuts if possible. Symmetrical tibial cuts act in both flexion and extension. It is rarely possible to modify the femoral cuts once the trial has been inserted, especially once the prosthetic posts have been
Fig. 8 Stability of the trials in flexion
drilled as insufficient bone stock would remain for fixation afterwards. Failure of bony contact with the femoral component can lead to the phenomenon of “pitching” during flexion and resultant loosening.
Definitive Implant Based on pre-operative planning and intra-operative assessment, the suitable tibial component (fixed or mobile, cemented or uncemented) can be determined (Fig. 9).
Fig. 9 Post-operative X-ray of a cemented UKA
176
Conclusion In the field of knee replacement, UKA is an attractive alternative to TKA in appropriate situations. Strict observance of indications, understanding of concepts and rigorous surgical technique, as discussed in this chapter, assist in obtaining excellent long term results.
References 1. Marmor L (1988) Unicompartmental arthroplasty of the knee with a minimum ten-year follow-up period. Clin Orthop Relat Res 228:171–177 2. Cartier P, Chaib S, Vanvooren P (1987) Prothèse unicompartimentale du genou. A propos de159 cas à un recul maximum de 10 ans. Rev Chir Orthop 73(suppl 2):130–133 3. Insall JN, Aglietti P (1980) A five to seven-year follow-up of unicondylar arthroplasty. J Bone Joint Surg Am 62:1329–1337 4. Laskin RS (1978) Unicompartmental tibiofemoral resurfacing arthroplasty. J Bone Joint Surg Am 60:182–185 5. Robertsson O, Borgquist L, Knutson K, Lewold S, Lidgren L (1999) Use of unicompartmental instead of tricompartmental prostheses were compared with 10, 624 primary medial or lateral unicompartmental prostheses. Acta Orthop Scand 70:170–175 6. Griffin T, Rowden N, Morgan D, Atkinson R, Woodruff P (2007) Madderng: unicompartmental knee arthroplasty for unicompartmental osteoarthritis: a systematic review. ANZ J Surg 77:214–221 7. Wood J (2006) Unicompartmental knee arthroplasty. Curr Opin Orthop 17:139–144 8. Kozinn S, Scott R (1989) Unicondylar knee arthroplasty: current concept review. J Bone Joint Surg 71A:145–150 9. Cartier P, Deschamps G Principes technique de l’arthroplastie unicompartimentale. In: Cartier P, Epinette J, Deschamps G, Hernigou P (ed) Prothèses unicompartimentales de genou. Cahier d’Enseignement de la SOFCOT – Expension scientifique publication, Paris, pp 145–151 10. Dejour D, Chatain F, Dejour H. Résultats clinique sd ela prothèse unicompartiementale HLS. In Cartier P, Epinette J, Deschamps G, Hernigou P (ed) Prothèses unicompartimentales de genou. Cahier d’Enseignement de
S. Lustig et al. la SOFCOT – Expension scientifique publication, Paris, pp 126–132 11. Deschamps G (1995) Arthroplastie unicompartimentale et système ligamentaire. Cahiers d’enseignement de SOFCOT 65:152–155 12. Goodfellow J, O’connor J (1987) The anterior cruciate ligament in knee arthroplasty. A risk-factor with unconstrained meniscal prostheses. Clin Orthop 222:239–248 13. Berger RA, Meneghini RM, Jacobs JJ, Sheinkop MB, Della Valle CJ, Rosenberg AG, Galante JO (2005) Results of unicompartmental knee arthroplasty at a minimum of ten years of follow-up. J Bone Joint Surg Am 87:999–1006 14. Tabor OB Jr, Tabor OB, Bernard M, Wan JY (2005) Unicompartmental knee arthroplasty: long-term success in middle-age and obese patients. J Surg Orthop Adv 14:59–63 15. Swienckowski J, Pennington D (2003) Unicomprtmental knee arthroplasty in patients sixty years of age or younger. J Bone Joint Surg 85:131–142 16. Deshmuckh RV, Scott RD (2001) Unicompartmental knee arthroplasty: long term results. Clin Orthop Relat Res 392:272–278 17. Engh GA (2002) Orthopedic crossfire: can we justify unicondylar arthroplasty as a temporizing procedure ? In the affirmative. J Arthroplasty 17(Suppl 1):54–55 18. Lustig S, Paillot JL, Servien E, Henry J, Ait Si Selmi T, Neyret P (2009) Cemented all polyethylene tibial insert unicompartimental knee arthroplasty: a long term follow-up study. Orthop Traumatol Surg Res 95:12–21 19. Levigne C (1991) Intérêt de l’axe épiphysaire dans l’arthrose. Analyse du groupe témoin. Communication aux 7émes Journées Lyonnaises du Genou. Les Gonarthroses, Lyon 20. Jenny JY, Boeri C, Ballonzoli L (2005) Difficulties and reproducibility of radiological measurement of the proximal tibial axis according to Levigne. Rev Chir Orthop 91:658–663 21. Pennington DW, Swienckowski JJ, Lutes WB et al (2006) Lateral unicomparmental knee arthroplasty. Survivorship and technical considerations at an average follow up of 12, 4 years. J Arthroplasty 21:13–17 22. Deschamps G, Lapeyre B (1987) La rupture du ligament croisé antérieur. Une cause d’échec souvent méconnue des prothèses unicompartimentales du genou. A propos d’une série de 79 prothèses lotus revues au dela de 5 ans. Rev Chir Orthop 73:544–551 23. Iesaka K, Tsumura H, Sonoda H et al (2002) The effects of tibial component inclination on bone stress after unicompartmental knee arthroplasty. J Biomech 35:969–974
Osteotomies Around the Knee Siegfried Hofmann, Philipp Lobenhoffer, Alex Staubli, and Ronald Van Heerwaarden
Introduction
Basic Biomechanics
The excellent mid- to long-term results of unicondylar replacements have decreased the indications for osteotomies in several countries even with younger and active patients [1]. Furthermore, because of fair results, relative high complication rates and several different surgical techniques, osteotomies were nearly abandoned in the AngloAmerican community. Nevertheless in Europe osteotomies around the knee for mono-compartment arthritis have remained still a standard approach for the last 30 years [2–6]. New knowledge of biomechanics, better patient selection, proper planning, standardized surgical technique and stable osteosynthesis, which allows early rehabilitation, have shown promising mid-term results [7–11]. The importance of combined cartilage and ligament reconstruction surgery with correction of mal-alignment have caused a new interest in osteotomy procedures [12]. In this review the basic principles of biomechanics, criteria for patient selection, planning, the most common surgical technique, rehabilitation, as well the clinical results for osteotomies around the knee for mono-compartment arthritis, will be described. The rare indications for three-dimensional deformity corrections [13] will not be covered in this consensus paper of the international AO knee expert group. Furthermore, for more information and details on osteotomies around the knee we refer to the more detailed description of this new concept [9, 14].
The biomechanical analysis may be performed according to five different parameters in the three planes (Table 1). Furthermore the hip and ankle joint should be always included in the deformity analysis also.
S. Hofmann () Head Knee Education Centre, Orthopaedic Department, General and Orthopaedic Hospital Stolzalpe, Stolzalpe 8852, Austria e-mail:
[email protected]
Frontal Alignment Based on a standardized long-leg, standing (three-joint) radiographs the six criteria for frontal alignment (Table 2) can be analysed [9, 15]. The deformity analysis of the frontal plane according to Paley [16] should include the mechanical axis of the femur and tibia, overall malalignment (varus or valgus), the lateral distal femoral angle (LDFA), medial proximal tibial angle (MPTA) and the joint line convergence angle (JLCA) (Figs. 1 and 2). Using the point where the weight-bearing line (Mikulicz line) crosses the tibia plateau width (TPW) the knee joint stress can be calculated (Table 3) [17]. This simple two-dimensional analysis has been confirmed with dynamic 3-D computersimulation models also [18].
Joint Line In the normal knee the joint line (middle of the base lines of femur/tibia) is in 3° of Varus in relation to the Mikulicz line (M-JL) and will be parallel to floor during the stance phase of the gait cycle (Fig. 2). Correction of the deformity on the wrong side of the bones (tibia and/or femur) will cause a pathological inclination of the joint line with significant shear forces on the cartilage [19–23]. The dogma of “varus is tibia” and “valgus is femur” is wrong. In a recent study using the deformity analysis described above, it could be shown, that in varus knees the deformity was in
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_14, © 2011 EFORT
177
178
S. Hofmann et al.
Table 1 Five parameters for the deformity analysis in mono-compartment arthritis 1. Frontal alignment 2. M-joint-line 3. Sagittal alignment 4. Patello-femoral joint 5. Rotational deformities (Mal-torsion)
Table 2 Biomechanical parameters and normal values for the frontal alignment 1. Mechanical axis of the leg (0 – 2° Varus) Varus 12°
2. Weight-bearing line (Mikulicz line) (8 ± 7 mm medial) 3. Cross-point of Mikulicz line through the tibia plateau width (0–5% TPW) 4. Lateral distal femoral angle (LDFA 88° – 85 – 90) 5. Medial proximal tibial angle (MPTA 87° – 85 – 90) 6. Joint-line convergence angle (JLCA 0 – 2°)
0°
50° 100°
59% the femur, 31% the tibia and in 10% on both sides. In valgus knees it was 45% in the tibia, 22% the femur and in 33% both [20].
Sagittal Alignment In the sagittal plane the anatomical posterior proximal tibia angle (aPPTA 81° [77–84°]) reflects the tibial slope and the anatomical posterior distal femur angle (aPDFA 83° [79–87°]) the flexion/extension positioning of the distal femur [16] (Fig. 3). Changes of the tibial slope might cause significant pathological forces and should be performed in selected cases only [24]. Pathological flexion/extension positioning of the distal femur and/or slope of the tibia might cause flexion contracture or recurvatum and should be corrected in selected cases only [25].
Patello-femoral Joint For analysis of the patello-femoral joint a special weightbearing axial view is helpful to identify patients with osteoarthritic changes and patellar mal-tracking [26]. Furthermore in the sagittal plane the patella height (baja or alta) reflects an important factor [15]. After osteotomy the biomechanical changes of the patello-femoral joint should be taken into account [27]. For further analysis of patella mal-tracking a special CT or MR-imaging is necessary to analyse the tibial
Fig. 1 Varus deformity 12°, mechanical axis femur and tibia, weight-bearing line (Mikulicz line) and cross-point with tibial plateau width (% TBW) (With permission from Hofmann et al. [9])
tubercle – trochlea groove (TT-TG) distance [15, 20, 28]. It is important to understand that rotational deformities of tibia and/or femur (mal-torsion syndromes) are the most common cause for patellar mal-tracking [9, 27].
Rotational Deformities Rotational deformities should be identified by clinical examination and need to be further investigated by CT or MR- imaging performing a torsional profile [15]. Mal-torsion syndromes must be included in the planning of osteotomies and compensation of patellar mal-tracking [23, 27].
179
Osteotomies Around the Knee
M-GL 90°
aPDFA = 83° (79°− 87°)
LDFA 96°
MPTA 82°
JLCA 3°
aPPTA = 81° (77°− 84°)
Fig. 3 Planning of sagittal alignment with aPDFA and aPPTA (With permission from Paley and Pfeil [21]) Fig. 2 Location of deformity of case Fig. 1 with LDFA 96°, MPTA 82° M-joint-line 90° and joint line congruence angle (JLCA) 3°, note the varus deformity is at the femur and tibia (With permission from Hofmann et al. [9])
Table 3 Medial tibial plateau stress with different varus deformities [17] Varusfehlstellung 0°
75%
Varusfehlstellung 5°
80%
Varusfehlstellung 10°
90%
Varusfehlstellung 15°
100%
Source: Modified according to Hsu et al. [17]
Patient Selection and Indication Patient selection for osteotomies should be performed in a standardized fashion. The ideal candidate is well-described by several criteria [29] (Table 4). Furthermore identification of risk factors might be very helpful for decision-making.
For the progression of the disease the mechanical axis of the leg is an important prognostic factor (1.5–2.0 higher risk compared to neutral alignment) [30]. The stress bone-marrow oedema (BME) in MR-imaging represents an important further risk factor for progression (4.5 higher risk) [30]. In case of pre-arthritic deformities (meniscus damage, instabilities and post-traumatic cartilage damage) or planned surgery with cartilage repair or ligament reconstruction, a deformity analysis of the leg with a long-leg standing radiograph should always be performed [12, 15, 31, 32]. The clinical significance of the patello-femoral joint for osteotomies remains still controversial. In principal the therapeutic approach to the patello-femoral joint should be based on biomechanics. Combinations of bony and soft tissue procedures for patello-femoral mal-tracking might be combined with correction osteotomies of the tibio-femoral joint [9]. The indication for an osteotomy remains based on a combination of morphological, functional as well as imaging criteria. Additional soft facts like patient’s expectations, compliance, occupation and sporting activities) should be also taken into account [9]. Length of rehabilitation is also an important social factor for the patient. After extensive
180
S. Hofmann et al.
Table 4 Patient selection for osteotomies in mono-compartment arthritis Ideal candidate
Possible candidate
No candidate
Isolated pain medial/lateral at joint level
Infection history
Contra-lateral arthritis and st.p. lateral meniscectomy
Age 40–60 years
Age <40 or >60
BMI < 30
BMI 30–40
Active patient but no running and jumping
Running and jumping sporting activities
Mal-alignment <15°
Mal-alignment >15° possible double osteotomy
Metaphyseal varus tibia and valgus femur
Metaphyseal varus femur and valgus tibia
Extra-articular deformity
Full ROM
Flexion contracture >15°
Flexion contracture >25°
No patello-femoral symptoms
Medium patello-femoral symptoms (grade 2–3)
Severe patello-femoral arthritis grade 4 and maltracking
Arthritis grade 1–3
Arthritis grade 4
Stable joint
Insufficient ACL or PCL
No smoker
Smoker
BMI > 40
Medio-lateral instability
Source: Modified according to ISAKOS guidelines [29]
explanation, the patient should be able to understand the principles and alternative options of the therapeutic concept.
Planning A standardized full-leg standing radiograph [9, 15, 23] is important for frontal plane and joint-line analysis (Table 2). Short radiographs allow the measurement of the tibio-femoral angle only and do not allow deformity analysis. [9, 23, 33]. The overall mal-alignment (varus or valgus), LDFA, MPTA, the tibia plateau stress (Table 3) as well as the M-joint line can be identified (Figs. 1 and 2). The location of the deformity (femur and/or tibia) can be easily identified by the LDFA and MPTA and will give important information where the correction osteotomy should be performed. Ligament laxity or medial cartilage loss causing a pathological JLCA should be included in the planning, otherwise overcorrection might occur [33]. On the lateral x-ray the tibia slope (aPPTA), flexion/ extension of the distal femur (aPDFA) and the patella height can be measured [15]. In the frontal plane the goal of the correction should be neutral to slightly overcorrection with a horizontal M-jointline. For the varus knee the individual correction should depend more on the additional pathologies of the knee (Table 5) [34] than by a certain point or area (i.e., 62% of TPW or 30–35% of the lateral tibia plateau) [33]. As in the normal
Table 5 Grade of correction in varus deformities Post-traumatic deformity without arthritis
0 – 2°
ACL insufficiency
0 – 2°
PCL insufficiency (and lateral instability)
2 – 4° (5°)
Cartilage repair with cartilage damage
3–5°
Arthritis grade I and II
2 – 4°
Arthritis grade III and IV
4 – 6°
Source: Modified according to Müller [34]
knee only 25% of the stress will be on the lateral compartment, in almost all cases correction to neutral will be enough for the valgus knee [20, 23]. The M-joint-line should be 90° ± 4° after the osteotomy, otherwise the patient will walk on an inclined joint-line with shear forces (see above). In about 10 –15% of the patients this goal cannot be achieved with one osteotomy only, as the deformity is both at the femur and tibia [22]. In younger and active patients a double osteotomy should be considered [20], otherwise the pathological M-joint-line might compromise the long-term result [7, 35]. In the sagittal plane the slope should be changed in selected cases only [25]. In cases with an insufficient anterior cruciate ligament the slope might be decreased and with an insufficient posterior cruciate the slope might be increased [12]. Full extension is an important goal of every osteotomy. Flexion contractures might be corrected by resection of anterior osteophytes, notchplasty or decreasing the slope of the tibia. Flexion contractures or hyperextensions, which
Osteotomies Around the Knee
are caused by a pathological distal femur (aPDFW) or pathological slope (aPPTA) should be corrected at the site of the deformity. The practical planning of the key parameters (location of deformity, type of osteotomy, location and angle of correction) can be made by conventional drawing or with computer assistance. The calculation of the bony resection (distractive) or opening height (additive) can be made by different methods [33]. One of the most common and conventional planning methods is the use
Fig. 4 Planning of tibial osteotomy according to Miniaci (With permission from Hofmann et al. [9])
181
of a simple parallelogram (Fig. 4) [23, 33]. The modern planning software allows easy planning of all parameters on the digitized, long-standing radiograph with the help of a computer. The different parameters and type of osteotomies might be changed several times until the optimal technique will be identified (Fig. 5). Furthermore the intra-operative control and accuracy of the planned correction might be achieved by conventional fluoroscopic controls or with the help of computer navigation systems [36, 37].
182 Fig. 5 Digital planning with the computer of three different osteotomies from the case of Fig. 1, Note the pathological M-joint-line for (a) single medial open wedge tibia and (b) single lateral closed wedge femur osteotomy and normal M-joint-line with (c) double osteotomy (With permission from Hofmann et al. [9])
S. Hofmann et al.
a
b
c
Axis 0°
Axis 0°
LDFA 93°
LDFA 83°
MPTA 95°
MPTA 84°
ML-JL 95°
ML-JL 84°
Wedge 14 mm
Wedge 13 mm
Axis 0° LDFA 90° MPTA 88° ML-JL 88° Wedges Femur 8 mm Tibia 6 mm
Surgical Techniques A ‘scope’ should be performed before every osteotomy to confirm the final indication and to allow repair of meniscus, cartilage and ligament damages [38]. Theoretically there are ten different possibilities for osteotomies around the knee: femur or tibia medial or
lateral, open (additive) or closed (subtractive), dome and double osteotomies [35]. De-rotation osteotomies in the third plane, which are performed for mal-torsion syndrome with patella mal-tracking, represent several further osteotomy options [27], but these complex osteotomies are not part of this review. In this paper some principal surgical rules can be described only. For a more detailed description, advantages, disadvantages as well
Osteotomies Around the Knee
183
as tips and tricks of the different techniques we have to refer to the special literature [4, 6, 9, 10, 39–42].
Tibial Osteotomies At the medial tibia, opening and closing and at the lateral tibia, closed wedge osteotomies may be performed [4, 10, 39, 41, 42]. For opening lateral osteotomies the rare posttraumatic impression fracture of the tibia plateau may be an indication only. The level and inclination of the osteotomy depends on the location, type and fixation. At the tibia the dome osteotomy and callus distraction with external fixation might be an alternative [6, 43, 44] but both methods have not been accepted for the mono-compartment arthritis as a routine procedure. For valgus correction at the tibia the lateral closedwedge osteotomy has been the preferred surgical technique for a long time [4]. The exact three-dimensional correction with two bone cuts is difficult. A further problem of the classical lateral closed-wedge osteotomy is the deformation of the proximal tibia, which can be prevented by a modified technique using an ascending osteototomy [39, 41] (Fig. 6). A medial closed osteotomy will be used for the less common valgus knees, where the deformity is at the tibia. One of the advantages of the closed-wedge tibia osteotomies is the possible use of minimal osteosynthesis for fixation. With the new angle stable plates and more precise surgical techniques, the medial open wedge osteotomies have become the favourite surgical technique for many surgeons [10, 42] (Fig. 7). A comparison of the advantages and disadvantages of closed and open osteotomies is summarized in Table 6. From a technical point of view the open wedge osteotomy is easier and more safe for a three-dimensional correction because only one osteotomy has to be performed. Furthermore with this technique the medial superficial ligament can be balanced and the problem of patella baja might be addressed with an inverse anterior osteotomy at the tibial tubercle [45]. When using a medial open-wedge osteotomy an angle stable implant should be used, as other implants have shown a high failure rate [8]. This insufficient stability of non-angle stable osteosynthesis might be compensated with bone grafting [8, 46].
Femoral Osteotomies For many surgeons femoral osteotomies represent not a standard procedure and due to the more demanding surgical technique and higher complication rate femoral osteotomies are performed rarely. Nevertheless at the femur medial closed wedge for valgus and lateral closed wedge osteotomies for varus femur deformities should be a standard
Fig. 6 Lateral closed wedge osteotomy with ascending osteotomy cut (With permission from Hofmann et al. [9])
technique when dealing with mono-compartment arthritis [11, 40, 47]. In cases with a varus femur, a medial open or lateral closed tibia osteotomy will cause a pathological M-joint-line (Fig. 5). At the femur only closed osteotomies should be performed because of the high biomechanical stress and less biological healing capacity (osteotomy outside spongy bones). Only in those cases where leg-length shortening might be a problem, open wedge osteotomies should be performed [20]. For a more safe bone healing autologous bone grafts should be used for open-wedge osteotomies at the femur additionally. Distal descending osteotomy levels are more safe, because the osteotomy parts fit more closely to each other and the osteotomy is partially in the spongy bone [11]. Nevertheless femoral osteotomies require a very stable osteosynthesis with some compression at the osteotomy site. The classical angled plates allow a stable osteosynthesis, but are demanding and less forgiving procedures [11]. Retrograde nailing and external fixation have not been widely accepted for osteotomies at the femur. The new biplanar osteotomy technique in combination with long angle stable plates allow a stable osteosynthesis at the femur with a relatively easy and reproducible technique [40] (Fig. 8).
184
S. Hofmann et al.
Double Osteotomies To achieve correct frontal alignment and a M-jointline within the limits of 90° ± 4° in about 10–15% a double osteotomy will be required [19, 20, 22, 35] (Fig. 5). When performing a double osteotomy, the natural M-jointline (87° ± 3°) should be the goal of correction. This demanding procedure might be offered to the more young and active patients up to 50 years of age. In older patients a slightly inclined M-joint-line can be accepted as a compromise [20]. In all patients with planned single level osteotomies the M-joint-line should be analysed also (see above) [35].
Rehabilitation
Fig. 7 Medial open wedge biplanar osteotomy with an angle stable plate and no bony interposition graft (With permission from Hofmann et al. [9])
Table 6 Comparison open versus closed wedge tibial osteotomies Criteria
Closed
Open
Technique
Two correction cuts
One correction cut only
Precise correction
Difficult
Simple
Fibula problem
Yes
No
Detachment of muscles
Tibialis loge
No
Nerve damage
Peroneal possible
No
Slope change
Reduction possible
Increasing possible
Ligament balancing
No
Medial possible
Patalla baja
Secondary possible Primary possible
Deformity proximal tibia
Possible
No
Bone transplant
No
Exceptional cases only
To deal with the changed biomechanics the knee needs to develop considerable compensation mechanism to adapt to the new situation. In general a stable osteosynthesis is necessary to allow early rehabilitation since a longer immobilization will further damage the arthritic joint. In patients with healthy bone, healing might be expected within 6–12 weeks after a closed or an open osteotomy with autologous bone grafting at the tibia. In open osteotomies without bone grafting, closed osteotomies at the femur and osteotomies in smokers the bony healing takes significantly longer [46]. Bracing is not necessary for stable osteosynthesis at the tibia but is recommended after femoral osteotomies. Immediately after surgery daily physiotherapy should be performed with muscle-strengthening and flexion exercises. Partial weight-bearing limited by pain can be performed from the beginning. Radiographic controls should be performed after 6–12 weeks and depending on the osteotomy and bony healing the further mobilisation should be performed on an individual basis. Depending on the occupation the patient can go back to work after 12 weeks on average. In patients with delayed bony healing the rehabilitation might be significantly longer.
Clinical Results The primary goal of reducing pain and gaining better function can be achieved with proper patient selection and surgical technique in more than 90% of patients. The peri- operative risk of under- or over-correction ranges from 5–20% in the literature [48, 49]. The long-term results of osteotomies show a great variability. In a meta-analysis of 24 publications (2,255 cases) the average survival rate
185
Osteotomies Around the Knee Fig. 8 Medial closed wedge biplanar femur osteotomy with an angle stable plate (With permission from Hofmann et al. [9])
(TKA as the end-point) was 73% (50–90) after 5 years and 52% (17–70) after 10 years. Some reports with selected patients (younger then 50 and with little arthritis) have shown a much better survival with 90% after 10 years. In a sub-collection of 12 studies (1,016 patients) the clinical results of osteotomies showed excellent and good scores in 58% only after 10 years [2]. The clinical and functional results in all studies were worse compared with unicondylar replacements [1]. Less data are available for going back to work and sporting activities after osteotomies. For osteotomies in general, the sporting levels are the same as before surgery, which is higher than that following unicondylar replacements [50]. Several prognostic factors for the long-term survival have been identified in a study with 217 patients after a mean follow-up of 9 years [2] (Table 7). Knees with good post-operative frontal alignment (femoro-tibial angle 179– 184° – no long standing radiographs available) showed a failure rate of 2% only and the clinical results were excellent and good in 75% of the cases. For this well-aligned
Table 7 Risk factors and parameters for successful osteotomies Correction at the location of deformity (femur and/or tibia) M-joint-line horizontal 90° ± 3° Post-operative frontal mechanical axis Pre-operative cartilage damage Age at time of surgery BMI
group radiographic progression occurred in 8% medial and 7% lateral compartments only. In this study a statistically significant difference (p < 0.01) was found compared to the non-well-aligned group with under- or over-correction. Frontal alignment as an important prognostic factor was identified by other authors with long-term studies [48, 49]. The constitutional tibia vara with typical metaphyseal varus deformity had been described as “Tibial Bone Varus Angle (TBVA)” in the French literature [2, 8, 33]. Patients with a TBVA > 5° (metaphyseal varus) showed a significant
186
better outcome compared with normal TBVA (no metaphyseal varus). The assumption that patients with varus deformities and normal TBVA show the deformity due to the cartilage loss in the medial compartment only is wrong. In a study using deformity analysis according to Paley [16] it could be shown that in patients with varus knees and normal TBVA the varus deformity was not due to medial cartilage damage only [20]. In these cases a normal TBVA (<5°) did correlate with a normal MPTA (85–90°) but the varus was at the femur with a pathological LDFA (>90°). The relatively common varus deformity at the femur will not be identified when using short radiographs and TBVA analysis only. In those cases with normal TBVA and osteotomy at the tibia the correction of the deformity had been made at the wrong side of the joint. This caused a pathological M-joint-line with significantly worse clinical results in these cases. Based on the above-mentioned principles the TBVA analysis should be replaced by the deformity analysis on a long standing radiograph using the LDFA and MPTA [8, 20, 21]. Pre-operative cartilage damage seems not to play an important prognostic role during the first 6 years [2]. Nevertheless after longer follow-up the clinical results were significantly worse (3.5 higher failure rate) in patients with higher grades of cartilage damage compared to lower grades [49]. The cartilage damage in the lateral compartment in a varus knee is still discussed controversially [38]. A clear guideline as to what amount of cartilage damage of the contra-lateral side is acceptable for an osteotomy still is not precise at the moment. An individual adaption of the amount of correction to the cartilage damage of the contralateral side might be a logical approach [38]. Age at the time of surgery seems to have an important influence on the long-term outcome. Patients less then age 50 have shown in several studies significantly better outcome compared to older patients [2]. Obesity represents a further patient-specific prognostic factor. In the abovementioned study patients with obesity (>30% above normal) have shown significant less excellent and good clinical results (20% vs 56%) compared to patients with normal weight after 10 years [2]. The long-term results of osteotomies are significantly worse compared to unicondylar replacements [1]. Never theless more recent results using the above new concept of osteotomies have shown not only in short and mid-term results [10], but also in a long-term follow-up study [7], very promising results. Using these new concepts, osteotomies around the knee have been shown to be an important alternative to unicondylar replacements. For active patients in the middle age group osteotomies should be considered still as the primary choice in case of mono-compartment arthritic disease of the knee.
S. Hofmann et al.
References 1. Newmann J, Pydisetty R, Ackroyd CE (2009) Unicompart mental or total knee replacement: the 15-year results of a prospective randomised controlled trial. J Bone Joint Surg Am 75:483–498 2. Bonnin M, Chambat P (2004) The closed wedge valgus osteotomy at the tibia for medial gonarthritis German. Orthopade 33:135–142 3. Coventry MB, Ilstrup DM, Wallrichs SL (1993) Proximal tibial osteotomy. A critical long-term study of eighty-seven cases. J Bone Joint Surg Am 75:196–201 4. Jakob RP, Jacobi M (2004) The closed tibia osteotomy for the treatment of the monocompartment arthritis of the knee German. Orthopade 33:143–152 5. Paley D, Maar DC, Herzenberg JE (1994) New concepts in high tibial osteotomy for medial compartment. Orthop Clin North Am 25:483–498 6. Pfeil J, Hasch E (2005) Transposition osteotomy on the knee joint German. Z Orthop Ihre Grenzgeb 143:43–64 7. Babis GC, An KN, Chao EY, Larson DR, Rand JA, Sim FH (2008) Upper tibia osteotomy: long term results – realignment analysis using OASIS computer software. J Orthop Sci 13:328–334 8. Brinkman JM, Lobenhoffer P, Agneskirchner JD, Staubli AE, Wymenga AB, van Heerwaarden RJ (2008) Osteotomies around the knee: patient selection, stability of fixation and bone healing in high tibial osteotomies. J Bone Joint Surg Br 90:1548–1557 9. Hofmann S, Lobenhoffer P, Staubli A, Van Heerwarden R (2009) Osteotomies around the knee German. Orthopade 38:755–770 10. Lobenhoffer P, Agneskirchner J, Zoch W (2004) The medial open wedge osteotomy at the tibia with a medial plate fixateur interne German. Orthopade 33:153–160 11. Stahelin T, Hardegger F (2004) Incomplete supracondylar femur osteotomy German. Orthopade 33:178–184 12. Imhoff AB, Linke RD, Agneskirchner J (2004) Corrective osteotomy in primary varus, double varus and triple varus knee instability with cruciate ligament replacement German. Orthopade 33:201–207 13. Marti RK, van Heerwaarden RJ (eds) (2008) Osteotomies for posttraumatic deformities. Thieme, Stuttgart 14. Lobenhoffer P, Agneskirchner J, Galla M (Hrsg) (2007) Osteotomies around the knee. Thieme, Stuttgart 15. Pietsch M, Hofmann S (2006) Radiographic imaging at the knee joint for orthopedic surgeons German. Radiologe 46:55–64 16. Paley D, Herzenberg JE, Tetsworth K, McKie J, Bhave A (1994) Deformity planning for frontal and sagittal plane corrective osteotomies. Orthop Clin N Am 25:425–465 17. Hsu RW, Himeno S, Coventry MB, Chao EY (1990) Normal axial alignment of the lower extremity and load-bearing distribution at the knee. Clin Orthop Relat Res 255:215–227 18. Heller M, Taylor W, Perka C, Duda G (2003) The influence of alignment on the musculo-skeletal loading conditions at the knee. Langenbecks Arch Surg 388:291–297
Osteotomies Around the Knee 19. Babis GC, An KN, Chao EY, Rand JA, Sim FH (2002) Double level osteotomy of the knee: a method to retain joint-line obliquity. Clinical results. J Bone Joint Surg Am 84-A:1380–1388 20. Hofmann S, Van Heerwaarden R (2007) Patient selection and indications for double osteotomies at the knee German. Orthop Praxis 3:143–146 21. Paley D, Pfeil C (2000) Principles of deformity corrections at the knee German. Orthopade 29:18–38 22. Saragaglia D, Mercier N, Colle PE (2009) Computerassisted osteotomies for genu varum deformity: which osteotomy for which varus? Int Orthop 34:24–30 23. Strecker W (2006) Planning analysis of knee-adjacent deformities. I. Frontal plane deformities. Oper Orthop Traumatol 18:259–272 24. Agneskirchner J, Hurschler C, Stukenborg-Colsman C, Imhoff AB, Lobenhoffer P (2004) Effect of high tibial flexion osteotomy on cartilage pressure and joint kinematics: a biomechanical study in human cadaveric knees. Arch Orthop Trauma Surg 3:3–9 25. Bonin N, Ait Si Selmi T, Dejour D, Neyret P (2004) Flexion and extension ostetomies at the knee in the adult German. Orthopade 33:193–200 26. Baldini A, Anderson J, Cerulli-Mariani P, Kalyvas J et al (2007) Patellofemoral evaluation after TKA: validation of new weight-bearing axial radiographic view. J Bone Joint Surg Am 89:1810–1817 27. Van Heerwaarden R, Van Der Haven J, Kooijman M, Wymenga A (2003) Derotation osteotomy for correction of congenital rotational lower limb deformities in adolescents and adults. Surg Tech Orthop Traumatol 55:575–585 28. Dejour D, Walch G, Nove-Josserand L, Guir C (1994) Factors of patellar instability: an anatomical radiographic study. Knee Surg Sports Traumatol Arthrosc 2:19–26 29. Rand JA, Neyret P (2005) ISAKOS meeting on management of osteoarthritis of the knee prior to total knee arthroplasty. ISAKOS Hollywood, Florida, 1–8 30. Felson DT, Chaisson CE, Hill CL, Totterman SM, Gale ME, Skinner KM, Kazis L, Gale DR (2001) The association of bone marrow lesions with pain in knee osteoarthritis. Ann Intern Med 134:541–549 31. Agneskirchner J, Lobenhoffer P (2007) Osteotomies and ligament instability: slope corrections and combined procedures at the knee joint German. In: Lobenhoffer P, Agneskirchner J, Galler M (eds) Osteotomies around the knee. Thieme, Stuttgart, pp 79–88 32. König U, Widmer H, Friederich NF (2004) The clinical significance for the valgus osteotomy at the tibia in combination with cartilage repair German. Arthroskopie 17:234–238 33. Pape D, Seil R, Adam F, Rupp S, Kohn D, Lobenhoffer P (2004) Imaging and planning for tibia osteotomies German. Orthopade 33:122–134 34. Müller W (2001) High tibial osteotomy. In: European Instructional Course Lectures EFORT; K. Thorngren, P. Soucacos, F. HJroan, J Scott (eds) 5:194–206
187 35. Van Heerwaarden R, Wagenaar F, Hofmann S (2006) Doppelosteotomien von Femur und Tibia. In: Lobenhoffer P, Agneskirchner J, Galla M (Hrsg) (2007) Osteotomies around the knee. Thieme, Stuttgart, pp 107–118 36. Kim SJ, Koh YG, Chun YM, Kim YC, Park YS, Sung CH (2009) Medial opening wedge high-tibial osteotomy using a kinematic navigation system versus a conventional method: a 1-year retrospective, comparative study. Knee Surg Sports Traumatol Arthrosc 17:128–134 37. Wiehe R, Becker U, Bauer G (2007) Computer-assisted openwedge osteotomy German. Z Orthop Unfall 145:441–447 38. Strecker W, Dickschas J, Harrer J, Muller M (2009) Arthroscopy prior to osteotomy in cases of unicondylar osteoarthritis. German. Orthopade 38:263–268 39. Baur W, Honle W, Schuh A (2005) Proximal tibial osteotomy for osteoarthritis of the knee with varus deformity. Oper Orthop Traumatol 17:326–344 40. Freiling D, Lobenhoffer P, Staubli A (2008) The closed wedge varus osteotomy at the femur for the treatment of the valgus gonarthritis German. Arthroskopie 21:6–14 41. Frey P, Muller M, Munzinger U (2008) Closing-wedge high tibial osteotomy with a modified Weber technique. Oper Orthop Traumatol 20:75–88 42. Hooper G, Leslie H, Burn J, Schouten R, Beci I (2005) Oblique upper tibial opening wedge osteotomy for genu varum. Oper Orthop Traumatol 17:662–673 43. Geiger F, Sabo D (2004) Osteotomies at the tibia with fixateur externe German. Orthopade 33:161–169 44. Hankemeier S, Paley D, Pape HC, Zeichen J, Gosling T, Krettek C (2004) The focal dome osteotomy at the knee German. Orthopade 33:170–177 45. Gaasbeek R, Sonneveld H, Van Heerwaarden R (2006) Distal tuberosity osteotomy in open wedge high tibial osteotomy can prevent patella infera: a new technique. Knee 11:457–461 46. Staubli AE, De SC, Babst R, Lobenhoffer P (2003) TomoFix: a new LCP-concept for open wedge osteotomy of the medial proximal tibia–early results in 92 cases. Injury 34(Suppl 2): 55–62 47. Franco V, Cipolla M, Gerullo G, Gianni E, Puddu G (2004) Open wedge osteotomy of the distal femur in the valgus knee German. Orthopade 33:185–192 48. Hernigou P, Medevielle D, Debeyre J, Goutallier D (1987) Proximal tibial osteotomy for osteoarthritis with varus deformity. A ten to thirteen-year follow-up study. J Bone Joint Surg Am 69:332–354 49. Jenny JY, Tavan A, Jenny G, Kehr P (1998) Long-term survival rate of tibial osteotomies for valgus gonarthrosis French. Rev Chir Orthop Reparatrice Appar Mot 84:350–357 50. Salzmann GM, Ahrens P, Naal FD, El-Azab H, Spang JT, Imhoff AB, Lorenz S (2009) Sporting activity after high tibial osteotomy for the treatment of medial compartment knee osteoarthritis. Am J Sports Med 37: 312–318
Total Knee Replacement for the Stiff Knee Philippe Massin
Introduction Total knee replacement (TKR) in patients with stiff knees presents a major surgical challenge. Flexion contracture, or limitation of knee extension, is defined by a passive deficit of extension greater than 20°. Limitation of flexion, or stiffness of knee flexion, is defined by a maximum flexion angle of less than 90°. These two types of stiffness may exist concurrently in some patients. The type and severity of stiffness influence the clinical outcome of TKR. Knee stiffness could be due to various causes, some of which may involve additional difficulties and risks in the treatment of degenerative arthritis of the knee. The goals of TKR may vary from patient to patient. In most cases of stiff knees the primary objective is likely to be pain relief whereas the stiffness itself may be considered a secondary complaint, particularly when moderate. However, for physically active patients the priority may be the recovery of the range of movement. Knee stiffness involves intra- and extra-articular factors that should be distinguished and treated specifically. The restoration of full mobility calls for special release techniques that may increase the risks of complications. There is no simple off-the-shelf solution for stiff knees. It is essential to carry out a pre-operative analysis of the causes of knee stiffness and evaluate the patient’s needs in order to formulate a prognosis anticipating functional results and complications. Here, we briefly examine the aetiology of osteoarthritis in stiff knees and discuss the anatomical causes of stiffness. Then, we discuss the pre-operative evaluation
P. Massin Service de Chirurgie Orthopédique, CHU Bichat Claude Bernard, 46 rue Henri Huchard, Université Paris Diderot, 75877, Cedex 18 Paris, France e-mail:
[email protected]
and planning of TKR. Finally, we review the recent literature concerning the results of TKR in patients with stiff knees.
Aetiology of Osteoarthritis in Stiff Knees Classical Aetiology of Osteoarthritis in Stiff Knees Osteoarthritis of various origins can lead to knee stiffness, the principal cause being idiopathic osteoarthritis of the knee. However, a higher risk of knee stiffness is associated with some aetiologies particularly post-traumatic osteoarthritis and haemophilia [1].
Post-traumatic Knee Stiffness In a multicentre series of 128 patients with stiff knees receiving a TKR, the stiffness was due to post-traumatic osteoarthritis in 42% of cases [1]. Most knees had been subjected to multiple operations leading to the formation of superficial and intra-articular scar tissue (Fig. 1). Previous surgery for the initial fracture had sometimes been followed by prolonged immobilization or complicated by infection. Intra-articular fractures were especially at risk, in particular if they had been incompletely reduced. In fact, the precise reduction of intra-articular fractures is essential to the treatment. However, in the case of complex fractures, optimal reduction is difficult to achieve without extensive surgery and the use of multiple hardware. The persistence of a slight default of reduction in the bearing zone of the joint, together with the subchondral haematoma provoked by the trauma, impair the mechanical properties of the cartilage and initiate degenerative changes. Lower limb deformities are often associated with a failure of reduction, resulting from femoral or tibial fracture mal-union. In elderly patients, frontal deformities often result from compression tibial plateau fractures [2]. Extra-articular mal-union of
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_15, © 2011 EFORT
189
190
P. Massin
particularly in patients not receiving appropriate knee re-education. The risk of deep adhesions is greater following invasive surgical treatment of the initial fracture involving open reduction and internal fixation. Deep adhesions are likely to occur when iterative surgery is used in cases of nonunion, whether septic or not. In any case, the treatment of knee arthritis is simpler if the knee joint itself has remained free of previous surgery.
Intra-articular Factors
Fig. 1 Knee with multiple anterior scars from previous surgery, which has a high risk of skin necrosis, which could compromise flexion recovery
diaphyseal or metaphyseal tibial or femoral fractures may generate coronal, sagittal and rotational deformities. TKR can be complicated by such deformities and these should be corrected before treating the stiff knee. Overall, the incidence of knee stiffness in post-traumatic osteoarthritis of the knee is about 26%. In their series of 152 cases of posttraumatic osteoarthritis of the knee following fractures of and about the knee, Massin et al. found that knee stiffness particularly affected intra-articular mal-unions (31% developing stiffness with limitation of flexion) compared to extra-articular mal-unions (19%) [3].
Haemophilia In addition to the conventional aetiology of osteoarthritis in stiff knees, haemophilia deserves a special mention. In a series of 128 cases of knee stiffness with limitation of flexion, Massin et al. found that haemophilia was involved in 16% of cases, of which 81% presented a combined stiffness of knee extension and knee flexion [1]. Repetitive intra-articular bleeding leads to cartilage damage as well as to synovial adhesions. Thus, with this particular aetiology of osteoarthritis, severe stiffness is relatively frequent and may be the primary reason for TKR [4].
Anatomical Causes of Knee Stiffness Extra-articular Factors Extra-articular factors affecting knee flexion typically include faulty healing of a diaphyseal femoral fracture leading to adhesions in the deep layers of the extensor apparatus,
Intra-articular factors in knee stiffness arise from articular fractures (Fig. 2). Synovial adhesions may develop after an inflammatory reaction due to bleeding following the initial surgery; in some cases, the synovial adhesions may be secondary to infection. Adhesions first appear in the suprapatellar bursa while osteophytes grow under the collateral ligaments. In the long run, this leads to retraction of the extensor mechanism, provoking extension contracture. Thus, the extra-articular factors become finally associated with the intra-articular factors of knee stiffness. The retraction of the extensor mechanism is difficult to treat since it requires extensive release of the extensor apparatus [5]. Nevertheless, if quadriceps release is decided upon, the intra-articular factors causing stiffness will have to be treated simultaneously. Indeed, the optimal friction of the bearing surfaces must be restored within the same procedure, i.e., by concomitant TKR. However, the risk of infection is greater than with the standard TKR procedure [3]. Therefore, it is usual to focus only on the intra-articular factors causing knee stiffness by replacing the joint, while the extra-articular factors are left untreated. Consequently, the results in terms of joint mobility and range of motion will be sub-optimal; these drawbacks must be made clear to the patient before obtaining informed consent for the procedure.
Flexion Contracture Flexion contracture occurs frequently due to intra-articular factors such as anterior or posterior osteophytes. Impingement between prominent anterior tibial osteophytes and femoral osteophytes filling the intercondylar notch limits knee extension. Posterior osteophytes growing from the posterior condyles provoke capsular retraction. Thus, restoration of full extension requires removal of all osteophytes as well as a notch-plasty, together with a complete release of the posterior capsule. Finally, the condylar grooves must be freed from all obstacles disturbing the course of the collateral ligaments in the flexion-extension range.
191
Total Knee Replacement for the Stiff Knee Fig. 2 Intra-articular factors causing stiffness. Posterior osteophytes (4 and 5) provoke capsular retraction and induce a flexion contracture. Adhesions in the deep layer of the quadriceps muscle (1 and 2) and to Hoffa’s ligament (3) also limit flexion. Lateral osteophytes, which grow in the condylar recesses (6) disturb the course of the collateral ligament, inducing a varus or valgus contracture (Figure 2a). Adhesions in the deep layer of these collateral ligaments (7) also contribute to the limitation of flexion (Figure 2b). (Drawings from Pr Philippe Burdin, France)
a
b 2 1
7
7
4 3 5 6
Pre-operative Evaluation of Patients for TKR The pre-operative evaluation of TKR involves identification of the aetiology of knee arthritis and the various factors of knee stiffness.
Grading the Osteoarthritic Process and Measuring Deformities Grading the severity of osteoarthritis requires antero-posterior views of the weight-bearing knee, as well as lateral and skyline views. Degenerative changes must be localized and graded. Ahlbäck grade 3 [6], defined as the complete narrowing of joint space, will generally call for partial or total replacement of the joint. Severe joint narrowing in any two of the three knee compartments will require TKR. Deformities must be measured before TKR since they must be corrected to ensure long-term survival of the prosthesis. Coronal deformations can be measured on long-leg standing radiographs but these may be biased by flexion or rotational deformities. Sophisticated equipment, such as Biospacemed’s EOS 2D/3D orthopedic imager, provides more accurate measurements, but is not yet available in everyday practice [7]. Comparative CT-scan measurements of tibial and femoral
torsion are required for detecting rotational deformities, which are often underestimated but occurs frequently following the widespread use of intra-medullary nails [8]. Intra-articular deformities are localized within the capsular envelope. After adequate release and joint exposure, these deformities can be reduced and the knee balanced without difficulty. By contrast, extra-articular deformities remain irreducible despite extensive articular release. Thus, they will eventually require additional corrective osteotomy in a single or a two-stage procedure. When corrective osteotomy appears necessary, the treatment of severe stiffness should be delayed and priority be given to restoration of the femoro-tibial alignment.
Assessing the Type and Severity of Knee Stiffness In cases of moderate limitation of knee flexion, i.e., with a pre-operative flexion range between 70o and 90°, the stiffness does not require any specific treatment. By contrast, even mild flexion contracture must be corrected to restore full extension at the end of the operation. In fact, functional results may be impaired by residual flexion contracture whereas they appear to be less influenced by the magnitude of the final flexion [9, 10].
192
P. Massin
History of Previous Knee Operations The history of earlier knee operations should be fully documented. The skin over the stiff knee should be thoroughly inspected and the scars resulting from previous knee operations carefully evaluated. The surgical approach should be planned preferably through an existing lateral scar to limit the risk of wound necrosis. Previous infections are at risk of recurrence and should be listed with dates of occurrence, inflammatory status, and bacteriological history.
Determining the Patient’s Needs The age, the bone mass index and the functional needs of the patient should be taken into account to determine the best indication for TKR while minimizing operative risks and responding to the patient’s expectations.
collateral ligaments. It may be preferable to first balance the flexion gap so as to determine the height of the extension gap that has to be restored. In the last resort, the distal femoral cut may be augmented by 2 mm or, exceptionally, 4 mm [9]. Some authors recommend the preservation of the flexion gap by means of a primary tibial over-resection associated with an increase in size of the femoral component [11]. In severe flexion contracture, i.e., greater than 60°, extensive soft tissue release may be combined with augmented bone cuts; this may require highly constrained implants, should the bone resection threaten the femoral insertion of the collateral ligaments [12]. In cases of stiffness affecting the range of extension as well as that of flexion, a moderate increase in the tibial cut will provide both extension and flexion gap widening. Whatever the technique used, it should restore full extension at the end of the operation [13].
Limitation of Flexion
Operative Planning of TKR Flexion Contracture Flexion contracture of the knee generally results from intraarticular causes and TKR requires a posterior release extended laterally toward the condylar recesses (Fig. 3). The posterior release is considered complete when the extension gap can be balanced with a tension of the capsule similar to that of the
a
Fig. 3 Example of a flexion contracture induced by posterior osteophytes (Figure 3a). At the end of the operation, full extension was regained after resection of the osteophytes, thus releasing the posterior capsule. No augmentation of the distal femoral bone cut was needed (Figure 3b).
Limitation of knee flexion generally does not call for the use of any specific technique before TKR other than that currently used for optimizing joint exposure. If the lack of flexion limits joint exposure, the anterior tibial tuberosity may be osteotomized, and then re-inserted at the end of the procedure; this will necessitate a 6 week period of protected-weight-bearing, with a splint maintaining the knee in full extension. In a series of 40 cases of post-traumatic knee stiffness with limitation of flexion, the anterior tibial tuberosity was osteotomized in 35% of cases [3].
b
193
Total Knee Replacement for the Stiff Knee
According to Judet et al. [5], the release of the quadriceps with extensive release of the extensor apparatus should only be planned in cases of severe stiffness, i.e., with a flexion less than 40°, and should not be combined with additional corrective osteotomy in a single-stage procedure since the constraints applied at the site of the osteotomy during reeducation could increase the risk of secondary displacement and subsequent non-union. The Judet operation involves complete release of the quadriceps muscle together with a complete articular release of the patellar retinacula and collateral ligaments. The technique calls for an extensive lateral femoral approach to release the extensor apparatus from the femur, the medial and lateral intermuscular septum, the anterior aspect of the hip capsule, and detachment of the upper insertion of the rectus femoris from the antero-inferior iliac spine. The best indicator of the patient’s potential is the intra-operative flexion against gravity with the capsule closed at the end of the procedure. The conditions for success are the absence of multiple superficial anterior scars together with the absence of any recent infection [1].
Special Situations Such as Ankylosis or Severe Knee Stiffness In patients with ankylosis or severe knee stiffness, i.e., flexion less than 40°, complications may be expected with TKR and require the planning of special procedures. In cases with extensive scar tissue over the joint, prior plastic reconstructive surgery should be considered, especially in the presence of convergent wounds from previous surgery. Free muscular flaps with microsurgical anastomosis may be envisaged for young patients. For elderly patients, local skin flaps are the only solution. The medial gastrocnemius flap is the easiest way of covering moderate anterior defects in the neighbourhood of the patellar tendon [14]. The distally-based vastus lateralis muscle flap has recently been used for addressing proximal skin defects over the patella [15]. However, apart from these methods, there seem to be few other therapeutic solutions available. Excessive patella infera, which is another major obstacle to knee exposure and to post-operative knee re-education [16], appears to compromise the final mobility score [17]. We have developed a special technique (to be published shortly) for lengthening the patellar tendon by means of inverted autologous quadricepsplasty connected to a patellar bone fragment. As with other techniques of patellar tendon reconstruction, post-operative extension lag should be avoided by tightening the plasty in full extension. This is why it may necessary to lower the distal expansions of the vastus lateralis and medialis. Ankylosis is a special situation, in which knee exposure requires extensive articular release such as the femoral peel, involving complete release of the proximal insertion
of both collateral ligaments. Moreover, the resection of heterotopic ossification embedding the medial collateral ligament may produce major laxity. This is why the procedure should not be undertaken unless a hinged prosthesis is available [18, 19].
Results TKR in Flexion Contracture Flexion contracture must be corrected before TKR since any serious residual flexion deformity could impair functional results [20]. Specific procedures for this correction have been described, with priority being given to soft tissue release followed by augmentation of bone cuts if needed. However, mild, residual post-operative flexion deformity often persists [10, 21]. The flexion gain may be measured as a percentage of the initial flexion defect. The rate of correction was found to be fairly constant in various cases, suggesting that moderate flexion contracture may be as difficult to correct as severe flexion contracture [10].
TKR in Limitation of Flexion In knees with limitation of flexion, TKR provides substantial flexion gains, estimated at an average of 30°. The final active flexion was reported to be significantly correlated to the pre-operative flexion [1]. The flexion gain was found to be negatively correlated with the pre-operative flexion, being greater in knees with severe pre-operative stiffness. In contrast, patients with moderate pre-operative stiffness, i.e., with a flexion about 90°, are more likely to lose some degrees of flexion post-operatively. Despite somewhat disappointing results in terms of mobility, the pain and walking scores improved significantly, affording patients a better quality of life. The final range of flexion did not appear to be correlated with the overall functional result [1, 22]. It is worth mentioning that TKR in post-traumatic stiffness yielded poorer functional results than in the case of other causes of knee stiffness, and had a higher complication rate [23]. As mentioned above, tibial tubercle osteotomy is often required to facilitate exposure of stiff knees. It may be planned before the intervention in case of severe knee stiffness but it may also be performed intra-operatively if the dislocation of the extensor apparatus appears too difficult. The osteotomy should help prevent patellar tendon avulsion, which might occur during or following surgery. However, in a recent series, patellar tendon rupture occurred in spite of tibial tubercle osteotomy but this was attributed to excessive traction during re-education, in particular during late
194
mobilization under general anesthesia [3]. Our opinion, in concordance with other authors [24], is that the manipulation of the joint after TKR in pre-operatively stiff knees is rather inefficient and possibly dangerous. Patients with a patella infera are particularly at risk. Apart from this complication, tibial tubercle osteotomy did not affect the range of motion (ROM) gain following total knee replacement [3]. In the presence of extra-articular deformities, corrective diaphyseal or metaphyseal osteotomy may be required before TKR to restore the femoro-tibial alignment. This does not reduce the ROM gain in comparison with TKR in stiff knees without corrective osteotomy [3]. Judet’s technique for quadriceps release, justified in cases with severe limitation of flexion, leads to significant improvement in flexion gain [5]. This is the only procedure available for treating intra- as well as extra-articular factors of stiffness. Extensive quadriceps release offers the additional advantage of making tibial tubercle osteotomy unnecessary for TKR. However, it is associated with a high complication rate when combined with TKR, causing patellar tendon avulsion, recurrence of infection, and skin necrosis, which may jeopardize the final result [1, 3]. Patellar tendon avulsion necessitates immediate repair; however, this may restrain the intensity of the post-operative reeducation. Therefore, it would be better to reserve this procedure for patients with no multiple anterior scars, and no recent infection. Finally, it is only indicated in knees with a well-balanced extensor apparatus after prior correction of patellar mal-tracking. Generally speaking, the more severe the pre-operative stiffness, the greater is the rate of complications, with the most challenging situation being complete knee ankylosis [25].
TKR in Combined Flexion Contracture and Limitation of Flexion Knee stiffness due to a combination of flexion contracture and limitation of flexion raises difficulties since it requires the simultaneous attenuation of both defects. Idiopathic osteoarthritis, haemophilia, and inflammatory arthritis are the main causes of combined knee stiffness. TKR provides an average ROM gain of 40° compared to pre-operative values [1, 26]. Extension ROM gains were about 20° and were greater on the average than flexion gains, attesting to the primary concern of surgeons in restoring full, active extension [1]. Complication rates as high as 50% have been reported, including infections, skin necrosis, patellar tendon avulsion and peroneal nerve palsy [27]. The complications, which occurred mainly in patients with severe knee stiffness, were more frequent in the combined cases than in patients with isolated flexion contracture or limitation of flexion. Patellar
P. Massin
tendon avulsion requires direct suture or trans-osseous fixation onto the tibial tubercle, if possible. It is essential to tighten the plasty in full extension so as to prevent residual extension lag. In cases with suture fragility (to be tested intra-operatively), there should follow a 6 week period of immobilization or at least restricted passive motion range to ensure safe healing. This procedure may be expected to restore full knee extension, with a lesser improvement of the final flexion range.
Conclusions The main objective of TKR in cases of stiff knees should be the recovery of extension. The technique of release used should ensure full extension at the end of the operation. Conversely, in cases with limitation of flexion, no specific techniques other than those required for wide joint exposure, are necessary. Care must be taken not to disrupt the extensor apparatus. Tibial tubercle osteotomy, which is widely indicated for knee exposure, does not compromise the final result. In such conditions, an average flexion gain of 30° may be expected. Whatever the ROM gains, the improvement of pain and functional scores are considerable, leading to a better quality of life for the patient. It is essential to establish a clear contract with patients at the outset, discussing the symptoms that are likely to be attenuated and the eventual drawbacks of the procedure. It should be noted that the functional range of flexion, i.e., greater than 110°, can rarely be restored. Such flexion gains will require a complete quadriceps release, which can only be carried out in the absence of any recent infection or multiple anterior scars resulting from trauma or prior surgery. Post-traumatic arthritis and haemophilia are the aetiologies at risk. Finally, in terms of mid-term survival, there seems to be little difference between the results of TKR in stiff knees compared with mobile knees.
References 1. Massin P, Lautridou C, Cappelli M, Petit A, Odri G, Ducellier F et al (2009) Total knee arthroplasty with limitations of flexion. Orthop Traumatol Surg Res 95(4 Suppl 1):S1–S6 2. Saleh KJ, Sherman P, Katkin P, Windsor R, Haas S, Laskin R et al (2001) Total knee arthroplasty after open reduction and internal fixation of fractures of the tibial plateau: a minimum five-year follow-up study. J Bone Joint Surg Am 83:1144–1148 3. Massin P, Bonnin M, Paratte S, Vargas R, Piriou P, Deschamps G (2011) Total knee replacement in
Total Knee Replacement for the Stiff Knee p ost-traumatic arthritic knees with limitation of flexion. Orthopaedics & Traumatology: Surgery & Research 97: 28–33 4. Augereau B, Travers V, Le Balch T, Witvoet J (1987) Total hip and knee arthroplasties in hemophilia. Apropos of 27 cases. Rev Chir Orthop Reparatrice Appar Mot 73:381–394 5. Judet R, Judet J, Lagrange J (1956) A technique for freeing the extensor apparatus in cases of stiffness of the knee. Mém Acad Chir 82:944–947 6. Ahlbäck S (1968) Osteoarthrosis of the knee. A radiographic investigation. Acta Radiol Diagn Suppl 277:7–72 7. Schlatterer B, Suedhoff I, Bonnet X, Catonne Y, Maestro M, Skalli W (2009) Skeletal landmarks for TKR implantations: evaluation of their accuracy using EOS imaging acquisition system. Rev Chir Orthop Traumatol 95:2–11 8. Jaarsma RL, Pakvis DF, Verdonschot N, Biert J, van Kampen A (2004) Rotational malalignment after intramedullary nailing of femoral fractures. J Orthop Trauma 18: 403– 409 9. Bellemans J, Vandenneucker H, Victor J, Vanlauwe J (2006) Flexion contracture in total knee arthroplasty. Clin Orthop Relat Res 452:78–82 10. Massin P, Petit A, Odri G, Ducellier F, Sabatier C, Lautridou C et al (2009) Total knee arthroplasty in patients with greater than 20 degrees flexion contracture. Orthop Traumatol Surg Res 95(4 Suppl 1):S7–S12 11. Whiteside LA, Mihalko WM (2002) Surgical procedure for flexion contracture and recurvatum in total knee arthroplasty. Clin Orthop Relat Res 404:189–195 12. Lu H, Mow CS, Lin J (1999) Total knee arthroplasty in the presence of severe flexion contracture: a report of 37 cases. J Arthroplasty 14:775–780 13. Firestone TP, Krackow KA, JDt D, Teeny SM, Hungerford DS (1992) The management of fixed flexion contractures during total knee arthroplasty. Clin Orthop Relat Res 284:221–227 14. Ries MD, Bozic KJ (2006) Medial gastrocnemius flap coverage for treatment of skin necrosis after total knee arthroplasty. Clin Orthop Relat Res 446:186–192 15. Auregan JC, Begue T, Tomeno B, Masquelet AC (2010) Distally-based vastus lateralis muscle flap: a salvage
195 a lternative to address complex soft tissue defects around the knee. Orthop Traumatol Surg Res 96:180–184 16. Caton J, Deschamps G, Chambat P, Lerat JL, Dejour H (1982) Patella infera. A propos de 128 cases. Rev Chir Orthop Reparatrice Appar Mot 68:317–325 17. Gandhi R, de Beer J, Leone J, Petruccelli D, Winemaker M, Adili A (2006) Predictive risk factors for stiff knees in total knee arthroplasty. J Arthroplasty 21:46–52 18. Kelly MA, Clarke HD (2003) Stiffness and ankylosis in primary total knee arthroplasty. Clin Orthop Relat Res 416:68–73 19. Thienpont E, Schmalzried T, Bellemans J (2006) Ankylosis due to heterotopic ossification following primary total knee arthroplasty. Acta Orthop Belg 72:502–506 20. Ritter MA, Lutgring JD, Davis KE, Berend ME, Meneghini RM (2007) The role of flexion contracture on outcomes in primary total knee arthroplasty. J Arthroplasty 22: 1092–1096 21. Berend KR, Lombardi AV Jr, Adams JB (2006) Total knee arthroplasty in patients with greater than 20 degrees flexion contracture. Clin Orthop Relat Res 452:83–87 22. Meneghini RM, Pierson JL, Bagsby D, Ziemba-Davis M, Berend ME, Ritter MA (2007) Is there a functional benefit to obtaining high flexion after total knee arthroplasty? J Arthroplasty 22:43–46 23. Gerich T, Bosch U, Schmidt E, Lobenhoffer P, Krettek C (2001) Knee joint prosthesis implantation after fractures of the head of the tibia. Intermediate term results of a cohort analysis. Unfallchirurg 104:414–419 24. Fox JL, Poss R (1981) The role of manipulation following total knee replacement. J Bone Joint Surg Am 63:357–362 25. Bhan S, Malhotra R, Kiran EK (2006) Comparison of total knee arthroplasty in stiff and ankylosed knees. Clin Orthop Relat Res 451:87–95 26. Aglietti P, Windsor RE, Buzzi R, Insall JN (1989) Arthroplasty for the stiff or ankylosed knee. J Arthroplasty 4:1–5 27. Bae DK, Yoon KH, Kim HS, Song SJ (2005) Total knee arthroplasty in stiff knees after previous infection. J Bone Joint Surg Br 87:333–336
Part X Foot, Ankle and Leg
Surgical Treatment of Displaced Calcaneal Fractures Zvi Cohen, Gershon Volpin, and Haim Shtarker
Introduction Calcaneal fractures (2% of all fractures) are usually the result of high energy injuries, falls from a height and road traffic accidents [1]. These fractures are the most common tarsal fractures and 60–75% of them are displaced intraarticular fractures. Calcaneal fractures are more common in males (90%), mostly industrial workers, 41–45 years of age. Ten per cent have associated fractures of the spine and 25% have other extremity injuries. The economic impact is enormous since about 20% of the patients are totally incapacitated for 3–5 years [1–3]. Despite advances in imaging, surgical techniques and surgical devices the functional results of displaced intra-articular fractures are not optimal and the literature still reveals controversy surrounding classification and treatment [4]. The purpose of this review is to present the anatomical and radiological structure of the calcaneus and various clinical aspects and surgical modalities for these types of fractures.
Anatomy and Physiology The calcaneus, the largest of the tarsal bones, transmits the weight of the body to the ground, projecting backwards to provide a short lever for the muscles of the calf. Cuboidal in shape, its long axis is directed forward, upwards and laterally. On the lateral radiograph of the calcaneus traction and compression trabeculae can be seen radiating from the inferior cortex to support the posterior and anterior facet forming
G. Volpin () Department of Orthopaedic Surgery, Western Galilee Hospital, Nahariya, Israel e-mail:
[email protected]
the “neutral triangle” [5]. The superior surface consists of three articular facets with the talus: the posterior (the major weight bearing and the largest of the three), the medial or the sustentaculum tali (located on a shelf-like process) and the anterior, the calcaneal articulation with the cuboid. The motion of the calcaneus with respect to the talus (oblique sub-talar joint axis) is called supination (the calcaneus bends inward) and pronation (the calcaneus bends outward). Rotation of the calcaneus in the frontal plane (about the antero-posterior axis) is termed inversion [6].
Radiographic Evaluation The assessment of suspected calcaneal fractures and associated fractures of the spine and extremity should start with simple radiographs. We use five different views to assess calcaneal fractures: the lateral and the Broden view (the foot is in neutral position, the leg is in internal rotation of 30°, the beam is over the lateral malleolus and X-rays are taken in 40°, 30°, 20°, and 10° towards the head of the patient) to assess any incongruity, compression or rotation of the posterior facet [7]; the axial or the Harris view to assess any deformation or widening of the tuberosity; and the oblique and anterio-posterior views to assess the anterior process of the calcaneus and the calcaneo-cuboid joint [8]. Two important angles to evaluate any compression of the posterior facet are used on the simple lateral radiograph of the calcaneus (Fig. 1). The first angle, the angle of Böhler, is formed by two lines that are drawn from the tip of the anterior process of the calcaneus to the tip of the posterior facet and the second line is drawn tangential to the superior edge of the calcaneal tuberosity, creating an angle of 20 –40° [9]. The second angle, the angle of Gissane, is formed by two lines that are drawn along the lateral cortex of the posterior facet and anteriorly to the tip of the anterior process of the calcaneus, creating an angle of 100° [10]. Any decrease in the Böhler angle and increase in the
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_16, © 2011 EFORT
199
200
Z. Cohen et al.
a
b
Fig. 1 Demonstrating a 35 year-old male who fell from a roof 4 m high with comminuted fracture of his left calcaneus. The Bohler’s angle on the right side is 28° whereas on the left side it
decreased to 4° (a). The Guissane’s angle on the right side is 95° and on the left deformed calcaneus 35° (b)
Gissane angle indicates a complete posterior facet compression fracture; in the case of an isolated lateral part compression fracture of the posterior facet, the Böhler angle does not change, and double cortical lines are seen. Any calcaneal fracture and in particular intra-articular fracture should be assessed by computerised tomography, the coronal (perpendicular to posterior facet) and axial views [4, 11–15].
fracture lines that divided the calcaneus in medial and lateral fragments and in anterior and posterior fragments. Based on these findings the authors developed the theory of medial and lateral columns [11]. The most common classifications of calcaneal articular fracture are the Crosby-Fitzgibbons [12] and Sanders classifications [4]. The Essex-Lopresti classification is less common. Apparently, Crosby and Fitzgibbons were the first authors who correlated clinical outcome with a CT scan of calcaneal displaced articular fracture classification. The classification is based on three types: type I, undisplaced, type II, displaced and type III, comminuted. The Sanders classification is based on the coronal projection of the posterior facet and the sustentaculum tali, the number and the location of the fragments (Fig. 2). The posterior facet is divided into three parts: medial, central and lateral, and the fourth part is the sustentaculum tali. In type I fracture, (undisplaced fracture), the prognosis is good and conservative treatment is recommended unless the body of the calcaneus is displaced. Type II fractures involve a twopart fracture of the posterior facet that is divided into three sub-types – A, B and C, which are based on the location of the fracture line, from lateral to medial and correlate to prognostic outcomes because of the difficulty in obtaining anatomical reduction in the medial fracture. Type III fracture involves a three-part fracture of the posterior facet divided, once again, into three sub-types – A, B and C,
Classification Two types of calcaneal fracture may occur, extra-articular and intra-articular. The extra-articular fractures are the result of direct low energy trauma, twisting, muscular or ligament avulsion. The intra-articular fractures are the result of high energy injuries, falls from a height and road traffic accidents. The mechanism of injury is still controversial. Many authors have developed theories based on x-ray findings. Essex–Lopresti believed that the trauma energy hits from lateral to medial and then anterior and posterior according to the foot position [16]; he found two types, the tongue fractures and the joint depression type fractures [17]. Carr et al. created an experimental intra-articular fracture by using below-knee amputation specimens, and found two
201
Surgical Treatment of Displaced Calcaneal Fractures Fig. 2 Demonstrating the Sanders’ classification of fractures of calcaneus based on CT findings of number and location of fracture fragments as explained in the manuscript (printed with permission of Prof Roy Sanders [20])
based on the location of fracture line, from lateral to medial, and correlating with prognostic outcomes because of the difficulty in obtaining an anatomical reduction in medial fractures. Type IV fracture is formed by three fragments of the posterior facet that are usually displaced with addition of the sustentaculum tali. This type is usually comminuted with poor clinical outcome, and primary sub-talar fusion is recommended [4]. The Essex-Lopresti classification is based on the mechanism of injury and cannot predict clinical outcome: type A, Joint depression fracture, type B, Tongue-type fracture. The displaced articular surface fracture results in a tongue or jointdepression fragment. Essex-Lopresti suggested that tonguetype fractures will be reduced by percutaneous leverage and joint-depression fractures will be reduced by open reduction and internal fixation [17].
Treatment of Displaced Calcaneal Intra-Articular Fractures The principal goals in treatment of calcaneal fractures are restoration of the shape of the hindfoot, restoration of the sub-talar joint, and re-creation of a normal foot that is able to bear normal weight without pain. All treatment approaches to calcaneal fractures may be divided into two categories: conservative treatment and surgery. Until 20 years ago, calcaneal intra-articular fractures were treated mostly nonoperatively. The introduction of CT (computerised tomography) has contributed to fracture evaluation, classification, surgical technique and the development of anatomical devices for internal fixation. We present a review of the main surgical modalities used nowadays in this field.
202
Operative Treatment of Intra-articular Fractures These fractures should be treated by experienced surgeons who evaluate the patient’s ability to co-operate, age, lifestyle demands, life expectancy, possibility of diabetic neuropathy and smoking habits [18]. Elevation of the foot and a posterior padded splint in neutral position after the injury is mandatory. As soon as the swelling decreases (about 1–2 weeks, but not more than 3 weeks, as the fracture consolidates and it is difficult to obtain reduction) surgery is possible. Severe soft tissue injury, swelling or blisters can preclude surgical treatment. The severity of soft tissue injury is energy-dependent. Several methods for surgical fixation have been developed, including the extended lateral approach [4, 18–37], the minimally-invasive approach [3, 25] and the Ilizarov external fixation.
Open Reduction and Internal Fixation In the lateral approach, the patient is placed in a lateral or prone position. For the patient who has sustained a bilateral fracture and has hip anteversion (which, of course, should be checked prior to surgery), it is possible to use the supine position. A tourniquet is inflated up to 350 mm of mercury, and the foot is placed on several sheets to create an elevated “working table”. The calcaneus is approached through an “L”-shaped incision: the perpendicular incision runs anterior to the Achilles tendon sparing the Sural nerve, while the horizontal incision runs along the plantar haematoma line and curves up distally to the calcaneo-cuboid joint (Fig. 3). A full-thickness flap is elevated through the periosteum to prevent damage to the peroneal tendon and the sural nerve. When the sub-talar and calcaneo-cuboid joints are exposed, two Kirschner wires (KW) of 2 mm diameter are inserted in the talus and one KW in the cuboid, and bent upwards to be used as retractors (Fig. 4). The lateral wall is reflected or removed and the damage to the posterior facet is detected. The bone debris, haematoma and synovial tissue are removed, reduction of the articular fragments is performed under vision and fixation by KW that is pointed to the sustentaculum tali (inferior to the tip of medial malleolus), guided by fluoroscope. Fixation by canullated halfthreaded screw 3.2 mm is then performed. Correction of the compression and varus deformity of the calcaneal body is done by a 6.5-mm threaded pin that is inserted in the postero-inferior calcaneal body. The use of supplementary bone graft is controversial as it is bone-loss dependent and usually unnecessary [14, 19, 20, 38 –44]. Once the reduction is achieved, one of the prefabricated “calcaneal plates” available is inserted under fluoroscopy (Figs. 3–7). It is
Z. Cohen et al.
very important to reduce the lateral wall and any metal protrusion to prevent lateral impingement with the lateral malleolus.
Sub-talar Fusion When open reduction of a comminuted fracture is impossible or failed, a sub-talar fusion is performed. The sub-talar joint is prepared for fusion and supplementary bone graft from the iliac crest is applied in case of major bone loss. The reduction is held in placed by KW’s that are inserted under fluoroscopy. Two 7.3 mm cannulated half-threaded cancellous screws are inserted under fluoroscopy from posterior tuberosity into the anterior dome of the talus and the alignment is checked again under fluoroscopy (Fig. 7). The tourniquet is deflated, careful haemostasis is performed, and subcutaneous and cutaneous sutures are applied carefully to achieve perfect adaptation to avoid tension on the skin edges. A padded soft dressing is applied [4].
Post-operative Care The patient is instructed to prop his leg up on a pillow and to engage in non-weight bearing ambulation. He may be discharged from the hospital on the third postoperative day, after the wound has been observed for any edge necrosis or dehiscence (which is relatively common) and after a short-leg, non-weight bearing soft cast in neutral position has been applied which is retained for the next 3– 4 weeks. When the cast and the stitches are eventually removed, a removable boot is applied for another 4 weeks, and the patient is instructed to begin gentle passive and active motion. About 8–10 weeks postoperatively, X-rays are taken for bone healing verification and if healing has progressed as expected, progressive weight-bearing is allowed with physiotherapy to recover the range motion of the ankle, sub-talar and foot joints. About 3–6 months post-operatively, the patient is allowed to walk with normal shoes. Patients who have sub-talar fusion receive the same post-operative care, except for the sub-talar motion.
Results of Treatment Both Crosby and Fitzgibbons [45] and Kitaoka et al. [46, 47] have shown poor clinical results in patients treated non-operatively with displaced articular fractures and recommended operative treatment. The lateral approach is
203
Surgical Treatment of Displaced Calcaneal Fractures
a
b
c
d
Fig. 3 Demonstrating stages of operation as follows: The incisional line is in a form an L-shape (a) demonstrating exploration of the sural nerve, followed by open reduction and anatomical alignment of the bony fragments (b) and fixation by calcaneal plate (c, d)
used by most surgeons. Through this approach reduction of displaced articular and body fractures are possible. In his studies, Sanders [2, 4] found that clinical results are affected by a surgeon-dependent learning curve and it requires 35–50 cases or about 2 years’ experience. His radiographic and clinical outcome was based on CT followup and the Maryland Foot Score. Sanders achieved a good reduction of heel height, length, width, and Böhler and
Gissane angle that were almost normal regardless of fracture type. With type II fractures, 86% had radiographic anatomic reduction of the articular surface, 73% had good or excellent clinical outcome. In the remainder, 10% had fair clinical outcome and 17% were considered failures in which 50% of these required sub-talar fusion. With type III fractures, 60% had radiographic anatomic reduction of the articular surface, 70% had good or excellent clinical outcome.
204
Z. Cohen et al.
Fig. 4 Demonstrating a 32 year-old male with a comminuted fracture of his right calcaneus, grade II A (a–c) treated by open reduction and fixation by calcaneal plate (d, e)
a
c
b
d
e
As for the remainder, 10% had fair clinical outcome and 20% were considered failures of which 78% of required sub-talar fusion. With type IV fractures, no anatomic reduction was achieved, 27% had radiographic near-anatomic reduction of the articular surface, 18% had approximate reduction of the articular surface, 18% had no reduction of the articular surface. Nine percent had good or excellent
clinical outcome, 18% had fair clinical outcome and 73% were considered complete failures. Sanders et al. concluded that (1) anatomical articular reduction is mandatory to obtain excellent or good results, (2) anatomical articular reduction cannot ensure good or excellent results, probably because of injury to the cartilage at the time of impact, (3) reproducible operative
205
Surgical Treatment of Displaced Calcaneal Fractures
a
d
b
c
e
Fig. 5 Demonstrating a 47 year-old male with a comminuted fracture of his right calcaneus, grade III A (a–c) treated by open reduction and fixation by calcaneal plate (d, e)
technique is surgeon-dependent, (4) type IV fractures are so severe that a primary arthrodesis is indicated after reconstruction of the calcaneal shape, and (5) the results deteriorate over time as the number of articular fracture fragments increases (4). Similar results were also described by other authors [18, 32–37].
complications include arthritis of the sub-talar and the calcaneo-cuboid joints [29] mal-position due to varus deformation of the tuberosity [2], tendinitis or dislocation of the peroneal tendon caused by lateral impingement [2], heel pain due to the crush injury to the soft tissue. [2], mal-union of fractures that cause pain and disability and are treated by osteotomies [2], heel exostosis at the plantar aspect of the heel [2] and complex regional pain syndrome [2, 32].
Complications Injury to the sural nerve may occur using the lateral approach, while injury to the calcaneal branch of the posterior tibial nerve may occur using medial approach. The damage can cause neuroma or loss of sensation in the affected region. Nerve entrapment of the posterior tibial nerve can occur secondary to fracture mal-union [32]. The incidence of wound dehiscence and apical necrosis is 10 –13% and osteomyelitis is 1.3–2.5% in patients who undergo surgery [29, 33]. Other
Treatment of Calcaneal Fractures by the Ilizarov External Fixation Method The Ilizarov External fixation method for intra-articular calcaneal fractures is based on closed reduction and percutaneous fixation with a Ring Frame. It is a simple, easy and reliable method, with a very low complication rate and with comparable results with other methods of treatment.
206
Z. Cohen et al.
a
b
c
e
d
f
Fig. 6 Demonstrating a 54 year-old male with a comminuted fracture of his left calcaneus, grade IV (a–d) treated by open reduction and fixation by calcaneal plate (e, f )
207
Surgical Treatment of Displaced Calcaneal Fractures
The principal of fracture reduction into the frame is somewhat similar to treatment by skeletal traction, but the frame fixation allows the Orthopaedist to achieve stable fixation after reduction, more precise reduction including small
a
fragments and immediate mobilization of the patient the day after surgery with the possibility of weight-bearing depending upon how much pain the patient will tolerate (Figs. 8 and 9).
b
c Fig. 7 Demonstrating a 52 year-old male with severe comminuted fracture of his left calcaneus, grade IV (a, b) treated by open reduction and fixation by calcaneal plate and initial sub-talar fusion (c)
208
a
Z. Cohen et al.
b
d
c
e
f g Fig. 8 Demonstrating plain radiographs and CT of a 29 year-old male who fell from a building of 5 ms height and suffered a comminuted intra-articular fracture of both calcanei (a–f). The
patient was treated by closed reduction and fixation by Ilizarov frames of both lower limbs, including foot support with full weight-bearing for 2 months (g, h)
Surgical Treatment of Displaced Calcaneal Fractures
h
Fig. 8 (continued)
All procedures are performed under C-Arm X-Ray. Fixation of the proximal ring or 5/8 half-ring is done by one wire and two half-pins, followed by provisional fixation of the calcaneus to a foot frame with a single transverse KW passed through tuber calcanei as distal as possible. In some cases with severe comminution, even sub-periostal placement of a provisional wire may be acceptable. Afterwards, fixation of metatarsal bones to the foot frame is performed by additional transverse KW’s. It is important to fix those transverse pins with the foot frame during their bending in a bow-like shape. The concave side of the calcaneal pin should be towards the tuber calcanei, and the concave side of the metatarsal pin should be towards the toes. The tensioning of these two pins applies longitudinal traction to the foot and restoration of the initial length of the foot by ligamentotaxis. The next step consists of reduction of the Boehler Angle by distraction between foot frame and base; this will pull the calcaneus downward. At this stage manual reduction may be added by squeezing of calcaneus and its re-shaping. During those manipulations the sub-talar joint will be opened by distraction approximately 7–10 mm. In cases of displaced fractures and depression of bone fragments it is possible after initial reduction by ligamentotaxis, to elevate the depressed upper surface of bone, using a curved bone punch which is inserted from the lateral side into the calcaneal bone through a small incision. Additional fixation is performed by a KW through the upper anterior, almost sub-articular part of the calcaneus. If on a pre-operative CT scan considerable widening of calcaneus is noted, we use “olive wires”
209
with stoppers in order to decrease this deformity. When the desired reduction is achieved, final stabilization of calcaneal bone is performed by two or three oblique cross wires through the calcaneal body and through the tuber calcanei. If after elevation of depressed upper bone surface, a large cavity appears, then percutaneous intra-osseous injection of bone substitute is done, using Calcium Pyrophosphate. After completion of frame stabilization, additional threaded rods are inserted between the anterior arch of the foot frame to the base. Sterile dressing with Synthamycin ointment is applied around the pins and wires for 2 days. After 2 days all bandages are removed and the area is cleaned twice daily with a spray of 70% alcohol. Patients are allowed to shower and clean the skin around the pins and wires with Polydine scrub. The same treatment is maintained until frame removal. Physical therapy is started on the day after surgery. If the patient’s condition (according to presence of polytrauma) allows weight-bearing, partial weight-bearing is started to the pain-tolerance point and increased gradually. In order to achieve immediate mobilization of patients with bilateral calcaneal fractures we add an additional ring below the foot frame; this allows easier weight-bearing without direct contact between floor and heel. Leonard et al. described the use of skeletal traction through the calcaneus prior and during application of external fixation [48]. We found this unnecessary, since adequate reduction may be achieved easily by a provisional KW which connects to the most distal ring. There are instances in which ORIF, through any approach, may be contraindicated, such as severe comminution and soft-tissue compromise [48]. In such cases of severe comminution of the sub-talar joint, arthrodiastasis of this joint by Ilizarov system is very helpful, and only a few patients will need subtalar arthrodesis in future. Ilizarov EF allows the restoration of the shape of the calcaneal bone, and stable fixation of even small fragments once reduced [49]. This makes early weight-bearing possible. Early mobilization of patients with polytrauma or bilateral calcaneal fractures changes the rehabilitation period dramatically. According to Emara and Allam the functional and radiographic outcomes of this technique were similar to those of ORIF [50]. Paley and Fishgrund cited the period of prolonged non-weight-bearing as a major contributing factor, during which time the soft tissues, particularly the heel pad, become overly sensitive [51]. Ilizarov external fixation allows early weight-bearing, and helps to avoid oversensitivity of the heel pad as well as disuse osteoporosis of the foot and ankle.
210
Operative Treatment of Extra-articular Fractures Posterior Tuberosity Fracture Posterior tuberosity fracture is an avulsion fracture type, caused by excessive pull of the Achilles tendon. The fracture is considered to be a complete fracture when the entire bone insertion is detached or incomplete (open beak fracture) when there is an avulsion of the postero-superior
Z. Cohen et al.
f ragment without disconnection of the postero-inferior insertion of Achilles tendon. Diagnosis is made by a lateral radiograph view. Clinically the Thompson test should be performed to detect disruption of the Achilles insertion, and any posterior bone protrusion that can cause skin damage and give difficulty in shoe wear in the future. In cases with skin damage, posterior bone protrusion and Achilles tendon insufficiency, the fragment can be
a
b
c
d
Fig. 9 Demonstrating in the same patient post-operative radiographs and clinical appearances 6 months after injury and 4 months after removal of Ilizarov frames demonstrating fracture healing and almost normal bone alignment and
calcaneal reconstruction of both calcanei, with anatomical shape of foot. He had a full ROM and could stand on heels and toes (a–f )
211
Surgical Treatment of Displaced Calcaneal Fractures
e
f
Fig. 9 (continued)
reduced and fixed through an open or percutaneous approach with a cannulated half-threaded cancellous screw or tension band (Fig. 10). The patient is treated for 10 weeks with a non-weight-bearing removable cast or boot in a neutral position for 10 weeks until bone healing is detected on radiographs, and then gradual full weight-bearing, normal shoes and physiotherapy are allowed [2, 34].
Anterior Process Fracture The anterior process fracture is the most mis-diagnosed fracture of the calcaneus. Inversion and flexion injury of the foot cause an avulsion fracture of anterior process by the bifurcate ligament. There is localized pain over the anterior part of the calcaneus and sub-talar joint motion is limited. Diagnosis is made by lateral and medial oblique radiograph views or CT scan. Treatment is based on a nonweight-bearing removable cast or boot for 2–3 weeks with passive and active movement as tolerated. Full weightbearing is allowed 4 weeks later when swelling and pain subside. The fracture heals in 3 months. Large displaced fragments are treated by open reduction and internal fixation with cannulated small fragment cancellous screws using the lateral Ollier approach. The post-operative care is the same as the non-operative.
Triple arthrodesis is performed in cases of post-traumatic sub-talar arthritis [2].
Body Fracture (Extra-articular) Body fractures make up about 20% of calcaneal fractures. Diagnosis is made by lateral radiograph view and CT scan. Undisplaced fractures are treated with a non weight-bearing removable cast or boot for 8–10 weeks with passive and active movement as tolerated. The indications for surgery are proximal and medial displaced fractures, valgus deformity of 40° and 30° of varus [2].
Medial or Lateral Process Fractures Medial or lateral process fractures are infrequent and caused by direct injury. Diagnosis is made by lateral and axial radiographic view or CT scan. Undisplaced fractures are treated with a non-weight-bearing removable cast or boot for 8–10 weeks with passive and active movements tolerated. Displaced fragments are treated by open reduction and internal fixation with cannulated small fragment cancellous screws using a short infero-medial approach. The postoperative care is the same as for the non-operative [2].
212
a
Z. Cohen et al.
b
d c Fig. 10 Demonstrating a 49 year-old male with a “tongue-like” fracture of the calcaneus and minimal displacement of the posterior facet of the right calcaneus, (a–c) treated by closed reduction and percutaneous fixation by cannulated screws (d)
References 1. Eastwood DM, Langkamer VG, Atkins RM, Eastwood DM, Gregg PJ, Atkins RM (1993) Intra-articular fractures of the calcaneum. Part I: pathological anatomy and classification. J Bone Joint Surg 75-B(2):183–188 2. Sanders R (1999) Fractures and fracture-dislocations of the calcaneus. In: Mann R, Coughlin M (eds) Surgery of the foot and ankle, vol 2, 7th edn. Mosby, St. Louis, pp 1422–1464 3. Stulik J, Stehlik J, Rysavy M, Wozniak A (2006) Minimallyinvasive treatment of intra-articular fractures of the calcaneum. J Bone Joint Surg Br 88-B:1634–1641 4. Sanders R (2000) Displaced intra-articular fractures of the calcaneus. J Bone Joint Surg Am 82-A:225–250
5. Harty M (1973) Anatomic considerations in injuries of the calcaneus. Orthop Clin North Am 4:179–183 6. Stephen S (2005) Mechanics of the subtalar joint and its function during walking. Foot Ankle Clin 10:425–442 7. Brodén B (1949) Roentgen examination of the subtaloid joint in fractures of the calcaneus. Acta Radiol 31:85–91 8. Isherwood I (1961) A radiological approach to the subtalar joint. J Bone Joint Surg 43-B(3):566–574 9. Böhler L (1931) Diagnosis, pathology, and treatment of fractures of the os calcis. J Bone Joint Surg 13:75–89 10. Gissane W (1947) Discussion on “Fractures of the os calcis” (Proceedings of the British Orthopaedic Association). J Bone Joint Surg 29:254–255 11. Carr JB, Hamilton JJ, Bear LS (1993) Experimental intra-articular calcaneal fractures: anatomic basis for a new
Surgical Treatment of Displaced Calcaneal Fractures classification. Foot Ankle 10:81–87; J Bone Joint Surg Br 75-B:189–195 12. Crosby LA, Fitzgibbons T (1990) Computerized tomography scanning of acute intra-articular fractures of the calcaneus: a new classification system. J Bone Joint Surg 72-A:852–859 13. Segal D, Marsh JL, Leiter B (1985) Clinical application of computerized axial tomography (CAT) scanning of calcaneus fractures. Clin Orthop Relat Res 199:114–123 14. Stephenson JR (1983) Displaced fractures of the os calcis involving the subtalar joint: the key role of the superomedial fragment. Foot Ankle 4:91–101 15. Zwipp H, Tscherne H, Thermann H, Weber T (1993) Osteosynthesis of displaced intraarticular fractures of the calcaneus. Results in 123 cases. Clin Orthop Relat Res 290: 76–86 16. Essex-Lopresti P (1952) The mechanism, reduction, technique, and results in fractures of the os calcis. Br J Surg 39: 395–419 17. Essex-Lopresti P (1993) Surgical Treatment of Displaced Calcaneal Fractures. Clin Orthop Relat Res 290:3–16 18. Poeze M, Verbruggen J, Brink P (2008) The relationship between the outcome of operatively treated calcaneal fractures and institutional fracture load, a systematic review of the literature. J Bone Joint Surg 90-A:1013–1021 19. Leung KS, Yuen KM, Chan WS (1993) Operative treatment of displaced intra-articular fractures of the calcaneum. Medium-term results. J Bone Joint Surg 75-B(2):196–201 20. Sanders R, Fortin P, Dipasquale T, Walling A (1993) Operative treatment in 120 displaced intraarticular calcaneal fractures. Results using a prognostic computed tomography scan classification. Clin Orthop Relat Res 290:87–95 21. Benirschke SK, Sangeorzan BJ (1993) Extensive intraarticular fractures of the foot. Surgical management of calcaneal fractures. Clin Orthop Relat Res 292:128–134 22. Bèzes H, Massart P, Delvaux D, Fourquet JP, Tazi F (1993) The operative treatment of intraarticular calcaneal fractures. Indications, technique, and results in 257 cases. Clin Orthop Relat Res 290:55–59 23. Buckley RE, Meek RN (1992) Comparison of open versus closed reduction of intraarticular calcaneal fractures: a matched cohort in workmen. J Orthop Trauma 6:216–222 24. Eastwood DM, Langkamer VG, Atkins RM (1993) Intraarticular fractures of the calcaneum. Part II: open reduction and internal fixation by the extended lateral transcalcaneal approach. J Bone Joint Surg 75-B(2):189–195 25. Fernandez DL, Koella C (1993) Combined percutaneous and “minimal” internal fixation for displaced articular fractures of the calcaneus. Clin Orthop Relat Res 290:108–116 26. Hutchinson F III, Huebner MK (1994) Treatment of os calcis fractures by open reduction and internal fixation. Foot Ankle Int 15:225–232 27. Melcher G, Bereiter H, Leutenegger A, Ruedi T (1991) Results of operative treatment for intra-articular fractures of the calcaneus. J Trauma 31:234–238 28. Melcher G, Degonda F, Leutenegger A, Ruedi T (1995) Tenyear follow-up after operative treatment for intra-articular fractures of the calcaneus. J Trauma 38:713–716
213 29. Sanders R (1992) Intra-articular fractures of the calcaneus: present state of the art. J Orthop Trauma 6:252–265 30. Sanders R, Gregory P (1995) Operative treatment of intraarticular fractures of the calcaneus. Orthop Clin North Am 26:203–214 31. Thordarson DB, Krieger LE (1996) Operative vs. nonoperative treatment of intra-articular fractures of the calcaneus: a prospective randomized trial. Foot Ankle Int 17:2–9 32. Myerson M, Quill GE Jr (1993) Late complications of fractures of the calcaneus. J Bone Joint Surg 75:331–341 33. Sangeorzan BJ, Benirschke SK, Carr JB (1995) Surgical management of fractures of the os calcis. In: Instructional course lectures, American Academy of Orthopaedic Surgeons, vol 44. American Academy of Orthopaedic Surgeons, Rosemont, pp 359–370 34. Rowe CR, Sakellarides HT, Freeman PA, Sorbie C (1963) Fractures of os calcis. A long term follow-up study of one hundred fortysix patients. JAMA 184:920–923 35. Herscovici D, Widmaier J, Scaduto JM, Sanders R, Walling A (2005) Operative treatment of calcaneal fracture in elderly patients. J Bone Joint Surg 87-A:1260–1264 36. Potter MQ, Nunly JA (2009) Long – Term functional outcome after operative treatment for intra-articular fracture of the calcaneus. J Bone Joint Surg 91-A:1854–1860 37. Rammelt S, Zwipp H (2004) Calcaneal fracture: facts, controversies and recent developments. Injury 35:443–461 38. Palmer I (1948) The mechanism and treatment of fractures of the calcaneus. Open reduction with the use of cancellous grafts. J Bone Joint Surg 30-A:2–8 39. Letournel E (1984) Open reduction and internal fixation of calcaneal fractures. In: Spiegel P (ed) Topics in orthopedic surgery. University Park Press, Baltimore, pp 173–192 40. Letournel E (1993) Open treatment of acute calcaneal fractures. Clin Orthop Relat Res 290:60–67 41. Stephenson JR (1987) Treatment of displaced intraarticular fractures of the calcaneus using medial and lateral approaches, internal fixation, and early motion. J Bone Joint Surg 69-A:115–130 42. Stephenson JR (1993) Surgical treatment of displaced intraarticular fractures of the calcaneus. A combined lateral and medial approach. Clin Orthop Relat Res 290:68–75 43. Leung KS, Chan WS, Shen WY, Pak PP, So WS, Leung PC (1989) Operative treatment of intraarticular fractures of the os calcis - the role of rigid internal fixation and primary bone grafting: preliminary results. J Orthop Trauma 3: 232–240 44. O’Farrell DA, O’Byrne JM, McCabe JP, Stephens MM (1993) Fractures of the os calcis: improved results with internal fixation. Injury 24:263–265 45. Crosby LA, Fitzgibbons TC (1996) Open reduction and internal fixation of type II intra-articular calcaneus fractures. Foot Ankle Int 17(5):253–258 46. Kitaoka HB, Alexander IJ, Adelaar RS, Nunley JA, Myerson MS, Sanders R (1994) Clinical rating systems for the ankle hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int 15:349–353 47. Kitaoka HB, Schaap EJ, Chao EYS, An K-N (1994) Displaced intra-articular fractures of the calcaneus treated non-operatively. Clinical results and analysis of motion and
214 ground-reaction and temporal forces. J Bone Joint Surg 76-A:1531–1540, 17: 253–258, 1996 48. Leonard M, Talarico DPM, Vito GR et al (2004) Management of displaced intraarticular calcaneal fractures by using external ring fixation, minimally invasive open reduction, and early weightbearing. J Foot Ankle Surg 43(1):43–50 49. Gupta V, Kapoor S, Clubb S, Davies M, Blundell C (2005) Treatment of bilateral open calcaneal fractures with Ilizarov frames. Injury 36:1488–1490
Z. Cohen et al. 50. Emara KM, Allam MF (2005) Management of calcaneal fracture using the Ilizarov technique. Clin Orthop Relat Res 439:215–220 51. Paley D, Fishgrund J (1993) Open reduction and circular external fixation of intraarticular calcaneal fractures. Clin Orthop Relat Res 290:125–131
Forefoot Deformities Wolfgang Schneider
Introduction Diagnosis of different forefoot deformities primarily is a clinical process. Mostly the diagnosis of forefoot deformities initiates the discussion concerning possible surgical solutions. This selection of operative procedures necessitates radiographic techniques, mostly plain radiographs. So in most cases clinical examination and radiographic assessment run parallel. Success in forefoot surgery depends on perfect technical realisation of pre-operative planning, and this needs deep understanding of how a certain procedure can influence pathologic conditions in biomechanics of the foot. But the basis of all surgery is primarily to select the right surgical procedure. As a result of the complex anatomy of bony structures, soft tissues and biomechanical function of forefoot, hindfoot and ankle, together with the need for permanent weight-bearing, the foot is one of the most complex structures in Orthopaedics. So, selecting the right operative procedure is the crucial point for success in forefoot surgery.
Decision-Making in Hallux Valgus Surgery Most algorithms for choosing an appropriate surgical method for the treatment of hallux valgus deformities include only very few parameters like intermetatarsal angle 1–2, first metatarsophalangeal angle or signs of degeneration of the first MTP-joint [1–3]. To meet the complexity of hallux valgus deformity, much more has to be considered:
W. Schneider Herz-Jesu Hospital Vienna, A-1030 Baumgasse 20A, Vienna, Austria e-mail:
[email protected]
Different radiographic (see Table 1) and clinical parameters (see Table 2) are assessed concerning their importance in decision-making for hallux valgus surgery:
Radiographic Parameters As one of the main parameters for the biomechanical function of the first ray, the first intermetatarsal angle [4, 5] is most important for selecting an operative procedure to realign a symptomatic foot. On the one hand, the corrective effect of an osteotomy reducing the intermetatarsal angle corresponds to the site of osteotomy: The more proximal the osteotomy, the more aggressive the amount of correction required. From a purely trigonometrical point of view, distal osteotomies like Chevron (= Austin) [6 –9], Scarf [10, 11] or Mitchell [12] are limited at a certain intermetatarsal angle. This trigonometrical corrective effect increases with midshaft osteotomies like the Ludloff procedure [13, 14] and has the most pronounced effect with proximal osteotomies like opening – or closing-wedge or crescentic osteotomies [15–17]. A further increase in corrective effect of the first intermetatarsal angle can be accomplished with a corrective arthrodesis of the first tarsometatarsal joint [18, 19]. On the other hand, the first intermetatarsal angle can be reduced using the articular function of the first tarsometatarsal joint using a distal soft tissue procedure. This normal mobility of the first TMT joint is a prerequisite for a correct working distal soft tissue procedure, approximating the first and the second metatarsal by repositioning the first metatarsal head onto the sesamoid complex [20]. On the contrary, an immobile first tarso-metatarsal joint will hinder the reduction of first intermetatarsal angle using a soft tissue procedure and thus necessitates a potent bony corrective procedure. Hypermobility of the first tarsometatarsal angle either will require a stabilizing distal soft tissue procedure [21] or fusion of the TMT joint [22]. So the two parameters intermetatarsal angle and mobility of the first TMT joint always have to be considered together.
G. Bentley (ed.), European Instructional Lectures. European Instructional Lectures 11, DOI: 10.1007/978-3-642-18321-8_17, © 2011 EFORT
215
216
W. Schneider
Table 1 Radiographic parameters and their implications on hallux surgery
Table 2 Clinical diagnostic criteria and their implications on hallux valgus surgery
Radiographic parameter
Surgical implication
Clinical parameter
Surgical implication
First IM angle
Site of metatarsal osteotomy (distal – midshaft – proximal – TMT-joint)
ROM of MTP joint
Necessity of shortening of the first metatarsal
First MTP angle
Necessity of distal soft tissue procedure
First IP angle
Necessity of phalangeral osteotomy or IP fusion
Width of the first metatarsal head
Possible amount of lateral displacement Necessity of fixation of osteotomy
Osteoarthrosis first MTP joint
Necessity of shortening of the first metatarsal Necessity of Cheilectomy (or further hallux rigidus procedures)
Congruency of first MTP joint
Site of osteotomy (metatarsal – phalangeal) Necessity of distal soft tissue procedure
Sesamoid luxation grade
Necessity of distal soft tissue procedure
Proximal articular set angle PASA
Necessity of correction of PASA Selection of suitable metatarsal osteotomy
Distal articular set angle
Necessity of phalangeral osteotomy
Metatarsal index 1–2
Possible amount of shortening of first metatarsal
Bone quality
Site and type of metatarsal osteotomy Necessity of fixation of osteotomy After treatment
Primary or revision surgery
Site and type of metatarsal osteotomy
The first metatarsophalangeal angle in a certain way corresponds to the first intermetatarsal angle, so a normalisation of the first IM angle will diminish the MTP angle automatically; the complete correction of the deformity has to be done with the soft tissue procedure. The width of the first metatarsal head limits the possible amount of lateral displacement of the head and therefore the corrective effect of different procedures. The lateral displacement in distal metatarsal osteotomies can be increased
Necessity of Cheilectomy (or further hallux rigidus procedures) ROM of IP joint
Prerequisite for fusion of MTP joint
Mobility of first TMT-joint
Efficacy of distal soft tissue procedure Necessity of proximal metatarsal osteotomy (in immobile TMT joint)
Level of activity
Stability of procedure
Shoe wear habits
Possible problems with fusion of MTP joint
Pathologies of second ray
Special need for stabilizing first ray
Instability or luxation of MTP joint
Shortening of second metatarsal with consequence for first ray surgery
by correct techniques of osteosynthesis allowing displacement of more than half of the width [10] or even displacement of complete width of the metatarsal head (Bösch [23, 24], SERI [25], Kramer [26], Stoffela [27]). In such a case of bony overcorrection, sometimes the procedure will work even with undercorrection of the soft tissues. The sesamoid luxation grade is a measure of the necessity of a distal soft tissue procedure. The goal of the soft tissue procedure is to reposition the metatarsal head onto the sesamoid complex: with a correct technique and the correct bony procedure, this will result in a correction of the first IM angle. Incongruency of first metatarsophalangeal joint has to be corrected to obtain a congruent joint. This needs the correct selection of metatarsal or phalangeal osteotomy and soft tissue procedure. On the other hand, a congruent joint may not be worsened into an incongruent joint by a wrong procedure. The proximal articular set angle PASA (= distal metatarsal articular angle DMAA) is assessed as an important parameter in selection of metatarsal osteotomies, as most proximal osteotomies worsen pathologically-high PASA values and most distal osteotomies can improve PASA. But scientific discussion could not really prove the importance of this parameter. Besides, the radiographic assessment of this value has a bad correlation with the clinical or anatomic situation [28–30]. The first metatarsal index (relative length of the first and second metatarsal) gives information about necessary
217
Forefoot Deformities
or possible shortening of the first metatarsal. Shortening of the first metatarsal makes correction of hallux valgus deformity easier and can help to improve range of motion in early cases of concomitant hallux rigidus. The distal articular set angle gives information about the necessity of osteotomies of the base of the phalanx, as does the first interphalangeal angle. Radiographic signs of degeneration of the first MTP joint will influence the selection of procedure and has to be assessed together with parameters like clinical symptoms, range of motion and first metatarsal index. Pathologies of the second ray have to be considered as far as the situation of the first ray is concerned. Especially instability, subluxation or luxation of the second tarsometatarsal joint have to be addressed, and these conditions require special care for the first ray to unload the second ray. Radiographic signs of poor bone quality have to be taken into account to ensure sufficient osteosynthesis or adjusted mobilisation in cases of general osteoporosis or to select an appropriate surgical procedure in cases of local cysts.
Clinical Parameters From a clinical point of view, the quantity and quality of complaints of the patient have to be analysed (exertional pain or pain even at rest, shoe wear problems, limp, problems on uneven surfaces, walking distance, need of support) to assess grade and reason of subjective problems. These clinical parameters are included in most clinical scoring systems [31, 32]. For decision-making, clinical valgus deformity of the great toe and presence of a bunion corresponds to radiographic measurements of first intermetatarsal angle and metatarsophalangeal angle. The range of motion of first MTP joint gives a good clinical indication of possible degenerative joint disease of the joint. Most authors require a minimum total range of motion of at least 50° [dorsal- plus plantarflexion] to consider a joint-preserving technique. A decreased range of motion may be an indication for metatarsal shortening to reduce the pressure in the first MTP joint and to facilitate correction of higher grade valgus deformity. The possibility of shortening has to be confirmed radiographically and should be avoided in cases of a short first metatarsal. Normal mobility of first tarsometatarsal joint has to be checked clinically, as this physiologic TMT mobility is a pre-requisite for a functioning distal soft tissue procedure, and pathologic conditions like hypermobility or immobility have to be addressed by a suitable surgical procedure. Calluses under the lesser metatarsal head as sign of metatarsal overload indicate the necessity to recreate a
load bearing first ray by correction of hallux valgus deformity, or special techniques of plantar displacement of the first metatarsal head. Similar evidence for metatarsal overload is the presence of hammer toes, or instability of the lesser metatarsophalangeal joints. The possibility of revision surgery may influence the operating technique – mostly to avoid the same site of osteotomy (in cases where the correct method was applied, but failed, or in cases where an inappropriate method was used primarily). The age of the patient was part of some older algorithms in decision-making for hallux surgery, but various publications showed no worse results depending on the age of the patient [33–35]. The decision should rather be influenced by bone quality, level of activity, local circulatory disturbances or other concomitant local or general diseases.
Surgical Procedures for Hallux Valgus Deformity Arthrodesis of first IP-joint has its indication in cases of degenerative or especially arthritic changes of the IP joint or neuropathic disorders [36] combined with valgus deformity of this most distal joint. Various types of Akin osteotomy [37, 38] (distal, midshaft, proximal) have their place in cases with deformity of the proximal phalanx. A distal Akin osteotomy is a perfect solution only in cases with valgus deformity near the IP joint with an intact joint itself. A proximal Akin osteotomy is indicated in hallux valgus cases with the deformity located at the base of the proximal phalanx combined with a congruent MTP joint [39]. Various types of distal metatarsal Osteotomies (Austin or Chevron, Mitchell) are indicated in mild to moderate hallux valgus deformities [9, 34, 35]. From a pure trigonometrical point of view [40 – 43], the corrective effect of all distal metatarsal osteotomies is limited by the width of the metatarsal head, but the indication for all these osteotomies can be stretched by combining them with an appropriate distal soft tissue procedure [20]: this helps to reposition the metatarsal head onto the sesamoid complex (provided there is normal mobility of the first tarsometatarsal joint) (Fig. 1). Correct surgical technique provided, the combination of distal metatarsal osteotomies and distal soft tissue procedures will not increase the risk of osteonecrosis of the metatarsal head [44]. Distal metatarsal osteotomies allow easy shortening of the first metatarsal [45], as well as correction of pathologic PASA angles; most distal osteotomies even allow variable plantar displacement of the metatarsal head.
218
A distal soft tissue procedure should be part of surgery in all cases with soft tissue pathologies around the first MTP joint (which is the case in the vast majority of hallux valgus deformities!). The aim of a correct distal soft tissue procedure is to release the contracted ligamentous structures on the lateral side of the MTP joint (especially the lateral collateral ligament and the lateral suspensory ligament [i.e. lateral metatarso-sesamoid ligament]) and to reconstruct the elongated ligamentous structures on the medial side of the joint (the medial collateral ligament and the medial suspensory ligament [i.e. medial metatarso-sesamoid ligament]) with the effect of reposition the first metatarsal head onto the sesamoid complex normalizing the first intermetatarsal angle. This type of distal soft tissue procedure increases the corrective effect of all types of first metatarsal osteotomies [46]. The Scarf osteotomy [10, 11, 47] has similar indications as most distal metatarsal osteotomies, only the bony correction can be extended to higher corrective lateral displacement due to stable osteosynthesis. On the other hand, shortening of the metatarsal is more demanding, and PASA correction is very limited.
Fig. 1 Hallux Valgus, Metatarsalgia: Indication for distal first metatarsal osteotomy (Austin/Chevron, Scarf, …) with distal soft tissue procedure; metatarsalgia should be addressed by correction and stabilisation of first ray only
W. Schneider
The Ludloff osteotomy [13, 14] has become more and more popular since its “re-invention” [48, 49] two decades ago due to its high potency for correcting the first intermetatarsal angle [50–52]. But with its broader use, this procedure showed certain problems of stability even with good osteosynthesis [53]. Proximal metatarsal osteotomies have the highest potency to correct the first intermetatarsal angle. This advantage stands in contrast to some disadvantages like worsening of PASA in most cases, lengthening of the metatarsal (in cases of opening wedge osteotomies), high demands on fixation [54] and post-operative immobilisation. Proximal opening wedge osteotomies fixed with new generation locking plates promised easy-to-perform unlimited correction of first intermetatarsal angle, but the experience of the last years showed these procedures to be highly demanding and not risk-free [54–56]. The indication for proximal metatarsal osteotomies are hallux valgus deformities with higher intermetatarsal angles (more than 17–20°), and especially cases with an immobile tarsometatarsal joint, that prevents correct functioning of the distal soft tissue procedure (Fig. 2). Special attention has to be paid to avoid dorsal displacement of the
Fig. 2 Hallux Valgus: Indication for proximal first metatarsal osteotomy with distal soft tissue procedure due to immobile but otherwise asymptomatic first tarsometatarsal joint
219
Forefoot Deformities
first metatarsal head [57], using correct surgical technique, fixation and immobilisation. Fusion of the first tarsometatarsal joint has – together with proximal metatarsal osteotomies – the most pronounced corrective effect on first intermetatarsal angle [58, 59]. It shares most of surgical risks with proximal metatarsal osteotomies. These are risk of elevation of the first ray (even higher due to shortening after fusion), risk of mal-union [60], worsening of PASA, high demands on the osteosynthesis, and necessity for non-weight-bearing mobilisation. The indication for first TMT fusion should be restricted to highly increased first intermetatarsal angles or cases with painful degenerative or arthritic changes of the first TMT joint or in cases of instability of this joint. Resection arthroplasty of first MTP joint should be restricted to resection of the base of the proximal phalanx. This technique according to Keller [61] and Brandes [62] was successfully modified by adding a distal soft tissue procedure according LeLievre [20]. Resection arthroplasty still has its indications in cases of concomitant degenerative changes of the joint in less active patients [33, 63]. Fusion of the MTP joint is indicated in cases of high grade deformity combined with degenerative changes in the joint. Fusion of the joint is the safest way to stabilize not only the first ray, but also the rest of the foot. It is especially indicated in cases of rheumatoid arthritis [64, 65] or in neuropathic deformities. In the last years, minimally-invasive techniques have gained some interest in foot surgery in the wake of MIS techniques in hip and knee surgery. In hallux surgery, most of these techniques are modifications of phalangeal osteotomies like the Akin osteotomy and of distal metatarsal osteotomies (Bösch [23, 24, 66], SERI [25]). The advantage of quicker surgery with reduced soft tissue trauma stands in conflict with relinquishing of concomitant soft tissue procedures.
Decision-Making in Hallux Rigidus The main criteria in decision making for hallux rigidus surgery are degree and location of clinical symptoms and radiographic stage of degenerative joint disease. These two parameters of course show certain correlations but have to be assessed independently.
Radiographic Parameters Degenerative changes in the first metatarsophalangeal joint are indicated by narrowing of the joint space,
irregularity of the joint line, sclerosis of subchondral bone, osteophytes and the condition of the metatarso-sesamoid joint. These parameters are graded according to Regnauld [67] on a scale of 0 (normal joint) to 3 (end-stage osteoarthrosis). These degenerative changes have to be assessed in detail and in combination with clinical symptoms: Osteophytes alone will probably need only simple removal or cheilectomy. In combination with destruction of the joint cheilectomy alone will not be successful. In those more pronounced cases of degenerative joint disease shortening and/or wedge osteotomies will be necessary to decrease tension in the joint and to optimize range of motion. Severe degeneration or destruction of the joint necessitate fusion, replacement or resection of the joint. Selection in detail has to be made according to other clinical and radiographic symptoms. Degenerative changes in the first IP joint are essential to assess, because normal function of the IP joint is pre-requisite for fusion of the MTP joint and should therefore be judged thoroughly radiographically and clinically. A symptomatic IP joint has to be addressed independently or in combination with concomitant symptoms. Malalignment of the first ray has to be considered in terms of the potency of the selected surgical procedure not only to address the rigidus problem but also to treat the valgus component.
Clinical Assessment Location of pain has to be assessed to distinguish between tenderness of single osteophytes – especially in shoes, dorsal impingement during dorsiflexion or general arthritic pain, implicating simple removal of osteophytes, cheilectomy or more invasive surgery, respectively. Actual and desired level of activity has to be assessed to select a surgical method to meet the requirements of the patient. Shoe-wear habits, especially the desire to wear higher than average heels have to be discussed with the patient, especially in cases where fusion of the joint would be indicated from a medical point of view.
Surgical Procedures for Hallux Rigidus Simple removal of osteophytes is indicated in cases of local tenderness of osteophytes with only minor degenerative changes of the first MTP joint. This has to be assessed primarily clinically by sufficient and pain-free range of motion. In cases with impingement during dorsiflexion of
220
the great toe, Cheilectomy is indicated in early stages [68–70], as long as the joint is pain-free within its free range of motion. It can be seen as an interim solution to gain time until more aggressive procedures are necessary. Wedge osteotomies for treatment of hallux rigidus aim to shift the pre-existing range of motion of the MTP joint towards dorsiflexion to improve gait pattern, but are unable to increase the range of motion. Moberg’s osteotomy [71] contains of removal of a dorsally-based wedge of the base of the proximal phalanx. Watermann’s osteotomy [72] tries to achieve the same effect by removal of a dorsallybased wedge in the metatarsal head. Due to the nature of a subcapital osteotomy Watermann’s procedure has a high risk of avascular necrosis of the metatarsal head or dorsal displacement of the metatarsal head with concomitant transfer metatarsalgia [73]. Shortening osteotomies try to improve range of motion of the MTP joint by reducing the pressure in the joint (Fig. 3). The Youngswick osteotomy [45] combines a
Fig. 3 Hallux rigidus: Indication for first metatarsal shortening osteotomy (Youngswick) due to relatively good range of motion, missing arthritic symptoms and overlong first metatarsal
W. Schneider
V-shaped distal metatarsal osteotomy (like Austin or Chevron osteotomy) with removal of a bony slice in the dorsal part of the osteotomy: This leads to shortening and plantar displacement of the metatarsal. The amount of shortening depends of the thickness of the bony slice, and plantar displacement depends on the direction of the plantar osteotomy. In cases of concomitant valgus deformity, the metatarsal head can be displaced laterally, as in a typical Austin osteotomy. Shortening of the proximal phalanx in the technique by Regnauld [74] cannot be recommended any more due to the high risk of avascular necrosis. In cases of advanced degeneration or destruction of the MTP joint, preservation of the joint will not be possible. In this case, fusion, joint replacement or resection arthroplasty are the remaining surgical options. Joint replacement can be divided into procedures replacing only the articular surface of the base of the proximal phalanx or the metatarsal head and into total joint replacement. Materials used in artificial replacement are silicone, metal, polyethylene and ceramics. Silicone arthroplasty failed in terms of a long-lasting solution due to massive foreign body reaction with severe osteolysis [75, 76]. Metal and ceramics [77] showed better results, but in most publications these implants failed in the long run [78, 79]. The best results published are for distal metal hemi-arthroplasty [80]. In general, results following artificial joint replacement are disappointing in comparison to contemporary hip and knee replacement [81]. Resection arthroplasty has become unpopular due to published problems in revision cases [82, 83]. Today, resection of the metatarsal head is obsolete due the high risk of instability of the great toe with high rate of transfer metatarsalgia. Resection of the base of the proximal phalanx according the technique of Keller still can be seen as a valuable procedure in older patients with lower demands [84] (Fig. 4). Fusion of the MTP joint is the most stable procedure for high demand patients (Fig. 5), but a certain revision rate due to mal-union or misalignment has to be considered [85]. Fusion of the joint will be the safest way to stabilize not only the first ray, but also the rest of the foot, especially in cases of instability of the complete forefoot, for example in rheumatoid arthritis [64, 65]. Fusion of the MTP joint needs an intact IP joint, otherwise the painful problem will only be shifted distally. Fusion is the safest way to revise failed hallux valgus and hallux rigidus procedures [83, 86]. Results of different procedures for hallux rigidus are summarized in Table 3.
Forefoot Deformities
Fig. 4 Hallux rigidus: Indication for resection arthroplasty (Keller) due to end-stage osteoarthrosis and clinical symptoms, old age of the patient, circulatory disturbances and low demand
Decision-Making in Metatarsalgia Surgery First of all, the diagnosis of metatarsalgia has to be defined more exactly, as “metatarsalgia” gives no evidence of the pathogenetic mechanism. Clinically the quality of pain has to be determined exactly, as the clinical and radiographic diagnosis in most cases implies the surgical treatment.
Clinical Assessment Static metatarsalgia as an overload syndrome in a clinically and radiographically otherwise normal foot is no indication for surgery and should be treated conservatively. In cases where static metatarsalgia can be seen as consequence of instability of the first ray, pathology in the first ray has to be addressed primarily, which is the case in most hallux valgus patients. In these cases stabilization of the first ray normally leads to reduction of metatarsalgia without treatment of the lesser rays themselves (Fig. 1).
221
Fig. 5 Hallux rigidus: Indication for fusion of MTP joint despite only minor degeneration of the joint, but severe synovitic pain, young age and very high demands for sporting activities
Plantar displacement of metatarsal heads can be suspected by clinical examination and should be verified radiographically. Plantar displacement of single metatarsals need surgical procedures to selectively elevate the metatarsal head. Dynamic metatarsalgia as sign of instability or arthritic irritation of the MTP joint has to be addressed at the site of MTP joint itself by synovectomy. The instability of the MTP joint has to be addressed by PIP-joint fusion or tendon surgery to regain flexor function in the MTP joint. In selected cases shortening of the metatarsal can be helpful to stabilize the joint. Tendinitis of the flexor tendons is a rare condition and should be treated primarily conservatively or according to the underlying cause. Intermetatarsal bursitis necessitates only bursectomy, but cases with sole intermetatarsal bursitis as reason for metatarsalgia are rare. Nerve compression syndromes like Morton’s Neuroma or tarsal tunnel syndrome have to be treated surgically with decompression – or resection in the
222
W. Schneider
Table 3 Rate of revisions and clinical outcome following hallux rigidus surgery Surgical procedure
Rate of revisions%
AOFAS score
Follow-up (years)
Number of cases
Easley et al. [69]
Cheilectomy
4.4
85
5.4
68
Feltham et al. [87]
Cheilectomy
7.0
80
5.2
53
Lombardi [88]
Arthrodesis
4.8
47.8
2.4
21
Lau and Daniels [89]
Cheilectomy + wedge osteotomy
12.0
n.a.
2.4
24
DeFrino et al. [90]
Arthrodesis
0
81
2.8
10
Coughlin and Shurnas [68]
Cheilectomy
7.5
90
9.6
93
Coughlin and Shurnas [68]
Arthrodesis
6.6
89
6.7
30
Ettl et al. [91]
Arthrodesis
0
53
4.3
36
Raikin et al. [79]
Hemiarthroplasty
23.8
72
6.5
21
Raikin et al. [79]
Arthrodesis
0
84
2.5
27
Wassink and van den Oever [92]
Arthrodesis
4.6
n.a
5.8
109
Brewster et al. [77]
Total joint replacement
6.3
74
4.3
32
Schneider et al. [84]
Resection arthroplasty
5.7
82.5
22.8
87
case of Morton’s neuroma – when conservative treatment has failed. Affections of the metatarsal bone itself like Köhler’s disease or bone marrow oedema syndrome and stress fractures of the metatarsal shaft have to be confirmed radiographically, in the early stages using MRI or bone scan in selected cases. Treatment depends on the stage of disease and subjective clinical symptoms.
Radiographic Assessment Radiographically, alignment of the metatarsal heads has to be evaluated to quantify length discrepancies of single metatarsals: Shortening of a single metatarsal in cases of symptomatic isolated overlength is indicated; special attention has to be dedicated to correct dorsoplantar alignment during surgery. Minor divergences of metatarsal lengths from an ideal alignment (according the perfect metatarsal parabola [93]) may not be of clinical significance and should not be treated by surgery automatically. Simple radiographic techniques allow quantification of dorsal/plantar displacement of metatarsal heads. Isolated plantar displacement as cause of metatarsalgia can be treated by isolated elevation of the metatarsal. Osteonecrosis of metatarsal heads should be treated according the stage of disease, beginning non-surgically in early cases. As osteonecrosis of metatarsal head usually affects the dorsal part of the joint, major
deformities can be treated with dorsally-closing wedge osteotomies. Fractures have to be treated according location of the fracture conservatively or surgically.
Surgical Procedures for Metatarsalgia The main aim in the treatment of metatarsalgia is to treat the primary cause of metatarsalgia and not to primarily treat the consequence of a problem not directly located in the region of metatarsal heads. Pes cavus, pes equinus, transfer metatarsalgia caused by hallux valgus, overload of the lateral column in cases of painful hallux rigidus or nerve compression syndromes not located in the forefoot are examples of cases of metatarsalgia that require treatment of the primary cause first without touching the metatarsals themselves. Instability of the joint will require stabilization of the metatarsophalangeal joint. As direct repair of ruptured plantar plate is impossible in most cases, soft tissue procedures are used to improve flexor forces in the MTP joint: PIP-joint fusion [94, 95] tendon transfer (Flexor-proExtensor-transfer) [96, 97] or tenodesis of flexor tendon. The usual side-effect of Weil’s osteotomy – arthrofibrosis of the joint [98] – can be used for additional stabilization of MTP joint. Weil osteotomy [99, 100] is the most widespread procedure to shorten the metatarsal. This is indicated especially in cases of luxation or subluxation of the
223
Forefoot Deformities
shortening. This procedure should be performed only as combined surgery on metatarsals 2–4, not as a single ray procedure. Metatarsal head resection is only justified in cases of severe destruction of the metatarso-phalageal joint or metatarsal head especially in cases of rheumatoid disease [64, 104, 105]. Otherwise, preservation of metatarsal heads should be preferred.
Decision-Making in Hammer Toe Surgery Clinical Assessment
Fig. 6 Hallux valgus, Metatarsalgia: Indication for Weil-2 due to luxation of second MTP-joint; indication for Weil-3 to preserve metatarsal alignment; indication for shortening first metatarsal osteotomy (shortening Scarf, Youngswick, etc.) to compensate for planned shortening of second metatarsal
metatarsophalangeal joint (Fig. 6); isolated metatarsal overlength will also need shortening of the metatarsal, but special attention has to focus on necessary elevation of the metatarsal head by resection of a bony slice or wedge to avoid plantar displacement during shortening. This type of osteotomy – as an intra-articular procedure – is known to develop a certain amount of arthrofibrosis with stiffening of the joint [98], a side-effect that can be taken advantage of to avoid recurrence of luxation. Pure elevation of the metatarsal head needs other techniques like a V-shaped osteotomy (Chevron-type dorso-plantar osteotomy) [101] distally, or closing wedge osteotomy (BRT osteotomy [102]) proximally. For these newer techniques, reliable reports about outcome are still not available. Helal’s osteotomy [103] was reported to produce unpredictable results due to malalignment or mal-union and has to be indicated carefully in only selected cases; the main biomechanical effect is elevation of the metatarsal head, combined with minor
The description of hammer toe deformity begins with an exact definition of the deformity in terms of flexion – or extension-deformity in DIP- , PIP- and MTP joints. The deformity has to be defined as flexible or contracted in all three joints. Contraction deformities mostly need a surgical approach directly to the affected joint. Varus – or valgus deviation may indicate soft tissue procedures or bony corrective osteotomies. Instability or luxation has to be assessed especially in the MTP-joint. Calluses show longer-lasting overload, especially under metatarsal heads and dorsally at PIP- and DIP-joints. Extensor tendon lengths have to be tested in the standing and load-bearing foot or using the “push-up-test” intra-operatively. Clinical signs of shortening necessitate lengthening of the tendons and/or bony shortening.
Radiographic Assessment Subluxation or luxation of MTP joints indicate the need for stabilisation of the MTP joint, in most cases combined with shortening of the metatarsal. Medio-lateral displacement requires soft tissue procedures or bony corrective procedures whilst degenerative or arthritic changes of PIP or DIP joints require fusion or resection arthroplasty, in the MTP mostly by any type of arthroplasty.
Surgical Procedures for the Treatment of Hammer Toes Resection arthroplasty (Hohmann’s procedure) was the standard procedure for hammer toes in the past and still is a valuable procedure for contracted uncomplicated hammer toes [106, 107] without the need for further stabilisation of the MTP joint. According the prevalence, this procedure is used mostly for the PIP joint. PIP-joint fusion
224
[94, 95] (using K-wire, special fixation devices or special screws [108]) not only treats the hammer toe deformity, but also influences the pathogenetic loss of flexor force in the MTP joint by converting the pathologic extensor function of the lesser toe flexors into flexion at the level of MTP joint and thus stabilizing the toe. Similar effects are achieved by tendon transfers (flexor-to-extensor-transfer [96, 97]) or flexor tenodesis (using transosseous suture or screw). Shortening of the metatarsal will be indicated in cases of instability or especially subluxation or luxation of the MTP joint.
Conclusion Selection of the right surgical procedure for forefoot disorders needs meticulous assessment of both clinical and radiographic parameters. As hallux disorders or deformities, hammer toes and metatarsalgia have a mutual effect on each other, the decision for a certain surgical procedure has to consider this complex system of interdependence. In many cases, complete restoration of anatomy and function cannot be achieved, so in these cases the aim of surgery is to achieve the best compromise in respect to the patient’s needs.
References 1. Bordelon RL (1987) Evaluation and operative procedures for hallux valgus deformity. Orthopedics 10:38–44 2. Mann RA (1990) Bunion surgery: decision making. Orthopedics 13:951–957 3. Wülker N (1997) Hallux Valgus – Hallux Rigidus. In: Bücherei des Orthopäden, Band 67, Enke-Stuttgart 84–85 4. Schneider W, Knahr K (1998) Metatarsophalangeal and intermetatarsal angle: different values and interpretation of postoperative results dependent on the technique of measurement. Foot Ankle Int 19:532–536 5. Schneider W, Csepan R, Knahr K (2003) Reproducibility of the radiographic metatarsophalangeal angle in hallux surgery. J Bone Joint Surg 85-A:494–499 6. Austin DW, Leventeen EO (1981) A new osteotomy for hallux valgus. A horizontally directed “V” displacement osteotomy of the metatarsal head for hallux valgus and primus varus. Clin Orthop Relat Res 157:25–30 7. Corless JR (1976) A modification of the Mitchell procedure. J Bone Joint Surg 58-B:138 8. Johnson KA, Cofield RH, Morrey BF (1979) Chevron osteotomy for hallux valgus. Clin Orthop 142:44–47 9. Steinböck G (1996) The horizontally directed “V” displacement osteotomy of the metatarsal head according to Austin for the treatment of hallux valgus. Orthopade 25:308–316
W. Schneider 10. Barouk LS (2005) Scarf osteotomy of the first metatarsal. In: Barouk LS (ed) Forefoot reconstruction. Springer, Paris, pp 25–73 11. Patton GW, Zelichovsky JE (1994) Middiaphyseal osteotomies. In: Hetherington VJ (ed) Hallux valgus and forefoot surgery. Churchill Livingstone, New York, pp 215–225 12. Mitchell CL, Fleming JL, Allen R, Glenney C, Sanford GA (1958) Osteotomy-bunionectomy for hallux valgus. J Bone Joint Surg 40-A:41–60 13. Ludloff K (1913) Hallux valgus-operation. ZBL Chir 40:306–307 14. Ludloff K (1918) Die Beseitigung des Hallux valgus durch die schräge plantar-dorsale Osteotomie des Metatarsus. Arch Klin Chir 110:364–387 15. Mann RA, Coughlin MJ (1992) Adult hallux valgus. In: Mann RA, Coughlin MJ (eds) Surgery of the foot and ankle. Mosby, St. Louis, Baltimore, pp 167–296 16. Tordason DB, Leventen EO (1992) Hallux valgus correction with proximal metatarsal Osteotomy: two-year followup. Foot Ankle 13:321–326 17. Zettl R, Trnka HJ, Easley M, Salzer M, Ritschl P (2000) Moderate to severe hallux valgus deformity: correction with proximal crescentic osteotomy and distal soft-tissue release. Arch Orthop Trauma Surg 120:397–402 18. Lapidus PW (1934) Operative correction of metatarsus varus primus in hallux valgus. Surg Gynecol Obstet 58:183–191 19. Lapidus PW (1960) The author’s bunion operation from 1931 to 1959. Clin Orthop 16:119–135 20. LeLievre J (1967) Technique chirurgicale de l’avant pied. Arthroplasie – Cerclace Fibreux. In: Pathologie du Pied. Paris: Masson et Cie. 809–812 21. Kim JY, Park JS, Hwang SK, Young KW, Sung IH (2008) Mobility changes of the first ray after hallux valgus surgery: clinical results after proximal metatarsal chevron osteotomy and distal soft tissue procedure. Foot Ankle Int 29:468–472 22. Klaue K, Ungersböck A (1996) Tarsometatarsalarthrodese des ersten Strahles bei Hypermobilität und Hallux valgus. Operat Orthop Traumatol 8:221–231 23. Bösch P, Markowski H, Rannicher V (1990) Technik und erste Ergebnisse der subcutanen distalen Metatarsale-1-Osteotomie. Orthop Praxis 26:51–56 24. Bösch P, Wanke S, Legenstein R (2000) Hallux valgus correction by the method of Bosch: a new technique with a seven-to-ten-year follow-up. Foot Ankle Clin 5:485–498 25. Maffulli N, Longo UG, Marinozzi A, Denaro V (2010) Hallux valgus: effectiveness and safety of minimally invasive surgery. A systematic review. Br Med Bull. doi:10.1093/ bmb/ldq027 26. Kramer J (1990) Die Kramer-Osteotomie zur Behandlung des Hallux valgus und des Digitus quintus varus. Operat Orthop Traumatol 2:14–18 27. Stoffela R (1998) Neue Osteosynthesetechnik zur subcapitalen Metatarsalosteotomie beim Hallux valgus. Operat Orthop Traumatol 10:317–328 28. Amarnek D, Mollica A, Jacobs A, Oloff L (1986) A statistical analysis on the reliability of the proximal articular set angle. J Foot Surg 25:39–42 29. Evans RD, Lile LN (2000) Proximal articular set angle. Radiographic versus intraoperative measurement. J Am Podiatr Med Assoc 90:199–202
Forefoot Deformities 30. Sullivan BT, Robison JB, Palladino SJ (1988) Interevaluator variablity in the measurement of proximal articular set angle. J Foot Surg 27:466–468 31. Kitaoka HB, Alexander IJ, Adelaar RS, Nunley JA, Myerson MS, Sanders M (1994) Clinical rating systems for the anklehindfoot, midfoot, hallux, and lesser toes. Foot Ankle 15:349–353 32. Schneider W, Knahr K (1998) Scoring in forefoot surgery: a statistical evaluation of single variables and rating systems. Acta Orthop Scand 69:498–504 33. Schneider W, Knahr K (2002) Keller procedure and chevron osteotomy in hallux valgus: five-year results of different surgical philosophies in comparable collectives. Foot Ankle Int 23:321–329 34. Schneider W, Aigner N, Punggera O, Knahr K (2004) Chevron osteotomy in hallux valgus: minimum 10-year results of 112 cases. J Bone Joint Surg 86-B:1016–1020 35. Trnka HJ, Zembsch A, Easley ME, Salzer M, Ritschl P, Myerson MS (2000) The chevron osteotomy for correction of hallux valgus. Comparison of findings after two and five years of follow-up. J Bone Joint Surg 82-A:1373–1378 36. Moyer J, Lowery C, Knox J (2004) Hallux IPJ fusion. Clin Podiatr Med Surg 21:51–64 37. Akin OF (1925) The treatment of hallux valgus: a new operative procedure and its results. Med Sentinel 33:678–679 38. Frey C, Jahss M, Kummer FJ (1991) The Akin procedure: an analysis of results. Foot Ankle 12:1–6 39. Mann RA (1989) The great toe. Orthop Clin North Am 20:519–533 40. Badwey TM, Dutkowsky JP, Graves SC, Richardson EG (1997) An anatomical basis for the degree of displacement of distal chevron osteotomy in the treatment of hallux valgus. Foot Ankle 18:213–215 41. Graves SC, Dutkovsky JP, Richardson EG (1993) The chevron bunionectomy: a trigonometric analysis to predict correction. Foot Ankle 14:90–96 42. Kummer FJ (1989) Mathematical analysis of the first metatarsal osteotomies. Foot Ankle 9:281–289 43. Sarrafian SK (1985) A method of predicting the degree of functional correction of the metatarsus primus varus with a distal lateral displacement osteotomy in hallux valgus. Foot Ankle 5:322–326 44. Jones KJ, Feiwell LA, Freedman EL, Cracchiolo A (1995) The effect of chevron osteotomy with lateral capsular release on the blood supply to the first metatarsal head. J Bone Joint Surg 77-A:197–204 45. Youngswick FD (1982) Modifications of the Austin bunionectomy for treatment of metatarsus primus elevatus associated with hallux limitus. J Foot Surg 21:114–116 46. Granberry WM, Hickey CH (1995) Hallux valgus correction with metatarsal osteotomy: effect of a lateral distal soft tissue procedure. Foot Ankle Int 16:132–138 47. Petersen W, Seide HW (2000) Early outcome of correction of hallux valgus with the Scarf osteotomy. Z Orthop Ihre Grenzgeb 138:258–264 48. Saxena A, McCammon D (1997) The Ludloff osteotomy: a critical analysis. J Foot Ankle Surg 36:100–105 49. Trnka HJ, Parks BG, Invanic G, Chu IT, Easley ME, Schon LC, Myerson MS (2000) Six first metatarsal shaft osteotomies: mechanical and immobilisation comparisons. Clin Orthop Relat Res 381:256–265
225 50. Chiodo CP, Schon LC, Myerson MS (2004) Clinical results with the Ludloff osteotomy for correction of adult hallux valgus. Foot Ankle Int 25:532–536 51. Hofstaetter SG, Gruber F, Ritschl P, Trnka HJ (2006) The modified Ludloff osteotomy for correction of severe metatarsaus primus varus with hallux valgus deformity. Z Orthop Ihre Grenzgeb 144:141–147 52. Petroutsas J, Trnka HJ (2005) The Ludloff osteotomy for correction of hallux valgus. Oper Orthop Traumatol 17:102–117 53. Robinson AH, Bhatia M, Eaton C, Bishop L (2009) Prospective comparative study of the scarf and Ludloff osteotomies in the treatment of hallux valgus. Foot Ankle Int 30:955–963 54. Vora AM, Myerson MS (2005) First metatarsal osteotomy non-union and malunion. Foot Ankle Clin 10:35–54 55. Hofstaetter SG, Glisson RR, Alitz CJ, Trnka HJ, Easley ME (2008) Biomechanical comparision of screws and plates for hallux valgus opening-wedge and Ludloff osteotomies. Clin Biomech 23:101–108 56. Scott AT, DeOrio JK, Montijo HE, Glisson RR (2010) Biomechanical comparison of hallux valgus correction using the proximal chevron osteotomy fixed with a medial locking plate and the Ludloff osteotomy fixed with two screws. Clin Biomech 25:271–276 57. Okuda R, Kinoshita M, Morikawa J, Yasuda T, Abe M (2005) Proximal metatarsal osteotomy: relation between 1- to greater than 3-years results. Clin Orthop Relat Res 435:191–196 58. Fuhrmann RA (2005) Arthrodesis of the first tarsometatarsal joint for correction of the advanced splayfoot accompanied by a hallux valgus. Oper Orthop Traumatol 17:195–200 59. Shi K, Hayashida K, Tomita T, Tanabe M, Ochi T (2000) Surgical treatment of hallux valgus deformity in rheumatoid arthritis: clinical and radiographic evaluation of modified Lapidus technique. J Foot Ankle Surg 39:376–382 60. Thompson IM, Bohay DR, Anderson JG (2005) Fusion rate of first tarsometatarsal arthrodesis in the modified Lapidus procedure and flatfoot reconstruction. Foot Ankle Int 26:698–703 61. Keller WL (1904) The surgical treatment of bunions and hallux valgus. N Y Med J 80:741–742 62. Brandes M (1929) Zur operativen Therapie des Hallux valgus. Zentralbl Chir 56:2434 63. Hamilton W, O’Malley M, Thompson F, Kovatis P (1997) Capsular interposition arthroplasty for severe hallux rigidus. Foot Ankle Int 18:68–70 64. Coughlin MJ (2000) Rheumatoid forefoot reconstruction. A long-term follow-up study. J Bone Joint Surg 82-A:322–341 65. Mann RA, Thompson FM (1984) Arthrodesis of the first metatarsophalangeal joint for hallux valgus in rheumatoid arthritis. J Bone Joint Surg 66-A:687–692 66. Portaluri M (2000) Hallux valgus correction by the method of Bosch: a clinical evaluation. Foot Ankle Clin 5:499–511 67. Regnauld B (1986) Hallux rigidus. In: Regnauld B (ed) The foot. Springer, Berlin, pp 342–349 68. Coughlin MJ, Shurnas PS (2003) Hallux rigidus. Grading and long-term results of operative treatment. J Bone Joint Surg 85-A:2072–2088 69. Easley ME, Davis WH, Anderson RB (1999) Intermediate to long-term followup of medial-approach dorsal cheilectomy for hallux rigidus. Foot Ankle Int 20:147–152
226 70. Hattrup SJ, Johnson KA (1988) Subjective results of hallux rigidus following treatment with cheilectomy. Clin Orthop Relat Res 226:182–191 71. Moberg E (1979) A simple operation for hallux rigidus. Clin Orthop Relat Res 142:55–56 72. Watermann H (1927) Die Arthritis deformans des Großzehengrundgelenkes als selbständiges Krankheitsbild. Orthopade 48:346–355 73. Kilmartin TE (2005) Phalangeal osteotomy versus first metatarsal decompression osteotomy for the surgical treatment of hallux rigidus: a prospective study of age-matched and condition-matched patients. J Foot Ankle Surg 44:2–12 74. Regnauld B (1986) Disorders of the great toe. In: Regnauld B (ed) The foot. Springer, Berlin, pp 249–290 75. Cracchiolo A, Weltmer JB, Lian G, Dalseth T, Dorey F (1992) Arthroplasty of the first metatarsophalangeal joint with a double stem silicone implant. J Bone Joint Surg 74-A:552–563 76. Papagelopoulos PJ, Kitaoka HB, Ilstrup DM (1994) Survivorship analysis of implant arthroplasty for the first metatarsophlanageal joint. Clin Orthop Relat Res 302:164–172 77. Brewster M, McArthur J, Mauffrey C, Lewis AC, Hull P, Ramos J (2010) Moje first metatarsophalangeal replacement – a case series with functional outcomes using the AOFAS-HMI score. J Foot Ankle Surg 49:37–42 78. Notni A, Fahrmann M, Fuhrmann RA (2001) Early results of implantation of an unconstrained metatarsophalangeal joint prosthesis of the first toe. Z Orthop Ihre Grenzgeb 139:326–331 79. Raikin SM, Ahmad J, Pour AE, Abidi N (2007) Comparison of arthrodesis and metallic hemiarthroplasty of the hallux metatarsophalangeal joint. J Bone Joint Surg 89-A:1979–1985 80. Townley CO, Taranow WO (1994) A metallic hemiarthroplasty resurfacing prosthesis for the hallux metatarsophalangeal joint. Foot Ankle Int 15:575–580 81. Esway JE, Conti SF (2005) Joint replacement in the hallux metatarsophalangeal joint. Foot Ankle Clin 10:97–115 82. Coughlin MJ, Mann R (1987) Arthrodesis of the first metatarsophalageal joint as salvage for the failed Keller procedure. J Bone Joint Surg 69A:68–75 83. Machacek F Jr, Easley ME, Gruber F, Ritschl P, Trnka HJ (2004) Salvage of a failed Keller resection arthroplasty. J Bone Joint Surg 86-A:1131–1138 84. Schneider W, Kadnar G, Kranzl A, Knahr K (2006) Keller’s resection arthroplasty for Hallux Rigidus: minimum 20-year results of 87 cases. In: Proceedings of the 6th international congress of the European foot and ankle society, Prag, 2006 85. Suckel A, Wülker N (2006) Arthrodesis of the first metatarsophalangeal joint. Orthopade 35:443–449 86. Hecht PJ, Gibbons MJ, Wapner KL, Cooke C, Hoisington SA (1998) Arthrodesis of the first metatarsophalangeal joint to salvage failed silicone implant arthroplasty. Foot Ankle Int 19:59–60 87. Feltham GT, Hanks SE, Marcus RE (2001) Age-based outcomes of cheilectomy for the treatment of hallux rigidus. Foot Ankle Int 22:192–197 88. Lombardi CM, Silhanek AD, Connelly FG, Dennis LN, Keslonsky AJ (2001) First metatarsophalangeal arthrodesis for treatment of hallux rigidus: a retrospective study. J Foot Ankle Surg 40:137–143
W. Schneider 89. Lau JT, Daniels TR (2001) Outcomes following cheilectomy and interposition arthroplasty in hallux rigius. Foot Ankle Int 11:462–470 90. DeFrino PF, Brodsky JW, Pollo FE, Crenshaw SJ, Beischer AD (2002) First metatarsophalangeal arthrodesis: a clinical, pedobarographic and gait analysis study. Foot Ankle Int 23:496–502 91. Ettl V, Radke S, Gaertner M, Walter M (2003) Arthrodesis in the treatment of hallux rigidus. Int Orthop 27:382–385 92. Wassink S, van den Oever M (2009) Arthrodesis of the first metatarsophalangeal joint using a single screw: retrospective analysis of 109 feet. J Foot Ankle Surg 48:653–661 93. Barouk LS (2005) The metatarsal parabola. In: Barouk LS (ed) Forefoot reconstruction. Springer, Paris, pp 365–366 94. Edwards WH, Beischer AD (2002) Interphalangeal joint arthrodesis of the lesser toes. Foot Ankle Clin 7:43–48 95. Lehmann DE, Smith RW (1995) Treatment of symptomatic hammertoe with a proximal interphalangeal joint arthrodesis. Foot Ankle Int 16:535–541 96. Girdlestone GR, Spooner RHJ (1939) A new operation for hallux valgus and hallux rigidus. J Bone Joint Surg 19:30–35 97. Taylor R (1951) The treatment of clawtoes by multiple transfers of flexor into extensor tendons. J Bone Joint Surg 33:529 98. Giese K, Follak N, Merck H (2004) Problemanalyse nach 200 Weil Osteotomien. Orthop Praxis 40:24–25 99. Barouk LS (2005) The Weil lesser metatarsal osteotomy. In: Barouk LS (ed) Forefoot reconstruction. Springer, Paris, pp 115–138 100. Barouk LS (1996) Die Metatarsalosteotomie nach Weil zur Behandlung der Metatarsalgie. Orthopade 25:338–344 101. Kitaoka HB, Patzer G (1998) Chevron osteotomy of lesser metatarsals for intractable plantar callosities. J Bone Joint Surg 80-B:516–518 102. Barouk LS (2005) The BRT proximal metatarsal osteotomy. In: Barouk LS (ed) Forefoot reconstruction. Springer, Paris, pp 139–154 103. Helal B, Greiss M (1982) Die Metatarsal-Verschiebeosteotomie zur Behandlung der Überlastungs-Metatarsalgie. Orthopade 11:214–228 104. Grifka J, Oest O (1989) Verlaufsbeobachtung nach Mittelfußköpfchenresektion nach Lelievre. Z Orthop 127:561–565 105. Justen HP, Berger W, Leeb I, Pilhofer C, Wessinghage D (2000) Longterm outcome of metatarsal head resection in rheumatoid arthritis. Z Rheumatol 59:101–107 106. Coughlin MJ, Dorris J, Polk E (2000) Operative repair of the fixed hammertoe deformity. Foot Ankle Int 21:94–104 107. O’Kane C, Kilmartin T (2005) Review of proximal interphalangeal joint excisional arthroplasty for the correction of second hammertoe deformity in 100 cases. Foot Ankle Int 26:320–325 108. Caterini R, Farsetti P, Tarantino U, Potenza V, Ippolito E (2004) Arthrodesis of the toe joints with an intramedullary cannulated screw for correction of hammertoe deformità. Foot Ankle Int 25:256–261
