Joint replacement technology, New Developments

Joint replacement technology

© 2008, Woodhead Publishing Limited

Related titles: Bioceramics and their clinical applications (ISBN 978-1-84569-204-9) Bioceramics are potentially suitable for a wide range of important applications within the medical device industry. Bioceramics and their clinical applications provides an authoritative review of this highly active area of research, written by leading academics from around the world. Chapters in the first section of the book discuss issues of significance to a range of bioceramics, such as their structure, mechanical properties and biological interactions. The second part reviews the fabrication, microstructure and properties of specific bioceramics and glasses, concentrating on the most promising materials. The final group of chapters reviews the clinical applications of bioceramics. Surfaces and interfaces for biomaterials (ISBN 978-1-85573-930-7) This book presents our current level of understanding on the nature of a biomaterial surface, the adaptive response of the biomatrix to that surface, techniques used to modify biocompatibility, and state-of-the-art characterisation techniques to follow the interfacial events at that surface. The first part of the book reviews the way biomaterial surfaces form. The second part discusses ways of monitoring and characterising surface structure and behaviour. The final two parts of the book look at a range of in vitro and in vivo studies of the complex interactions between biomaterials and the body. Medical modelling: The application of advanced design and development techniques in medicine (ISBN 978-1-84569-138-7) Medical modelling is an increasingly important tool in surgery and rehabilitative medicine. This authoritative book describes the key steps in modelling from acquisition of medical scan data, transfer and translation of data formats, methods of utilising the data and finally using the information to produce physical models using rapid prototyping techniques. Technologies are fully described, highlighting their key characteristics, advantages and disadvantages. A series of case studies illustrates a broad range of medical applications in surgery or prosthetic rehabilitation. Details of these and other Woodhead Publishing books, as well as books from Maney Publishing, can be obtained by: · visiting our web site at www.woodheadpublishing.com · contacting Customer Services (e-mail: [email protected]; fax: +44 (0) 1223 893694; tel.: +44 (0) 1223 891358 ext. 130; address: Woodhead Publishing Limited, Abington Hall, Granta Park, Great Abington, Cambridge CB21 6AH, England) If you would like to receive information on forthcoming titles, please send your address details to: Francis Dodds (address, tel. and fax as above; e-mail: [email protected]). Please confirm which subject areas you are interested in. Maney currently publishes 16 peer-reviewed materials science and engineering journals. For further information visit www.maney.co.uk/journals.

© 2008, Woodhead Publishing Limited

Joint replacement technology Edited by Peter A. Revell

Woodhead Publishing and Maney Publishing on behalf of The Institute of Materials, Minerals & Mining CRC Press Boca Raton Boston New York Washington, DC

© 2008, Woodhead Publishing Limited

Woodhead Publishing Limited and Maney Publishing Limited on behalf of The Institute of Materials, Minerals & Mining Published by Woodhead Publishing Limited, Abington Hall, Granta Park, Great Abington, Cambridge CB21 6AH, England www.woodheadpublishing.com Published in North America by CRC Press LLC, 6000 Broken Sound Parkway, NW, Suite 300, Boca Raton, FL 33487, USA First published 2008, Woodhead Publishing Limited and CRC Press LLC ß 2008, Woodhead Publishing Limited The authors have asserted their moral rights. This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. Reasonable efforts have been made to publish reliable data and information, but the authors and the publishers cannot assume responsibility for the validity of all materials. Neither the authors nor the publishers, nor anyone else associated with this publication, shall be liable for any loss, damage or liability directly or indirectly caused or alleged to be caused by this book. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming and recording, or by any information storage or retrieval system, without permission in writing from Woodhead Publishing Limited. The consent of Woodhead Publishing Limited does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from Woodhead Publishing Limited for such copying. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress. Woodhead Publishing Limited ISBN 978-1-84569-245-2 (book) Woodhead Publishing Limited ISBN 978-1-84569-480-7 (e-book) CRC Press ISBN 978-1-4200-7962-3 CRC Press order number WP7962 The publishers' policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp which is processed using acid-free and elementary chlorine-free practices. Furthermore, the publishers ensure that the text paper and cover board used have met acceptable environmental accreditation standards. Project managed by Macfarlane Book Production Services, Dunstable, Bedfordshire, England (e-mail: [email protected]) Typeset by Godiva Publishing Services Limited, Coventry, West Midlands, England Printed by TJ International Limited, Padstow, Cornwall, England

© 2008, Woodhead Publishing Limited

Contents

Contributor contact details

xv

Preface

xxi

Part I Introduction 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7

2 2.1 2.2 2.3 2.4 2.5 2.6

3

Biomechanics of joints

G R J O H N S O N , Newcastle University, UK

Introduction Introduction to biomechanics Key aspects of biomechanics of major joints The upper limb Summary Sources of further information and advice References

3 3 14 20 29 29 29

Tribology in joint replacement

31

Introduction Theoretical tribological studies Experimental tribological studies Issues of tribology for joint replacements and future trends Sources of further information and advice References and further reading

31 38 43 49 50 50

Z J I N and J F I S H E R , University of Leeds, UK

Biomaterials and the chemical environment of the body K J B U N D Y , Tulane University, USA

3.1 3.2

3

Introduction Chemical environment for joint replacement

© 2008, Woodhead Publishing Limited

56 56 57

vi

Contents

3.3 3.4 3.5 3.6 3.7

Surfaces and interfaces Corrosion Conclusion Sources of further information and advice References

4

4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10

5

Materials for joint replacement

K S K A T T I , D VE R M A a n d D R K A T T I , North Dakota State University, USA

Introduction Materials criteria for total joint replacement History of materials used in joint replacement Traditional materials Bone cement materials Composite materials and new nanocomposite systems Natural materials Summary Acknowledgments References

Regulatory issues affecting joint replacement: the case of the UK

E D A M I E N , MHRA, UK, B P A U L , Kyiv Medical Academy, Ukraine and S M DA M I E N a n d C S D A M I E N , QMUL, UK 5.1 5.2 5.3 5.4 5.5

Introduction and background The regulatory process Planning for the regulatory approval of a product Summary References and useful websites for further information

61 64 76 78 79

81

81 81 83 84 90 91 93 94 94 94

105

105 105 108 111 111

Part II Material and mechanical issues 6

6.1 6.2

Metals for joint replacement

115

Introduction General requirements for biomaterials

115 116

Y T K O N T T I N E N , Helsinki University Central Hospital, Finland, I M I L O SÏ E V , JosÏef Stefan Institute, Slovenia, R T R E B SÏ E , Orthopaedic Hospital Valdoltra, Slovenia, P R A N T A N E N and R L I N D E N , National Agency of Medicines, Finland, V M T I A I N E N , ORTON Orthopaedic Hospital of the Invalid Foundation, Finland and S V I R T A N E N , University of Erlangen-Nuremberg, Germany

© 2008, Woodhead Publishing Limited

Contents 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13 6.14

7

7.1 7.2 7.3 7.4 7.5 7.6

8

8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8

9

Examples of currently valid European Union standards Overview of metals Biomechanical properties Corrosion Corrosion testing Metals used in joint replacements Particle disease Clinical success of metals used in joint replacement surgery Future trends Acknowledgements Sources of further information and advice: useful websites References

Ceramics for joint replacement

D K L U E S S , W M I T T E L M E I E R and R B A D E R , University of Rostock, Germany

118 122 125 130 130 134 141 142 147 155 155 155

163

Introduction Material and mechanical properties of ceramics Ceramics in total hip replacement Ceramics in total knee replacement Summary References

163 164 165 170 172 173

Joint bearing surfaces and replacement joint design

176

Introduction Articulating surfaces in natural joints Demands for the bearing surfaces Different solutions available Special concepts and designs for bearing surfaces Comparison of bearing surface solutions Future trends References

176 176 177 178 185 186 187 188

R L A P P A L A I N E N and M S E L E N I U S , University of Kuopio, Finland

Cementless fixation techniques in joint replacement

M J C R O S S and J S P Y C H E R , The Australian Institute of Musculoskeletal Research, Australia

9.1 9.2 9.3

vii

Introduction Cementless fixation Initial stability

© 2008, Woodhead Publishing Limited

190

190 191 192

viii

Contents

9.4 9.5 9.6 9.7 9.8

Osseous integration of cementless implants Mechanical properties of the implant Why do you still use cement? Future trends References

193 197 199 203 205

10

Bone cement fixation: acrylic cements

212

10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 10.10 10.11 10.12 10.13 10.14 10.15 10.16 10.17 10.18 10.19

Introduction Acrylic bone cements ± history and evolution Clinical application and function Composition Polymer powder/liquid monomer ratio Polymerisation reaction Polymerisation heat Polymerisation shrinkage Molecular weight and sterilisation Residual monomer and monomer release Viscosity and handling properties Antibiotics in poly(methylmethacrylate) bone cement Radiopacifier in poly(methylmethacrylate) bone cement Mechanical properties Mixing methods Joint replacement cementing technique Problems with acrylic cements Summary References

212 212 213 214 215 216 219 220 221 221 222 225 226 227 230 233 240 242 242

11

Bone±cement fixation: glass-ionomer cements

252

11.1 11.2 11.3 11.4 11.5

12

12.1 12.2

J-S W A N G , Lund University, Sweden, and N D U N N E , Queen's University of Belfast, UK

P V H A T T O N , V K E A R N S and I M B R O O K , University of Sheffield, UK Introduction Structure and properties of glass±ionomer cements Biological evaluation Future trends References

Failure mechanisms in joint replacement

M B U R K E and S G O O D M A N , Stanford University Medical Center, USA Introduction Wear and debris

© 2008, Woodhead Publishing Limited

252 252 253 259 259

264

264 264

Contents 12.3 12.4 12.5 12.6 12.7 12.8 12.9

13 13.1 13.2 13.3 13.4 13.5

Implant or bone fracture Dislocation Stress shielding Comment on surgical failure Summary Future trends References

Predicting the lifetime of joints: clinical results

L R Y D , Karolinska University Hospital/Huddinge, Sweden Introduction National joint replacement registries Radiostereometric analysis Future trends References

ix 267 276 280 281 281 282 283

286 286 287 299 306 306

Part III The device biological environment 14

The healing response to implants used in joint replacement

315

Introduction Immediate response to prosthesis placement Remodelling of bone around implants The cemented joint prosthesis The uncemented prosthetic joint component Bioactive surfaces on prostheses Adjunctive methods or treatments and their effects Summary References

315 316 318 323 329 333 338 343 344

Biological causes of prosthetic joint failure

349

P A R E V E L L , University College London, UK 14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8 14.9

15 15.1 15.2 15.3 15.4 15.5 15.6 15.7 15.8

P A R E V E L L , University College London, UK

Introduction Infection Aseptic loosening The isolation and characterisation of wear particles The cellular reaction to particulate wear debris The role of macrophages and multinucleate giant cells Bone resorption and wear debris: osteoclasts, macrophages and multinucleate giant cells Lymphocytes, sensitisation and aseptic loosening

© 2008, Woodhead Publishing Limited

349 350 356 357 363 368 371 372

x

Contents

15.9 15.10

Evidence for immunological processes in loosening Wear particles and corrosion products in distant organs: systemic effects Summary and conclusions References

382 384 385

Using drug delivery systems to enhance joint replacement

397

15.11 15.12

16

D P P I O L E T T I , Ecole Polytechnique FeÂdeÂrale de Lausanne, Switzerland 16.1 16.2 16.3 16.4 16.5

17 17.1 17.2 17.3 17.4 17.5

Why do we need to improve the outcome of orthopaedic implants? What is the clinical situation for orthopaedic implants used as drug delivery systems? Is the research for orthopaedic drug delivery systems advanced enough to translate it to clinical applications? Will drug delivery systems be the future for orthopaedic implants? References

Sterilization of joint replacement materials

A I A N U Z Z I and S M K U R T Z , Exponent, Inc., USA Introduction Sterilization techniques and their suitability Issues with sterilization of joint replacement materials Conclusions References

374

397 398 399 402 403

407 407 411 417 424 424

Part IV Specific joints 18

Hip replacement: tribological principles, materials and engineering

D D O W S O N , University of Leeds, UK 18.1 18.2 18.3 18.4 18.5 18.6 18.7

Introduction Millennium prostheses Introduction to the tribology of total hip replacements Hard-on-hard total hip joint tribology Wear particles and metal ions Summary References

© 2008, Woodhead Publishing Limited

431 431 437 439 449 453 456 458

Contents

19 19.1 19.2

Hip replacement: clinical perspectives

xi

462

M R E V E L L and E T D A V I S , Royal Orthopaedic Hospital, UK

19.3 19.4 19.5 19.6 19.7

Introduction Problems with hip replacement at the beginning of the 21st century Specific complications Current solutions Computer navigation Conclusions References

20

Knee replacement: clinical perspectives

20.1 20.2 20.3 20.4 20.5

Introduction Kinematics and knee joint prosthesis design Analysis of the kinematics of total joint prostheses Summary References

481 482 502 509 510

21

Intervertebral disc joint replacement technology

515

21.1 21.2

515

21.3 21.4 21.5 21.6 21.7

Introduction Orthopedic materials and methodology available for use in intervertebral disc replacements Early intervertebral disc replacement designs Current designs Clinical concerns Conclusions References

22

Replacing temporomandibular joints

22.1 22.2 22.3 22.4 22.5 22.6 22.7 22.8

J D B L A H A , University of Michigan Medical School, USA

N H A L L A B , Rush University Medical Center, USA

J V A N L O O N and L G M D E B O N T , University of Groningen, The Netherlands and G J V E R K E R K E , University of Groningen and University of Twente, The Netherlands Introduction Temporomandibular joint prosthesis criteria Design Development and test procedures First clinical application Conclusions Sources of further information and advice: useful websites References

© 2008, Woodhead Publishing Limited

462 465 466 471 472 476 476

481

519 524 525 536 544 544

549

549 553 554 559 562 565 565 565

xii

Contents

23

Replacing ankle joints

23.1 23.2 23.3 23.4 23.5 23.6 23.7 23.8 23.9 23.10

Introduction: short history of ankle replacement Anatomical, biomechanical and biological features of the normal ankle joint Pathologies leading to degeneration of the ankle joint Contraindications for ankle replacement Materials used to replace the ankle Fixation of ankle prostheses The interrelationship between the ankle and the hindfoot Long-term results of uncemented current designs Future trends References

24

Replacing shoulder joints

24.1 24.2 24.3 24.4 24.5 24.6 24.7 24.8

Introduction Biomechanics of total shoulder arthroplasty Indications for total shoulder arthroplasty Surgical technique Complications Prognostic factors for clinical outcome Summary References

579 580 588 595 601 602 604 604

25

Replacing elbow joints

611

25.1 25.2 25.3 25.4 25.5 25.6 25.7 25.8 25.9

Introduction Materials and device design Indications and contraindications Surgical technique overview Clinical results Complications Revision surgery Summary References

611 611 615 616 619 623 625 628 628

26

Replacing joints with pyrolytic carbon

26.1 26.2

H K O F O E D , Federiksberg Hospital, Denmark

L D E W I L D E , University Hospital of Ghent, Belgium

J S A N C H E Z - S O T E L O , Mayo Clinic, USA

J S T A N L E Y , J K L A W I T T E R and R M O R E , Wrightington Hospital, UK Introduction What is pyrolytic carbon?

© 2008, Woodhead Publishing Limited

569 569 571 573 573 574 575 575 576 576 577

579

631

631 631

Contents 26.3 26.4 26.5 26.6 26.7 26.8 26.9

History of pyrolytic carbon use Review of pyrolytic carbon joint clinical history/performance Design and testing of pyrolytic carbon joint replacement implants Hemi-joint arthroplasty Conclusion Forward-looking statement with respect to pyrolytic carbon in orthopaedics References

© 2008, Woodhead Publishing Limited

xiii 637 638 642 644 653 653 654

Contributor contact details

Editor P. A. Revell University College London Eastman Dental Institute London WC1X 8LD UK

Chapter 2 Z. Jin* and J. Fisher Institute of Medical and Biological Engineering University of Leeds Leeds LS2 9JT UK E-mail: [email protected] [email protected]

High Ridge 7 Exmouth Road Budleigh Salterton Devon EX9 6AF UK E-mail: [email protected]

Chapter 3 K. J. Bundy IrisvaÈgen 88 72246 VaÈsteraÊs Sweden E-mail: [email protected]

Chapter 1 G. R. Johnson Centre for Rehabilitation and Engineering Studies (CREST) School of Mechanical and Systems Engineering Stephenson Building Newcastle University Newcastle upon Tyne NE1 7RU UK E-mail: [email protected]

Chapter 4 K. S. Katti,* D. Verma and D. R. Katti North Dakota State University Department of Civil Engineering CIE 201B 1410 14th Avenue North Fargo, ND 58105 USA E-mail: [email protected]

(* = main contact)

© 2008, Woodhead Publishing Limited

xvi

Contributor contact details

Chapter 5 E. Damien* MHRA London SW8 5NQ UK E-mail: [email protected] [email protected]

P. Rantanen and R. Linden National Agency of Medicines Mannerheimintie 103b 00301 Helsinki Finland E-mail: [email protected] [email protected]

S. M. Damien and C. S. Damien QMUL London E1 4NS UK

V. M. Tiainen Anders LangenskioÈld Research Laboratory Tenholantie 10 ORTON Orthopaedic Hospital of the Invalid Foundation Helsinki FIN-00280 Finland

B. Paul Kyiv Medical Academy Trauma & Orthopaedics Kyiv Ukraine Chapter 6 Y. T. Konttinen* Department of Medicine Institute of Clinical Medicine Helsinki University Central Hospital Biomedicum Helsinki Haartmaninkatu 8 (PO Box 700) FIN-00029 HUS Finland E-mail: [email protected] I. MilosÏev JosÏef Stefan Institute Department of Physical and Organic Chemistry Jamova 39 1000 Ljubljana Slovenia E-mail: [email protected] R. TrebsÏe Orthopaedic Hospital Valdoltra Jadranska C.31 6280 Ankaran Slovenia E-mail: [email protected] © 2008, Woodhead Publishing Limited

S. Virtanen Department of Material Sciences University of Erlangen-Nuremberg Mrtenstr. 7 D-91058 Erlangen Germany E-mail: [email protected] Chapter 7 D. Kluess, W. Mittelmeier and R. Bader* Department of Orthopaedics University of Rostock Doberaner Strasse 142 D-18057 Rostock Germany E-mail: [email protected] Chapter 8 R. Lappalainen* and M. Selenius University of Kuopio Dept of Physics PO Box 1627 70211 Kuopio Finland E-mail: [email protected]

Contributor contact details Chapter 9 M. J. Cross* and Dr J. Spycher The Australian Institute of Musculoskeletal Research 286 Pacific Highway Crows Nest NSW Australia 2065 E-mail: [email protected] Chapter 10 J. S. Wang* C12, BMC Department of Orthopedics Clinical Sciences Lund University S-221 84 Lund Sweden E-mail: [email protected] N. Dunne Polymer Research Cluster School of Mechanical and Aerospace Engineering Queen's University of Belfast Ashby Building Stranmillis Road Belfast BT9 5AH Northern Ireland UK E-mail: [email protected] Chapter 11 P. V. Hatton*, V. Kearns and I. M. Brook Centre for Biomaterials and Tissue Engineering School of Clinical Dentistry University of Sheffield Sheffield S10 2TA UK E-mail: [email protected]

© 2008, Woodhead Publishing Limited

xvii

Chapter 12 M. Burke and S. Goodman* Department of Orthopaedic Surgery Stanford University Medical Center 300 Pasteur Drive Stanford, CA 94305 USA E-mail: [email protected] Chapter 13 L. Ryd Department of Orthopaedics Karolinska University Hospital/ Huddinge S-141 86 Stockholm Sweden E-mail: [email protected] Chapters 14 and 15 P. A. Revell University College London Eastman Dental Institute London WC1X 8LD UK High Ridge 7 Exmouth Road Budleigh Salterton Devon EX9 6AF UK E-mail: [email protected] Chapter 16 D. P. Pioletti Laboratory of Biomechanical Orthopedics EPFL-HOSR Institute of Translational Biomechanics Ecole Polytechnique FeÂdeÂrale de Lausanne Station 15 1015 Lausanne Switzerland E-mail: [email protected]

xviii

Contributor contact details

Chapter 17 A. Ianuzzi and S. M. Kurtz* Exponent, Inc. 3401 Market Street Suite 300 Philadelphia, PA 19130 USA E-mail: [email protected]

Chapter 21 N. Hallab Department of Orthopedic Surgery Rush University Medical Center 1735 W. Harrison St Chicago, IL 60612 USA E-mail: [email protected]

Chapter 18 D. Dowson Institute of Engineering Thermofluids, Surfaces and Interfaces School of Mechanical Engineering University of Leeds Leeds LS2 9JT UK E-mail: [email protected]

Chapter 22 J. P. Van Loon and L. G. M. De Bont TMJ Research Group Department of Oral and Maxillofacial Surgery University Medical Center Groningen University of Groningen PO Box 196 9700 AD Groningen The Netherlands

Chapter 19 M. Revell* and E. T Davis Royal Orthopaedic Hospital The Woodlands Bristol Road South Birmingham B31 2AP Email: [email protected]; [email protected] [email protected] Chapter 20 J. D. Blaha Department of Orthopaedic Surgery University of Michigan Medical School (UMMS) 1500 E. Medical Center Drive Ann Arbor, MI 48109-0328 USA E-mail: [email protected]

© 2008, Woodhead Publishing Limited

G. J. Verkerke* Department of Biomedical Engineering University Medical Center Groningen University of Groningen PO Box 196 9700 AD Groningen The Netherlands E-mail: [email protected] and Department of Biomechanical Engineering University of Twente Enschede The Netherlands

Contributor contact details Chapter 23 H. Kofoed Orthopaedic University Clinic Federiksberg Hospital 57 Ndr DK-21000 Denmark E-mail: [email protected] Chapter 24 L. De Wilde Department of Orthopaedic Surgery Physical Medicine and Rehabilitation University Hospital of Ghent De Pintelaan 185 B-9000 Belgium E-mail: [email protected]

© 2008, Woodhead Publishing Limited

xix

Chapter 25 J. Sanchez-Sotelo Department of Orthopedic Surgery Mayo Clinic 200 First Street SW Rochester, MN 55905 USA E-mail: [email protected] Chapter 26 J. Stanley,* J. Klawitter and R. More Wrightington Wigan and Leigh NHS Trust Wrightington Hospital Hall Lane Appley Bridge Wigan WN6 9EP UK E-mail: [email protected]

Preface

Joint replacement has become one of the major successes of modern medical treatment. The expectations of patients as well as their surgeons continue to rise in terms of the ability to restore function, relieve pain and provide long-term performance with the devices implanted. Success in one particular joint, the hip, has led in fairly short order to similarly good outcomes in the knee. The outlook is then that restorative implantation surgery will provide answers for those debilitated by disease in other joints such as those of the hand and foot, the ankle and the upper limb joints, as well as the spine. The temporomandibular joint is mostly forgotten when replacement surgery is being considered. Each of these joints has its own unique characteristics in terms of the requirements for the successful development of a replacement device. Solutions provided for one anatomical site will not necessarily have relevance to another. There may also be a real, and perfectly natural, tendency for those involved in developing and implanting a replacement for a particular joint to focus on that problem and not to look at the solutions provided for other sites. In terms of the outcome for the patient and the high level of competence of the surgeon, such specialisation in orthopaedics has great benefits. The same must be true for the material scientists, engineers and industrial partners involved in artificial joint development. However, this highly focused approach may have an accompanying disadvantage, namely, that there is less awareness in one community of what is happening in terms of the innovation, or even improvement by gradual evolution, in other fields. The aim of this book is to provide an update on progress in the technology of joint replacement for the medical and scientific world. The book should be useful not only to the engineering and materials scientific communities but also to the surgeons seeking the best treatment for their patients. The need for different scientific skills in solving the problems of prosthetic implant development has long been recognised, and there are those, the editor being included, who feel that true interdisciplinary research is the key to success. By this is meant the real and close collaboration between clinicians and researchers across many different areas in science and medicine. This approach brings great © 2008, Woodhead Publishing Limited

xxii

Preface

rewards, but also presents certain potential problems, not least of which is the need for effective communication, the balance between the appropriate use of technical language on the one hand and simplification with the avoidance of specialist terms on the other being sometimes difficult to achieve. As the writer knows only too well, one person's technical language is just jargon to another, while being careful to explain the meaning of something adequately may at the same time seem facile and even patronising. The concept from the outset was to provide a book that included introductory sections with chapters explaining basic principles alongside parts about ways in which the understanding of such fundamentals had changed. This would be followed by a section describing where developments had occurred in the production of suitable materials for joint replacement and for fixing the surgical implant to the skeleton. Even though some failure mechanisms are still much as they were in the time of the pioneers, there has been much advance in the understanding of joint failure from the point of view of both the mechanical and biological processes involved. Other biological considerations include the healing and repair around the artificial joint, ideas around the orthopaedic implant as used for drug delivery and the important effects of sterilisation procedures on the biological properties of prosthetic materials. Evaluation of clinical performance is essential if real improvements are to be made. A book dealing with all these aspects, it was felt, would be useful, but then to have contributions giving some idea of the state of the art and the current issues in the development of replacements for all the different joints individually was considered to be both important and worthwhile. Deciding on whom to invite as authors on this project was an interesting challenge and although various names came to mind, the editor also used bibliometrics at the outset to make the process less subjective and to find those who had made significant contributions in their specialist areas. Canvassing of opinions as to likely authors was also carried out by the team at Woodhead. In the end, a wish list was drawn up of those who should be invited to join the project. Some turned out to be friends or acquaintances and colleagues from the past, while others were completely unknown to the editor at a personal level. All have provided clear accounts of the area within joint replacement technology on which they were asked to write. While there is some overlap, but not a great deal, between parts of some contributions, this has the advantage either of providing a different viewpoint on an issue or giving emphasis to an aspect by the occurrence of the repetition. Where possible, cross-referencing has been provided between chapters and sections within the book. Acting as editor has been an interesting and rewarding task. Thoughtprovoking ideas and statements are present throughout and these reflect the challenges being met by the community involved. Some examples follow, chosen more or less at random. Thus, for the hip and knee joints, it is usually assumed that an individual takes one million steps (cycles in each joint) in a © 2008, Woodhead Publishing Limited

Preface

xxiii

year, but younger and more active individuals may reach up to five million cycles per year for each lower limb joint. It is this requirement for greater functional capacity that has contributed to the recent developments in hip joint prostheses. Another important factor in bringing change to the materials used in articulating surfaces resulted from the recognition of the degradation of polyethylene mainly with storage and although this has been addressed in current prostheses, large numbers of ultra-high molecular weight polyethylene (UHMWPE) components that were air-sterilised and then stored remain in patients, since it is estimated that two million individuals were implanted with this material between 1990 and 1995 in the United States alone. Returning to the number of loading cycles in a joint, it is pointed out by another contributor that with an average of two million strides and 125 000 significant bends per annum, a lumbar spine implant may be subjected to over 100 million cycles in its expected duration of implantation. When the complexity of spinal anatomy, biomechanics and kinematics is added to this picture, the challenge of replacement and repair in this part of the body can begin to be appreciated. Similar insights into the various other aspects of joint replacement technology in this book cannot even be summarised here, and no doubt individual contributors would choose different aspects of their writings from those favoured by the editor. Next, it must be said that no matter how good the technology and the science, the ultimate success of joint replacement depends on the skill of the surgeon, as pointed out by the authors of Chapter 22, namely, that `a prosthesis will function properly only when implanted correctly'. It is said that the accurate collection of right data is essential to good science, but such data need to be analysed appropriately before there is any generation of real information. In turn, information must be processed and evaluated in the light of other information before it contributes to knowledge. The use of accumulated knowledge refined in the light of practical experience provides a depth of knowledge and understanding which might be referred to as wisdom. This book not only provides a great deal of information and knowledge, but also the specialist wisdom of those who have contributed in a way that cannot be provided in scientific papers or even review articles. Hopefully, as a group we have made available a resource that will prove valuable to a wider readership than might have been reached in our own individual fields of expertise. It is the hope that this book will provide for an increased understanding of joint replacement. Peter A Revell Budleigh Salterton, Devon

© 2008, Woodhead Publishing Limited

Part I

Introduction

© 2008, Woodhead Publishing Limited

1

Biomechanics of joints

G R J O H N S O N , Newcastle University, UK

1.1

Introduction

The well-documented rise in the numbers of older people is creating an everincreasing demand for total joint replacement. At the same time, the increasing health and activeness of these people creates demand for long-lasting reliable joints, minimising the need for costly revision surgery. The design and development of new joint replacements are highly interdisciplinary activities, calling for the combination of sound biomechanical understanding, detailed knowledge of anatomy and surgical experience and insight. The purpose of this chapter is to provide a solid biomechanical background to the material to be presented in the later chapters. It provides initially an overview of basic mechanics ± both kinematic and kinetics followed by basic stress analysis. The second part of the chapter applies the basic mechanical principles to some of the major joints with a particular emphasis on the functional kinematics and of the roles of the major muscles and ligaments.

1.2

Introduction to biomechanics

1.2.1

Defining the biomechanical properties of a joint: degrees of freedom and constraints

Almost without exception, human joints have more than one axis of rotation. The joints of the fingers, while they may superficially be viewed as hinge joints, allow small out-of-plane rotations and translations. Therefore, while the anatomical conventions suffice for clinical discussion, there is a need for a more rigorous set of definitions for biomechanical analysis. In general, the movement of a body is composed of two types: rotation, in which a defined point in the body rotates about a defined axis, and translation, in which motion occurs along a line. Considering first a simple hinge joint, then, only a single quantity is needed to define the position (e.g., the angle of flexion of a finger joint). However, if translation also takes place (perhaps due to ligamentous laxity) then a second © 2008, Woodhead Publishing Limited

4

Joint replacement technology

quantity is required to define the relative position of the two bones. These quantities are degrees of freedom which may be defined as the number of independent quantities required to define a position. Thus, a single uncoupled rigid body in three-dimensional space, capable of three translations and three rotations, has six degrees of freedom. Any constraint applied to the rigid body ± these constraints may take the form of geometric features (e.g., the approximate ball and socket construction of a hip) or external connecting structures such as a ligament ± will reduce the number of degrees of freedom from this maximum of six. It should further be pointed out that the coupling between degrees of freedom (e.g., the translations that accompany flexion/extension of the knee) are kinematic constraints and reduce the number of independent movements. Furthermore, in many cases, the simplified view of a human joint may suggest perhaps a single degree of freedom (e.g., knee) but more detailed studies reveal further movements which are rather smaller but, nevertheless, may be clinically important.

1.2.2

Forces and moments

Basic Newtonian mechanics According to Newton's First Law of Motion, a body will continue to move at a constant velocity unless acted upon by a force. Thus, a force may be defined as an action which causes acceleration of a mass. Force is a vector quantity ± that is to say that it must be specified in terms of both magnitude and (three-dimensional) direction. According to Newton's Second Law of Motion, the acceleration of a body is proportional to and occurs in the direction of an applied force: F ˆ ma where force (F) is measured in newtons, mass (m) is in kilograms (kg) and acceleration (a) is in m/s2. Vectors and equilibrium Figure 1.1 shows a system of forces acting on a particle (i.e., a rigid body having no physical size). The resultant force corresponding to a combination of forces can be found as the vector sum ± shown graphically in Fig. 1.1; this shows that there is a net force acting on the particle, i.e. it is not in equilibrium. For the particle to be in equilibrium the resultant of the forces must be zero and so the result of the graphical summation of the vectors must be a closed figure (Fig. 1.2). The solution of the majority of biomechanics problems involves the analysis of equilibrium and a clear understanding is necessary to understand a wide range of problems involving external, joint and muscle forces. It should also be noted that this vector approach can be used `in reverse' so that a vector may be broken © 2008, Woodhead Publishing Limited

Biomechanics of joints

5

1.1 Summation of force vectors acting on a particle ± vectors are added `head to tail'. Resultant vector is from first tail to final head.

1.2 Equilibrium of force vectors ± the rules for addition are identical to those in Fig. 1.1. However, in this situation the end point of the vector addition coincides with the start point so that there is zero net resultant.

down in to components (usually mutually perpendicular). This is particularly useful for solving some equilibrium problems. Dynamics In situations where the forces are not in equilibrium, then the particle will experience an acceleration, according to Newton's Second Law. The acceleration will have a magnitude dependent upon its mass and a direction corresponding to that of the resultant force. Using vector notation: X ~ F ˆ m~ a This analysis of dynamics is key to the understanding of biomechanical motion. For instance, the detailed calculation of the loading of the lower limb during gait requires this approach. Rotations and moments If a system of forces acts on a finite rigid body, then it is important to consider both translation and rotation. In particular, it is possible that, while a set of © 2008, Woodhead Publishing Limited

6

Joint replacement technology

1.3 Moment produced by a force acting at a distance from the centre of rotation ± note that the moment is equal to the magnitude of force F and the perpendicular distance h.

forces has a zero resultant force, the points of application are such that they cause a rotation. Similarly, where there is a rotational degree of freedom, then the net resultant force may not pass through the centre of rotation and will produce a moment. Moments, which may be thought of as the turning effect of a force, are of particular importance to the mechanics of joints since these are the actions of muscles, e.g. quadriceps at the knee. Mechanically, the moment of a force about a point is defined as the magnitude of the force multiplied by the perpendicular distance between the point and the line of action of the force. Moments have units of newton±metre (N m). The generation of a moment is shown in Fig. 1.3 illustrating a simplified joint acted upon by a single force which does not pass through the centre of rotation. This leads to a moment about the joint centre equal to F (the magnitude) of the force multiplied by h, the perpendicular distance between the centre of rotation and the line of action of the force.

1.2.3

Equilibrium of a joint: role of joint structures, muscles and ligaments

An arthrodial joint consists of joint surfaces of known (but to some degree variable) geometry, and is crossed by both ligaments and muscles/tendons. For a © 2008, Woodhead Publishing Limited

Biomechanics of joints

7

joint to be in equilibrium after the application of external loads, then the appropriate forces and moments must be produced by these crossing structures. Using the representation of Fig. 1.3 it is now possible to look at the procedure for determining the system of forces acting on a body, e.g. a bone. Equilibrium of forces must be achieved across the joint and the external moment must be balanced by an equal and opposite moment produced by muscle(s). To understand this clearly, it is important to separate the two halves of the joint and to consider free body diagrams of the two bones. It is important to distinguish between the joint contact forces and the external loads. A free body diagram of the ball section of the joint is shown in Fig. 1.4. If we assume that there is no friction at the joint (this is usually realistic for human joints where the coefficients of friction are remarkably small), then the reaction force between the ball and socket must pass though the centre of rotation. In addition, for equilibrium, there must be an external moment M on the joint to balance the moment created by the other forces (which are not collinear). The major role of muscles is to produce joint moments ± the ability to do this is measured by the moment arm which may be defined as the moment produced by a force of 1 N in the muscle. For most joints and muscles, the moment arms are relatively small, so that large muscle forces are commonly required to produce the necessary moments.

1.4 Free body diagram to calculate external forces and moments ± the joint shown in Fig. 1.3 has been `disarticulated' so that the forces acting on a single component can be analysed. © 2008, Woodhead Publishing Limited

8

1.2.4

Joint replacement technology

Applications to joint mechanics

Elbow flexion Figure 1.5 is a free body diagram of the forearm in order to determine the force in a flexor muscle acting across the elbow. There are two external loads acting on the forearm ± the weight of the forearm mf g and a mass being held in the hand mg. At the centre of the elbow joint there are two force vectors Fx and Fy representing the force transmitted across the elbow joint. The vector Fm represents the muscle force which has a moment arm equal to the perpendicular distance h. In order to calculate the muscle and joint forces it is necessary to calculate the conditions for equilibrium of the forearm. This requires the satisfaction of three conditions ± equilibrium in x direction, equilibrium in y direction and equilibrium of the moments generated about the centre of the elbow joint. Mathematically this is as follows: · Resolving forces vertically: Fmy ‡ Fy ÿ mg ÿ mf g ˆ 0 · Resolving forces horizontally: Fx ÿ Fmx ˆ 0 · Taking moments about elbow centre: Fm  h ÿ mf g  xf ÿ mg  x ˆ 0

1.5 Free body diagram of forearm when supporting a hand-held weight. Note the force vectors representing the weight carried and the weight of the forearm. The vector triangle illustrates how the muscle force line of action may be broken down into two components corresponding with the coordinate axes. © 2008, Woodhead Publishing Limited

Biomechanics of joints

9

Using approximate values for the masses and dimensions as follows: x ˆ 300 mm xf ˆ 150 mm m ˆ 10 kg mf ˆ 2 kg h ˆ 30 mm Fmy ˆ 0.94Fm Fmx ˆ 0.34Fm yields: Muscle force Fm ˆ 1079 N Fx ˆ 367 N Fy ˆ ÿ896.5 N (i.e. this force acts downwards on the forearm) q  Resultant join force F ˆ

F2x ‡ F2y ˆ 968:7 N

Note that these forces are much larger than the load being carried (98.1 N). This results from the fact that the moment arm of the flexor muscle is very much smaller than the length of the forearm. Hip ± single legged stance A good example of the importance of joint moments is the need for equilibrium of the hip while standing on one leg (a necessity for unaided gait). Figure 1.6 shows a simplified two-dimensional view of the hip joint while standing on one leg (McLeish and Charnley, 1970). In this situation, a moment about the hip arises because of its distance from the line of action of the ground reaction force. Equilibrium at the hip is achieved by the abductor muscles. A further consideration of equilibrium is required to calculate the resulting joint contact force at the hip. Some important conclusions emerge from this analysis: · The muscle forces contribute to the joint contact force. · Since the muscle line of action lies close to the joint centre (i.e., the moment arm is small), then the muscle forces required to achieve a given moment are likely to be large. · The consequence of the above is that joint forces are likely to be considerably larger than body weight (for instance, we know from experimental and modelling studies that the contact forces at the hip can be in excess of four times the body weight).

1.2.5

Materials science and engineering: stress, strain, failure and fatigue

Both biological and non-biological materials can be characterised by their behaviour under load. Considering first metallic materials, then these all obey © 2008, Woodhead Publishing Limited

10

Joint replacement technology

1.6 Moment at the hip when standing on one leg. Note how the resultant of body weight (excluding the weight of the supporting leg) acts at a much larger distance from the centre of rotation of the hip than do the abductor muscles. © 2008, Woodhead Publishing Limited

Biomechanics of joints

11

1.7 Diagram illustrating applied direct and shear stresses applied to a surface.

Hooke's Law ± that is to say that, under the action of a load, they will exhibit a deformation that is proportional to the applied load. If this statement is generalised, so that force/area ˆ stress, and proportional deformation ˆ strain, then we may write:  ˆ Ee where  ˆ stress (N/m2), e ˆ strain (dimensionless) and E ˆ Young's modulus (N/m2). There are two types of stress: normal stress in which a load is transmitted normal to a surface and shear stress where load is transmitted parallel to a surface (see Fig. 1.7). In fact, in virtually all applications, materials are subject to both types of stress simultaneously.

1.2.6

Stresses due to bending and torsion

While, in some cases, these stresses may result from the direct application of a force (e.g., tension in a tendon), bending and/or torsion are the most common causes. It has already been shown that muscle forces act to create moments at joints. Similarly, they can act to produce bending moments in long bones such as the femur and particularly in hip prostheses having inadequate proximal support. Torsion on a structure leads to shear stresses. A good biomechanical example is the incidence of tibial fractures in skiing accidents which can be largely prevented by the use of appropriate bindings. Although metals obey Hooke's law within a limited range of stress, it is necessary to look at the stress/strain graph of a material to gain a full understanding of its behaviour under load (see Fig. 1.8). Figure 1.8(a) shows the stress±strain graph for a typical metallic material used for total joint replacement. It can be seen that, as the stress is increased, there is an increasing strain (deformation) which is proportional to the stress up until the limit of proportionality ± this is linear elastic behaviour. In this region, the gradient of the graph © 2008, Woodhead Publishing Limited

12

Joint replacement technology

1.8 (a) Stress±strain diagram showing ductile behaviour in a tensile test. (b) Brittle behaviour in which fracture occurs before yield, i.e. there is no limit of proportionality.

is a measure of material stiffness measured as Young's modulus. Some typical values of this parameter are shown in Table 1.1. After this point, as the strain continues to increase, the stress is increasing more slowly. This latter part of the graph represents yield in which there is permanent deformation. It should be noted that, while in the elastic region all deformation will be lost on the removal of the stress, after yield has occurred then the material will not fully recover. This yield (or plastic) deformation is frequently regarded as a desirable property in that, if a component is overloaded, then permanent deformation rather than fracture will occur. Examination of the stress±strain graph readily provides important design information. In particular it is important to look at Fig. 1.8(b) showing a material in which fracture occurs before yield. This is a brittle material. In such a material, fracture can occur without warning and there is Table 1.1 Physical properties of some important structural materials Material

Density (mg/m3)

Mild steel Stainless steel High strength steel Aluminum alloy Titanium alloy Compact bone Ultra-high molecular weight polyethylene (UHMWPE) Poly(methylmethacrylate)

© 2008, Woodhead Publishing Limited

7.8 7.8 7.8 2.7 4.5 2.0 0.93 1.1

Young's modulus (GPa) 210 210 210 70 100 14 0.725 2.0

Yield stress (MPa)

Ultimate tensile strength (MPa)

200 240 1240 500 910 100

380 590 1550 570 950 100

23 ±

53 30

Biomechanics of joints

13

no opportunity for energy to be absorbed in yield, meaning that in the context of orthopaedic implants, there is a risk of catastrophic failure. It is important to make the point that brittle fractures occur more commonly under tensile stresses ± brittle materials are stronger in compression than in tension. Fatigue In many applications (including biomechanical), components are subjected to a cyclically varying stress, e.g. the bending stress on a total hip replacement. After a large number of repetitions, this cyclical loading can lead to fatigue failure, which takes the form of a crack propagating through the structure until it is no longer strong enough to carry the applied load. The number of cycles leading to such failure is a function of material static properties, the type of loading, the rate of application and any features which may lead to local stress concentrations. This behaviour is normally represented by an S±N curve showing the relationship between the applied stress amplitude and the number of cycles to failure. Biological and non-metallic materials Biological and non-metallic materials differ from metals in two important ways ± they no longer have a linear stress/strain relationship (i.e., they may not obey Hooke's law) and second, their stress/strain behaviour is frequently influenced by the rate of strain. Figure 1.9 shows the stress/strain behaviour of a commonly used biomedical polymer and cortical bone (at different strain rates).

1.9 Illustration of viscoelastic behaviour. Note that when a stress is applied instantaneously, there is a time delay in the resulting strain. The same effect occurs when the stress is removed. © 2008, Woodhead Publishing Limited

14

Joint replacement technology

1.3

Key aspects of biomechanics of major joints

1.3.1

Lower limb ± hip, knee and ankle

Forces and moments during walking The major functional activity of the lower limb is, of course, that of walking and so it is important to look first at the external forces and moments during this activity. Typical forces are shown in Fig. 1.10. As a result of the ground reaction force, there are external forces and moments produced at the hip, knee and ankle. As discussed above, the need for the muscles to achieve equilibrium at each of the joints leads to the internal joint forces which are of major importance to designers of joint replacements. The associated joint moments are shown in Fig. 1.11.

1.3.2

Hip joint

Basic anatomy and kinematics For almost all biomechanical analysis, the hip may be considered as a three degrees of freedom ball and socket joint. The ball and socket arrangement is further strengthened by a strong ligamentous band between the femur and the

1.10 Ground reaction forces during normal walking (data from Winter, 1991). © 2008, Woodhead Publishing Limited

Biomechanics of joints

15

1.11 Hip, knee and ankle moments during walking (data from Winter, 1991). © 2008, Woodhead Publishing Limited

16

Joint replacement technology

pelvis. There is, in addition, an internal ligament ± the fovea. The socket is deep and so dislocation of the hip in adults is relatively rare. Muscles and forces The hip joint is controlled by large muscles, some of which also cross the knee. In some cases a muscle itself may cross the joint, but in other situations, there will be a tendon attachment. The actions of the major muscles at the hip are summarised in Table 1.2. The internal joint forces at the hip during walking have been predicted by Paul (1966) and more recently by Stansfield et al. (2003) who was able to compare them with the in vivo loads measured using instrumented implants (Bergmann et al., 2001). While Paul's work was predicting peak loads of around Table 1.2 Actions of major muscles at the hip (from Palastanga et al., 2006) Direction

Muscles

Flexion

Psoas major Iliacus Pectineus Rectus femoris Sartorius

Extension

Gluteus maximus Hamstrings (semitendinosus, semimembranosus, biceps femoris)

Abduction

Gluteus maximus Gluteus medius Gluteus minimus Tensor fascia lata

Adduction

Adductor magnus Adductor longus Adductor brevis Gracilis Pectineus

Internal rotation

Gluteus medius (anterior part) Gluteus minimus (anterior part) Tensor fascia lata Psoas major Iliacus

External rotation

Gluteus maximus Piriformis Gemellus superior Gemellus inferior Quadratus femoris Obturator externus

© 2008, Woodhead Publishing Limited

Biomechanics of joints

17

4 body weight (BW), the in vivo studies showed rather smaller forces of 2.4 BW during level walking at 4 km/h. These latter data were, of course, recorded from patients with joint replacements rather than a normal healthy hip.

1.3.3

Knee joint

Basic anatomy and kinematics The basic anatomy of the knee is shown in Fig. 1.12. While, at the most basic level, the knee may be thought of as a single degree of freedom hinge in the sagittal plane, the kinematics are rather more subtle. Understanding of the sagittal kinematics depends on examining the geometry of the joint surfaces together with the arrangement of the cruciate ligaments. The manner in which this leads to a four bar linkage has been discussed in detail by Zavatsky and O'Connor (1992a,b). This resulting motion consists of a combination of rotation and translation (two degrees of freedom) which are coupled by a kinematic constraint leading to a single degree of freedom movement ± that is to say the position of the femur with respect to the tibia can be completely defined by a

1.12 Diagram showing the major biomechanical structures at the knee. © 2008, Woodhead Publishing Limited

18

Joint replacement technology

single measurement ± usually joint angle. It should be noted that the geometry of the tibial plateau is such that there would be little anterior posterior constraint without this ligamentous arrangement. The relatively complex kinematics of the knee make it essential to define the degrees of freedom carefully. In particular, while there may be kinematic coupling in the healthy knee, injury or pathology may reduce or destroy these constraints and so effectively increase the available independent degrees of freedom. Many of the clinical tests in routine use (e.g., anterior drawer test) are intended to identify and quantify these additional degrees of freedom. When designing total joint replacements it is essential, at the design stage, to decide on the amount of constraint or degrees of freedom to be incorporated into the design. It is also important to realise that this is an oversimplification, particularly in three dimensions, when tibial rotation about the long axis must be taken into consideration. The amount of available rotation is related to the angle of flexion and the configuration of the collateral ligaments. Major muscles, ligaments and forces The knee has a large range of motion (predominantly two dimensional) and is able to support large moments ± particularly flexion moments, for instance when descending into a deep squat. Because of its largely two-dimensional nature, the muscles can be divided into two groups ± flexors and extensors. Perhaps of more importance are the passive structures of the knee ± the menisci and the ligaments. The need for a large range of flexion leads to the use of a highly non-conforming geometry ± at the simplest level the tibial plateau may be regarded as a flat surface. This geometry implies a very small contact area between the plateau and the curved femoral condyles which, bearing in mind the high loads to be transmitted, would lead to high stresses in the articular cartilage. This problem is largely overcome in the knee by the presence of menisci (see in Fig. 1.12), which are saucer-shaped structures of fibrocartilage allowing the transmission of the compressive joint force as a tensile stress. The menisci can also slide on the tibial plateau to accommodate the kinematics discussed above. As was mentioned above, the tibial plateau is such that it cannot, in its interaction with the surface of the femur, transmit significant shear (anterior±posterior, AP) forces. Therefore, these loads must be transmitted by the cruciate ligaments. To summarise, the major active and passive stabilisers of the knee are shown in Table 1.3.

1.3.4

Patellofemoral joint

The extensor muscles of the knee terminate at a sesamoid bone, the patella, which attaches to the tibia by a short ligament. This arrangement allows the production of high extension moments by transmission of high loads around the © 2008, Woodhead Publishing Limited

Biomechanics of joints

19

Table 1.3 Active and passive stabilisers of the knee (from Palastanga et al., 2006) Direction

Active

Passive

Flexion

Hamstrings Gastrocnemius Gracilis Sartorius

Extension

Quadriceps Tensor fascia lata

Internal rotation

±

Collateral ligaments (at full extension)

External rotation

±

Collateral ligaments (at full extension)

Valgus

±

Medial collateral ligament

Varus

±

Lateral collateral ligament

Anterior±posterior

±

Cruciate ligaments

joint. The resulting patellofemoral joint is a synovial articulation in which the geometry of the patella allows it to slide in the intercondylar groove of the femur. This relatively conforming joint is required to transmit patellofemoral contact loads which can be as high as 1.6 kN (2.3 BW approx.) (Singerman et al., 1994) when loading a flexed knee ± for instance in a squat. These loads act on a small contact area leading to particular technical challenges in the design of patellar replacements.

1.3.5

Ankle joint

Anatomy and kinematics Rather than an individual joint, the ankle should be thought of as a joint complex consisting of the talocrural joint and the subtalar joint. Both of these joints have effectively single axes both of which are inclined obliquely with respect to the standard anatomical axes (Mann and Inman, 1964). The talocrural axis is inclined by approximately 6ë to the mediolateral direction and by approximately 8ë in the frontal plane. The subtalar joint (see Fig. 1.13) lies at around 23ë from the A±P direction in the horizontal plane and at 42ë in the sagittal plane. This joint has been described as a mitre hinge joint by Mann and colleagues; this description explains clearly the manner in which internal rotation of the lower leg can result in supination of the foot and vice versa. As a result of the arrangement of the joints complex, the ankle can be seen to have two degrees of freedom. While the axes of the joints do not coincide with preferred anatomical axes, the resulting motion of the ankle complex can be regarded as a combination of inversion/eversion and plantar/dorsiflexion. © 2008, Woodhead Publishing Limited

20

Joint replacement technology

1.13 Diagram showing the major biomechanical structures at the ankle.

Major muscles, ligaments and forces The talocrural joint is of a tenon and mortise structure with strong medial and lateral collateral ligaments capable of withstanding the significant moments which can result from support ground reaction forces on the inverted or everted foot. These ligaments are organised in such a way as not to obstruct plantar or dorsiflexion. The greatest moments at the ankle during gait are in dorsiflexion requiring a plantar flexion moment to be generated by forces in the Achilles tendon. In fact this moment, which occurs in late stance in normal walking, is the largest joint moment in the lower limb throughout the gait cycle. The major muscles acting at the ankle are listed in Table 1.4.

1.4

The upper limb

While the mechanics and loading of the lower limb are largely prescribed by a single activity ± walking ± the loading of the upper limb is considerably more varied. Furthermore, the need to perform a wide range of tasks calls for a large © 2008, Woodhead Publishing Limited

Biomechanics of joints

21

Table 1.4 Actions of major muscles at the ankle (from Palastanga et al., 2006) Direction

Muscle

Plantarflexion

Gastrocnemius Soleus Plantaris Peroneus longus Flexor digitorum longus Flexor hallucis longus

Dorsiflexion

Tibialis anterior Extensor digitorum longus Extensor hallucis longus Peroneus tertius

Inversion

Tibialis posterior Tibialis anterior

Eversion

Peroneus longus Peroneus brevis Peroneus tertius

range of motion of the hand. This is achieved, particularly, by the large range of motion at the shoulder complex (Murray and Johnson, 2004). Although the external loading is highly task dependent, it is useful to summarise the external loading at the shoulder and elbow during some everyday tasks (Table 1.5).

1.4.1

Shoulder

Anatomy and kinematics The shoulder joint should be considered as a joint complex rather than a single joint ± the required large movements of the upper arm relative to the trunk are achieved by the combined movements of the glenohumeral and scapulothoracic joints. The kinematics are further constrained by the clavicle providing a link between the acromion and the thorax. The particularly unusual feature of the shoulder complex is the controlled kinematic relationship (scapulohumeral rhythm) between the humerus, scapula and thorax. This has been studied by a number of researchers; while early radiographic studies suggested a linear relationship between scapula and humeral angles, more recent work, using instrumented palpation, has demonstrated a non-linear three-dimensional relationship (Barnett et al., 1999). The glenohumeral joint which has a range of motion of approximately 120ë should be thought of as a ball and saucer rather than a ball and socket joint. Although some constraint is provided by the labrum around the glenoid saucer, joint stability is achieved largely by the rotator cuff muscles, particularly for the © 2008, Woodhead Publishing Limited

Table 1.5 Ranges of motion and external moments at the shoulder and elbow during a range of tasks of daily living (Murray and Johnson, 2004). Shoulder Range of motion (degrees) Moments (N m) Elbow Range of motion (degrees) Moments (N m)

Flexion 14.7 (7.6) 0

Abduction

111.9 (7.4) +14.3 (1.4)

ÿ20.1 (9.2) ÿ3.7 (1.2)

Flexion 15.6 (6.6) ÿ2.8 (0.9)

39.7 (6.9) ‡4.2 (1.8)

Pronation

164.8 (8.0) 5.8 (0.5)

ÿ53.7 (12.6) ÿ0.026 (0.028)

© 2008, Woodhead Publishing Limited

WPTF3007

65.3 (8.2) 0.025 (0.026)

Internal rotation 18.7 (7.8) ÿ85.9 (11.7) 0 ‡3.9 (0.6) Internal rotation ± ÿ0.8 (0.1)

± 0.2 (0.1)

Biomechanics of joints

23

prevention of superior migration. The overall range of motion of the scapula on the thorax is approximately 50ë. Muscles and forces Because of the complexity of the shoulder complex and the interactions between glenohumeral, scapulohumeral and thoracohumeral muscles, it is not appropriate to present a table of the actions of each muscle; details of these muscles are presented in Johnson et al. (1996). Motion of the upper arm is achieved largely through combined contributions of the deltoid muscle attaching at the distal end of the humerus and the rotator cuff muscles attaching to the proximal humerus close to the humeral head, and to the scapula (Fig. 1.14). Modelling studies

1.14 The bony anatomy of the shoulder complex. © 2008, Woodhead Publishing Limited

24

Joint replacement technology

suggest that deltoid is of key importance during abduction but demonstrates the vital role of the rotator cuff muscles ± infraspinatus and subscapularis for other movements (Charlton and Johnson, 2006). While there are loads transmitted by all of the components of the shoulder complex, the loading of the glenohumeral joint is of the greatest importance from the viewpoint of joint replacement. The loads at this joint during activities of daily living have been predicted in a number of modelling studies, Poppen and Walker (1978), van der Helm (1994) and Charlton and Johnson (2006) all suggesting loads of 0.5±0.75 BW during scapular plane abduction. Only recently, in vivo data are becoming available from studies using instrumented prostheses (Bergmann et al., 2007), which have reported loads of 0.9 BW during similar movements and appear to be in general agreement with the model predictions. However, much further work of this kind is required for confidence in the available models. Clearly, much higher loading is to be expected during more strenuous sporting activities, e.g. baseball pitching.

1.4.2

Elbow

Anatomy and kinematics At the basic level, the elbow may be considered as a single degree of freedom hinge joint. However, the anatomy is complicated by the need to accommodate articulations with both ulna and radius. Because of this arrangement, it is best to consider the elbow as a two degree of freedom mechanism allowing elbow flexion/extension and forearm pronation/supination. Internally, there are three separate synovial joints ± humero-ulnar, humero-radial and radio-ulnar with subtle interactions. Of particular interest is the humero-radial joint in which there occurs a combination of relative motions ± elbow flexion (shared with the ulna) and axial rotation of the radius accompanying forearm pronation/supination. The basic geometry of the three joints is shown in Fig. 1.15. Muscles and forces The muscles acting across the elbow joint (brachialis, biceps brachii, brachioradialis and triceps) all produce flexion or extension moments. Pronation is produced by forearm muscles (pronator teres, pronator quadratus and flexor carpi radialis). Supination is achieved by a combination of supinator (in forearm) and biceps brachii which, because of its attachment to the ulna, provides a strong supination moment. The muscles acting at the elbow are listed in Table 1.6. The contact forces at the individual joints have been predicted using modelling approaches. Chadwick and Nicol (2000) have calculated for a range © 2008, Woodhead Publishing Limited

Biomechanics of joints

25

1.15 The bony anatomy of the elbow.

of tasks predicting loads of 1600 N (2.3 BW approx.) in the humero-ulnar joint and 800 N (1.1 BW approx) in the humero-radial joint. In earlier studies of patients with rheumatoid arthritis (Amis et al., 1979), forces in the humero-ulnar joint of up to 0.65 kN in isometric extension and humero-coronoid forces of 1.49 kN have been described during isometric flexion. The corresponding forces in the humero-radial joint were 1.44 kN and 1.41 kN respectively.

Table 1.6 Actions of major muscles at the elbow (from Palastanga et al., 2006) Direction

Muscle

Elbow flexion

Brachialis Biceps brachii Brachioradialis

Elbow extension

Triceps brachii

Forearm pronation

Pronator teres Pronator quadratus Flexor carpi radialis

Forearm supination

Supinator Biceps brachii

© 2008, Woodhead Publishing Limited

26

1.4.3

Joint replacement technology

Temporomandibular joint

Anatomy and kinematics This joint complex between the jaw and the skull is unusual in a number of ways. The individual joints, which can be considered a partially constrained ball and socket, have a unique configuration. The joint is a synovial joint containing a fibro-cartilage disc. While the condyle of the jaw is curved to allow angular motion against the disc, the skull socket is relatively flattened so that, with the ligamentous arrangement, it can allow forward and backward translation. Because of the flexible nature of the disc and the ill-conforming joints, it is difficult to define exactly the available degrees of freedom. However, it is suggested that the principal movements are two degrees of freedom of rotation combined with a single translation, i.e. three degrees of freedom (Fig. 1.16). When considering the mechanics of the assembled jaw, it is necessary to look at the mechanism resulting from the essentially rigid connection of the two joints. From the point of view of the kinematics, it is probably reasonable to assume that each individual joint has four degrees of freedom. Since the rigid bony connection imposes rigid constraints, the resulting mechanism can be seen to have three degrees of freedom ± opening (depression) and closing (elevation), forward/backward translation (protraction/retraction) and angular rotation about the vertical axis causing side to side movements of the jaw. Muscles and forces Because of its inherent laxity, movements of the temporomandibular joint are limited by three ligaments ± lateral ligament, sphenomandibular ligament and

1.16 Illustration of the kinematics of the temporomandibular joint. In particular, it should be noted how the translation available at each side make available a further rotational degree of freedom of the jaw. © 2008, Woodhead Publishing Limited

Biomechanics of joints

27

stylomandibular ligament. The movements of the jaw are achieved by the masticatory muscles ± masseter, temporalis, and medial and lateral pterygoid. The greatest moments available are those for closing the mouth and chewing produced by the combined action of masseter, medial pterygoid and temporalis. Bite forces for normal men have been reported to be 300 N (May et al., 2001) with associated joint forces of 250 N.

1.4.4

Intervertebral joints

A brief discussion of the mechanics of the intervertebral joint is included here for completeness and to demonstrate a different approach to an articulation. The intervertebral joint is considered as a unit consisting of two vertebrae connected by an intervertebral disc. This arrangement is not an arthrodial joint but the connection of two bones (vertebral bodies) by a flexible intervertebral disc having special biomechanical properties. The joint is remarkable further because there are additional synovial joint surfaces (zygapophysial joints) which transmit load only under particular circumstances ± types of loading or posture. For instance, if the upper disc rocks backwards, then loads can be transmitted by the articular processes of these synovial joints (Fig. 1.17). Similarly, an axial load on the unit will be shared between the disc and the articular processes (Fig. 1.18). The intervertebral disc itself may be considered as a pressure vessel in which a fibrous outer sack contains a viscoelastic gel (nucleus pulposus). From the

1.17 Vertebral anatomy illustrating the way in which extension of the spine may lead to load transmission by articular processes. © 2008, Woodhead Publishing Limited

28

Joint replacement technology

1.18 Sharing of load between articular processes and intervertebral disc under application of an axial load.

viewpoint of kinematics, flexion and extension (forward or lateral) are permitted by this flexible disc structure. Axial applied load can be supported by two mechanisms ± hydrostatic pressure in the disc and axial loading of the fibrous structure. While it is entirely possible for the disc to carry the necessary loads imposed on the spinal column, the zygapophysial (synovial) joints are engaged and can then transmit axial loads. The degree of load bearing by the zygapophysial joints in the lumbar spine has been variously reported as between 16% and 40% of the total load. Ligaments also play an important role in determining the behaviour of the intervertebral joint. If the joint is regarded as having three (rotational) degrees of freedom, the ranges of motion of the unit are limited either by ligaments or by zygapophysial joints. In summary, the intervertebral joint is a unique structure. The combination of the intervertebral disc and the vertebrae allows it to transmit high loads while providing a high degree of flexibility. The spine can, of course, suffer injury and © 2008, Woodhead Publishing Limited

Biomechanics of joints

29

pathology which is difficult to manage. Because of this, there is considerable interest in the development of artificial discs ± hence the inclusion in this chapter.

1.5

Summary

The purpose of this chapter has been to provide a refresher on basic mechanics and to illustrate the application of these principles to the major candidate joints for replacement. Inevitably, much detail has been omitted. With regard to the biomechanics, then the reader is recommended to study the texts listed below. The detail aspects of the individual joints are, of course, covered in the following chapters.

1.6

Sources of further information and advice

Bogduk N (1997) Clinical Anatomy of the Lumbar Spine and Sacrum, 3rd edn. Churchill Livingstone. Nigg B M and Herzog W (2007) Biomechanics of the Musculo-skeletal System, 3rd edn. John Wiley & Sons Palastanga N, Soames R W and Field D (2006) Anatomy and Human Movement: Structure and Function, 5th edn. Butterworth Heinemann Smith L K, Weiss E L and Lehmkuhl L D (1996) Brunnstrom's Clinical Kinesiology, 5th edn. F.A. Davis. Winter D A (1991) The Biomechanics and Motor Control of Human Gait: Normal, Elderly and Pathological. University of Waterloo Press. Zatsiorsky V M (1998) Kinematics of Human Motion. Human Kinetics

1.7

References

Amis, A. A., Hughes, S., Miller, J. H., Wright, V., and Dowson, D. 1979, `Elbow joint forces in patients with rheumatoid arthritis', Rheumatol. Rehabil., vol. 18, no. 4, pp. 230±234. Barnett, N. D., Duncan, R. D., and Johnson, G. R. 1999, `The measurement of three dimensional scapulohumeral kinematics ± a study of reliability', Clin. Biomech. (Bristol., Avon.), vol. 14, no. 4, pp. 287±290. Bergmann, G., Deuretzbacher, G., Heller, M., Graichen, F., Rohlmann, A., Strauss, J., and Duda, G. N. 2001, `Hip contact forces and gait patterns from routine activities', J. Biomech., vol. 34, no. 7, pp. 859±871. Bergmann, G., Graichen, F., Bender, A., Kaab, M., Rohlmann, A., and Westerhoff, P. 2007, `In vivo glenohumeral contact forces ± measurements in the first patient 7 months postoperatively', J. Biomech., vol. 40, no. 10, pp. 2139±2149. Chadwick, E. K. and Nicol, A. C. 2000, `Elbow and wrist joint contact forces during occupational pick and place activities', J. Biomech., vol. 33, no. 5, pp. 591±600. Charlton, I. W. and Johnson, G. R. 2006, `A model for the prediction of the forces at the glenohumeral joint', Proc. Inst. Mech. Eng [H.], vol. 220, no. 8, pp. 801±812. Johnson, G. R., Spalding, D., Nowitzke, A., and Bogduk, N. 1996, `Modelling the muscles of the scapula morphometric and coordinate data and functional © 2008, Woodhead Publishing Limited

30

Joint replacement technology

implications', J. Biomech., vol. 29, no. 8, pp. 1039±1051. Mann, R. and Inman, V. T. 1964, `Phasic activity of intrinsic muscles of the foot', J. Bone Joint Surg. Am., vol. 46, pp. 469±481. May, B., Saha, S., and Saltzman, M. 2001, `A three-dimensional mathematical model of temporomandibular joint loading', Clin. Biomech. (Bristol., Avon.), vol. 16, no. 6, pp. 489±495. McLeish, R. D. and Charnley, J. 1970, `Abduction forces in the one-legged stance', J. Biomech., vol. 3, no. 2, pp. 191±209. Murray, I. A. and Johnson, G. R. 2004, `A study of the external forces and moments at the shoulder and elbow while performing everyday tasks', Clin. Biomech. (Bristol., Avon.), vol. 19, no. 6, pp. 586±594. Palastanga N, Soames R W and Field D (2006) Anatomy and Human Movement: Structure and Function, 5th edn. Butterworth Heinemann Paul, J. P. 1966, `Biomechanics. The biomechanics of the hip-joint and its clinical relevance', Proc. R. Soc. Med., vol. 59, no. 10, pp. 943±948. Poppen, N. K. and Walker, P. S. 1978, `Forces at the glenohumeral joint in abduction', Clin. Orthop. Relat Res., no. 135, pp. 165±170. Singerman, R., Berilla, J., Kotzar, G., Daly, J., and Davy, D. T. 1994, `A six-degree-offreedom transducer for in vitro measurement of patellofemoral contact forces', J. Biomech., vol. 27, no. 2, pp. 233±238. Stansfield, B. W., Nicol, A. C., Paul, J. P., Kelly, I. G., Graichen, F., and Bergmann, G. 2003, `Direct comparison of calculated hip joint contact forces with those measured using instrumented implants. An evaluation of a three-dimensional mathematical model of the lower limb', J. Biomech., vol. 36, no. 7, pp. 929±936. van der Helm, F. C. 1994, `Analysis of the kinematic and dynamic behavior of the shoulder mechanism', J. Biomech., vol. 27, no. 5, pp. 527±550. Winter, D. A. 1991, The Biomechanics and Motor Control of Human Gait: Normal, Elderly and Pathological. University of Waterloo Press. Zavatsky, A. B. and O'Connor, J. J. 1992a, `A model of human knee ligaments in the sagittal plane. Part 1: Response to passive flexion', Proc. Inst. Mech. Eng [H.], vol. 206, no. 3, pp. 125±134. Zavatsky, A. B. and O'Connor, J. J. 1992b, `A model of human knee ligaments in the sagittal plane. Part 2: Fibre recruitment under load', Proc. Inst. Mech. Eng [H.], vol. 206, no. 3, pp. 135±145.

© 2008, Woodhead Publishing Limited

2

Tribology in joint replacement Z J I N and J F I S H E R , University of Leeds, UK

2.1

Introduction

2.1.1

Tribology

Tribology is defined as `the study of friction, wear and lubrication, and design of bearings, science of interacting surfaces in relative motion' (Concise Oxford Dictionary, `tribo-' is derived from the Greek word `tribos', meaning rubbing and friction). It was only introduced into English literature in 1966 in the Jost Report (Lubrication (tribology) Education and Research, Department of Education and Science, HMSO, 1966) and was formally defined as `The science and technology of interacting surfaces in relative motion and the practices related thereto'. It encompasses a number of basic engineering subjects such as solid mechanics, fluid mechanics, lubricant chemistry, material science and heat transfer. Important considerations in tribology include surfaces, both microscopic surface topographies and macroscopic bearing geometries, bearing materials, relative motion and loading as well as lubricants. The transient nature of tribology processes should be pointed out, since the loading and motion involved are often dynamic and the wear of the bearing surfaces can modify the geometry both microscopically and macroscopically. Tribology plays an important role in the functioning of artificial joints. Hip joints are subjected to a large dynamic load during normal walking, up to a few times bodyweight, and yet often accompanied with a large range of motions. Friction played an important role in the design of original Charnley low-friction arthroplasty. Wear is important, not only from the integrity of the prosthetic component point of view, but also from that of wear debris which can cause adversely biological reactions. Lubrication can be the most effective means to reduce both friction and wear.

2.1.2

Surfaces and roughness

Tribology is mainly concerned with the surfaces in relative motion. Therefore the surface profile, texture and topography are all important. For example, in an © 2008, Woodhead Publishing Limited

32

Joint replacement technology

2.1 Design and manufacturing parameters associated with the bearing surfaces of artificial hip joints.

artificial hip joint, the important design parameters include the radii of the femoral head and the acetabular cup, or the diametral clearance between the femoral head and the acetabular cup as shown in Fig. 2.1. The important manufacturing features include sphericity and roughness. The terms, definitions and texture parameters are given in the ISO standard (4287: 2000 Geometrical product specification (GPS) ± Surface texture: Profile method). Surface texture is often divided into waviness, with widely spaced irregularities, as a result of vibration in the machining process, and roughness with fine irregularities, as a result of the process itself such as machining and polishing. Surface roughness can be quantified using a profile method through either a contacting stylus such as Talysurf (Taylor Hobson) or a non-contacting interferometry technique using either a white light or a laser source. The most commonly used roughness parameters are the arithmetical mean deviation (or average roughness or centre line average, Ra) and the root-mean-square roughness (Rq). However it should be pointed out that both of these parameters only refer to the roughness height and generally do not provide spacing information. Other parameters relating to wavelength and shape are often required as well. The definitions of different surface roughness parameters and their application to artificial hip joints can be found elsewhere (Hall et al., 1997; Affatato et al., 2006).

2.1.3

Contact mechanics

Contact mechanics refers to the mechanics when two bodies are brought into contact. Contact mechanics was first studied by Hertz in 1880s with references to optical lenses (http://en.wikipedia.org/wiki/Heinrich_Rudolf_Hertz). Contact mechanics in engineering has been reviewed comprehensively by Johnson (1985). The output from a contact mechanics study generally includes contact © 2008, Woodhead Publishing Limited

Tribology in joint replacement

33

stresses, both at the bearing surfaces (also known as the contact pressure) and within the component, and the contact area. The common approach to the study of contact mechanics is either through experimental measurement or computational prediction. The experimental approaches include using engineering blue for contact area measurement (El-Deen et al., 2006) and thin film transducers for both contact pressure and area measurements such as pressuresensitive film (Fuji prescale film) and electrical resistance sensor (TekScan) (Bachus et al., 2006). It has been shown by these authors that the advantages of using TekScan include producing real-time data and the ability to evaluate a wider range of loads with greater accuracy and reliability. However, one of the major limitations associated with both of these methods is the thickness of the film or sensor which is of the order of 100 m and therefore neither method should be suitable for close conforming metal-on-metal hip implants. Computational simulation is often carried out using either the finite difference method (Jin et al., 2000) or increasingly more often the finite element method (Liu et al., 2005; Udofia et al., 2007). Determination of contact stresses is in general not a trivial task, either experimentally or computationally. However, the following simple relationship can be used to relate the average contact pressure (P) and the contact area (A) under an applied load (W): W 2:1 A Therefore, an increase in the contact area generally leads to a decrease in the predicted contact pressure. The study of contact mechanics in artificial joints is important for the following reasons. The contact parameters are closely linked to the tribology of the bearing surfaces and often used as input conditions to the overall tribological studies. The contact stresses are important considerations in the design of both hip and knee joint replacements (Bartel et al., 1985, 1995). Pˆ

2.1.4

Friction

Friction generally refers to the resistance to motion. The importance of friction in the design of artificial hip joints was first recognised by the late Sir John Charnley in his low-friction arthroplasty. The mechanical loosening often observed in the early McKee±Farrar metal-on-metal hips promoted him to look for alternative bearing materials. As a result, poly(tetrafluoroethylene) (PTFE) was selected for its lowest frictional coefficient, although massive wear was subsequently found with the cups made of this material. Now, it is generally accepted that the high friction observed in the first generation metal-on-metal hip bearings was mainly a result of poor design and manufacturing. Nevertheless, the frictional torque in metal-on-metal bearings is still much higher than that in other bearings, particularly under a prolonged period of loading and there © 2008, Woodhead Publishing Limited

34

Joint replacement technology

2.2 Schematic diagram of a hip implant with radii of the femoral head (Rhead) and the outside of the cup (Rfix).

remain concerns in relation to large diameter metal-on-metal hip resurfacing prostheses (Wimmer et al., 2006). The following three laws of dry friction are often defined: 1. The force of friction (F) is directly proportional to the applied load (W). 2. The force of friction (F) is independent of the apparent area of contact. 3. The kinetic force of friction (F) is independent of the sliding speed (V). A non-dimensional ratio, known as coefficient of friction and denoted by , is defined from the first law of friction: F or F ˆ W 2:2 W The kinetic coefficient of friction is generally less than or equal to the static coefficient of friction. The friction at the bearing surfaces directly affects the stresses transmitted through the fixation interface. This can readily be demonstrated through a simple analysis as illustrated in Fig. 2.2. The frictional force (S) at the fixation interface between the outside of the acetabular cup and the underlying support (either cement or bone) is: ˆ

Sˆ

WRhead Rfix

2:3

where Rhead and Rfix are the radii of the femoral head and the outside of the acetabular cup respectively. Therefore, to reduce the frictional force transmitted to the fixation interface, it is important not only to minimise the friction coefficient, but also to reduce the femoral head radius and to increase the outside radius of the acetabular cup. These are essentially the design features considered in the Charnley low-friction arthroplasty.

2.1.5

Wear

Wear is defined as progressive loss of substance from the operating surface of a body occurring as a result of relative motion at the surface. The importance of © 2008, Woodhead Publishing Limited

Tribology in joint replacement

35

wear in artificial joints is manifested not merely by the loss of the accuracy of the bearing geometry, which can subsequently decrease tribological and kinematics functions. However, the importance of wear in artificial joints has become more evident recently as a result of recognition of wear debris induced adverse biological reactions. It is now generally accepted that wear particles liberated from artificial joints can cause adverse tissue reactions, osteolysis and loosening (Ingham and Fisher, 2005). Different terms are often used to describe the wear phenomenon in artificial joint replacements. These include pitting, scratching, burnishing and delamination on retrieved total condylar knee joint replacements (Hood et al., 1978). However, the following five wear mechanisms are usually used to describe the fundamental wear process (Jin et al., 2006b): 1. Abrasive ± the displacement of materials by hard particles. 2. Adhesive ± the transference of material from one surface to another during relative motion by the process of solid-phase welding. 3. Fatigue ± the removal of materials as a result of cyclic stress variations. 4. Erosive ± the loss of material from a solid surface due to relative motion in contact with a fluid that contains solid particles. This is often subdivided into impingement erosion and abrasive erosion. If no solid particles are present, erosion can still take place such as rain erosion and cavitation. 5. Corrosive ± a process in which chemical or electrochemical reactions with the environment dominates, such as oxidative wear. Pitting and delamination are usually related to fatigue wear, while burnishing and scratching are different degrees of abrasive wear. Understanding the wear mechanism is important to design appropriate strategies to reduce wear in artificial joints. For example, abrasive wear can be minimised using hard, smooth bearing surfaces such as alumina ceramics as well as effective cleaning during surgery and possible sealing of the whole joint to prevent hard particles from entering the articulating surfaces. Fatigue wear mainly depends on the contact stresses and the bearing material, which in turn depend on the prosthesis design. It is important to minimise the contact stresses in order to avoid shortterm fatigue failure and breakage of the components, particularly for thin plastic cups or tibial inserts. Effective lubrication, in terms of both boundary and fluidfilm lubrication, is the key to minimising adhesive wear in metal-on-metal bearings for artificial hip joints. Corrosive wear mainly depends on the choice of the metallic materials and for this reason generally similar metallic materials (cobalt±chromium alloy) are used as the bearing surfaces for metal-on-metal hip implants. However, the taper connection between a cobalt chromium alloy head and a titanium femoral stem may elevate corrosive wear (Urban et al., 2004). Wear volume (V) is generally found to increase proportionally to the normal load (W) and the sliding distance (x) as follows: © 2008, Woodhead Publishing Limited

36

Joint replacement technology V ˆ kWx

2:4 3

where k is a wear factor, usually with a unit of mm /(N m). Wear of artificial joints is usually studied experimentally through joint simulators, although simple screening tests such as pin-on-disc and pin-on-plate machines are often used to rank different bearing materials. A joint simulator is usually designed according to the ISO standard (14242-1:2000), consisting of a dynamic vertical load up to 3 kN and three angular motions of flexion± extension, adduction±abduction and inward±outward rotations. However, various versions are possible, ranging from simple patterns of a single motion to three-dimensional full physiological simulators. It should be pointed out that it is not just the wear volume, but also the wear particles, in terms of their size distribution and morphology, are equally important. This reinforces the close coupling between tribological studies of wear debris and biological studies of tissue reactions.

2.1.6

Lubrication

Lubrication generally refers to the presence of a lubricant between the two bearing surfaces of artificial joints. Synovial fluid is generally present in healthy natural joints. After joint replacements, a pseudo-periprosthetic synovial fluid is found to be similar to those from patients with osteoarthritis (Saari et al., 1993; Delecrin et al., 1994). The lubricant used for simulator testing is usually bovine serum, diluted to various concentrations, although according to the ISO standard (14242-1:2000), 25% bovine serum is recommended. Although the viscosity of the lubricant plays an important role in the fluid film lubrication of artificial joints, it should be pointed out that the boundary constituents of these biological lubricants in terms of proteins and lipids are probably more important under boundary lubrication conditions as described below. In engineering, lubrication is usually divided into three regimes, fluid film, mixed and boundary lubrication, as illustrated schematically in Fig. 2.3. The tribological characteristics associated with each lubrication regime are listed below: 1. Fluid-film lubrication: a complete separation is achieved between the two bearing surfaces. The most important lubricant parameter is viscosity. Under the fluid film lubrication regime, both friction and wear are minimised. However, a complete elimination of friction and wear is impossible in artificial joints due to the viscous shearing of the lubricant and the breakdown of fluid film lubrication associated with start-up and stop motions. 2. Boundary lubrication: extensive asperity contacts occur and both wear and friction are significantly increased. Boundary lubricating films play an important role in this lubrication regime, which depend on both the physical and chemical properties of the lubricant. © 2008, Woodhead Publishing Limited

Tribology in joint replacement

37

2.3 Schematic diagram of three different lubrication regimes.

3. Mixed lubrication: this lubrication regime consists of a mixture of both fluid film and boundary lubrication regions. The tribological characteristics in this lubrication regime depend on the relative contribution of the fluid film and boundary lubrication. The lubrication regime can be assessed either experimentally or theoretically. The experimental assessment is achieved through indirect measurement of friction in the so-called Stribeck diagram as shown in Fig. 2.4. The alternative experimental measurement of lubrication is directly through separation techniques. The principle of the separation technique is based either on the resistivity (Dowson et al., 2000) or ultrasound measurements (Brockett, 2007). The theoretical assessment is based on the determination of the lambda ratio defined as: ˆ

hmin ˆ hÿ Ra

hmin
2 ÿ
2 i1=2 Ra_head ‡ Ra_cup

2.4 Typical friction factors and associated lubrication regimes. © 2008, Woodhead Publishing Limited

2:5

38

Joint replacement technology

where hmin is the minimum film thickness predicted based on the assumption of smooth bearing surfaces and Ra is average roughness. Therefore, if a representative minimum film thickness is estimated and the surface roughness parameters are measured, the lambda ratio and the corresponding lubrication regimes can be determined accordingly: · Boundary lubrication:   1 · Mixed lubrication: 1 <  < 3 · Fluid film lubrication:   3

2.2

Theoretical tribological studies

Theoretical studies of the tribological problems in artificial joints offer a number of advantages over experimental approaches. These include a relatively short period of time to develop the theoretical model and cost effectiveness to perform a theoretical analysis. Therefore, theoretical models are particularly suitable for screening analyses of design parameters and identifying the underlying mechanism when combined with experimental studies. It is also useful to explore the scenarios that cannot be readily addressed by experimental studies. For example, a simulator testing up to 50 million cycles may take up to 5±10 years to complete and this is not practical. However, computational wear modelling may offer an alternative. Theoretical modelling has been increasingly used for artificial joints recently owing to significant development of computing power and availability of commercial finite element software. Although there are a number of advantages of the theoretical models, experimental studies are equally important, not only providing input parameters required for the theoretical models but also validation. Integrated experimental and theoretical studies are essential. In particular, theoretical analyses involving contact mechanics, lubrication and wear have received significant attention and are reviewed in this section. Numerical methods are usually employed to solve the theoretical models. Although the finite difference method can be used, the finite element method is particularly useful in considering the complex geometry and material properties. There are a number of commercial finite element packages available, including ABAQUS. However, for complex lubrication problems, which generally involve coupling between solid and fluid mechanics and cannot be readily dealt with by either the finite element or the finite different method, a combination of these two methods has been found to be feasible and robust (Jagatia and Jin, 2001).

2.2.1

Contact mechanics

A ball-in-socket configuration is usually adopted for the contact mechanics analysis of artificial hip joints. As a first approximation, an equivalent ball-on© 2008, Woodhead Publishing Limited

Tribology in joint replacement

39

2.5 (a) Ball-in-socket and (b) equivalent ball-on-plane models.

plane model shown in Fig. 2.5 can be used based on the Hertz contact theory, with the following equivalent radius defined as: Rˆ

Rcup Rhead Rhead Rhead ˆ Rcup ÿ Rhead d=2

2:6

where d denotes the diametral clearance between the head and the cup. For a conforming ball-in-socket configuration and complex structural supports in artificial hip joints, the finite element method is usually adopted. The major challenge involved is the modelling of the two contacting surfaces, which is generally non-linear and time-consuming. Contact mechanics modelling of artificial hip joints has been considered, for example, by Bartel et al. (1985), Jin et al. (1999) and Plank et al. (2007), on the design parameters of femoral head radius, clearance between the femoral head and the acetabular cup and polyethylene thickness, and by Korhonen et al. (2005), on the implantation angle. For artificial knee joints, it is usually possible to use an equivalent ellipsoid-onplane geometry as a first approximation (Jin et al., 1995a,b). However, for complex geometries, the finite element method is usually used (Bartel et al., 1995). The predicted maximum contact pressure is summarised in Table 2.1 for artificial hip joint replacements with various bearing surfaces. Table 2.2 summarises the predicted maximum contact pressure in different forms of metalTable 2.1 Maximum contact pressure prediction for total artificial hip joints with various bearing surfaces under a load between 2500 and 3000 N Major design features

Diameter (mm)

UHMWPE22±46 on-metal Metal-on28 metal Ceramic-on28 ceramic

© 2008, Woodhead Publishing Limited

Liner thickness (mm)

Diametral clearance (m)

Max contact pressure (MPa)

3±14

100±400

10±25

7

60

50

5

80

80

Reference

Jin et al. (1999); Plank et al. (2007) Jagatia and Jin (2001) Mak and Jin (2002)

40

Joint replacement technology

Table 2.2 Maximum contact pressure prediction for various metal-on-metal hip implants under a load of 2500 N Implant design features

Diameter (mm)

Diametral clearance (m)

Max contact pressure (MPa)

28 28 28 28 35 50

60 120 60 120 158 145

55 90 35 44 20 18

Thick cup (> 7 mm) Thick cup (> 7 mm) Taper-connected cup Sandwich cup McKee±Farrar Resurfacing

Reference

Jagatia and Jin (2001) Jagatia and Jin (2001) Besong et al. (2001) Liu et al. (2003) Yew et al. (2003) Liu et al. (2005)

Table 2.3 Maximum contact pressure prediction for various knee implants under a load of 2500 N Major design features

Large equivalent radius (mm)

Small equivalent radius (mm)

Max contact pressure (MPa)

Conforming Unconforming

300±500 100±300

 200 50±100

15±20 20±30

Reference

Stewart et al. (1995) Stewart et al. (1995)

on-metal hip joint replacements. The contact pressure in artificial knee joint replacements depends on flexion angle, medial/lateral load sharing, ultra-high molecular weight polyethylene (UHWMPE) thickness and the bearing geometry, all of which can be expected to vary significantly. Typical predicted maximum contact pressures are summarised in Table 2.3, mainly depending on the equivalent radii of the bearing surfaces.

2.2.2

Lubrication

For the equivalent ball-on-plane model, the minimum film thickness formulae developed in engineering, such as for ball-bearings, can be directly adopted (Hamrock and Dowson, 1978; Jin et al., 1997):  u 0:65  W ÿ0:21 hmin ˆ 2:8 0 2:7 R ER E0 R2 where the equivalent radius R in equation (2.7) can be calculated from equation (2.6) and  denotes the viscosity of the lubricant. The entraining velocity u can be calculated from the angular velocity of the femoral head !: uˆ

!Rhead 2

© 2008, Woodhead Publishing Limited

2:8

Tribology in joint replacement

41

Finally the equivalent elastic modulus E0 is given by: E0 ˆ 

2 1 ÿ  cup 2 1 ÿ  head ‡ Ehead Ecup 2



2:9

where E and  denote elastic modulus and Poisson ratio of the bearing materials. For the ball-in-socket model, numerical methods are generally required. This is usually achieved through a combination of the finite difference method for the hydrodynamic lubrication analysis and the finite element method for the elastic deformation calculation (Jagatia and Jin, 2001). The versatility of the finite element method allows different bearings of artificial hip joints to be readily analysed. The predicted minimum film thickness is compared in Table 2.4 for different bearings. The importance of the fluid film lubrication contribution to the overall tribological performance of metal-on-metal bearings has been recognised recently. Table 2.5 compares the predicted minimum lubricant film thickness Table 2.4 Theoretical estimation of in vivo minimum lubricant film thickness and corresponding lubrication regimes in various hip implants with different bearing surfaces based on the  ratio, using the Hamrock and Dowson formulae (Rhead ˆ 14 mm; W ˆ 1:5  2:5 kN;  ˆ 0:0025 Pa s; ! ˆ 1:5 rad/s). Bearing couples

Minimum film thickness (nm)

Composite Ra roughness (nm)

 ratio (lubrication regime)

83

50±1000

36

14±28

24

7

0.08±1.7 (boundary to mixed) 1.3±2.6 (mixed to fluid-film) 3.4 (fluid-film)

UHMWPEon-metal Metal-on-metal Ceramic-onceramic

Table 2.5 Comparison of minimum film thickness prediction between different metal-on-metal hip implants ( ˆ 0:0025 Pa s, W ˆ 2:5 kN and ! ˆ 1:5 rad/s) Major design features

Diameter (mm)

Thick cup (> 7 mm) Sandwich cup McKee±Farrar Resurfacing (average wall thickness = 4 mm)

28 28 35 50

© 2008, Woodhead Publishing Limited

Diametral clearance (m)

Min. film thickness (m)

60 120 158 145

0.023 0.02 0.028 0.06

References

Jagatia and Jin (2001) Liu et al. (2004) Yew et al. (2004) Liu et al. (2006)

42

Joint replacement technology

for different designs and forms of metal-on-metal hip joint replacements. Compared with the hip joint, relatively little research has been carried out for the lubrication analysis in artificial knee joint replacements.

2.2.3

Wear

Although wear is usually investigated experimentally, theoretical prediction has also been carried out recently following the pioneering work by Maxian et al. (1996). The computational wear prediction has focused on the screening analysis of different design parameters such as the femoral head radius and clearance in UHMWPE-on-metal combinations (Maxian et al., 1996, 1997). More recently, more complex computational wear models have been developed, particularly for metal-on-metal bearings (Harun et al., 2007). This requires taking into account not only the complex kinematics of two bearing surfaces, which can both wear, but also the consideration of lubrication-dependent wear required for metal-onmetal bearings. Such a theoretical approach is particularly useful for simulating long-term wear up to 50 or even to 100 million cycles (Kang et al., 2006). Wear prediction is largely based on Archard's law in equation (2.4) or expressed in terms of linear wear (l) and contact pressure (p): l ˆ kpx

2:10

where x is the sliding distance. Such an approach usually requires the experimental input of a wear factor. The contact pressure can be predicted from the finite element method as outlined in Section 2.2.1. This enables the linear wear to be predicted from equation (2.10), which can then be used to update the bearing geometry, which can in turn affect the contact pressure. Therefore, the essence of the computational wear modelling is the coupling between contact mechanics and wear. It should be pointed out that the predicted wear volume is fixed for a chosen wear factor, for a given set of kinematics and particular loading conditions. However, both the linear wear and wear scar are independent parameters and can provide further useful information and validation. The key to the computational wear prediction is the wear factor. A simple pin-on-plate machine, which is usually used to obtain wear factors, may not replicate the lubrication condition, which can be important for some bearing surfaces such as metal-on-metal combinations. Such a problem can be partially overcome by using different wear factors, determined from full simulator studies by matching the predicted wear volume with the experimental measurements in both runningin and steady-state phases (Kang et al., 2006). A long-term wear prediction can then be carried out assuming that the wear factor during the steady-state phase is unchanged. A comprehensive consideration of lubrication in computational wear modelling in metal-on-metal bearing is still lacking. Direct wear prediction without experimental input of wear factors would be very difficult if not impossible. © 2008, Woodhead Publishing Limited

Tribology in joint replacement

43

Computational wear prediction in artificial knee joints largely follows the methodology adopted for the hip joints (Fregly et al., 2005; Laz et al., 2006; Knight et al., 2007). However, the major issues to be addressed in the knee joint include complex kinematics and cross-shear motion. In particular, the latter has been shown to be important, particularly for UHMWPE-on-metal material combinations (Hamilton et al., 2005).

2.3

Experimental tribological studies

Surface roughness, friction and wear are usually measured experimentally and these are reviewed in this section, together with the limited experimental studies of lubrication which are available.

2.3.1

Surface topography of bearing surfaces used for artificial joints

According to ISO standard (7206-2), for UHMWPE-on-metal hip joints, the spherical articulating surfaces of metallic and ceramic femoral components should have Ra values no greater than 0.05 m and 0.02 m respectively. For the plastic cup, the spherical articulating surface of the implant should have an Ra value no greater than 2 m. For the knee implants (ISO 7207-2), the metallic or ceramic femoral components, when measured in accordance with ISO 468, the articulating surface should have an Ra value no greater than 0.1 m, while for the plastic tibial and patellar components, the articulating surfaces should have an Ra value no greater than 2 m. All the above specifications are for a cut-off value of 0.08 mm. However, the typical values achieved with current manufacturing standards from the majority of orthopaedic manufacturers are far less than those specified for different artificial hip joints, as shown in Table 2.6.

Table 2.6 Typical average roughness values for various bearing surfaces used in current artificial hip joints and their composite (Ra ) values Bearings

Femoral

Ra_head (m)

Acetabular

Ra_cup (m)

Composite Ra* (m)

UHMWPEon-metal Metal-onmetal Ceramicon-ceramic

Cobalt± chrome Cobalt± chrome Alumina

0.01±0.025

UHMWPE

0.1±2.5

0.1±2.5

0.005±0.025

Cobalt± chrome Alumina

0.005± 0.025 0.005± 0.01

0.0071±0.035

0.005±0.01

*Note: the composite roughness is defined as: Ra ˆ

© 2008, Woodhead Publishing Limited

q ÿ
2 ÿ
2 Ra_head ‡ Ra_cup

0.0071±0.014

44

2.3.2

Joint replacement technology

Friction and lubrication

Significant efforts have been made in the literature to determine the coefficients of friction for various biomaterials in engineering. Typical coefficients of friction are shown in Table 2.7. For hip implants, friction torque is usually measured in a functional friction hip simulator, usually with a vertical dynamic load and a horizontal angular velocity (Scholes and Unsworth, 2000; Brockett et al., 2006). The frictional torque measured (T) is then used to calculated the friction factor, f (equivalent to coefficient of friction when the load is transmitted through a point): f ˆ

T WRhead

2:11

Typical friction factor values in different bearings for hip implants have been measured, mainly from the universities of Durham (Scholes and Unsworth, 2000) and Leeds (Brockett et al., 2006), and are summarised in Table 2.8. The experimental assessment of lubrication is usually achieved through measuring either the resistance between the articulating surfaces (Dowson et al., 2000) or the gap with ultrasound (Brockett, 2007). Although the resistivity technique is relatively simple and straightforward, it is only possible to detect whether or not there is a lubricant film between the two bearing surfaces. The ultrasound method is capable of quantifying the film thickness but is, however, limited to relatively thick films. Table 2.7 Typical coefficients of friction for clean materials in dry contact in the presence of air (taken from Dowson and Wright, 1981) Material combination Steel-on-steel Polyethylene-on-steel Polyethylene-on-polyethylene PTFE-on-PTFE PTFE-on-steel

Coefficient of friction 0.6±0.8 0.3 0.2±0.4 0.04±0.2 0.04±0.2

Table 2.8 Typical friction factors for various bearings for artificial hip joints in the presence of bovine serum Bearing UHMWPE-on-metal UHMWPE-on-ceramic Metal-on-metal Ceramic-on-ceramic Ceramic-on-metal

© 2008, Woodhead Publishing Limited

Friction factor 0.06±0.08 0.04±0.08 0.10±0.20 0.002±0.07 0.002±0.07

Tribology in joint replacement

2.3.3

45

Wear

The most complex tribological problem in artificial joints is wear. Wear depends on many factors. The effect of a parameter is often masked by slight changes in other parameters. This is made particularly difficult since different conditions are usually employed in different studies. Direct comparison of wear is not trivial, and may not be advisable. The wear factors determined from simple screening devices, which generally provide a unidirectional rotation or a reciprocating motion, are only useful for comparative studies and ranking of different materials (for example, composition, structure, processing). It should be noted that the difference in wear factors observed is often not valid, since lubrication regimes are not fully replicated in these simple machines. The addition of a rotational motion to a linear reciprocating motion causes a multidirectional motion which can be important for both UHMWPE-on-metal and metal-on-metal material combinations. It is interesting to note that in the case of the UHMWPE-on-metal combination, cross-shear motion elevates wear (Galvin et al., 2006), while in the case of a metal-on-metal bearing, multi-direction promotes a self-polishing action and reduces wear (Tipper et al., 1999; Scholes and Unsworth, 2001). Typical wear factors obtained from simple screening devices are shown in Table 2.9. These wear factors can be considered as values under boundary lubrication conditions. Volumetric wear rates are usually defined as wear volume divided by either the number of cycles or years. It is usually assumed that 1 year is equivalent to 1 million cycles. However, more recent studies by Schmalzried et al. (1998) and Goldsmith et al. (2001) have shown that the most active patients can walk 3.2 million steps on average and reach up to 5 million steps per year. Therefore, the volumetric wear rate and linear wear rate both defined in relation to the number of cycles are more appropriate (Fisher et al., 2006). Ultra-high molecular weight polyethylene: crosslinking Crosslinked UHMWPE cups have been extensively studied and introduced recently. A remarkable reduction in wear volume of crosslinked UHMWPE has been reported in simulator studies. However, the amount of wear reduction has been found to be quite variable, ranging from zero (or even negative, which may Table 2.9 Representative wear factors, k, for various material combinations tested in pin-on-plate machines Material combination

Wear factor (mm3/N m)

UHMWPE-on-metal Metal-on-metal Ceramic-on-ceramic

 10ÿ7  10ÿ7  10ÿ8

© 2008, Woodhead Publishing Limited

46

Joint replacement technology

be due to fluid absorption) (Muratoglu et al., 2001) to 5 mm3/million cycles, an eight-fold reduction compared with conventional polyethylene (Fisher et al., 2006). Clinically retrieved samples have shown creep, loss of original machining marks and wear (Digas et al., 2003; Martell et al., 2003; Bradford et al., 2004). The effect of increased head size on the wear of crosslinked UHMWPE cups is contradictory; for one study by Muratoglu et al. (2001) has shown no apparent change, while another study by Fisher et al. (2006) has shown the increase of wear to 10 mm3/million cycles when the head diameter is increased from 28 to 36 mm. Furthermore, the crosslinked UHMWPE wear particles are generally smaller and may be more reactive biologically, and consequently the overall functional biological activity is only improved by three- to four-fold, compared with conventional polyethylene (Fisher et al., 2006). Ceramic-on-ceramic: importance of microseparation The wear in ceramic-on-ceramic hip implants is generally very low, under standard simulator testing conditions. For example, the wear rate in the 28 mm diameter ceramic-on-ceramic hips has been measured to be 0.1 mm3/million cycles. However, these low wear rates have not been observed on retrieved components. In clinically retrieved components, stripe wear is usually observed, which is thought to be related to the micro-separation (Nevelos et al., 2001). Simulator testing considering micro-separation resulted in not only the stripe wear, but also a wear rate of 1:4  0:2 mm3/million cycles (Nevelos et al., 2001; Stewart et al., 2001). Furthermore, ceramic wear particles have been shown to be less bioreactive, resulting in a substantial lower overall functional biological activity (Fisher et al., 2006). Metal-on-metal: lubrication dependent Wear of metal-on-metal bearings for hip implants is generally quite complex. There are a number of important factors related to bearing materials and lubrication. Two distinct wear phases are usually observed, the initial running-in phase with a relatively high wear rate, which is followed by a steady-state phase with a much reduced wear rate. The design parameters of the metal-on-metal bearings have a large effect on the lubrication and hence wear. These include both the head diameter and the diametral clearance between the head and the cup. It is generally accepted that high-carbon (>0.2%) cobalt±chromium alloy produces less wear than low-carbon materials. For example, the overall wear rate has been shown to be increased from 0.1 mm3/million cycles to 0.6 mm3/ million cycles when low-carbon cobalt±chromium alloy was used (Fisher et al., 2006). However, differences in the raw materials (either cast or wrought) or © 2008, Woodhead Publishing Limited

Tribology in joint replacement

47

processing routes (hot isostatic pressing and solution annealing or as cast) do not appear to make much difference in the wear of metal-on-metal bearings, particularly under more realistic simulator testing conditions (Chan et al., 1999; Dowson et al., 2004a). Similar conclusions have also been reached from clinical studies by DoÈrig et al. (2006) and Miloev et al. (2006) who found that 10-year survivorships were 98.3% and 91% respectively for the high-carbon and lowcarbon bearings respectively. The design parameters such as the femoral head radius and the clearance between the head and the cup can have a significant influence on the wear generated in the metal-on-metal bearings. The femoral head radius not only affects the sliding distance, as in the case of UHMWPE bearings, but also the sliding velocity and consequently the lubrication. If the increase in the femoral head radius shifts the lubrication regime to a predominantly fluid film region, the adverse effect of increasing sliding distance becomes less of an issue. For example, an increase in the femoral head diameter from 16 to 22.225 mm resulted in an increase in the wear rate, consistent with boundary lubrication and UHMWPE-on-metal bearings (Smith et al., 2001a,b). However, a further increase in the femoral head radius beyond 22.225 mm resulted in a significant wear reduction, presumably due to improved lubrication and increased fluid-film contribution, and reduced asperity contacts. Understanding of the lubrication mechanism in metal-on-metal bearings has provided a theoretical basis for the extensive introduction of large diameter metal-on-metal hip resurfacing prostheses. The benefit of large femoral heads in wear reduction has been well documented in these alternative forms of metal-on-metal hip resurfacing prostheses (Dowson et al., 2004b; Rieker et al., 2005; Fisher et al., 2006). Clearance also plays an important role in the wear of metal-on-metal hip implants, since it directly affects the conformity between the articulating surfaces and consequently the lubrication. Theoretically, a smaller clearance increases the conformity and should result in a better lubrication. On the other hand, if the clearance is too large, the contact pressure can be transmitted only through a small contact area and consequently is significantly increased. This also leads to a deterioration of the lubrication, moving the lubrication regime further towards the boundary lubrication region. Consequently, wear can be significantly increased, particularly when this is coupled with a large diameter metal-on-metal bearing for hip resurfacing prostheses, owing to the adverse effect of the increased sliding distance. For example, it has been shown by Fisher et al. (2006) that for a large diameter 55 mm bearing, an increase in the radial clearance from 51 to 150 m almost doubled the wear. However, the clearance should not be designed below a certain limit, owing to manufacturing and potential component deformation during implantation (Jin et al., 2006a). If the clearance is too small, contact between the two bearing surfaces may occur at the edge of the cup, not only leading to stress concentration, but also blocking lubricant entry and causing lubricant starvation, which increases wear signifi© 2008, Woodhead Publishing Limited

48

Joint replacement technology Table 2.10 Typical volumetric and linear wear rates for different bearings for hip implants (Jin et al., 2003) Bearing couples UHMWPE-on-metal UHMWPE-on-ceramic Crosslinked UHMWPE Metal-on-metal Ceramic-on-ceramic

Volumetric wear rate (mm3/million cycles) 30±100 15±50 5±10 0.1±1 0.05±1

Linear wear rate (m/million cycles) 100±300 50±150 15±30 2±20 1±20

cantly. It has also been shown that a negative clearance leads to erratic and high wear (Liao and Hanes, 2006). Therefore there appears to be an optimum range, but the optimum clearance appears to depend upon the bearing systems (Rieker et al., 2004, 2005). Kinematic and loading conditions can also affect the lubrication of metal-onmetal bearings and hence wear (Firkins et al., 2001a). Furthermore, start-up and stopping (Roter et al., 2002), micro-separation (Williams et al., 2004), stumbling (Bowsher et al., 2002) and fast jogging (Bowsher et al., 2006) have all been shown to result in increased wear. The effect of swing phase load was considered by Williams et al. (2006). It was shown by these authors that a small decrease in the swing load from 280 N (according to the ISO 14242-1, 2000) to 100 N could lead to a 10-fold increase in overall wear rates. Typical volumetric and linear wear rates for different hip implants with different bearings are compared in Table 2.10. Studies of biological reactions to wear debris are equally important to the tribological considerations of the bearing surfaces. Table 2.11 summarises typical particle sizes and biological responses in different bearings for hip implants (Ingham and Fisher, 2005).

Table 2.11 Typical particle sizes and biological responses in different bearings for hip implants (Ingham and Fisher, 2005) Bearing couples

Dominant particle diameters (m)

Biological responses

UHMWPE-onmetal/ceramic

UHMWPE, 0.01±1

Macrophages/osteoclasts/ osteolysis

Metal-on-metal

Metallic, 0.02±0.1

Low osteolysis, cytotoxicity

Ceramic-on-ceramic

Ceramic, 0.01±0.02 Ceramic, 0.1±10

Bio-inert, low cytotoxicity Macrophages/osteoclasts/ osteolysis

© 2008, Woodhead Publishing Limited

Tribology in joint replacement

2.4

49

Issues of tribology for joint replacements and future trends

The clinical success of total joint replacements in the older patient group has resulted in an increasing and widespread use of these devices in the younger and more active patients. However, younger and more active patients, who have longer life expectancies with the prosthetic joint likely to be in place beyond 20 years coupled with higher levels of activity with up to a few million steps per year, mean that the lifetime tribological demand may increase to 100±200 million steps, up to a 10-fold increase on the tribological demand (Fisher et al., 2006). Furthermore, there are increased interests in the use of larger diameter femoral heads to provide a greater range of motion and improve stability. The clinical limitation of current UHMWPE hip implants means that revision will be generally required for younger patients with further life expectancies after surgery in excess of 20 years. Currently, one of the major strategies to avoid revision is to improve total joint replacements by using novel bearing couples to reduce wear and wear particle generation and this should extend the clinical lifetime of the implant. In addition to the bearing couples discussed so far, other materials combinations have also been developed and are in current clinical trials, including surface engineered thick CrN coating for metal-on-metal bearings (Fisher et al., 2004) and ceramic-on-metal (Firkins et al., 2001b). In the meantime, the increasingly rigorous ethical and regulatory environment is demanding more extensive preclinical studies, as part of the translation of any new technology to the patient. Coupled studies of the tribology of the bearing surfaces and biological reactions to wear debris have contributed significantly to such drives, to developments of novel bearing material combinations and to successes of total hip joint replacements. Further important considerations to be addressed in future include patient-specific issues, more realistic and long-term wear simulation and an interaction between tribology and fixation. Wear improvement of alternative bearings to polyethylene in artificial hip joints means that longer simulator testing is required. The majority of current simulator testing is conducted for only 5±10 million cycles. Longer simulator testing well beyond 10 million cycles is required, particularly for alternative bearings such as metal-on-metal to investigate the change of the bearing geometry due to wear and the potential long-term effect on tribology. Such a consideration is particularly important when a more realistic pattern of daily activities is considered (Morlock et al., 2001). Computational wear modelling beyond 100 million cycles may provide a useful alternative to current simulator testing (Liu et al., 2008). The majority of tribological studies on knee implants have focused on the femoral±tibial contact. The increasing use of a patellar component means that wear simulation of the patellar±femoral contact is equally important (Ellison et al., 2007). © 2008, Woodhead Publishing Limited

50

Joint replacement technology

Owing to the ever-increasing introduction of minimally invasive and conservative prosthetic designs in order to delay the end stage total joint replacements, the interaction between the fixation of the prosthetic components and the tribology of the bearing surfaces becomes important. For example, the underlying supports to the bearing may well affect the tribology, and equally the tribology at the bearing surfaces may affect how the stresses are transmitted to the fixation and the underlying materials. Simulator testing with a more realistic biological environment and fixation may be important for investigating novel conservative designs. Furthermore, in cartilage substitution devices, cartilage becomes one of the bearing surfaces. The tribology of cartilage in combination with a wide range of biomaterials as well as its potential degradation has not been extensively studied. Of particular importance is the biphasic lubrication of the articular cartilage, where the loading time history has a marked effect on friction (Forster and Fisher, 1996; Ateshian, 1997; Jin et al., 2000; Muller et al., 2004) and the potential for degradation and wear. When substituting part or whole of one cartilage surface, there is the potential to markedly alter the tribological lubrication regime found in the natural joint with two articular cartilage surfaces. Such considerations are equally important for tissue engineered articular cartilage, which has been developed extensively in the past 5±10 years. However, it is only recently that the functional requirements of tissue-engineered cartilage have been addressed, such as tribological characteristics (Lima et al., 2006; Morita et al., 2006; Plainfosse et al., 2006).

2.5

Sources of further information and advice

Tribological studies of artificial joints have received significant attention from clinical organisations such as the Orthopaedic Research Society (ORS, http:// www.ors.org/web/index.asp) and the British Orthopaedic Research Society (BORS, http://www.borsoc.org.uk/). Tribology of artificial joints is addressed in clinical meetings such as the annual ORS and BORS as well as in bioengineering meetings such as the World Congress of Biomechanics and the World Congress of Biomaterials. Such topics are also addressed in engineering conferences such as the World Congress of Tribology; Wear of Materials and Leeds±Lyon Symposium on Tribology.

2.6

References and further reading

Affatato S, Bersaglia G, Junqiang Y, Traina F, Toni A and Viceconti M (2006) The predictive power of surface profile parameters on the amount of wear measured in vitro on metal-on-polyethylene artificial hip joints. J Eng Med, Proc Inst Mech Engrs, 220(3), 457±64. Ateshian, G.A. (1997) A theoretical formulation for boundary friction in articular cartilage. J Biomech Eng, 119(1), 81±6. Bachus KN, DeMarco AL, Judd KT, Horwitz DS and Brodke DS (2006) Measuring © 2008, Woodhead Publishing Limited

Tribology in joint replacement

51

contact area, force, and pressure for bioengineering applications: using Fuji film and TekScan systems. Med Eng Phys, 28(5), 483±8. Bartel DL, Burstein AH, Toda MD and Edwards DL (1985) The effect of conformity and plastic thickness on contact stresses in metal-backed plastic implants. J Biomech Eng, 107(3), 193±9. Bartel DL, Rawlinson JJ, Burstein AH, Ranawat CS and Flynn WF Jr (1995) Stresses in polyethylene components of contemporary total knee replacements. Clin Orthop Relat Res, 317, 76±82. Besong AA, Farrar R and Jin ZM (2001) Contact mechanics of a novel metal-on-metal total hip replacement. J Eng Med, Proc Inst Mech Engrs, 215, 543±8. Bowsher JG, Nevelos J, Pickard J and Shelton JC (2002) Hip simulator testing ± the next generation? Proc. Int. Conf. Engineers and Surgeons Joined at the Hip, IMechE, C601/021/2002. Bowsher JG, Hussain A, Williams PA, Shelton JC (2006) Metal-on-metal hip simulator study of increased wear particle surface area due to `severe' patient activity. J Eng Med, Proc Inst Mech Engrs, 220(2), 279±87. Bradford L, Baker DA, Graham J, Ries M and Pruitt LA (2004) Wear and surface cracking in early retrieved highly cross-linked polyethylene acetabular liners. J Bone Joint Surg Am, 86, 1271±82. Brockett C (2007) Tribology of large diameter metal-on-metal hip resurfacing replacements, PhD thesis, School of Mechanical Engineering, University of Leeds. Brockett C, Williams S, Jin ZM, Isaac G and Fisher J (2006) Friction of total hip replacements with different bearings and loading conditions, J Biomed Mater Res, Part B: Appl Biomaterials, 13, 81B(2), 508±15. Chan FW, Bobyn JD, Medley JB, Krygier JJ and Tanzer M (1999) The Otto Aufranc Award. Wear and lubrication of metal-on-metal hip implants. Clin Orthop Relat Res, 369, 10±24. Delecrin J, Oka M, Takahashi S, Yamamuro T and Nakamura T (1994) Changes in joint fluid after total arthroplasty. Clin Orthop Relat Res, 307, 240±9. Digas G, KaÈrrholm J, Thanner J, Malchau H and Herberts P (2003). Highly cross-linked polyethylene in cemented THA. Clin Orthop Relat Res, 417, 126±38. DoÈrig MF, Schueler M and Odstrcilik E (2006) Ceramic-on-polyethylene versus metalon-metal. A prospective follow-up study, at least 10 years after primary implantation. European Hip Society, Domestic Meeting, Antalya, Turkey, 21±24 June, O-003. Dowson D and Wright V (1981) Introduction to the Biomechanics of Joints and Joint Replacements, Mechanical Engineering Publications Ltd, London. Dowson D, McNie CM and Goldsmith AAJ (2000) Direct experimental evidence of lubrication in a metal-on-metal total hip replacement tested in a joint simulator, J Mech Eng Sci, Proc Inst Mech Engrs, 214, 75±86. Dowson D, Hardaker C et al. (2004a) A hip joint simulator study of the performance of metal-on-metal joints. Part I: The role of materials. J Arthroplasty, 19(8, Supplement 1), 118±23. Dowson D, Hardaker C, Flett M and Isaac GH (2004b) A hip joint simulator study of the performance of metal-on-metal joints. Part II: Design. J Arthroplasty, 19(8, Supplement 1), 124±30. El-Deen M, Garcia-Finana M and Jin ZM (2006) Effect of ultra-high molecular weight polyethylene thickness on contact mechanics in total knee replacement. J Eng Med, Proc Inst Mech Engrs, 220(7), 733±42. Ellison P, Barton DC, Esler C, McEwen HM, Shaw DL, Stone MH and Fisher J (2007) © 2008, Woodhead Publishing Limited

52

Joint replacement technology

Wear and Creep of Replacement Patellofemoral Joints. Orthopaedic Research Society, Vol. 32, San Diego, CA, abstract no. 1880. Firkins PJ, Tipper JL, Ingham E, Stone MH, Farrar R and Fisher J (2001a) Influence of simulator kinematics on the wear of metal-on-metal hip prostheses. J Eng Med, Proc Inst Mech Engrs, 215(1), 119±21. Firkins PJ, Tipper JL, Ingham E, Stone MH, Farrar R and Fisher J (2001b) A novel low wearing differential hardness, ceramic-on-metal hip joint prosthesis. J Biomech, 34(10), 1291±8. Fisher J, Hu XQ, Stewart TD, Williams S, Tipper JL, Ingham E, Stone MH, Davies C, Hatto P, Bolton J, Riley M, Hardaker C, Isaac GH and Berry G (2004) Wear of surface engineered metal-on-metal hip prostheses. J Mater Sci Mater Med, 15(3), 225±35. Fisher J, Jin ZM, Tipper J, Stone M and Ingham E (2006) Presidential guest lecture ± Tribology of alternative beatings. Clin Orthop Relat Res, 453, 25±34. Forster H and Fisher J (1996) The influence of loading time and lubricant on the friction of articular cartilage. J Eng Med, Proc Inst Mech Engrs, 210, 109±19. Fregly BJ, Sawyer WG, Harman MK and Banks SA (2005) Computational wear prediction of a total knee replacement from in vivo kinematics. J Biomech, 38(2), 305±14. Galvin A, Kang L, Tipper J, Stone M, Ingham E, Jin ZM and Fisher J (2006) Wear of crosslinked polyethylene under different tribological conditions. J Mater Sci: Mater Med, 17(3), 235±43 Goldsmith AA, Dowson D, Wroblewski BM, Siney PD, Fleming PA, Lane JM, Stone MH and Walker R (2001) Comparative study of the activity of total hip arthroplasty patients and normal subjects. J Arthroplasty, 16(5), 613±19. Hall RM, Siney P, Unsworth A and Wroblewski BM (1997) The effect of surface topography of retrieved femoral heads on the wear of UHMWPE sockets. Med Eng Phys, 19(8), 711±19. Hamilton MA, Sucec MC, Fregly BJ, Banks SA and Sawyer WG (2005) Quantifying multidirectional sliding motions in total knee replacements. J Tribol, 127, 280±6. Hamrock BJ and Dowson D (1978) Elastohydrodynamic lubrication of elliptical contacts for materials of low elastic modulus. I: Fully flooded conjunction. Trans ASME, J Lubric Technol, 100(2), 236±45. Harun M, Wang FC, Jin ZM and Fisher J (2007) Development of computational wear simulation of metal-on-metal hip joint replacement. Trans Orthop Res Soc, 32, 1661. Hood RW, Wright TM and Burstein AH (1978) Retrieval analysis of total knee prostheses: a method and its application to 48 total condylar prostheses. J Biomed Mater Res, 17, 829±42. Ingham E and Fisher J (2005) The role of macrophages in the osteolysis of total joint replacement. Biomaterials, 26(11), 1271±86. Jagatia M and Jin ZM (2001) Elastohydrodynamic lubrication of metal-on-metal hip prosthesis under steady-state entraining motion. J Eng Med, Proc Inst Mech Engrs, 215, 531±41. Jin ZM (2000) A general axisymmetric contact mechanics model for layered surfaces, with particular reference to artificial hip joint replacements. J Eng Med, Proc Inst Mech Engrs, 214, 425±35. Jin ZM, Dowson D and Fisher J (1995a) Contact pressure prediction in total knee joint replacements. Part 1: General elasticity solution for elliptical contacts. J Eng Med, Proc Inst Mech Engrs, 209, 1±8. © 2008, Woodhead Publishing Limited

Tribology in joint replacement

53

Jin ZM, Dowson D and Fisher J (1995b) Contact pressure prediction in total knee joint replacements. Part 2: Application to the design of total knee joint replacements. J Eng Med, Proc Inst Mech Engrs, 209, 9±15. Jin ZM, Dowson D and Fisher J (1997) Analysis of fluid film lubrication in artificial hip joint replacements with surfaces of high elastic modulus. J Eng Med, Proc Inst Mech Engrs, 211, 247±56. Jin ZM, Heng SM, Ng HW and Auger DD (1999) An axisymmetric contact model of ultra high molecular weight polyethylene cups against metallic femoral heads for artificial hip joint replacements. J Eng Med, Proc Inst Mech Engrs, 213, 317±27. Jin ZM, Pickard JE, Forster H, Ingham E and J Fisher (2000) Frictional behaviour of bovine articular cartilage. Biorheology, 37, 57±63. Jin ZM, Medley JB and Dowson D (2003) Fluid film lubrication in artificial hip joints. In Tribological Research and Design for Engineering Systems, Ed. by Dowson D, Priest M, Dalmaz G and Lubrecht AA, Elesvier B.V., Sara Burgerhartstraat 25, P.O. Box 211, 1000 AE Amsterdan, The Netherlands, Proceedings of 29th Leeds±Lyon Symposium on Tribology, 237±56. Jin ZM, Meakins S, Morlock MM, Parsons P, Hardaker C, Flett M and Isaac G (2006a) Deformation of press-fitted metallic resurfacing cups. Part 1: Experimental simulation. J Eng Med, Proc Inst Mech Engrs, 220(2), 299±309. Jin ZM, Stone MS, Ingham E and Fisher J (2006b) Biotribology. Curr Orthop, 20, 32±40. Johnson KL (1985) Contact Mechanics. Cambridge University Press, Cambridge. Kang L, Jin ZM, Isaac G and Fisher J (2006) Long term wear modelling of metal-onmetal hip resurfacing prosthesis: effect of clearance. Trans Orthop Res Soc, 31, abstract no. 0501. Knight LA, Pal S, Coleman JC, Bronson F, Haider H, Levine DL, Taylor M and Rullkoetter PJ (2007) Comparison of long-term numerical and experimental total knee replacement wear during simulated gait loading. J Biomech, 40(7), 1550±8. Korhonen RK, Koistinen A, Konttinen YT, Santavirta SS and Lappalainen R (2005) The effect of geometry and abduction angle on the stresses in cemented UHMWPE acetabular cups ± finite element simulations and experimental tests. Biomed Eng Online, 4(1), 32. Laz PJ, Pal S, Halloran JP, Petrella AJ and Rullkoetter PJ (2006) Probabilistic finite element prediction of knee wear simulator mechanics. J Biomech, 39(12), 2303±10. Liao YS and Hanes M (2006) Effects of negative clearance on the wear performance of a modern metal-on-metal implants in a hip simulation study. Trans Orthop Res Soc, 31, abstract no. 0503. Lima E, Bain LM, Serebrov A, Mauck R, Byers B, Tuan R, Ateshian G and Hung C (2006) Measuring the frictional properties of tissue-engineered cartilage constructs. Trans Orthop Res Soc, 31, abstract no. 1501. Liu F, Jin ZM, Grigoris P, Hirt F and Rieker C (2003) Contact mechanics of metal-onmetal hip implants employing a metallic cup with an UHMWPE backing. J Eng Med, Proc Inst Mech Engrs, 217, 207±13. Liu F, Wang FC, Jin ZM, Hirt F, Rieker C and Grigoris P (2004) Steady-state elastohydrodynamic lubrication analysis of a metal-on-metal hip implant employing a metallic cup with an UHMWPE backing. J Eng Med, Proc Inst Mech Engrs, 218, 261±70. Liu F, Udofia1 IJ, Jin ZM, Hirt F, Rieker C, Roberts P and Grigoris P (2005) Comparison of contact mechanics between a total hip replacement and a hip resurfacing with a metal-on-metal articulation. J Mech Eng Sci, Proc Inst Mech Engrs, 219, 727±32. Liu F, Jin ZM, Roberts P and Grigoris P (2006) Importance of head diameter, clearance © 2008, Woodhead Publishing Limited

54

Joint replacement technology

and cup wall thickness on elastohydrodynamic lubrication analysis of metal-onmetal hip resurfacing prostheses. J Eng Med, Proc Inst Mech Engrs, 220(6), 695± 704. Liu L, Leslie I, Williams S, Fisher J and Jin ZM (2008) Development of computational wear simulation of metal-on-metal hip resurfacing replacements. J Biomechanics, 44(3), 68694. Mak MM and Jin ZM (2002) Analysis of contact mechanics in ceramic-on-ceramic hip joint replacements. J Eng Med, Proc Inst Mech Engrs, 216, 231±6. Martell JM, Verner JJ and 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, 55±60. Maxian TA, Brown TD, Pedersen DR and Callaghan JJ (1996) The Frank Stinchfield Award. 3-dimensional sliding/contact computational simulation of total hip wear. Clin Orthop Relat Res, 333, 41±50. Maxian TA, Brown TD, Pedersen DR, McKellop HA, Lu B and Callaghan JJ (1997) Finite element analysis of acetabular wear. Validation, and backing and fixation effects. Clin Orthop Relat Res, 344, 111±17. Miloev I, Trebe R, Kovac S, Cor A and Pisot V (2006) Survivorship and retrieval analysis of Sikomet metal-on-metal total hip replacements at a mean of seven years. J Bone Joint Surg Am, 88(6), 1173±82. Morita Y, Tomita N, Aoki H, Sonobe M, Wakitani S, Tamada Y, Suguro T and Ikeuchi K (2006) Frictional properties of regenerated cartilage in vitro. J Biomech, 39(1), 103±9. Morlock M, Schneider E, Bluhm A, Vollmer M, Bergmann G, Muller V and Honl M. (2001) Duration and frequency of every day activities in total hip patients. J Biomech, 34(7), 873±81. MuÈller LP, Degreif J, Rudig L, Mehler D, Hely H and Rommens PM (2004) Friction of ceramic and metal hip hemi-endoprostheses against cadaveric acetabula. Arch Orthop Trauma Surg, 124(10), 681±7. Muratoglu OR, Bragdon CR, O'Connor DO, Perinchief RS, Estok DM II, Jasty M and Harris WH (2001) Larger diameter femoral heads used in conjunction with a highly cross-linked ultra-high molecular weight polyethylene. J Arthroplasty, 16, 24±30. Nevelos JE, Ingham E, Doyle C, Nevelos AB and Fisher J (2001) Wear of HIPed and non-HIPed alumina±alumina hip joints under standard and severe simulator testing conditions. Biomaterials, 22(16), 2191±7. Plainfosse M, Katta J, Jin ZM, Fisher J, Hatton PV and Crawford A (2006) Tribology and mechanical properties of tissue engineered cartilage using bovine chondrocytes seeded on PGA scaffolds. 5th UK Society of Biomaterials Conference, Manchester, UK, 28±29 June, 76. Plank GR, Estok DM 2nd, Muratoglu OK, O'Connor DO, Burroughs BR and Harris WH (2007) Contact stress assessment of conventional and highly crosslinked ultra high molecular weight polyethylene acetabular liners with finite element analysis and pressure sensitive film. J Biomed Mater Res B Appl Biomater, 80(1), 1±10. Rieker CB, Schon R and Kottig P (2004) Development and validation of a secondgeneration metal-on-metal bearing: laboratory studies and analysis of retrievals. J Arthroplasty, 19(8 Suppl 3), 5±11. Rieker CB, Schon R, Konrad R, Liebentritt G, Gnepf P, Shen M, Roberts P and Grigoris P (2005) Influence of the clearance on in-vitro tribology of large diameter metal-onmetal articulations pertaining to resurfacing hip implants. Orthop Clin North Am, 36(2), 135±42. © 2008, Woodhead Publishing Limited

Tribology in joint replacement

55

Roter GE, Medley JB, Bobyn JD, Krygier JJ and Chan FW (2002) Intermittent Motion: A Novel Simulator Protocol for the Wear of Metal±Metal Hip Implants. Tribology Series 40, Elsevier B.V., Sara Burgerhartstraat 25, P.O.Box 211, 1000 AE Amsterdam, The Netherlands, 367±76. Saari H, Santavirta S, Nordstrom D, Paavolainen P and Konttinen YT (1993) Hyaluronate in total hip replacement. J Rheumatol, 20(1), 87±90. Schmalzried TP, Szuszczewicz ES, Northfield MR, Akizuki KH, Frankel RE, Belcher G and Amstutz HC (1998) Quantitative assessment of walking activity after total hip or knee replacement. J Bone Joint Surg Am, 80(1), 54±9. Scholes SC and Unsworth A (2000) Comparison of friction and lubrication of different hip prostheses. J Eng Med, Proc Inst Mech Engrs, 214(1), 49±57. Scholes SC and Unsworth A (2001) Pin-on-plate studies on the effect of rotation on the wear of metal-on-metal samples. J Mater Sci Mater Med, 12(4), 299±303. Smith SL, Dowson D and Goldsmith AAJ (2001a) The lubrication of metal-on-metal total hip joints: a slide down the Stribeck curve. J Eng Tribol, Proc Inst Mech Engrs, 215(J5), 483±93. Smith SL, Dowson D and Goldsmith AAJ (2001b) The effect of femoral head diameter upon lubrication and wear of metal-on-metal total hip replacements. J Eng Med, Proc Inst Mech Engrs, 215 (H2), 161±70. Stewart T, Jin ZM, Shaw D, Auger DD, Stone M and Fisher J (1995) Experimental and theoretical study of the contact mechanics of five total knee joint replacements. J Eng Med, Proc Inst Mech Engrs, 209, 225±31. Stewart T, Tipper JL, Streicher R, Ingham E and Fisher J (2001) Long-term wear or HIPed alumina on alumina bearings for THR under microseparation conditions. J Mater Sci Mater Med, 12, 1053±6. Tipper JL, Firkins PJ, Ingham E, Fisher J, Stone MH and Farrar R (1999) Quantitative analysis of the wear and wear debris from low and high carbon content cobalt chrome alloys used in metal on metal total hip replacements. J Mater Sci Mater Med, 10(6), 353±62. Udofia IT, Liu F, Jin ZM, Roberts P and Grigoris P (2007) Initial stability and contact mechanics analysis of press-fit hip resurfacings prostheses. J Bone Joint Surg, (B), 89(4), 549±56. Urban RM, Tomlinson MJ, Hall DJ and Jacobs JJ (2004) Accumulation in liver and spleen of metal particles generated at nonbearing surfaces in hip arthroplasty. J Arthroplasty, 19(8 Suppl 3), 94±101. Williams S, Isaac G, Hatto P, Stone MH, Ingham E and Fisher J (2004) Comparative wear under different conditions of surface-engineered metal-on-metal bearings for total hip arthroplasty. J Arthroplasty, 19(8 Suppl 3), 112±17. Williams S, Jalali-Vahid D, Jin ZM, Stone M, Ingham E and Fisher J (2006) Effect of swing phase load on metal-on-metal hip lubrication, friction and wear. J Biomechanics, 39(12), 2274±81. Wimmer MA, Nassutt R, Sprecher C, Loos J, Tager G and Fischer A. (2006) Investigation on stick phenomena in metal-on-metal hip joints after resting periods. J Eng Med, Proc Inst Mech Engrs, 220(2), 219±27. Yew A, Jagatia M, Ensaff H and Jin ZM (2003) Analysis of contact mechanics in McKee±Farrar metal-on-metal hip implants. J Eng Med, Proc Inst Mech Engrs, 217, 333±40. Yew A, Udofia I, Jagatia M and Jin ZM (2004) Analysis of elastohydrodynamic lubrication in McKee±Farrar metal-on-metal hip joint replacements. J Eng Med, Proc Inst Mech Engrs, 218(1), 27±34. © 2008, Woodhead Publishing Limited

3

Biomaterials and the chemical environment of the body K J B U N D Y , Tulane University, USA

3.1

Introduction

The chemical environment of the body that is in contact with implanted biomaterials can have a profound (and deleterious) effect on their performance as surgical devices. All classes of materials used in implants (polymers, ceramics, metals, and composites) will degrade to some extent owing to prolonged exposure to body fluids. This degradation both undercuts engineering performance of devices and undermines the biocompatibility of surgical implants. This chapter focuses on metallic biomaterials and their deterioration in the in vivo chemical environment due to corrosion.

3.1.1

Overview of factors affecting corrosion of implants

Corrosion is a multifactorial process involving metallurgical, microstructure, and composition variables, environmental solution chemistry, applied mechanical stresses, and geometrical factors. For implants that are placed within living organisms, there are additional factors related to biological and physiological conditions that impact on the corrosion situation.

3.1.2

Relation between corrosion, biocompatibility, and engineering failure of implants

In the field of engineering as a whole, corrosion represents a problem since it undermines the integrity of structures, leading to fractures, leaks, contamination of product, and other sorts of failure modes. Though, as discussed later, there are problems with implanted devices where failure in an engineering sense is caused or exacerbated by corrosion processes leading to fracture, this is not, however, the primary mode of implant failure induced by corrosion. The reasons for corrosion-related implant failure are primarily biological and physiological. Thus, failures of procedures using metallic implants most commonly arise from biocompatibility problems that are caused by adverse © 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

57

reactions of the body's defense systems provoked by release of implant degradation products. The nature of these reactions is discussed further below.

3.1.3

Organization of this chapter

This chapter treats the impact of the chemical environment of the body on the performance and durability of metallic implant devices, emphasizing those that are load-bearing. Section 3.2 describes the chemical composition of body fluids in terms of the species present that can affect corrosion attack of implant alloys, as well as active aspects of the biological environment that may influence corrosion of implanted devices. Corrosion is a surface phenomenon occurring at the metal surface/electrolyte interface. Accordingly, phenomena that affect the nature of the implant surface will also affect corrosion processes. These factors are discussed in Section 3.3. Section 3.4 provides background information about the electrochemical basis of corrosion processes. Various means for testing the corrosion resistance of implant materials are also covered, as are specific types of corrosion processes that can adversely affect implant performance. Finally, at the end of the chapter, likely future trends related to implant corrosion, as well as sources for further information on this topic are also presented.

3.2

Chemical environment for joint replacement

The chemistry of the body is complex and varies both with location and time. It is important to understand the nature of the chemical environment for joint replacement devices in order to see what factors influence corrosion of such implants. The environmental factors to be considered include both passive aspects related to the chemical composition of body fluids (ions, dissolved gases, and other components that are present) as well as active processes pertinent to physiological functioning of a living organism.

3.2.1

Ions

As far as their inorganic salt composition is concerned, body fluids are essentially dilute saltwater solutions (Bundy and Zardiackas, 2006). Table 3.1 shows the ionic compositions of extracellular fluid and blood plasma. The chloride ion is the most important of these constituents from the corrosion point of view, since it is the most potent ion present in terms of its ability to disrupt passive films, as discussed further below.

3.2.2

Dissolved gases

Besides the dissolved salts, gases present in body fluids have important physiological roles. Oxygen is essential for cellular function and also plays an © 2008, Woodhead Publishing Limited

58

Joint replacement technology Table 3.1 Concentrations of anions and cations in body fluids in contact with metallic biomaterials Blood plasma*

Extracellular fluid*

Anions Clÿ HCO3ÿ H2PO4ÿ HPO4ÿ2 SO4ÿ2

96±111 16±31 2 1±1.5 0.35±1.0

112±120 25.3±29.7 ± 1 0.4

Cations Na‡ K‡ Ca‡2 Mg‡2

131±155 3.5±5.6 1.9±3.0 0.7±1.9

141±145 3.5±4.0 1.4±1.55 1.3

* Concentration units: mM.

important part in corrosion reactions as a cathodic reactant (as discussed later). Carbon dioxide is an important regulator of body pH. Since the acidity of the chemical environment of the body has a great impact on the stability of passive films, CO2 also can influence corrosion of medical devices. The concentrations of these gaseous species in various fluids that could contact surgical implants are shown in Table 3.2. Note that concentrations of dissolved gases can be expressed in terms of their partial pressure.

3.2.3

Organic compounds

Many different organic compounds are found in various body fluids that can contact metallic implants, particularly blood. Table 3.3 shows the main organic components found in plasma (Orten and Neuhaus, 1975; Bundy, 1994). The total concentration of organics can exceed 80 g/l. These compounds are predominantly proteins, though other substances such as fatty acids, glucose, cholesterol, lactate, and urea are also present. Adsorption of organics onto the surfaces of implants can have important influences on corrosion processes, as described later. Table 3.2 Concentrations of dissolved gases in various body fluids in contact with metallic biomaterials Gas component O2 Dissolved O2 O2 combined with hemoglobin CO2

Arterial blood

Venous blood

Interstitial fluid

100 mm Hg 3 ml/l 200 ml/l

40 mm Hg 1.2 ml/l 154 ml/l

2±40 mm Hg ± ±

15±19 mm Hg

17±20 mm Hg

46 mm Hg

© 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

59

Table 3.3 Concentration of organic compounds found in blood plasma Substance Albumin

-globulins -globulins 1 -lipoproteins -globulins Fibrinogen Fatty acids Total cholesterol Glucose Lactate Urea

3.2.4

Concentration (g/l) 30±55 6.6±15 6±12 6±12 5±10 1.7±4.3 1.9±4.5 1.2±2.5 0.65±1.1 0.27±0.62 0.03±0.13

pH effects

Body pH is homeostatically regulated, but can vary considerably from location to location. Gastric juices, for example, can have a pH value as low as 1 (Black, 1992), while most body fluids are at near-neutral pH, often tending to be slightly alkaline (Bundy, 1994). The pH ranges for various fluids that the surface of orthopedic alloys might touch, are given in Table 3.4. In unusual situations, e.g. at an inflammation site as discussed below, the pH can drop to an abnormally acidic value. Also, when there are small volumes of fluid present just below implant surfaces (where diffusion and convection of ions to and from the bulk electrolyte are restricted), due to localized corrosion processes such as pitting or crevice corrosion, very acidic pH values can locally develop. The mechanisms underlying this phenomenon are discussed below.

3.2.5

Possible influences of active physiological processes

Besides the influences on implant corrosion that have been discussed above, various active cellular processes can possibly influence corrosion of implant Table 3.4 pH ranges for various body fluids that can touch orthopedic alloys Fluid Whole blood Interstitial fluid Blood serum or plasma Intracellular fluid Synovial fluid

© 2008, Woodhead Publishing Limited

Range 7.03±7.78 7.0±7.78 7.38±7.42 6.8±7.0 7.29±7.7

60

Joint replacement technology

devices. As will be seen later, corrosion is an electrochemical process. Some of these effects are electrical in nature and may result in electrical potentials being applied to surgical implants. In other cases, cellular activities can markedly locally alter the usual chemical environment of the body, and this may cause more aggressive conditions than are usually present. These aspects are discussed further below. Bioelectric effects Electrical potentials in tissues can arise due to the functioning of nerves, or in the case of some musculoskeletal tissues (bone and cartilage), they are caused by the application of mechanical forces to the tissues. With respect to load-bearing, various explanations for these potentials have been advanced over the years. Originally, the potentials were attributed to piezoelectric effects (Fukada and Yasuda, 1957), although more recently interest has focused upon streaming potentials (Guzelsu and Walsh, 1990). In any case, in principle it is possible that these potentials could polarize localized portions of an implant's surface. In a borderline passivity situation, this might spur acceleration of corrosive attack. Table 3.5 provides the range of potentials that could be expected from active physiological electrical activity. The original data sources for the table are provided in the review article by Bundy (1994). Inflammation Various factors associated with inflammation may have an effect on implant alloy corrosion processes. For example, pH values may shift as low as 4±5 at an inflammation site, and this may persist for many days if a hematoma is present (Laing, 1977). The vast majority of studies related to corrosion and inflammation have been aimed at investigating the biological reactions induced by implant corrosion, and their biocompatibility sequelae. On the other hand, active oxygen-containing species, such as the superoxide anion and others, are known to be produced by the activity of macrophages and other phagocytic cells (Coury, 2004). These have the potential to affect corrosion Table 3.5 Range of bioelectric potentials produced by physiological activity in various tissues Tissue Nerves (transmembrane) Cartilage Muscle Bone Nerves (measured via remote electrode)

© 2008, Woodhead Publishing Limited

Potential range (mV) ÿ90 to ‡35 ÿ1 to ‡22 1 to 12 ÿ2 to 7.6 ÿ0.4 to 1.4

Biomaterials and the chemical environment of the body

61

resistance of implants, and a few investigators have examined their influence on metallic biomaterials. So far, the situation regarding how these reactive species influence the corrosion behavior of biomaterials seems to be complex. The effect of these species seems to be to enhance passive film stability in some cases and to disrupt the passive layer in others. This is reminiscent of the influence (described later) of adsorbed protein on implant corrosion. For titanium alloys, Lin and Bumgardner (2004a) found that the released reactive chemical species from macrophages enhanced passive film integrity and decreased the rate of corrosion of Ti±6Al±4V. However, Mu et al. (2000) found that active oxygen species released by macrophages increased the corrosion rate of titanium. Lin and Bumgardner (2004b) also found that CoCrMo alloys, when cultured with macrophages, had lower corrosion rates, which they attributed to the same mechanism that applied to the titanium alloy that they tested.

3.3

Surfaces and interfaces

By nature, corrosion is a surface phenomenon that involves the interaction of the metal and the electrolyte it contacts at the interface between them. This section discusses various aspects of the surface of implanted materials that have a major impact on corrosion behavior.

3.3.1

Passive films

If the electrolyte were to be directly in contact with bare metal, the rate of corrosion of an implant would be quite rapid. However, many materials are in fact quite corrosion resistant, including those used in joint replacement. Though some materials made of very noble metals (gold, for example) have thermodynamic immunity to corrosion, the formation of passive films on the surface of implant biomaterials is, by far, the mechanism most responsible for their corrosion resistance. A passive film is a very thin oxide (or oxygen-containing) layer, whose thickness is of the order of nanometers, at the surface that screens off the substrate metal from contact with the electrolyte. The corrosion rate is not zero when a passive film is present, since the passive film can dissolve in the electrolyte to a certain extent, and it may have imperfections and porosity that allow a small portion of bare metal surface to be in direct contact with the electrolyte.

3.3.2

Surface finish and exposed area

All things being equal, the amount of metal released into the body is proportional to the surface area of the implant that is exposed to the electrolyte. For a device such as an artificial joint replacement, the true amount of area © 2008, Woodhead Publishing Limited

62

Joint replacement technology

exposed to the electrolyte per unit of nominal surface area can vary rather widely (by about an order of magnitude) because of the different surface finishes and topography that may be used on various portions of the device. For example, the bearing surface of an artificial joint replacement may be highly polished to minimize friction and wear, while other parts of the implant may have a rougher grit-blasted finish, with more true surface area per unit of nominal area. For some implants, a porous-coated layer is used to foster bone ingrowth at certain places on the device, and these surfaces will present the largest amount of true surface area to the electrolyte per nominal unit area of the implant.

3.3.3

Adsorbed protein

When an implant surface is placed in contact with the in vivo milieu, it readily becomes coated with a layer of adsorbed protein. This occurs for two reasons (Israelachvili and McGuiggan, 1988). First, attractive van der Waals forces exist between protein macromolecules and the surface. There can be electrostatic interactions as well, since the surface of a metal in an electrolyte has a net charge, and various functional groups in certain amino acids comprising the polypeptide chain of the protein are also charged, e.g. aspartate, glutamate, lysine, arginine, and histidine (Denniston et al., 2004). There have been various reports regarding the influence of the adsorbed protein layer on corrosion. According to some reports, the adsorbed layer can further protect the surface from contact with the electrolyte, thus acting in a sense as an additional passive layer. On the other hand, proteins can chelate metal atoms. In some cases, this can be a mechanism that accelerates corrosion.

3.3.4

Contact stresses

Since metal implants are made from strong materials, their main uses are in load-bearing applications for orthopedics and dentistry. Because of the forces applied to body tissues, such as bone and teeth, significant mechanical stresses can be created at the surface of implants. Similarly, in multi-component devices (such as a fracture fixation plate held in place with multiple screws), significant contact stresses can be developed at the surfaces of mating components. As discussed later, there are several forms of corrosion that are exacerbated by stresses applied to biomaterials, which lead to greater release of metal into the in vivo environment, as well as to fracture in some cases.

3.3.5

Crevice geometry

In many situations, two surfaces are in extremely close proximity in an electrolyte, such that only a very small gap exists between them. In the field of corrosion science, this gap is commonly known as a `crevice'. Crevice geometry © 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

63

may exist, for example, at the plate/screw interface mentioned in the previous section, or between the ball of a total hip replacement femoral component and the acetabular cup. In modular total hip replacement devices, the taper region between the ball and the shaft portions of the femoral component represents another example of crevice geometry (Collier et al., 1991). The area between an implant and overlying, or underlying, bone could also present crevice conditions, as might the area in a total joint replacement between the metallic component and the bone cement mantle, if loosening has occurred at the interface between the two. The presence of a crevice can lead to aggressive chemical conditions being established within it, a phenomenon known as crevice corrosion (which is discussed in more detail later). This effect may undermine the stability of passive films in the affected area, which can markedly increase the rate of corrosion. An example of such corrosion in a modular total hip replacement device is shown in Fig. 3.1.

3.1 Severe corrosion in retrieved modular total hip replacement crevice (photograph courtesy of Dr J. Collier). Source: Bundy (1994). © 2008, Woodhead Publishing Limited

64

3.3.6

Joint replacement technology

Materials used to achieve corrosion resistance

Only a few of the many thousands of existing alloy systems have been found over the years to be suitable for in vivo engineering service, because of the demanding performance requirements for load-bearing surgical implant materials, i.e. superior mechanical strength and durability properties, a very high degree of corrosion resistance, and good biocompatibility. Those that have been used include certain stainless steels, cobalt±base alloys, and titanium alloys. In the United States, these three types of alloy systems respectively make up about 70%, 20%, and 10% of the market for implanted metallic orthopedic devices. 316L stainless steel, Ti±6Al±4V ELI, and various Co±Cr±Mo alloys account for the vast majority of these devices. The `ELI' designation for the titanium alloy refers to `extra low interstitials'. This means that the alloy has a very low concentration of interstitial elements (i.e. very small atoms such as carbon and nitrogen whose presence can diminish the mechanical performance of the material). Only about 21 specific alloy compositions have been sufficiently well accepted for surgical implant use in the United States such that standards exist for them. The applicable ASTM standards for metallic orthopaedic implant materials are given in Table 3.6. A standard also exists for tantalum implants (F560). Such devices were rarely used for many years. However, in the past decade or so, a porous form of tantalum has been employed in a variety of orthopedic applications (Bobyn et al., 1999; Findlay et al., 2004; Chalkin and Minter, 2005; Macheras et al., 2006).

Table 3.6 ASTM standard designations for the major orthopedic implant alloy systems Stainless steels: F138, F1314, F1586, F2229 Cobalt±chromium alloys: F75, F90, F562 F563, F688, F799, F961, F1058, F1537 Titanium/Ti-based alloys: F67, F136, F1295, F1341, F1472, F1580, F1813, F2066, F2146

3.4

Corrosion

Corrosion is a complex, multifactorial electrochemical phenomenon that can unfold in many ways. Aspects of both thermodynamics and kinetics are needed to describe the events occurring during corrosion processes. Perhaps because of this complexity, a wide spectrum of experimental methods has been developed to measure and monitor corrosion of metals, including implant alloys. These aspects are discussed below.

© 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

3.4.1

65

Basic aspects of electrochemistry

Corrosion is an electrochemical process that is a type of oxidation±reduction reaction. For a corrosion process there is a minimum of two electrochemical reactions that are of vital importance ± the anodic reaction and the cathodic reaction. The anodic reaction represents oxidation in the chemical sense. For example, for a metal M that electrolytically dissolves in an electrolyte to form cations, the anodic reaction would be represented as: M ! M+z + zeÿ

3.1 -

where z is the oxidation state of the ion and e refers to an electron. The z electrons that are left behind at the surface as the metal dissolves are consumed by the cathodic reaction, which is a reduction reaction in the chemical sense, usually involving a gaseous species. Probably the most common cathodic reaction in vivo is reduction of dissolved oxygen, which under the near-neutral pH conditions generally prevalent in the body, would be: O2 ‡ 2H2O ‡ 4eÿ ! 4OHÿ

3.2

The electrical potential E (versus a reference electrode) at which an anodic or cathodic reaction is at dynamic equilibrium (i.e. progressing at an equal rate in the forward and reverse direction) is called the reversible electrode potential, Erep. The value of the reversible electrode potential is governed by the Nernst equation (Bockris and Reddy, 1970):     …aox spec †nox RT 3:3  log Erep ˆ Eo ‡ 2:3 …ared spec †nred zF where Eo is a constant termed the standard electrode potential, R is the universal gas constant, T is the absolute temperature, z refers to the number of electrons involved in the reaction, F is another constant (known as the Faraday constant, equal to 9:65  104 C/mole of charge), aox spec and ared spec refer to the activities (which for dilute solutions equal the concentrations) of the oxidized and reduced species, and nox and nred refer to the number of moles of the oxidized and reduced species involved in the reaction. For the cathodic reaction given by reaction [3.2], for example, the Nernst equation would be:     RT pO2 3:4 Erep ˆ Eo ‡ 2:3  log aOHÿ 4F As pointed out before, note that, for a gaseous species, the activity is equal to its partial pressure p (or more precisely to its fugacity). Values of Eo for many reactions are assembled in many reference works into a table called the electromotive series. From consulting this table (Bockris and Reddy, 1970), the standard electrode potential for the oxygen reduction reaction, for example, is +0.40 V versus the standard hydrogen electrode. The Nernst equation is the

© 2008, Woodhead Publishing Limited

66

Joint replacement technology

fundamental equation governing the thermodynamic status of an electrochemical reaction. Since the anodic reaction represents a flow of ions into the environment, there is a current associated with this charge flow, here denoted as Ic. This is often normalized to the area of the metal exposed to the electrolyte so that the flow is expressed as a current density, symbolized as ic. The fundamental equation governing electrochemical kinetics, i.e. the relationship between potential and current density, is known as the Butler±Volmer equation (Bockris and Reddy, 1970), which is a form of the Arrhenius equation. The Butler±Volmer equation is given by:      Fzan …1 ÿ an † ÿFzca  ca 3:5 i…† ˆ ic exp ÿ exp RT RT where  is the applied overpotential (i.e. E ÿ Ec ), i…† is the net current density as a function of , and ic refers to the free corrosion current density (i.e. the current density at the free corrosion potential Ec). F, z, R, and T have the same meaning as before. The subscripts `an' and `ca' respectively refer to the anodic and cathodic reactions, and is a constant termed the symmetry coefficient. At Ec, the current associated with the anodic reaction equals that for the cathodic reaction, and both reactions therefore occur at an equal rate.

3.4.2

Nature of corrosion products

As we saw in the previous section, metal ions are released into the electrolyte by the anodic reaction (reaction [3.1] above). The ions represent soluble corrosion products. They can, however, react further with other components in the in vivo electrolyte, hydroxide ions, for instance, to form compounds that can be insoluble. For a divalent cation, for example, this precipitation reaction would be: M+2 ‡ 2OHÿ ! M(OH)2 #

3.6

Another possibility is that ions may be complexed by proteins. This may denature the proteins in some cases, presenting a biocompatibility challenge. A final possibility is that ions may be incorporated into formed elements (cells) by passing through cell membranes or adsorbing to cell surfaces.

3.4.3

Means for assessing corrosion of surgical implants

A wide variety of approaches has been used to evaluate corrosion of implant materials. These differ mainly in terms of the type of corrosive environment used for study, as well as the types of data that are acquired and analyzed. Laboratory test environments, where the investigator selects the electrolyte chemistry, represent the most controlled, but least realistic, experimental conditions. Conversely, corrosion testing using animal models are the most © 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

67

realistic, but least controlled, approach. Other techniques are intermediate in terms of their control/realism mix. With regard to specific experimental techniques used to investigate implant corrosion processes, microscopic observation, surface chemical analysis, and electrochemical testing all play a role. Electrochemical testing is emphasized in this section because more elaborate discussion is needed to describe this approach compared with the other two. Both in terms of the equipment used and mathematical data analysis required, electrochemical methodology for assessing corrosion processes spans a wide range of sophistication, from quite basic to rather complex. Electrochemical testing in laboratory environments: electrolyte selection As previously indicated, body fluids that implants contact are basically saline solutions with dissolved gases that also contain various organic compounds. The most controlled means of corrosion assessment involves testing under laboratory conditions using aqueous solutions designed to mimic in vivo conditions (at least to some extent). There is no one standard solution that is used for this purpose. Some solutions that are commonly employed include simple saline solutions, Ringer's solution, Tyrode's solution, and Hank's solution. The detailed chemical composition of these solutions are provided in various works (e.g., Bundy, 1994; Fraker, 2005). Some investigators add protein to the laboratory test solution. Typically albumin, the most abundant plasma protein (see Table 3.3), is used for this purpose. Experimental techniques For materials that are not very corrosion resistant, gravimetric means can be used to assess the susceptibility for corrosion. For highly corrosion-resistant materials, however, as all implant alloys are, weight loss measurements are not practical, because of the very slow rates of corrosion involved (and thus the extremely long exposure times that would be required to build up an appreciable weight loss). Electrochemical methods are therefore almost always used to measure corrosion of surgical implant materials. These techniques involve the measurement of corrosion current, I, as a function of potential, E. Both DC and AC methods can be used. For the DC techniques, E is changed over time at a constant rate (to the order of mV/s). Here the waveform used is therefore a ramp function. For AC methods, on the other hand, sinusoidal variations of potential with time are used as the applied waveform (Bard and Faulkner, 1980). In either case, the device that applies these potentials is termed a potentiostat. Considering the DC methods first, in the linear polarization method a small amplitude voltage scan is used that falls roughly within about 25 mV of Ec. © 2008, Woodhead Publishing Limited

68

Joint replacement technology

Here there is a linear relationship between potential and current. The slope of the line,
E=
I is known as the polarization resistance, Rp. The corrosion current density can be obtained via the Stern±Geary equation (Fontana and Greene, 1978), which is a linearized form of the Butler±Volmer equation: ic ˆ

an ca 2:3Rp … an ca †

3:7

where an and ca are known as the anodic and cathodic Tafel slopes, respectively, and are equal to: an ˆ

2:3RT Fzan …1 ÿ an †

3:8

ca ˆ

2:3RT Fzca ca

3:9

Because of the low amplitude of the potential excursion about the free corrosion potential, the linear polarization method is a non-destructive technique that does not lead to appreciable surface attack during the conduct of the test. Accordingly, this method can be applied repeatedly in order to monitor the corrosion current over time. When large DC potential excursions away from Ec are used (of a volt or more, for example), the method is termed potentiodynamic polarization, and the E vs. I curve is called a potentiodynamic polarization curve. Generally this is plotted as E vs. log I because the current may change by orders of magnitude as a consequence of the potential range applied during the measurement. For a metal undergoing active corrosion (i.e. with no passive film formation), the potentiodynamic polarization curve plotted as explained above will be linear, when  is sufficiently large. This can be seen from the Tafel law: an ˆ an log…i=ic †

3:10

ca ˆ ÿ ca log…i=ic †

3:11

for anodic and cathodic polarization, respectively. The Tafel law can be obtained by solving the Butler±Volmer equation for , when the amount of anodic polarization is sufficiently large so that the cathodic polarization term is negligible, resulting in equation [3.10], and vice versa, resulting in equation [3.11]. From the potentiodynamic polarization curve, the Tafel slopes can be measured by determining the change in potential resulting from a decade (i.e., order of magnitude) change in the current density. For metals that show both active and passive behavior (depending on the value of E), insights can be obtained from this curve as to the potential range in which the passive film is stable. AC methods, variously known as electrochemical impedance spectroscopy (EIS) or the AC impedance (ACI) technique, utilize a somewhat different © 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

69

approach. Here the test electrode is stimulated by a low-amplitude sinusoidal potential (within 10 mV or so of Ec) and the phase delay, , between the applied E and the resultant I is measured. A wide range of frequencies (!) is typically involved in an EIS measurement. The  versus ! data is then analyzed mathematically using various circuit analysis techniques originally developed by electrical engineers. This provides a circuit analogy for the metal/electrolyte interface, i.e. a representation of its electrochemical behavior in terms of circuit elements such as resistors, capacitors, etc. (Bard and Faulkner, 1980). This technique provides a more powerful means for assessing corrosion. Not only can Rp be determined in this way, but also information can be obtained about electrolyte conductivity and the capacitance of the double layer at the metal/ electrolyte interface. From the capacitance, the true surface area of a material can be established. Changes in area of the metal exposed to the electrolyte due to pitting or cracking can also be monitored with the EIS method (Bundy et al., 1993). More elaborate circuit models can provide information regarding surface porosity, passive film structure, the influence of diffusion of reactants on the corrosion rate, and other aspects of electrochemical behavior. Retrieval analysis Retrieval analysis refers to the examination of implants removed from patients, for various reasons, usually after they have been in service in vivo for an extensive period of time. Retrieved devices are usually observed visually or with low magnification optical microscopy. Scanning electron microscopy and chemical analysis of implant surfaces are also often employed. Surrounding tissue at the implant site is commonly observed histologically, and may also be subjected to chemical analysis to determine its metal content. Though evaluation of implant corrosion is not the only objective of retrieval studies, much information about it has been gleaned from this type of research. All the forms of corrosion that are mentioned in Section 3.4.4 have been observed in implanted devices. Reports of corrosion are much more numerous for surgical implants made of stainless steel than for those fabricated from cobalt-based alloys. Titanium alloys are the most corrosion resistant of the three common orthopedic alloy systems. Though the number of reports of corrosion is, to a certain extent reflective of a biomaterial's share of the implant market, it is still clear that the stainless steels are the most corrosion-prone type of surgical implant alloy. In some studies, the percentage of retrieved implants exhibiting corrosion (by eye or low-power optical magnification) is 100%. These findings mirror the results obtained via electrochemical testing. The review article by Bundy (1994) has further information on retrieval analyses, including many examples of individual retrieval studies found in the literature.

© 2008, Woodhead Publishing Limited

70

Joint replacement technology

Cell culture Mammalian cells and bacteria (in the case of microbiological corrosion) can influence the corrosion rate of biomedical materials. The vast majority of studies related to cells and corrosion have been concerned with biocompatibility and have involved the exposure of the cells (or bacteria) to various implant corrosion products (usually over a range of concentration) and monitoring of various aspects of cellular response to them. A few studies, however, have combined electrochemical corrosion measurement techniques and cell culture methods. The investigations mentioned previously that examined the influence of macrophages on implant alloy corrosion rates are examples of such work. Using polarization and EIS methods, Hallab et al. (1995) investigated how implant biomaterial surface charge and corrosion potential affected fibroblast adhesion strength. Messer et al. (2006) describe a cell culture chamber in which the influence of cells on implant corrosion processes can be examined under conditions where the electrolyte flows over the biomaterial surface. Using cell culture models to investigate various aspects of how cells influence implant alloy corrosion is an area worthy of much more intensive investigation in the future. Animal models In vivo investigations in animal models have been employed to study implant corrosion, tissue reactions to such corrosion, and the concentrations of alloy degradation products that build up in body tissues and fluids because of it. A variety of animal species have been used in these tests including baboons, monkeys, sheep goats, dogs, rabbits, hamsters, and rats, and no consensus has developed as to which one is best to use (Bundy, 1994). In this body of research, significant efforts have been devoted to electrochemical testing, microscopic observation and chemical analysis of surface damage, and non-electrochemical means of corrosion testing (particularly for forms of corrosion involving interaction with mechanical loading). All of the electrochemical testing techniques mentioned in this chapter and more have been employed for corrosion testing of surgical implant materials in animal models. Generally, though not always, good correlation has been observed between in vivo tests in animals and in vitro laboratory testing or retrieval studies. DC electrochemical testing, however, can be susceptible to inaccuracies due to high solution or tissue resistance compared to laboratory environments (Bundy and Luedemann, 1989). This deficiency can be corrected by employing EIS methods for testing.

© 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

3.4.4

71

Types of corrosion relevant to artificial joint replacement

Though the fundamentals of corrosion science discussed above apply to all corrosion processes, the specific manner in which the corrosion attack damages the surface differs among them. A number of specific mechanisms of corrosive attack have been identified, most (but not all) of which have been observed in retrieved surgical implants. Those that apply to medical devices are described below. These mechanisms differ in terms of whether the attack is uniform or localized at specific spots on the surface, whether mechanical forces are involved or not, and whether or not modifications in electrolyte chemistry as a consequence of the corrosion are important. Uniform corrosion For a uniform corrosion process, the amount of metal leaving the surface per unit time is statistically the same, creating a uniform depth of surface attack at all points. The weight of metal, W, lost in exposure time t is given by Faraday's law: W ˆ kIc t

3:12

where Ic is the average corrosion current flowing over the time interval t, and k is a constant termed the electrochemical equivalent. The electrochemical equivalent refers to the weight of metal that must be lost by corrosion to produce one mole of electronic charge. The electrochemical equivalent can be calculated in the following manner: AW 3:13 zF where AW is the atomic weight of the metal, and z and F have the same meaning described before. Kˆ

Pitting Pitting occurs when there is a situation of borderline passivity that is due to mechanical damage to the surface or else local differences in alloy composition, metallurgical structure, environmental chemistry, etc. For this form of corrosion, most of the surface behaves in a corrosion-resistant fashion. However, in localized spots the passive film is disrupted. At these locations, attack of the underlying metal can then occur very rapidly, forming small holes or `pits' proceeding into the surface. Chemical changes in the electrolyte found in the pit (described in the next section) can exacerbate the corrosive attack. The severity of pitting can be monitored by finding the pitting density, i.e. the number of pits per unit area, or by determination of the pitting factor, PF: PF ˆ P=d

© 2008, Woodhead Publishing Limited

3:14

72

Joint replacement technology

where d is the average depth of uniform corrosion attack that has occurred, and P is the depth of the average pit. Besides the potential for adverse biocompatibility consequences, pitting corrosion can have negative consequences for mechanical behavior of loadbearing surgical implants. Since a pit can act as a stress concentration from which cracks can originate (Fraker, 2005), pitting corrosion can cause premature failure of implanted devices due to fatigue, corrosion fatigue and stress corrosion cracking. These corrosion mechanisms are discussed below. Potentiodynamic polarization curves are useful for assessing the susceptibility to pitting corrosion. As the potential is increased for an alloy in the passive region, eventually the breakdown potential Ebd is reached. Here the passive film is disrupted and further increases in potential produce a sharp increase in current. When the voltage scan is reversed (creating a cyclic potentiodynamic polarization curve), the value of potential on the back scan where the current again becomes equal to its original passive value, known as the repassivation potential Erp, can be measured. An example of such a curve for an implant alloy is shown in Fig. 3.2. By comparison of these two values of potential with the free corrosion potential Ec, insight into pitting susceptibility under given environmental conditions can be obtained: If Ec > Ebd, then the material is undergoing pitting corrosion. When Ec < Erp, the material is not susceptible to pitting (or to crevice corrosion, since the electrolyte chemistry is similar; see the next section). For Erp < Ec < Ebd, new pits will not form, but existing ones will propagate.

3.2 Cyclic potentiodynamic polarization curve for 316L stainless steel in 37 ëC Ringer's solution illustrating various electrochemical parameters. Source: Bundy (1994). SCE refers to the saturated calomel electrode, the reference electrode used for the measurement. © 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

73

Crevice corrosion Crevice corrosion, as the name would imply, is a form of localized corrosive attack that occurs due to the presence of a geometrical crevice. As mentioned above, in a pit or crevice the chemistry of the local environment changes and becomes more aggressive. To explain this, consider the following series of events (Fontana and Greene, 1978). When corrosion starts, the O2 initially present in the crevice is depleted after a period of time by oxygen reduction as shown in reaction [3.2]. However, in this case, because of the crevice geometry, the O2 level cannot be significantly replenished by convection and/or diffusion. This situation in the crevice will therefore favor the occurrence of the anodic reaction, reaction [3.1], since there will be insufficient oxygen in the crevice to drive the oxygen reduction reaction [3.2]. The build-up of positive ionic charge in the crevice due to the anodic reaction will attract anions (mainly chloride) into the crevice because of the electric fields and electrostatic forces that are created. For what follows below, consider that the metallic ions are univalent. Since the electrolyte in the crevice is an aqueous environment, water molecules are also present, so the chemical reactions in the crevice can be described as follows: M‡ ‡ Clÿ ‡ H2O ! M‡ ‡ Clÿ ‡ H‡ ‡ OHÿ

3.15

If the hydroxide is only sparingly soluble, then a reaction similar to [3.6] will occur, so: M‡ ‡ Clÿ ‡ H‡ ‡ OHÿ ! (MOH)# ‡ H‡ ‡ Clÿ

3.16

In other words, insoluble hydroxides and hydrochloric acid will be formed in a pit or crevice. This can be quite a substantial effect. In a near-neutral external environment, the pH in a pit or crevice can diminish down to a value of approximately 2, and the chloride level can be about 10 times as concentrated as in the bulk solution. Both low pH and the presence of Clÿ ions will disrupt passive films, a situation leading to significant corrosive attack in the crevice or pit. Intergranular corrosion As the name would imply, intergranular corrosion refers to a situation in a polycrystalline material where the grain boundaries are anodic to the interiors of the grains, and are rapidly attacked. Differences in chemical composition with distance away from the grain boundary (due to grain boundary precipitates, or other reasons) are usually the cause of intergranular corrosion. The main situation where this phenomenon has been observed in implants is with stainless steel devices that have been improperly heat treated, and the series of events described below has occurred (Fontana and Greene, 1978). Stainless steels must have at least 12 wt% chromium, in order to develop passive films. In certain temperature ranges, chromium-rich carbides can precipitate intergranularly by a © 2008, Woodhead Publishing Limited

74

Joint replacement technology

process of nucleation and growth. This will deplete the steel of Cr in the vicinity of the grain boundary, creating a narrow strip along the grain boundary where the %Cr is less than the critical value of 12%. If the steel is not then held at elevated temperature long enough for diffusion to eliminate the concentration gradient in Cr, these small areas in the vicinity of the grain boundary will rapidly corrode, first since they are deficient in chromium so will not be covered with a passive film, and secondly because they have much less surface area than does the grain interior. Since the current associated with the anodic and cathodic reaction is equal, the small areas depleted in Cr will therefore have a high anodic current density. Galvanic corrosion When two different metals are in physical contact, they are also in electrical contact. According to what is known as mixed potential theory (Vetter, 1967), two metals with different Ec values will polarize until they reach the same potential (called the mixed potential) at a value lying between the two Ec values. The material with the lower corrosion potential will therefore have a current density higher than it would in the absence of contact with the other metal. This effect is termed galvanic corrosion and leads to accelerated corrosive attack of the more anodic material. In the early days of orthopedic surgery, multicomponent implants were often made from more than one material (fracture fixation plates different from the screws for instance). When mixed potential theory was discovered, a long period followed in which mixed metals were never used. This approach was gradually relaxed, however, since it was thought by some, e.g. Lucas et al. (1981), that it was permissible to mix alloys if both were very corrosion resistant and had chemically stable passive films. However, experience with modular total hip replacements in the 1990s, some of which used mixed metals (CoCrMo alloy heads and Ti±6Al±4V ELI femoral stems) showed that pronounced corrosion could appear in the taper junction between the two components (Collier et al., 1991). Though in such situations, the degree to which fretting, crevice corrosion, and/or galvanic effects are responsible for the attack is not always clear, it is certain that the key to minimizing the corrosion of such devices is to truly ensure that the passive film does remain stable in the face of both the chemical environment of the body and the mechanical stresses to which the modular total hip replacement is subjected. Influence of applied stress Conjoint action of corrosive attack and mechanical forces is responsible for several mechanisms of corrosive attack. These all involve passive films and situations of borderline passivity, but differ from one another in terms of the © 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

75

type of loading involved and the specificity of the chemical conditions needed to initiate the corrosive attack. The three corrosion mechanisms involving interaction with mechanical stresses: fretting, corrosion fatigue, and stress corrosion cracking are described in more detail below. Fretting Fretting corrosion is a form of corrosion that, like crevice corrosion, involves corrosion where two different surfaces have an interface that brings them into quite close proximity. Here, though, it is not so much electrolyte chemistry that plays an influential role. Rather, mechanical load effects are the cause of fretting corrosion. The fracture fixation plate, previously mentioned, provides a good example of a situation where fretting corrosion might occur. Over the gait cycle, the forces applied to the implant device will cause the two surfaces at a screw/plate interface to rub against one another. This cyclically applied abrasion, or fretting, can damage the passive film. Because of this wear occurring at the interface, significant localized corrosive attack of the implant occurs. Another important example where this process can play a significant role is at the taper junction of modular total hip replacements, mentioned earlier. Many cases of frettingassisted crevice corrosion have been reported (e.g. Gilbert et al., 1993; Brown et al., 1995). Here, relative motion between the two surfaces at the taper causes fretting corrosion, and the attack that occurs causes changes in the chemistry of the electrolyte (see above) that has infiltrated into the taper. These changes are sufficient so that the crevice corrosion reactions (described previously) can be sustained over time. Corrosion fatigue Fatigue refers to the fracture of materials by repeated cyclic loads that are of insufficient magnitude to break the material in one load application. Fatigue behavior is often described by an S±N curve, a graph of the number of load cycles N required to fracture the material at a stress amplitude S (Callister, 2003). A fatigue process consists of two portions ± first initiation of a crack at the surface of the material where tensile stresses are applied, followed by propagation of the crack through the cross-section of the part. When the area bearing the load has been sufficiently diminished, the material fails by rapid fracture of the remaining ligament of material. Corrosion fatigue is a situation where there is premature failure due to the action of corrosion at the surface, which shortens the time required for the nucleation of a fatigue crack. This corrosion can cause stress concentrations where crack nucleation is favorable. Corrosion fatigue is manifested in the S±N curve by a diminished number of load cycles to failure for a given S value and a reduced S to cause © 2008, Woodhead Publishing Limited

76

Joint replacement technology

failure for a given number of load cycles. Corrosion fatigue is another example of a localized corrosion process, since there is little evidence of corrosion occurring except on the surfaces of the two sides of the crack. Stress corrosion cracking Stress corrosion cracking, or SCC, is another form of corrosion that involves both corrosive attack and mechanical stresses. Here, though, in contrast to corrosion fatigue, it is static tensile stresses that are involved (Bundy and Zardiackas, 2006). The stress may be residual or applied. As with corrosion fatigue, SCC is a gradual process and can take prolonged periods of time for both crack initiation and propagation. Besides the fact that static, as opposed to dynamic, stress is involved, another important difference between SCC and corrosion fatigue is that SCC only occurs for a restricted number of material/ environment combinations. Interestingly, the environments that cause SCC for given materials are not necessarily those that are chemically most aggressive. For example, certain stainless steels are susceptible to SCC in chloridecontaining environments. Corrosion fatigue, on the other hand, can occur (to a greater or lesser degree) for all materials in all electrolytes. Though it is a rare occurrence, SCC can occur in vivo (Bundy and Zardiackas, 2006). Most of the reports of in vivo SCC have been for stainless steel devices.

3.5

Conclusion

At this point it is worthwhile to recapitulate the main points associated with implant corrosion processes, as well as to consider what future developments may be in the offing in this field.

3.5.1

Summary

As we have seen, the chemical environment of the body is quite complex. This creates a challenging and aggressive environment as far as corrosion of metallic biomaterials is concerned. Many specific corrosion mechanisms have been found to be operative under in vivo conditions for artificial hip and other total joint replacements, as well as other sorts of orthopaedic surgical implants. Some of these mechanisms involve interactions between electrochemical attack and applied mechanical stresses in load-bearing implant devices. Corrosion of biomaterials is a surface phenomenon occurring at the metal/electrolyte interface. Besides aspects of the chemical environment of the body previously mentioned, many features and characteristics of the surface of the implant alloy itself play dominant roles in the corrosion of the material. Though engineering failures per se can result from corrosion of implants, adverse reactions of the tissues and fluids that contact the implant to various released corrosion products are the © 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

77

most common causes for the need to remove implants from patients. The field of biocompatibility is the branch of science that studies these biological reactions. Many types of experimental methods have been used to study the complex process of corrosion of metallic surgical implant biomaterials. Over the past century and more of experience with metallic devices in vivo, much knowledge has been amassed regarding corrosion of implanted materials. In turn, this knowledge has been applied toward the development of improved, more corrosion-resistant and biocompatible materials. We can certainly expect that this process will continue into the future. The next section examines likely future directions in this regard.

3.5.2

Future trends

Based upon the past history of the development of corrosion-resistant alloys and of the field of biomaterials as a whole, a number of trends appear likely to lead to even more improved implant materials over time. These developments will stem from advances in biomaterials science and input from other fields, based upon increased understanding in various branches of science pertinent to surgical implants, and their application to the field of biomedical materials. The trend in development of cleaner alloys and ones with modified chemical compositions, which has been occurring over time (particularly in recent years), can be expected to continue. This will eventually be reflected in even more rigorous standards for implant alloys and lead to materials that are even more corrosion resistant than the alloys used at present. Further improvements in coating technology, more effectively using diamond coatings and ion implantation, for example, and surface topography may also lead to enhanced resistance to both corrosion and mechanical deterioration of implant materials. Advanced fields of materials science as a whole, such as nanotechnology, smart materials, and self-repairing materials, can be expected to have salutary effects on the field of biomaterials in general and corrosion resistance of implant alloys in particular. Increasing use of non-metallic implants will also diminish problems due to implant corrosion, but might lead to different kinds of biocompatibility problems, as other means of implant deterioration become more common. Better mechanical design may serve to further lower the stresses that develop in implants, which will lead to a lessened frequency of fretting corrosion, corrosion fatigue, and stress corrosion cracking. This may come about due to better theoretical modeling because of increasing knowledge of the biomechanics pertinent to artificial joint applications or to designs more specifically tailored to the needs of individual patients, or both. Increased understanding of the field of biocompatibility almost certainly will lead to even more biocompatible implants. Here, too, an approach aimed not only at the population level but also directed toward biocompatibility at the individual © 2008, Woodhead Publishing Limited

78

Joint replacement technology

patient level could well serve to improve the reliability of implanted devices. Improvements in technology over the present state of the art will be needed to achieve this end, but can be expected to become available ultimately. For example, practical means for measuring corrosion rates in vivo can be expected to eventually make monitoring of the corrosion of implants in individual patients routine. As well, biosensor technology could be used to monitor the build up of metallic products in various body fluids of individual patients, in order to spot potential future biocompatibility problems while they are in a nascent and manageable stage (Bundy, 2004). Ultimately, indwelling sensors could be coupled with telemetry to allow virtually continuous monitoring. Engineering of surfaces for enhanced biocompatibility and using tissue engineering and other types of approaches may also help to produce improved metallic implants, with fewer instances of corrosion-provoked biocompatibility problems. These could include surfaces that could topically deliver antibacterial or anti-inflammatory agents, surfaces with intentionally designed protein or cell coatings, and surfaces with growth factors or other macromolecules attached for constructive stimulation of nearby cells. As well, surfaces with chelating agents on them might be able to tie up metallic elements that are released and so lessen the possibility that they would build up in body fluids and tissues in harmful forms. Although this is much more speculative, in the far future, it is even possible that implant corrosion could be intentionally used as a constructive treatment for deficiencies of trace certain trace elements. In any case, the realization of any and all of these improvements will require that the latest advances in corrosion science, materials science, surface science, and biocompatibility be translated into practical applications for enhancing the performance of medical devices made from surgical implant alloys.

3.6

Sources of further information and advice

Baboian R, Ed. (2005), Corrosion Tests and Standards: Application and Interpretation, 2nd Edn, West Conshohocken, PA, ASTM Int. Bockris J O'M and Reddy A K N (1970), Modern Electrochemistry, Vol. 2, New York, Plenum Press. Bundy K J (1994), `Corrosion and other electrochemical aspects of biomaterials', Crit Rev Biomed Eng 22(3/4), 139±251. Bundy K J and Zardiackas L D (2006), `Corrosion fatigue and stress corrosion cracking in metallic biomaterials,' in Cramer S D and Covino, Jr B S, Eds, ASM Handbook, Vol. 13C, Corrosion: Environments and Industries, Materials Park, OH, ASM International, 853±890. Fontana M G (1986), Corrosion Engineering, 3rd Edn, New York, McGraw-Hill. Gileadi E, Kirowa-Eisner E, and Penciner J (1975), Interfacial Electrochemistry ± An Experimental Approach, Reading, MA, Addison-Wesley. Park J B (1984), Biomaterials Science and Engineering, New York, Plenum Press. Pourbaix M (1973), Lectures on Electrochemical Corrosion, New York, Plenum Press. Scully J C (1975), The Fundamentals of Corrosion, 2nd Edn, Oxford, Pergamon Press. © 2008, Woodhead Publishing Limited

Biomaterials and the chemical environment of the body

79

Sedriks A J (1979), Corrosion of Stainless Steels, New York, Wiley. Williams D F (1981), `Electrochemical aspects of corrosion in the physiological environment', Chapter 2 in Williams D F, Ed., Fundamental Aspects of Biocompatibility, Vol. 1, Boca Raton, CRC Press, 11±42. Williams D F and Williams R L (2004), `Degradative effects of the biological environment on metals and ceramics', Chapter 6.3 in Ratner B D, Hoffman A S, Schoen F J, and Lemons J E, Eds., Biomaterial Science: An Introduction to Materials in Medicine, 2nd Edn, Amsterdam, Elsevier Academic Press, 430±439.

3.7

References

Bard A J and Faulkner L R (1980), Electrochemical Methods ± Fundamentals and Applications, New York, John Wiley. Black J (1992), Biological Performance of Materials ± Fundamentals of Biocompatibility, 2nd Edn, Chapter 2, `Introduction to the biological environment' and Chapter 4, `Corrosion and dissolution', New York, Marcel Dekker. Bobyn J D, Stackpool G J, Hacking, S A, Tanzer M and Krygier J J (1999), `Characteristics of bone ingrowth and interface mechanics of a new porous tantalum biomaterial', J Bone Joint Surg [Br], 81-B, 907±914. Bockris J O'M and Reddy A K N (1970), Modern Electrochemistry, Vol. 2, New York, Plenum Press. Brown S A, Eng D, Flemming C A C, Kawalec J S, Placko H E, Vassaux C, Merritt K, Payer J H and Kraay M J (1995), `Fretting corrosion accelerates crevice corrosion of modular hip tapers', J Appl Biomater, 6(1), 19±26. Bundy K J (1994), `Corrosion and other electrochemical aspects of biomaterials', Crit Rev Biomed Eng, 22(3/4), 139±251. Bundy K J (2004), `Relationships between biomaterials and biosensors', Chapter 23 in Yaszemski M J, Trantolo D J, Lewandrowski K-U, Hasirci V, Altobelli D E, and Wise D L, Eds, Tissue Engineering and Novel Delivery Systems, New York, Marcel Dekker, 505±520. Bundy K J and Luedemann R (1989), `Factors which influence the accuracy of corrosion rate determination of implant materials', Ann Biomed Eng, 17, 159±175. Bundy K J and Zardiackas, L D (2006), `Corrosion fatigue and stress corrosion cracking in metallic biomaterials', in Cramer S D, and Covino, Jr B S, Eds, ASM Handbook, Vol. 13C, Corrosion: Environments and Industries, Materials Park, OH, ASM International, 853±890. Bundy K J, Dillard J and Luedemann R (1993), `The use of AC impedance methods to study the corrosion behavior of implant alloys', Biomaterials, 14, 529±536. Callister Jr W D (2003), Materials Science ± An Introduction, 6th Edn, New York, John Wiley. Chalkin C and Minter J (2005), `Limb salvage and abductor reattachment using a custom prosthesis with porous tantalum components', J Arthroplasty, 20(1), 127±130. Collier J P, Surprenant V A, Jensen R E and Mayor M B (1991), `Corrosion at the interface of cobalt-alloy heads on titanium-alloy stems', Clin Orthop, 271, 305± 312. Coury A J (2004), `Chemical and biochemical degradation of polymers', Chapter 6.2 in Ratner B D, Hoffman A S, Schoen F J, and Lemons J E, Eds, Biomaterial Science: An Introduction to Materials in Medicine, 2nd Edn, Amsterdam, Elsevier Academic Press, 411±430.

© 2008, Woodhead Publishing Limited

80

Joint replacement technology

Denniston K J, Topping J J and Caret R L (2004), General, Organic and Biochemistry, 4th Edn, New York, McGraw-Hill, 564. Findlay D M, Welldon K, Atkins G J, Howie D W, Zannettino A C and Bobyn D (2004), `The proliferation and phenotypic expression of human osteoblasts on tantalum metal', Biomaterials, 25(12), 2215±2227. Fontana M G and Greene N D (1978), Corrosion Engineering, 2nd Edn, New York, McGraw-Hill. Fraker A C (2005), `Medical and dental', Chapter 79 in Baboian R., Ed., Corrosion Tests and Standards: Application and Interpretation, 2nd Edn, West Conshohocken, PA, ASTM Int., 834±845. Fukada E and Yasuda I (1957), `On the piezo-electric effect in bone', J Physical Soc Jap, 12(10), 1158±1162. Gilbert J L, Buckley C A and Jacobs, J J (1993), `In vivo corrosion of modular hip prosthesis components in mixed and similar metal combinations. The effect of crevice, stress, motion, and alloy coupling', J Biomed Mater Res, 27(12), 1533± 1544. Guzelsu N and Walsh W R (1990), `Streaming potential of intact wet bone', J Biomech, 23(7), 673±685. Hallab N J, Bundy K J, O'Connor K, Clark R and Moses R L (1995) `Cell adhesion to biomaterials: correlations between surface charge, surface roughness, adsorbed protein, and cell morphology', J Long Term Eff Med Implants, 5(3), 209±231. Israelachvili J N and McGuiggan P M (1988), `Forces between surfaces in liquids', Science, 241(4867), 795±800. Laing P G (1977), `Tissue reaction to biomaterials', in NBS Special Publication 472, Gaithersburg, MD, 31. Lin H-Y and Bumgardner J D (2004a), `In vitro biocorrosion of Ti±6Al±4V implant alloy by a mouse macrophage line', J Biomed Mater Res Part A, 68A(4), 717±724. Lin H-Y and Bumgardner J D (2004b), `In vitro biocorrosion of Co±Cr±Mo by macrophage cells', J Orthop Res, 22(6), 1231±1236. Lucas L C, Buchanan R A, and Lemons J E (1981), `Investigations on the galvanic corrosion of multialloy hip prostheses', J Biomed Mater Res, 15(5), 731±747. Macheras G A, Papagelopoulos P J, Kateros K, Kostakos A T, Baltas D and Karachalios T S (2006), `Radiological evaluation of the metal±bone interface of a porous tantalum monoblock acetabular component', J Bone Joint Surg [Br], 88-B(3), 304± 309. Messer R L W, Mickalonis J, Adams Y, and Tseng W Y (2006), `Corrosion rates of stainless steel under shear stress measured by a novel parallel-plate flow chamber', J Biomed Mater Res Part B, 76B(2), 273±280. Mu Y, Kobayashi K, Sumita M, Yamamoto A, and Hanawa T (2000), `Metal ion release from titanium with active oxygen species generated by rat macrophages in vitro', J Biomed Mater Res, 49(2), 238±243. Orten L M, and Neuhaus O W (1975), Human Biochemistry, 9th Edn, C. V. Mosby. Vetter K J (1967), Electrochemical Kinetics ± Theoretical Aspects, New York, Academic Press.

© 2008, Woodhead Publishing Limited

4

Materials for joint replacement

K S K A T T I , D V E R M A and D R K A T T I , North Dakota State University, USA

4.1

Introduction

The materials and devices used in orthopedic applications are designed to sustain the load bearing function of human bones for the duration of the patient's life. Orthopedic applications include numerous products for the rehabilitation and reconstruction required as a result of various diseases of the musculoskeletal system as well as aging. The worldwide market for materials used for orthopedic applications is estimated to be $14 billion in 2002. Also, about $12 billion is spent on joint replacements (Hallab et al., 2004). This chapter provides an overview of the current and future materials used for joint replacement. Key physical and mechanical properties are discussed in addition to mechanics, degradation, and biocompatibility issues associated with specific materials. New advances in the use of novel nanocomposite systems and natural materials are also discussed.

4.2

Materials criteria for total joint replacement

The primary function of orthopedic materials is to bear load and provide structural integrity to the human body. Table 4.1 shows the mechanical properties of bones. Structural integrity implies a combination of fracture toughness, strength, ductility, and hardness, and also time-dependent properties and fatigue resistance. In addition, the human body provides a fairly corrosive environment (Ratner et al., 2004) and thus biocompatibility and corrosion resistance are also important requirements of these materials. In addition, the deterioration products of the orthopedic materials such as from a joint replacement implant should not adversely affect the bodily environment. Thus, the combination of biocompatibility requirements over and above the mechanical properties expected under the corrosive environment of the human body makes for very stringent requirements on the materials design of orthopedic materials. Wear resistance and corrosion resistance properties of various biomaterials are shown in Tables 4.2 and 4.3. © 2008, Woodhead Publishing Limited

82

Joint replacement technology Table 4.1 Mechanical properties of bones (adapted from Black, 1992; Currey, 1984; Fratzl et al., 1998; Katti, 2004) Tissue

Tibia Femur Radius Humerus Cervical Lumbar

Compressive strength (MPa)

Tensile strength (MPa)

Elastic modulus (GPa)

159 167 114 132 10 5

140 121 149 130 3.1 3.7

18.1 17.2 18.6 17.2 0.23 0.16

Table 4.2 Wear rate of materials used in orthopedics Materials

Wear rate (mm3/million cycles)

UHMWPE/zirconia (n ˆ 3) Cobalt chrome/cobalt chrome (n ˆ 3) Alumina/alumina (n ˆ 3) Alumina/UHMPE Alumina/crosslinked UHMWPE CoCrMo/CoCr/Mo

References

31  4.0 1.23  0.5

Tipper et al. (2001) Tipper et al. (2001)

0.05  0.02 51  11 5.62  3.5

Tipper et al. (2001) Essner et al. (2005) Essner et al. (2005)

6.30  10.3

Essner et al. (2005)

* UHMWPE ˆ ultra-high moleculat weight polyethylene.

Table 4.3 Electrochemical properties of implant metals in 0.1 M NaCl at pH 7. Higher corrosion potential, lower passive current density, and higher breakdown voltage represent better corrosion resistance (adapted from Ratner et al., 2004) Alloy

Stainless steel CoCrMo CPTi Ti±6Al±4V Ti±5Al±2.5Fe Ni±45Ti

© 2008, Woodhead Publishing Limited

Corrosion potential (mV)

Passive current density (mA/cm2)

Breakdown potential (mV)

ÿ400 ÿ390 ÿ90 to ÿ630 ÿ180 to ÿ510 ÿ530 ÿ430

0.56 1.36 0.72±9.0 0.9±2.0 0.68 0.44

200±770 420 >2000 >1500 >1500 890

Materials for joint replacement

4.3

83

History of materials used in joint replacement

One of the earliest treatments to address painful hip joints consisted of simply removing the acetabular and femoral diseased bone. These procedures were attempted as early as the 1820s. During the 1830s to 1880s, wooden blocks and animal soft tissues were also placed between the acetabular and femoral components in order to ease pain in the hip joints. A prosthetic replacement for hip joints was first attempted in 1890 using a carved ivory structure to replace the femoral head with the use of plaster of Paris and pumice-based bone cement. Prosthetic replacement of femoral heads using ivory and rubber was extensively popularized in the late 1800s and early 1900s. Using various materials such as wood, gold foil, and animal soft tissues as an interpositional membrane continued until the early 1900s. These procedures were not very successful in relieving pain, and a quest for a solution to the replacement of joints continued. One of the earliest attempts at replacement of hip joints with synthetic materials was only as recent as 1925 wherein mold arthroplasty was attempted. This method used a molded piece of glass in the shape of a hollow hemisphere that fitted over the ball of the hip joint. This attempt was made by Dr SmithPeterson, a surgeon at Massachusetts General Hospital in Boston. The primary reason for the failure of this device was the poor mechanical performance of glass. Several attempts at using other materials, with similar biocompatibility properties as glass but superior mechanical properties such as stainless steel, were fabricated into hollow hemispheres. In the quest for better materials that were more suitable, the next breakthrough in biomaterials was in 1936 with the fabrication of cobalt±chromium alloys. Many attempts were made in using these new alloys in mold arthroplasty but these devices did not adequately satisfy the need to cure a variety of painful deformities of the hip resulting from arthritis and other conditions. The next major breakthrough was the type of hip replacement called hemiarthroplasty which consisted of replacing the entire ball of the hip but not the socket. This procedure of hip replacement consisted of a long metal stem placed in the femur connected to a metal ball that sat in the hip socket. This was the state of the art in the 1950s. This procedure often resulted in loosening of the implant. New bone cement fixation techniques were also pursued around the same time. One of the pioneers of total hip replacement was the surgeon Dr John Charnley who first attempted replacement of the diseased hip socket. He used Teflon and polyethylene. In the late 1950s Dr Charnley performed many successful hip replacement surgeries, which resulted in his eventual knighting by Queen Elizabeth II. He used a steel femoral component and a plastic socket cup. The use of replacement of both the hip socket and femoral heads has since been extensively popularized and a variety of materials such as polymers, metals, composites, and biological materials are being used by surgeons and many more studied by researchers in the quest for a replacement of © 2008, Woodhead Publishing Limited

84

Joint replacement technology

a hip joint. The primary advantage of this method is the ability to design anatomically sized femoral heads, stems, and sockets. Attempts to replace the knee joint were also being made concurrently with those of the hip joint. The earliest attempts consisted of hinges that were fixed to the bones through the hollow bone marrow cavity. In the 1950s metal spacers were placed between the knee bones to prevent the rubbing of bone against bone. Later in the 1960s, Frank Gunston, an orthopedist at the Sir John Charnley Hip Center, designed a metal on polymer knee joint that was attached using a bone cement. But the first total knee replacement was attempted by Dr John Insall in New York in 1972, which consisted of replacement for surfaces of all three surfaces of the knee, the femur, tibia, and kneecap. This method remains a prototype for current knee replacement methodologies.

4.4

Traditional materials

4.4.1

Metals

Metallic implants are the primary biomaterials used for joint replacement and becoming increasingly important. The metallic implants used for orthopedic applications can be categorized as stainless steel, CoCr alloys, and Ti and Ti alloys. These metallic materials have several properties such as high strength, high fracture toughness, hardness, corrosion resistance and biocompatibility, which make them an excellent choice for total joint replacement. The disadvantage with metallic implants is their high elastic modulus, which causes stress shielding. Toxic effects caused by ions released from metallic implants are also a major concern. Stainless steel alloys were the first metals to be used for orthopedics. Stainless steel alloys contain carbon, chromium, nickel, molybdenum, and manganese, phosphorus, sulfur, and silicon as trace elements. These components affect the mechanical properties of steel by alteration of its microstructure. A high nickel content (10±14%) in stainless steel can cause toxicity. This has prompted research in the development of Ni-free stainless steel alloys. Cobalt-based alloys are the other metallic implants used for joint replacement. CoCrMo and CoNiCrMo are the two main cobalt based alloys generally used in orthopedics. Especially for joint replacement where low frictional resistance is desired, CoCrMo alloys are preferred over CoNiCrMo alloys. Commercially pure Ti (CPTi) and Ti±6Al±4V are two dominant Ti-based materials used in joint replacement. Ti-based implants have excellent corrosion resistance and biocompatibility. And the credit goes to the oxide layer, which spontaneously forms in the presence of oxygen. It has been shown that Ti-based alloys promote osteoblast activities. Therefore for uncemented joint replacement, Ti-based alloys are preferred over other metallic implants. One drawback with Ti alloys is that they are relatively softer than stainless steel alloys and cobalt-based alloys. This makes them more susceptible to wear where articulation is required. © 2008, Woodhead Publishing Limited

Materials for joint replacement

85

Zirconium is classified as a refractory metal because of its high temperature resistance properties. It also provides high chemical resistance because of the formation of a highly stable oxide layer, which is about 5 m in thickness. Several zirconium-based alloys have been developed for use in joint prostheses. One such alloy is Oxinium. Oxinium possesses excellent wear resistance (approximately 10 times that of Co and Ti-based alloys). Also, Zr-based implants have the advantage of both ceramic and metallic implants. Their metallic core provides high fracture toughness and the oxide layer provides excellent wear resistance and biocompatibility. The only disadvantage associated with these alloys is the high cost of forming and machining. Implants interact with the body through their surfaces. Further, wear and corrosion are initiated at the surface. Surface characteristics of an implant decide its fate in the body. Therefore, for proper surface response, several strategies have been proposed recently (Lappalainen and Santavirta, 2005). Surface response depends on both surface topography and chemical composition. Surface topography can be modified by grit or sand blasting or plasma treatments. Large varieties of coatings have been used, such as hydroxyapatite (Capello et al., 1998), titanium oxide and nitride (Teresa Raimondi and Pietrabissa, 2000), zirconium oxide (Patel and Spector, 1997) and diamond-like carbon coatings (Affatato et al., 2000; Lappalainen et al., 1998) to improve surface characteristics. Several studies have also discussed coating implants with growth factor (Cole et al., 1997; Lind et al., 2000), collagen (Roehlecke et al., 2001), RGD peptides (De Giglio et al., 2000), and fibronectin (Degasne et al., 1999). Moreover, implants are also being coated with osteoblast cells (Frosch et al., 2003). Table 4.4 shows the mechanical properties of various metals and alloys used in orthopedic applications.

4.4.2

Ceramics

Ceramic materials possess several useful properties, which make them excellent materials for orthopedic implants. They exhibit high stiffness, inert behavior under physiological environment, and superior wear resistance as compared with metallic and polymeric bearing surfaces. One limiting property of ceramic materials is their brittleness. Since the mechanical properties of ceramic materials are highly dependent on their density, small voids left in the implant during processing severely affect their longevity. Alumina was the first ceramic material used for joint replacement (Boutin, 1972). Table 4.5 shows the mechanical properties of various ceramics used in orthopedic applications. The wear rate of alumina is reported to be 20 times lower than that of ultra-high molecular weight polyethylene (UHMWPE). Femoral heads for hip replacements and wear plates in knee replacements have been fabricated using alumina. One of the concerns with alumina implants was its low fracture toughness, which was overcome later by increasing purity, lowering porosity, grain size and © 2008, Woodhead Publishing Limited

86

Joint replacement technology Table 4.4 Mechanical properties of alloys in total joint replacement (Davidson and Georgette, 1987; Davis, 2003; Katti, 2004; Long and Rack, 1998) Material

Ti±Zr Ti±6Al±4V Ti±6Al±7Nb Ti±5Al±2.5Fe Ti±3Al±2.5V Ti±13Nb±13Zr Ti±15Mo±5Zr±3Al Ti±12Mo±6Zr±2Fe Ti±15Mo±2.8Nb±3Al Ti±35Nb±5Ta±7Zr (TNZT) Ti±15Mo±2.8Nb±0.2Si±0.3O Ti±35Nb±5Ta±7Zr±0.4O Ti±15Mo Ti±16Nb±10Hf CPTi (>>98% Ti) Co±Cr±Mo Co±Cr alloys Stainless steel 316L

Tensile strength (MPa)

Modulus (MPa)

900 960±970 1024 1033 690 1030 882±975 1060±1100 812 590 1020 1010 795 851 785 600±1795 655±1896 465±950

± 110 105 110 100 79 75 74±85 82 55 82 66 78 81 105 200±230 210±253 200

improving manufacturing techniques (Boehler et al., 2000; Fritsch and Gleitz, 1996). In hip replacements, alumina is also used as the femoral head with a metallic femoral stem and UHMWPE as an acetabular cup opposing articulating surface. In February 2003, the United States Food and Drug Administration Table 4.5 Mechanical properties of ceramic materials used in orthopedics (Davis, 2003; Katti, 2004; Ramakrishna et al., 2001; Schmitt, 1985) Ceramic

Zirconia Alumina Bioglass C (graphite) C (vitreous) C (low-temperature isotropic carbon (LTI) pyrolytic) C (silicon alloyed LTI) C ultra-low temperature isotropic carbon (ULTI) Hydroxyapatite Apatite-Wollastonite (AW) glass ceramic

© 2008, Woodhead Publishing Limited

Compressive strength (MPa)

Modulus (GPa)

2000 4000 1000 138 172 900

220 380 75 25 31 28

± ±

28±41 14±21

600 1080

117 118

Materials for joint replacement

87

(FDA) approved the first ceramic-on-ceramic articulated hip implant for marketing in the United States. Later, several studies focused on other materials as an alternate for alumina. The first paper on zirconia as a potential material for application in orthopedics was published in 1969 (Helmer and Driskell, 1969) and the first publication on design of zirconia ball heads for total hip replacement was reported in 1988 (Christel et al., 1988). Because of its better mechanical properties over alumina, zirconia has attracted considerable research interest. Zirconia femoral heads have been gaining market share, and more than 300 000 zirconia femoral heads have already been implanted (Chevalier et al., 1997; Hench and Wilson, 1993). One problem with zirconia implants is low-temperature degradation (Yoshimura et al., 1987). Zirconia ceramics are polycrystals of tetragonal phase, stabilized by yttria, and commonly referred to as yttria-stabilized±tetragonal-zirconia polycrystals (Y-TZP). The Y-TZP material slowly undergoes phase transformation to monoclinic form at room temperature, which is accompanied by deterioration in its mechanical properties. Composites of ceria-stabilized± zirconia-tetragonal polycrystal (C-TZP) with alumina polycrystals have shown improved resistance to low-temperature degradation (Tanaka et al., 2002, 2003). However, this composite showed no improvement in bone bonding ability. Recent studies have shown that bioactivity of these composites can be improved by surface chemical treatments (Takemoto et al., 2005; Uchida et al., 2002). Zirconia-toughened alumina prostheses have also shown superior properties over currently used alumina implants for hip replacement (Insley and Streicher, 2004; Wang and Stevens, 1989). Also, calcium phosphate materials and bioglass ceramics are being investigated as alternatives to poly(methylmethacrylate) (PMMA) for bone cement applications. Their osteophilic characteristics make them excellent candidates for orthopedic applications. A better match between the bulk material properties of the implant and the bone it replaces can decrease some of the problems such as stress shielding currently associated with metallic implants. This is often achieved by coating the metallic implants with bioactive materials such as hydroxyapatite (HAP), tricalcium phosphate (TCP), and bioglass. Tricalcium phosphate (TCP) (Ca3(PO4)2) and HAP (Ca10(PO4)6(OH)2) are both biocompatible materials and have the ability to bond directly to bone. Several researchers have attempted to develop high-strength consolidated HAP bodies (Bagambisa et al., 1993; Hench and Wilson, 1993). Bending strength as high as 90 MPa has been achieved by colloidal processing of HAP (Hench and Wilson, 1993). However, these materials still suffer from poor mechanical properties such as low strength and limited fatigue resistance, which limit their application as load bearing biomaterials. Mechanical properties of ceramic biomaterials are shown in Table 4.5. Several methods have been developed to coat implants with hydroxyapatite and other calcium phosphate materials. Among them, thermal spraying has produced most promising results © 2008, Woodhead Publishing Limited

88

Joint replacement technology

(Chu, 2007; Heimann, 2006). Simply soaking an implant in simulated body fluid to coat it with apatite has also been investigated (Li, 2003). Initially, the formation of amorphous calcium phosphate layer at the interface of implant was causing coating failure (Ji et al., 1992; Park and Condrate, 1999; Park et al., 1998). However, using a `bond coat' on implant surface or annealing the implant after coating beyond 900 ëC improved adhesion of coating significantly (De Groot et al., 1987; Gross et al., 1998). Improvement in adhesion due to annealing occurs with the formation of a several micrometer thick layer of Ca± Ti±oxide at the interface. The important parameters that control the success of a coating are composition and crystallinity (Fazan and Marquis, 2000). Crystallinity also affects the dissolution of the apatite. Although dissolution of apatite is necessary for bonding of coating with the surrounding bone tissues, dissolution in excessive amounts may cause inflammation due to changes in local pH (Chou et al., 1999; LeGeros et al., 1991).

4.4.3

Polymeric materials

Several polymers have been used for orthopedic applications such as acrylic, nylon, silicone, polyurethane, UHMWPE, and polypropylene (PP) (Davidson and Georgette, 1987). Mechanical properties of these polymers are shown in Table 4.6. UHMWPE is one of the most preferred polymers as an orthopedic implant because of its high mechanical strength, low wear rate, and biocompatibility (Costa and Brach del Prever, 2000; Kelly, 2002). Although UHMWPE has been used for over 30 years, osteolysis caused by wear debris is still a concern (Goldring et al., 1983; Sinha et al., 1998; Willert and Semlitsch, 1977). Several studies have been conducted to understand the wear mechanism and also the osteolysis caused by the wear debris (Ingham and Fisher, 2005; Ren et al., 2006; von Knoch et al., 2005; Wedemeyer et al., 2007). Osteolysis is the Table 4.6 Mechanical properties of polymeric materials in orthopedics (Katti, 2004; Ramakrishna et al., 2001; Schmitt, 1985) Polymer

Tensile strength (MPa)

UHMWPE Polyacetal Polysulfone Polyurethane Silicone tubber Polyetheretherketones (PEEK) Polytetrafluoroethane (PTFE) Polyethylene terephthalate (PET) Poly(methylmethacrylate) (PMMA)

21 67 75 35 7.6 139 28 61 21

© 2008, Woodhead Publishing Limited

Modulus (GPa) 1 2.1 2.67 0.02 0.008 8.3 0.4 2.85 4.5

Materials for joint replacement

89

resorption of bone surrounding the implant, which occurs in association with the formation of vascularized granuloma at the interface of implant and the bone (Granchi et al., 2005). The formation of granuloma is a body response to clean up the wear particles. The response to wear particles is dealt with in detail by Revell in Chapter 15. Since the main problems associated with the use of UHMWPE as acetabular cups is not the degradation in mechanical properties of cups but weakening of the interfacial adhesion between tissue and implant (due to wear debris) considerable efforts have been made to improve the wear resistance of UHMWPE (Affatato et al., 2005; Bell et al., 2001; McEwen et al., 2005). It has been observed that increasing crystallinity and crosslinking density improve wear resistance of UHMWPE (Endo et al., 2001). The crosslinking of UHMWPE is usually achieved by exposing implant to irradiation. Although, crosslinking improves wear resistance, but at the same time also degrade tensile strength, fracture toughness and fatigue crack propagation resistance (Baker et al., 2003; Gomoll et al., 2002). Increasing crystallinity of UHMWPE also improves its wear resistance, elastic modulus and resistance to crack propagation (Champion et al., 1994). Recently, it has been suggested that crosslinking of UHMWPE in combination with higher crystallization can improve wear resistance and fatigue fracture resistance (Simis et al., 2006). One other reason for failure of UHMWPE in implants is its oxidation during the sterilization process (Fisher et al., 2004; Premnath et al., 1996). Oxidation of UHMWPE is minimized by sterilizing the implant in an inert atmosphere and adopting other sterilization procedures such as gas plasma and ethylene oxide (Kurtz et al., 1999), but exposure to high-intensity radiation causes formation of free radicals in the crystalline phase of UHMWPE. These free radicals react with dissolved oxygen and cause oxidative embrittlement and subsequently hamper the mechanical properties of the implant. To prevent oxidative embrittlement of UHMWPE, addition of vitamin E is suggested as an antioxidant (Parth et al., 2002; Reno and Cannas, 2006; Tomita et al., 1999). Polyurethanes are another class of polymers that has been considered for joint replacement materials. Recent studies have shown that polyurethane provides a lower coefficient of friction than UHMWPE bearings (Quigley et al., 2002). There are different types of polyurethane and they are identified from the type of linkages they have, such as polyesterurethanes (Coury et al., 1984; Mandarino and Salvatore, 1960) which incorporate ester linkages, polyetherurethanes (PEUs) (Lamba et al., 1998; Zdrahala, 1996) which incorporate ether moieties, and polycarbonateurethanes (PCUs) which incorporate carbonate linkages (Hoffman et al., 1993; Lemm, 1984). Polyesterethane and polyetherurethane are prone to hydrolytic degradation. So for biomedical applications such as joint replacement where long-term stability is required, polycarbourethanes are being investigated (Gunatillake et al., 2003; Khan et al., 2005). Other materials which have been studied as an artificial cartilage are water swollen hydrogels (Oka et © 2008, Woodhead Publishing Limited

90

Joint replacement technology

al., 2000), poly(vinyl alcohol) cryogel (PVA-c) (Szkowski et al., 2006), hyaluronan esters (Zhang and James, 2005) and multilayer polyelectrolyte films (Pavoor et al., 2006). Multilayer polyelectrolyte films showed 33% reduction in wear compared with UHMWPE bearings.

4.5

Bone cement materials

The primary functions of bone cement are to secure the orthopedic implants to bone and transfer mechanical loads from the implant to the bone. The femoral stem and acetabular cups are cemented, screwed or press fit into place. Approximately 50% of all orthopedic implants utilize bone cement to achieve implant fixation. PMMA is the most commonly used bone cement. PMMAbased bone cements are mainly two-part formulations. The first part contains pre-polymerized PMMA, an initiator and a radiopacifier. The second part contains mostly liquid MMA, an accelerator and a stabilizer to prevent premature polymerization. These bone cements have shown high success rate, averaging 90% after 15 years (Murray et al., 1995; Nafei et al., 1996). Seven main drawbacks associated with PMMA-based bone cements have been identified (Lewis, 1997), as follows: 1. Local tissue damage, which occurs due to the exothermic nature of the cement setting reaction (Liu et al., 1987). The temperature goes up as high as from 60 to 120 ëC depending on formulation of the cement (Kindt-Larsen et al., 1995; Wang et al., 1995). 2. The release of the unreacted MMA, which causes chemical necrosis of the bone (Kindt-Larsen et al., 1995). 3. The high shrinkage of the cement after polymerization which is about 21% (Thompson et al., 1979). 4. The stiffness mismatch between bone and the cement. 5. The cement does not bond chemically with either of bone and implant and acts in such as way as to bring about `weak link zones' (Bragdon et al., 1995; Harrigan et al., 1992). 6. Cement particle mediated osteolysis of the bone. 7. Bacterial infection is also associated with bone cements. Several studies have focused on solving the problems associated with PMMA-based cements outlined above. Partial replacement of MMA with 2,2bis [4(2-hydroxy-3-methacryloxypropoxy) phenyl] propane caused significant improvement in volume shrinkage (Vallo and Schroeder, 2005). Efforts have also been made to improve the mechanical and biological properties of PMMAbased cements. Several studies have investigated the effect of additives such as carbon (Friis et al., 1996), graphite (Knoell et al., 1975), aramid (Pourdeyhimi et al., 1986), titanium (Timmie Topoleski et al., 1992), UHMWPE (Gilbert et al., 1994; Pourdeyhimi and Wagner, 1989). To improve the bioactivity of PMMA© 2008, Woodhead Publishing Limited

Materials for joint replacement

91

based cements, additives such as hydroxyapatite have also been included (Kim et al., 2004; Serbetci et al., 2004; Vallo et al., 1999). Bacterial infection issues have been addressed by incorporation of antibiotic within the bone cements. Some studies also suggested use of combination of antibiotics. Bone cements are also dealt with in detail in Chapter 9.

4.6

Composite materials and new nanocomposite systems

In spite of tremendous success achieved by currently used bone implants for joint replacement, there is still need for development of materials that are more biocompatible and last longer in the body. Ceramic polymer composites have shown more superior properties than either of their components used as total hip replacement materials (Wang et al., 1996). Specifically, HAP-based polymer composites have received significant research interest. Table 4.7 shows the mechanical properties of composites used in orthopedic applications. HAP is a natural component of bone and is thus highly biocompatible, with superior bondforming ability. Hence several studies have been conducted on polymer composite where HAP is used as the ceramic filler component (Boanini et al., 2006; Boduch-Lee et al., 2004; Causa et al., 2006; Higashi et al., 1986; Kikuchi et al., 1997; Yoshida et al., 2006; Zhitomirsky and Pang, 2006). Use of nanosized HAP particles and various techniques for modifying HAP±polymer interfaces have been explored (Sinha et al., 2007; Verma et al., 2007). Specially, the HAP particles having high aspect ratio (whisker or fiber) significantly improves the modulus with a lower loading wt%. Thus, several attempts have been made to synthesize whisker-like HAP particles (Converse et al., 2007; Roeder et al., 2003; Viswanath and Ravishankar, 2005; Yue and Roeder, 2006; Zhang et al., 2002). One study showed ultimate strength, elastic modulus and elongation at break of composite based on poly( -hydroxyalkanoates) (PHA) with HAP similar to similar to bone and is being investigated as a potential material for total hip replacement (Galego et al., 2000). Carbon fibers have good biocompatibility and excellent mechanical properties. They have been used to reinforce ultra high molecular weight polyethylene in total hip replacement components. The composites of carbon fiber with PMMA (Woo et al., 1974), polypropylene and polysulphone (Christel et al., 1980; Claes et al., 1997), polyethylene, polybutylene terephthalate, and PEEK (Gillett et al., 1986; Jockisch et al., 1992; Rushton and Rae, 1984) have all been investigated for potential applications in load-bearing applications. Multilayered laminated composites of carbon fibers with PMMA and PEEK have also been investigated (Fujihara et al., 2003; Sorrell et al., 2000). HAP is a bioactive material. Several studies have focused on HAP-containing composites to improve bioactivity and mechanical properties of composites for orthopedic applications: metal and ceramic fiber reinforcement of HAP (Ehsani et al., 1995; © 2008, Woodhead Publishing Limited

92

Joint replacement technology

Table 4.7 Mechanical properties of composites for orthopedics Composites

Strength (MPa)

Modulus (GPa)

Reference

Poorly crystalline carbonateapatite + tetracalcium phosphate + collagen

6.08±11 (tensile)

0.66±2.24

Du et al. (2000)

Direct mineralized collagen composite (0±39% calcium phosphate)

34±53 (tensile)

0.44±2.82

Wahl and Czernuszka (2006)

Decalcified bone composite (10±15% calcium phosphate)

44.87 (tensile)

0.68

Wahl and Czernuszka (2006)

PHB/HAP (30% HAP)

67

2.52

Galego et al. (2000)

Polyacrylic acid/HAP (40±70% HAP)

20±60 (compressive)

1±1.8

Katti et al. (2008)

UHMWPE±collagen± hydroxyapatite (23±40% HAP)

11.0±17.0 (tensile)

0.11±0.17

163 (compressive)

2.06

Verma et al. (2007)

Chitosan/hydroxyapatite composite (50% HAP)

74.08

1.02

Verma et al. (2007)

Chitosan/hydroxyapatite (70% HAP)

120 (compressive)

Zhang et al. (2007)

Self-hardening chitosan/ hydroxyapatite

26.2 0.88±4.29 (compressive)

Lu et al. (2007)

Chitosan±polygalacturonic acid-hydroxyapatite (50% HAP)

Chemically coupled PE/HAP 18.34±20.67 Biphasic calcium phosphate/polylactic acid Polylactic acid/HAP

30±60

Roy Chowdhury et al. (2007)

Wang and Bonfield (2001) 0.296±2.48

Bleach et al. (2002)

0.66±2.24

Ignjatovic et al. (1999a)

Ruys et al., 1991), HAP/polyethylene (Bonfield, 1993; Wang et al., 1994), HAP/ polyethyl ester (Liu et al., 1997), HAP/polyphosphasone (Reed et al., 1996), HAP/polylactide (Ignjatovic et al., 1999a) and HAP/alumina composites (Li et al., 1995) have all been described. To improve the mechanical properties and bone-bonding properties of PMMA bone cements, the composites of PMMA with HAP and bioglass have been investigated. The addition of these materials showed enhancement of © 2008, Woodhead Publishing Limited

Materials for joint replacement

93

osteogenic properties of the implants as well as improvement in mechanical properties (Moursi et al., 2002; Vallo et al., 1999). In general the polymer/HAP interfaces are known to have an important role on the resulting mechanical properties (Ignjatovic et al., 1999b). Interfaces play an important role in deciding the overall mechanical properties of a composite. A weak interface may lead to a deterioration of the mechanical properties of the composite. Some studies have shown that chemical modification of an interface results in improvement in mechanical properties of composites. For example chemically coupled HAP±polyethylene composites (Wang and Bonfield, 2001), chemically formed HAP±Ca poly(vinyl phosphonate) composites (Greish and Brown, 2001) and polylactic acid HAP fiber composites (Kasuga et al., 2001) have shown improvement in mechanical properties.

4.7

Natural materials

A new class of materials has been developed which have a similar structure or composition to bone or are synthesized by following basic principles of biomineralization. These are called biomimetic materials. A key step in composite synthesis is the growth of minerals on an organic matrix in aqueous media (Du et al., 2000). Several polymers of both synthetic and natural origin have been used for synthesis of biomimetic HAP/polymer composites (Bakos et al., 1999; Bigi et al., 1998; Itoh et al., 2002; Katti et al., 2008; Teng et al., 2006; Wan et al., 2006; Zhao et al., 2002). This strategy has shown significant potential for development of materials for bone substitution. Collagen and calcium phosphate minerals, being the natural component of bone, are the natural choices for the development of these composites (Lawson and Czernuszka, 1998). Collagen constitutes 20% of bone and provides toughness. Collagen is also known for its bone formation ability. Several recent studies have focused on the development of collagen/hydroxyapatite composites (Wahl and Czernuszka, 2006). An additional advantage with collagen/hydroxyapatite composite is that it can easily be remodeled by the body (Du et al., 1998). The mechanical properties of these composites lie between those of cancellous and cortical bone (Clarke et al., 1993; Mathers and Czernuszka, 1991). A recent study indicated comparable wear resistance of hydroxyapatite±collagen±hyaluronic acid with UHMWPE (Roy Chowdhury et al., 2007). Chitosan-based composites prepared by biomimetic methods have also gathered significant research attention for possible application in load-bearing applications (Lu et al., 2007; Zhang et al., 2007). Chitosan, a polysaccharide, is biocompatible, biodegradable and exhibits antigenic properties. Incorporation of polyanionic polymer to chitosan-based composites has shown improvement in mechanical responses (Li et al., 2007; Rodrigues et al., 2003; Verma et al., 2007; Zhang et al., 2004).

© 2008, Woodhead Publishing Limited

94

4.8

Joint replacement technology

Summary

Consistent with the current advancements in materials science and engineering, especially in the realm of advanced materials design, the medical community has significantly benefited by the applications of many advanced materials and composites for orthopedic applications, especially for joint replacements. This chapter describes an overview of the various different polymeric, ceramic, metallic, composite, and natural-based materials used for joint replacement applications. This review also describes current and new advancements in composites research with the use of nano-reinforcements for use for implant applications. Mechanical properties of these materials and their merits and demerits for implant applications are also described here. The current applications of materials for orthopedic applications have relied heavily on experimental research. Many testing-based experimental studies have been the basis for the advancement in the field until recently. Only recently with the availability of fast, large and expansive parallel computing capabilities is materials design starting to be aided by computational simulations. Simulationbased design is certainly the way of the future in advanced materials research and it is to be expected that medical research will benefit from such advancements in simulations of tissue±biomaterial interactions over the lifetime of implants at length scales ranging from the molecular to the macroscopic.

4.9

Acknowledgments

This work is supported in part with a grant from National Science Foundation (NSF CAREER # 0132768). D Verma would like to acknowledge support from North Dakota State University, Graduate School doctoral dissertation award.

4.10

References

Affatato, S., Frigo, M., and Toni, A. 2000, `An in vitro investigation of diamond-like carbon as a femoral head coating', Journal of Biomedical Materials Research, vol. 53, no. 3, pp. 221±226. Affatato, S., Bersaglia, G., Rocchi, M., Taddei, P., Fagnano, C., and Toni, A. 2005, `Wear behaviour of cross-linked polyethylene assessed in vitro under severe conditions', Biomaterials, vol. 26, no. 16, pp. 3259±3267. Bagambisa, F. B., Joos, U., and Schilli, W. 1993, `Mechanisms and structure of the bond between bone and hydroxyapatite ceramics', Journal of Biomedical Materials Research, vol. 27, no. 8, pp. 1047±1055. Baker, D. A., Bellare, A., and Pruitt, L. 2003, `The effects of degree of crosslinking on the fatigue crack initiation and propagation resistance of orthopedic-grade polyethylene', Journal of Biomedical Materials Research ± Part A, vol. 66, no. 1, pp. 146±154. Bakos, D., SoldaÂn, M., and HernaÂndez-Fuentes, I. 1999, `Hydroxyapatite±collagen± hyaluronic acid composite', Biomaterials, vol. 20, no. 2, pp. 191±195.

© 2008, Woodhead Publishing Limited

Materials for joint replacement

95

Bell, J., Tipper, J. L., Ingham, E., Stone, M. H., and Fisher, J. 2001, `The influence of phospholipid concentration in protein-containing lubricants on the wear of ultrahigh molecular weight polyethylene in artificial hip joints', Proceedings of the Institution of Mechanical Engineers [H.], vol. 215, no. 2, pp. 259±263. Bigi, A., Panzavolta, S., and Roveri, N., 1998, ` Hydroxyapatite±gelatin films: a structural and mechanical characterization', Biomaterials, vol. 19, nos 7±9, pp. 739±744. Black, J. 1992, Biological Performance of Materials: Fundamentals of Biocompatibility. CRC Press. Bleach, N. C., Nazhat, S. N., Tanner, K. E., Kellomaki, M., and Tormala, P. 2002, `Effect of filler content on mechanical and dynamic mechanical properties of particulate biphasic calcium phosphate±polylactide composites', Biomaterials, vol. 23, no. 7, pp. 1579±1585. Boanini, E., Torricelli, P., Gazzano, M., Giardino R., and Bigi, A. 2006, `Nanocomposites of hydroxyapatite with aspartic acid and glutamic acid and their interaction with osteoblast-like cells', Biomaterials, vol. 27, no. 25, pp. 4428±4433 Boduch-Lee, K. A., Chapman, T., Petricca, S. E., Marra, K. G., and Kumta, P. 2004, `Design and synthesis of hydroxyapatite composites containing an mPEG-dendritic poly(Llysine) star polycaprolactone', Macromolecules, vol. 37, no. 24, pp. 8959±8966. Boehler, M., Plenk, H., and Salzer, M. 2000, `Alumina ceramic bearings for hip endoprostheses: the Austrian experiences', Clinical Orthopaedics and Related Research, no. 379, pp. 85±93. Bonfield, W. 1993, `Design of bioactive ceramic±polymer composites', An Introduction to Bioceramics, vol. 1, pp. 299±303. Boutin, P. 1972, `Total arthroplasty of the hip by fritted aluminum prosthesis. Experimental study and 1st clinical applications', Revue de Chirurgie Orthopedique et Reparatrice de l'Appareil Moteur, vol. 58, no. 3, pp. 229±246. Bragdon, C. R., Biggs, S., Mulroy, W. F., Kawate, K., and Harris, W. H. 1995, `Defects in the cement mantle: a fatal flaw in cemented stems for THR', Transactions of the 21st Annual Meeting of the Society of Biomaterials, p. 314. Capello, W. N., D'Antonio, J. A., Manley, M. T., and Feinberg, J. R. 1998, `Hydroxyapatite in total hip arthroplasty: Clinical results and critical issues', Clinical Orthopaedics and Related Research no. 355, pp. 200±211. Causa, F., Netti, P. A., Ambrosio, L., Ciapetti, G., Baldini, N., Pagani, S., Martini, D., and Giunti, A. 2006, `Poly--caprolactone/hydroxyapatite composites for bone regeneration: in vitro characterization and human osteoblast response', Journal of Biomedical Materials Research ± Part A, vol. 76, no. 1, pp. 151±162. Champion, A. R., Li, S., Saum, K., Howard, E., and Simmons, W. 1994, `The effect of crystallinity on the physical properties of UHMWPE', Transactions of the Orthopedic Research Socoety, vol. 19, pp. 585±589. Chevalier, J., Drouin, J. M., and Cales, B. 1997, `Low temperature ageing behaviour of zirconia hip joint heads', Bioceramics, vol. 10, pp. 135±138. Chou, L., Marek, B., and Wagner, W. R. 1999, `Effects of hydroxylapatite coating crystallinity on biosolubility, cell attachment efficiency and proliferation in vitro', Biomaterials, vol. 20, no. 10, pp. 977±985. Christel, P., Meunier, A., Dorlot, J. M., Crolet, J. M., Witvoet, J., Sedel, L., and Boutin, P. 1988, `Biomechanical compatibility and design of ceramic implants for orthopedic surgery', Annals of the New York Academy of Sciences, vol. 523, pp. 234±256. Christel, P. S., Leray, I. L., Sedal, L., and Morel, E. 1980, Mechanical Evaluation and Tissue Compatibility of Materials for Composite Bone Plates in Mechanical Properties of Biomaterials, Wiley, New York, pp. 367±377. © 2008, Woodhead Publishing Limited

96

Joint replacement technology

Chu, P. K. 2007, `Plasma-treated biomaterials', IEEE Transactions on Plasma Science, vol. 35, no. 2 I, pp. 181±187. Claes, L., Hutter, W., and Weiss, R. 1997, `Mechanical properties of carbon fiber reinforced polysulfone plates for internal fixation', Biological and Biomechanical Performance of Biomaterials, pp. 81±86. Clarke, K. I., Graves, S. E., Wong, A. T. C., Triffitt, J. T., Francis, M. J. O., and Czernuszka, J. T. 1993, `Investigation into the formation and mechanical properties of a bioactive material based on collagen and calcium phosphate', Journal of Materials Science: Materials in Medicine, vol. 4, no. 2, pp. 107±110. Cole, B. J., Bostrom, M. P. G., Pritchard, T. L., Sumner, D. R., Tomin, E., Lane, J. M., and Weiland, A. J. 1997, `Use of bone morphogenetic protein 2 on ectopic porous coated implants in the rat', Clinical Orthopaedics and Related Research, no. 345, pp. 219±228. Converse, G. L., Yue, W., and Roeder, R. K. 2007, `Processing and tensile properties of hydroxyapatite-whisker-reinforced polyetheretherketone', Biomaterials, vol. 28, no. 6, pp. 927±935. Costa, L. and Brach del Prever, E. M. 2000, UHMWPE for arthroplasty. Edizioni Minerva Medica, Turin. Coury, A. J., Cobian, K. E., Cahalan, P. T., and Jevne, A. H. 1984, `Biomedical uses of polyurethanes', Advances in Urethane Science and Technology, vol. 9, pp. 130±168. Currey, J. D. 1984, The Mechanical Adaptations of Bones. Princeton University Press. Davidson, J. A. and Georgette, F. S. 1987, `State-of-the-art materials for orthopaedic prosthetic devices: On implant manufacturing and material technology', Proceedings of the Society of Manufacturing Engineering. EM87-122, 122126. Davis, J. R. 2003, Handbook of Materials for Medical Devices. ASM International. De Giglio, E., Sabbatini, L., Colucci, S., and Zambonin, G. 2000, `Synthesis, analytical characterization, and osteoblast adhesion properties on RGD-grafted polypyrrole coatings on titanium substrates', Journal of Biomaterials Science, Polymer Edition, vol. 11, no. 10, pp. 1073±1083. De Groot, K., Geesink, R., Klein, C. P. A. T., and Serekian, P. 1987, `Plasma sprayed coatings of hydroxylapatite', Journal of Biomedical Materials Research, vol. 21, no. 12, pp. 1375±1381. Degasne, I., Basle, M. F., Demais, V., Hure, G., Lesourd, M., Grolleau, B., Mercier, L., and Chappard, D. 1999, `Effects of roughness, fibronectin and vitronectin on attachment, spreading, and proliferation of human osteoblast-like cells (Saos-2) on titanium surfaces', Calcified Tissue International, vol. 64, no. 6, pp. 499±507. Du, C., Cui, F. Z., Feng, Q. L., Zhu, X. D., and De Groot, K. 1998, `Tissue response to nano-hydroxyapatite/collagen composite implants in marrow cavity', Journal of Biomedical Materials Research, vol. 42, no. 4, pp. 540±548. Du, C., Cui, F. Z., Zhang, W., Feng, Q. L., Zhu, X. D., and De Groot, K. 2000, `Formation of calcium phosphate/collagen composites through mineralization of collagen matrix', Journal of Biomedical Materials Research, vol. 50, no. 4, pp. 518±527. Ehsani, N., Ruys, A. J., and Sorrell, C. C. 1995, `Thixotropic casting of FeCr alloy fibrereinforced hydroxyapatite', Key Engineering Materials, vol. 104±107, no. 1, pp. 373±380. Endo, M. M., Barbour, P. S., Barton, D. C., Fisher, J., Tipper, J. L., Ingham, E., and Stone, M. H. 2001, `Comparative wear and wear debris under three different counterface conditions of crosslinked and non-crosslinked ultra high molecular weight polyethylene', Biomedical Materials Engineering, vol. 11, no. 1, pp. 23±35. © 2008, Woodhead Publishing Limited

Materials for joint replacement

97

Essner, A., Sutton, K., and Wang, A. 2005, `Hip simulator wear comparison of metal-onmetal, ceramic-on-ceramic and crosslinked UHMWPE bearings', Wear, vol. 259, no. 7±12, pp. 992±995. Fazan, F. and Marquis, P. M. 2000, `Dissolution behaviour of plasma-sprayed hydroxyapatite coating', Journal of Materials Science: Materials in Medicine, vol. 11, no. 12, pp. 787±792. Fisher, J., McEwen, H. M. J., Barnett, P. I., Bell, C., Stone, M. H., and Ingham, E. 2004, `Influence of sterilizing techniques on polythene wear', The Knee, vol. 11, no. 3, pp. 173±176. Fratzl, P., Misof, K., Zizak, I., Rapp, G., Amenitsch, H., and Bernstorff, S. 1998, `Fibrillar structure and mechanical properties of collagen', Journal of Structural Biology, vol. 122, no. 1±2, pp. 119±122. Friis, E. A., Kumar, B., Cooke, F. W., and Yasuda, H. K. 1996, `Fracture toughness of surface treated carbon fiber reinforced compositer bone cement', Transactions of the Annual Meeting of the Society for Biomaterials in conjunction with the International Biomaterials Symposium, vol. 1, p. 913. Fritsch, E. W. and Gleitz, M. 1996, `Ceramic femoral head fractures in total hip arthroplasty', Clinical Orthopaedics and Related Research, no. 328, pp. 129±136. Frosch, K. H., Sondergeld, I., Dresing, K., Rudy, T., Lohmann, C. H., Rabba, J., Schild, D., Breme, J., and Stuermer, K. M. 2003, `Autologous osteoblasts enhance osseointegration of porous titanium implants', Journal of Orthopaedic Research, vol. 21, no. 2, pp. 213±223. Fujihara, K., Huang, Z. M., Ramakrishna, S., Satknanantham, K., and Hamada, H. 2003, `Performance study of braided carbon/PEEK composite compression bone plates', Biomaterials, vol. 24, no. 15, pp. 2661±2667. Galego, N., Rozsa, C., Sanchez, R., Fung, J., Analia Vazquez, A., and Santo, T. 2000, `Characterization and application of poly( -hydroxyalkanoates) family as composite biomaterials', Polymer Testing, vol. 19, no. 5, pp. 485±492. Gilbert, J. L., Ney, D. S., and Lautenschlager, E. P. 1994, `Self-reinforced PMMA composites: strength, fatigue and fracture toughness evaluation', Transactions of the 20th Annual Meeting of the Society for Biomaterials, p. 141. Gillett, N., Brown, S. A., Dumbleton, J. H., and Pool, R. P. 1986, `The use of short carbon fibre reinforced thermoplastic plates for fracture fixation', Biomaterials, vol. 6, p. 113. Goldring, S. R., Schiller, A. L., and Roelke, M. 1983, `The synovial-like membrane at the bone-cement interference in loose total hip replacements and its proposed role in bone lysis', Journal of Bone and Joint Surgery ± Series A, vol. 65, no. 5, pp. 575± 584. Gomoll, A., Wanich, T., and Bellare, A. 2002, `J-integral fracture toughness and tearing modulus measurement of radiation cross-linked UHMWPE', Journal of Orthopaedic Research, vol. 20, no. 6, pp. 1152±1156. Granchi, D., Amato, I., Battistelli, L., Ciapetti, G., Pagani, S., Avnet, S., Baldini, N., and Giunti, A. 2005, `Molecular basis of osteoclastogenesis induced by osteoblasts exposed to wear particles', Biomaterials, vol. 26, no. 15, pp. 2371±2379. Greish, Y. E. and Brown, P. W. 2001, `Chemically formed HAp-Ca poly(vinyl phosphonate) composites', Biomaterials, vol. 22, no. 8, pp. 807±816. Gross, K. A., Gross, V., and Berndt, C. C. 1998, `Thermal analysis of amorphous phases in hydroxyapatite coatings', Journal of the American Ceramic Society, vol. 81, no. 1, pp. 106±112. Gunatillake, P. A., Martin, D. J., Meijs, G. F., McCarthy, S. J., and Adhikari, R. 2003, © 2008, Woodhead Publishing Limited

98

Joint replacement technology

`Designing biostable polyurethane elastomers for biomedical implants', Australian Journal of Chemistry, vol. 56, no. 6, pp. 545±557. Hallab, N. J., Jacobs, J. J., Katz, J. L. 2004, `Chapter 7.7 Orthopedic Applications', in Biomaterials Science: An Introduction to Materials in Medicine, Elsevier Academic Press. Harrigan, T. P., Kareh, J. A., O'Connor, D. O., Burke, D. W., and Harris, W. H. 1992, `A finite element study of the initiation of failure of fixation in cemented femoral total hip components', Journal of Orthopaedic Research, vol. 10, no. 1, pp. 134±144. Heimann, R. B. 2006, `Thermal spraying of biomaterials', Surface and Coatings Technology, vol. 201, no. 5, pp. 2012±2019. Helmer, J. D. and Driskell, T. D. 1969, `Research on bioceramics', Proceedings of the Symposium on Use of Ceramics as Surgical Implants. South Carolina, Clemson University, USA. Hench, L. L. and Wilson, J. 1993, An Introduction to Bioceramics, World Science Publishing Company, Singapore. Higashi, S., Yamamuro, T., and Nakamura, T. 1986, `Polymer±hydroxyapatite composites for biodegradable bone fillers', Biomaterials, vol. 7, no. 3, pp. 183±187. Hoffman, D., Gong, G., Pinchuk, L., and Sisto, D. 1993, `Safety and intracardiac function of a silicone±polyurethane elastomer designed for vascular use', Clinical Materials, vol. 13, no. 1±4, pp. 95±100. Ignjatovic, N. L., Plavsic, M., Miljkovic, M. S., Zivkovics, L. M., and Uskokovic, D. P. 1999a, Journal of Microscopy, vol. 196, p. 23. Ignjatovic, N., Tomic, S., Dakic, M., Miljkovic, M., Plavsic, M., and Uskokovic, D. 1999b, `Synthesis and properties of hydroxyapatite/poly-L-lactide composite biomaterials', Biomaterials, vol. 20, no. 9, pp. 809±816. Ingham, E. and Fisher, J. 2005, `The role of macrophages in osteolysis of total joint replacement', Biomaterials, vol. 26, no. 11, pp. 1271±1286. Insley, G. M. and Streicher, R. M. 2004, `Next generation ceramics based on zirconia toughened alumina for hip joint prostheses', Key Engineering Materials, vol. 254± 256, pp. 675±678. Itoh, S., Kikuchi, M., Koyama, Y., Takakuda, K., Shinomiya, K., and Tanaka, 2002, `Development of an artificial vertebral body using a novel biomaterial, hydroxyapatite/collagen composite', Biomaterials, vol. 23, no. 19, pp. 3919±3926. Ji, H., Ponton, C. B., and Marquis, P. M. 1992, `Microstructural characterization of hydroxyapatite coating on titanium', Journal of Materials Science: Materials in Medicine, vol. 3, no. 4, pp. 283±287. Jockisch, K. A., Brown, S. A., Bauer, T. W., and Merritt, K. 1992, `Biological response to chopped-carbon-fibre-reinforced peek', Journal of Biomedical Research, vol. 26, pp. 133146. Kasuga, T., Ota, Y., Nogami, M., and Yoshihiro, A. 2001, `Preparation and mechanical properties of polylactic acid composites containing hydroxyapatite fibers', Biomaterials, vol. 22, p. 1923. Katti, K. S. 2004, `Biomaterials in total joint replacement', Colloids and Surfaces B: Biointerfaces, vol. 39, no. 3, pp. 133±142. Katti, K. S., Turlapati, P. K., Verma, D., Bhowmik, R., Gujjula, P. K., and Katti, D. R. 2008, `Static and dynamic mechanical behavior of hydroxyapatite-polyacrylic acid composites under simulated body fluid', American Journal of Biochemistry and Biotechnology, vol. 2, no. 2, pp. 73±79. Kelly, J. M. 2002, `Ultra-high molecular weight polyethylene', Journal of Macromolecular Science ± Polymer Reviews, vol. 42, no. 3, pp. 355±371. © 2008, Woodhead Publishing Limited

Materials for joint replacement

99

Khan, I., Smith, N., Jones, E., Finch, D. S., and Cameron, R. E. 2005, `Analysis and evaluation of a biomedical polycarbonate urethane tested in an in vitro study and an ovine arthroplasty model. Part I: Materials selection and evaluation', Biomaterials, vol. 26, no. 6, pp. 621±631. Kikuchi, M., Suetsugu, Y., Tanaka, J., and Akao, M. 1997, `Preparation and mechanical properties of calcium phosphate/copoly-L-lactide composites', Journal of Materials Science: Materials in Medicine, vol. 8, no. 6, pp. 361±364. Kim, S. B., Kim, Y. J., Yoon, T. L., Park, S. A., Cho, I. H., Kim, E. J., Kim, I. A., and Shin, J. W. 2004, `The characteristics of a hydroxyapatite±chitosan±PMMA bone cement', Biomaterials, vol. 25, no. 26, pp. 5715±5723. Kindt-Larsen, T., Smith, D. B., and Jensen, J. S. 1995, `Innovations in acrylic bone cement and application equipment', Journal of Applied Biomaterials, vol. 6, no. 1, pp. 75±83. Knoell, A., Maxwell, H., and Bechtol, C. 1975, `Graphite fiber reinforced bone cement ± an experimental feasibility investigation', Annals of Biomedical Engineering, vol. 3, no. 2, pp. 225±229. Kurtz, S. M., Muratoglu, O. K., Evans, M., and Edidin, A. A. 1999, `Advances in the processing, sterilization, and crosslinking of ultra-high molecular weight polyethylene for total joint arthroplasty', Biomaterials, vol. 20, no. 18, pp. 1659±1688. Lamba, N. M. K., Woodhouse, K. A., and Cooper, S. L. 1998, Polyurethanes in Biomedical Applications. CRC Press, New York. Lappalainen, R. and Santavirta, S. S. 2005, `Potential of coatings in total hip replacement', Clinical Orthopaedics and Related Research, no. 430, pp. 72±79. Lappalainen, R., Anttila, A., and Heinonen, H. 1998, `Diamond coated total hip replacements', Clinical Orthopaedics and Related Research, no. 352, pp. 118±127. Lawson, A. C. and Czernuszka, J. T. 1998, `Collagen±calcium phosphate composites', Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine, vol. 212, no. 6, pp. 413±425. LeGeros, R. Z., Orly, I., Gregoire, M., and Daculsi, G. 1991, The Bone±Biomaterial Interface, University of Toronto Press, Toronto, pp. 76±88. Lemm, W. 1984, `Biodegradation of polyurethanes', Polyurethanes in Biomedical Engineering, pp. 103±108. Lewis, G. 1997, `Properties of acrylic bone cement: state of the art review', Journal of Biomedical Materials Research, vol. 38, no. 2, pp. 155±182. Li, J., Fartash, B., and Hermansson, L. 1995, `Hydroxyapatite±alumina composites and bone-bonding', Biomaterials, vol. 16, no. 5, pp. 417±422. Li, P. 2003, `Biomimetic nano-apatite coating capable of promoting bone ingrowth', Journal of Biomedical Materials Research ± Part A, vol. 66, no. 1, pp. 79±85. Li, Q. L., Chen, Z. Q., Darvell, B. W., Zeng, Q., Li, G., Ou, G. M., and Wu, M. Y. 2007, `Biomimetic synthesis of PEC-HA composite analogous to bone', Key Engineering Materials, vol. 336±338 II, pp. 1699±1702. Lind, M., Overgaard, S., Song, Y., Goodman, S. B., Bunger, C., and Soballe, K. 2000, `Osteogenic protein 1 device stimulates bone healing to hydroxyapaptite-coated and titanium implants', Journal of Arthroplasty, vol. 15, no. 3, pp. 339±346. Liu, Q., De Wijn, J. R., and Van Blitterswijk, C. A. 1997, `Nano-apatite/polymer composites: mechanical and physicochemical characteristics', Biomaterials, vol. 18, no. 19, pp. 1263±1270. Liu, Y. K., Park, J. B., Njus, G. O., and Stienstra, D. 1987, `Bone-particle-impregnated bone cement: an in vitro study', Journal of Biomedical Materials Research, vol. 21, no. 2, pp. 247±261. © 2008, Woodhead Publishing Limited

100

Joint replacement technology

Long, M. and Rack, H. J. 1998, `Titanium alloys in total joint replacement ± a materials science perspective', Biomaterials, vol. 19, no. 18, pp. 1621±1639. Lu, X., Zheng, B., Chen, N., and Yuan, H. 2007, `Preparation and evaluation of selfhardening bone-rehabilitative composite with natural hydroxyapatite/chitosan', Key Engineering Materials, vol. 334±335 II, pp. 1197±1200. Mandarino, M. P. and Salvatore, J. E. 1960, `A polyurethane polymer (Ostumer): its use in fractured and diseased bones', AMA Arch Surg, vol. 80, pp. 623±627. Mathers, N. J. and Czernuszka, J. T. 1991, `Growth of hydroxyapatite on type I collagen', Journal of Materials Science Letters, vol. 10, no. 17, pp. 992±993. McEwen, H. M., Barnett, P. I., Bell, C. J., Farrar, R., Auger, D. D., Stone, M. H., and Fisher, J. 2005, `The influence of design, materials and kinematics on the in vitro wear of total knee replacements', Journal of Biomechanics, vol. 38, no. 2, pp. 357±365. Moursi, A. M., Winnard, A. V., Winnard, P. L., Lannutti, J. J., and Seghi, R. R. 2002, `Enhanced osteoblast response to a polymethylmethacrylate-hydroxyapatite composite', Biomaterials, vol. 23, no. 1, pp. 133±144. Murray, D. W., Carr, A. J., and Bulstrode, C. J. 1995, `Which primary total hip replacement?' Journal of Bone and Joint Surgery ± Series B, vol. 77, no. 4, pp. 520±527. Nafei, A., Kristensen, O., Knudsen, H. M., Hvid, I., and Jensen, J. 1996, `Survivorship analysis of cemented total condylar knee arthroplasty: a long-term follow-up report on 348 cases', Journal of Arthroplasty, vol. 11, no. 1, pp. 7±10. Oka, M., Ushio, K., Kumar, P., Ikeuchi, K., Hyon, S. H., Nakamura, T., and Fujita, H. 2000, `Development of artificial articular cartilage', Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine, vol. 214, no. 1, pp. 59±68. Park, E. and Condrate, S. 1999, `Graded coating of hydroxyapatite and titanium by atmospheric plasma spraying', Materials Letters, vol. 40, no. 5, pp. 228±234. Park, E., Condrate, S., Hoelzer, D. T., and Fischman, G. S. 1998, `Interfacial characterization of plasma-spray coated calcium phosphate on Ti±6Al±4V', Journal of Materials Science: Materials in Medicine, vol. 9, no. 11, pp. 643±649. Parth, M., Aust, N., and Lederer, K. 2002, `Studies on the effect of electron beam radiation on the molecular structure of ultra-high molecular weight polyethylene under the influence of -tocopherol with respect to its application in medical implants', Journal of Materials Science: Materials in Medicine, vol. 13, no. 10, pp. 917±921. Patel, A. M. and Spector, M. 1997, `Tribological evaluation of oxidized zirconium using an articular cartilage counterface: a novel material for potential use in hemiarthroplasty', Biomaterials, vol. 18, no. 5, pp. 441±447. Pavoor, P. V., Gearing, B. P., Muratoglu, O., Cohen, R. E., and Bellare, A. 2006, `Wear reduction of orthopaedic bearing surfaces using polyelectrolyte multilayer nanocoatings', Biomaterials, vol. 27, no. 8, pp. 1527±1533. Pourdeyhimi, B. and Wagner, H. D. 1989, `Elastic and ultimate properties of acrylic bone cement reinforced with ultra-high-molecular-weight polyethylene fibers', Journal of Biomedical Materials Research, vol. 23, no. 1, pp. 63±80. Pourdeyhimi, B., Wagner, H. D., and Schwartz, P. 1986, `A comparison of mechanical properties of discontinuous Kevlar 29 fibre reinforced bone and dental cements', Journal of Materials Science, vol. 21, no. 12, pp. 4468±4474. Premnath, V., Harris, W. H., Jasty, M., and Merrill, E. W. 1996, `Gamma sterilization of UHMWPE articular implants: an analysis of the oxidation problem', Biomaterials, vol. 17, no. 18, pp. 1741±1753. © 2008, Woodhead Publishing Limited

Materials for joint replacement

101

Quigley, F. P., Buggy, M., and Birkinshaw, C. 2002, `Selection of elastomeric materials for complaint-layered total hip arthtopology', Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine, vol. 216, no. 1, pp. 77±83. Ramakrishna, S., Mayer, J., Wintermantel, E., and Leong, K. W. 2001, `Biomedical applications of polymer-composite materials: a review', Composites Science and Technology, vol. 61, no. 9, pp. 1189±1224. Ratner, B. D., Hoffman, A. S., Schoen, F.J., Lemons A. E. 2004, Biomaterials Science: An Introduction to Materials in Medicine, Elsevier Academic Press. Reed, C. S., TenHuisen, K. S., Brown, P. W., and Allcock, H. R. 1996, `Thermal stability and compressive strength of calcium-deficient hydroxyapatite-poly[bis(carboxylatophenoxy)phosphazene] composites', Chemistry of Materials, vol. 8, no. 2, pp. 440±447. Ren, W. P., Markel, D. C., Zhang, R., Peng, X., Wu, B., Monica, H., and Wooley, P. H. 2006, `Association between UHMWPE particle-induced inflammatory osteoclastogenesis and expression of RANKL, VEGF, and Flt-1 in vivo', Biomaterials, vol. 27, no. 30, pp. 5161±5169. Reno, F. and Cannas, M. 2006, `UHMWPE and vitamin E bioactivity: an emerging perspective', Biomaterials, vol. 27, no. 16, pp. 3039±3043. Rodrigues, C. V. M., Serricella, P., Linhares, A. B. R., Guerdes, R. M., Borojevic, R. , Rossi, M. A., Duarte, M. E. L., and Farina, M. 2003, `Characterization of a bovine collagenhydroxyapatite composite scaffold for bone tissue engineering', Biomaterials, vol. 24, no. 27, pp. 4987±4997. Roeder, R. K., Sproul, M. M., and Turner, C. H. 2003, `Hydroxyapatite whiskers provide improved mechanical properties in reinforced polymer composites', Journal of Biomedical Materials Research ± Part A, vol. 67, no. 3, pp. 801±812. Roehlecke, C., Witt, M., Kasper, M., Schulze, E., Wolf, C., Hofer, A., and Funk, R. H. W. 2001, `Synergistic effect of titanium alloy and collagen type I on cell adhesion, proliferation and differentiation of osteoblast-like cells', Cells Tissues Organs, vol. 168, no. 3, pp. 178±187. Roy Chowdhury, S. K., Kulkarni, A. C., Basak, A., and Roy, S. K. 2007, `Wear characteristic and biocompatibility of some hydroxyapatite±collagen composite acetabular cups', Wear, vol. 262, no. 11±12, pp. 1387±1398. Rushton, N. and Rae, T. 1984, `The intra-articular response to particulate carbon fibre reinforced high density polyethylene and its constituents: an experimental study in mice', Biomaterials, vol. 5, no. 6, pp. 352±356. Ruys, A. J., Ziegler, K. A., Brandwood, A., Milthorpe, B. K., Morrey, S., and Sorrell, C. C. 1991, `Reinforcement of hydroxyapatite with ceramic and metal fibres', Bioceramics, vol. 4, pp. 281±286. Schmitt, F. O. 1985, `Adventures in molecular biology', Ann. Rev. Biophys. Biophys. Chem., vol. 4, pp. 122. Serbetci, K., Korkusuz, F., and Hasirci, N. 2004, `Thermal and mechanical properties of hydroxyapatite impregnated acrylic bone cements', Polymer Testing, vol. 23, no. 2, pp. 145±155. Simis, K. S., Bistolfi, A., Bellare, A., and Pruitt, L. A. 2006, `The combined effects of crosslinking and high crystallinity on the microstructural and mechanical properties of ultra high molecular weight polyethylene', Biomaterials, vol. 27, no. 9, pp. 1688±1694. Sinha, A., Das, G., Sharma, B. K., Roy, R. P., Pramanick, A. K., and Nayar, S. 2007, `Poly(vinyl alcohol)±hydroxyapatite biomimetic scaffold for tissue regeneration', © 2008, Woodhead Publishing Limited

102

Joint replacement technology

Materials Science and Engineering: C, vol 27, no. 1, pp. 70±74. Sinha, R. K., Shanbhag, A. S., Maloney, W. J., Hasselman, C. T., and Rubash, H. E. 1998, `Osteolysis: cause and effect', Instructional Course Lectures, vol. 47, pp. 307±320. Sorrell, C. C., Marques, A. T., and Jeronimidis, G. 2000, `Design of a controlled-stiffness composite proximal femoral prosthesis', Composites Science and Technology, vol. 60, no. 4, pp. 559±567. Szkowski, W., Ku, D. N., Bersee, H. E. N., and Kurzydlowski, K. J. 2006, `An elastic material for cartilage replacement in an arthritic shoulder joint', Biomaterials, vol. 27, no. 8, pp. 1534±1541. Takemoto, M., Fujibayashi, S., Suzuki, J., Kokubo, T., and Nakamura, T. 2005, `Bonebonding ability of Ce-TZP/Al2O3 nanocomposite with a microporous surface and calcium phosphate coating', Key Engineering Materials, vol. 284±286, pp. 987± 990. Tanaka, K., Tamura, J., Kawanabe, K., Nawa, M., Oka, M., Uchida, M., Kokubo, T., and Nakamura, T. 2002, `Ce-TZP/Al2O3 nanocomposite as a bearing material in total joint replacement', Journal of Biomedical Materials Research, vol. 63, no. 3, pp. 262±270. Tanaka, K., Tamura, J., Kawanabe, K., Nawa, M., Uchida, M., Kokubo, T., and Nakamura, T. 2003, `Phase stability after aging and its influence on pin-on-disk wear properties of Ce-TZP/Al2O3 nanocomposite and conventional Y-TZP', Journal of Biomedical Materials Research ± Part A, vol. 67, no. 1, pp. 200±207. Teng, S., Shi, J., Peng, B., and Chen, L. 2006, `The effect of alginate addition on the structure and morphology of hydroxyapatite/gelatin nanocomposites', Composites Science and Technology, vol. 66, no. 11±12, pp. 1532±1538. Teresa Raimondi, M. and Pietrabissa, R. 2000, `The in-vivo wear performance of prosthetic femoral heads with titanium nitride coating', Biomaterials, vol. 21, no. 9, pp. 907±913. Thompson, V. P., Williams, E. F., and Bailey, W. J. 1979, `Dental resins with reduced shrinkage during hardening', Journal of Dental Research, vol. 58, no. 5, pp. 1522± 1534. Timmie Topoleski, L. D., Ducheyne, P., and Cuckler, J. M. 1992, `The fracture toughness of titanium-fiber-reinforced bone cement', Journal of Biomedical Materials Research, vol. 26, no. 12, pp. 1599±1617. Tipper, J. L., Firkins, P. J., Besong, A. A., Barbour, P. S. M., Nevelos, J., Stone, M. H., Ingham, E., and Fisher, J. 2001, `Characterisation of wear debris from UHMWPE on zirconia ceramic, metal-on-metal and alumina ceramic-on-ceramic hip prostheses generated in a physiological anatomical hip joint simulator', Wear, vol. 250±251, no. PART 1, pp. 120±128. Tomita, N., Kitakura, T., Onmori, N., Ikada, Y., and Aoyama, E. 1999, `Prevention of fatigue cracks in ultrahigh molecular weight polyethylene joint components by the addition of vitamin E', Journal of Biomedical Materials Research, vol. 48, no. 4, pp. 474±478. Uchida, M., Kim, H. M., Kokubo, T., Nawa, M., Asano, T., Tanaka, K., and Nakamura, T. 2002, `Apatite-forming ability of a zirconia/alumina nano-composite induced by chemical treatment', Journal of Biomedical Materials Research, vol. 60, no. 2, pp. 277±282. Vallo, C. I. and Schroeder, W. F. 2005, `Properties of acrylic bone cements formulated with bis-GMA', Journal of Biomedical Materials Research ± Part B Applied Biomaterials, vol. 74, no. 2, pp. 676±685. Vallo, C. I., Montemartini, P. E., Fanovich, M. A., Porto, L., and Cuadrado, T. R. 1999, © 2008, Woodhead Publishing Limited

Materials for joint replacement

103

`Polymethylmethacrylate-based bone cement modified with hydroxyapatite', Journal of Biomedical Materials Research, vol. 48, no. 2, pp. 150±158. Verma, D., Katti, K. S., Katti, D. R., and Mohanty, B. 2007, `Mechanical response and multilevel structure of biomimetic hydroxyapatite/polygalacturonic/chitosan nanocomposites', Materials Science and Engineering: C, vol. 28, pp. 399405. Viswanath, B. and Ravishankar, N. 2005, `Biphasic composite of tricalcium phosphate reinforced with hydroxyapatite whiskers', Materials Research Society Symposium Proceedings, vol. 898, pp. 80±85. von Knoch, F., Heckelei, A., Wedemeyer, C., Saxler, G., Hilken, G., Brankamp, J., Sterner, T., Landgraeber, S., Henschke, F., Loer, F., and von Knoch, M. 2005, `Suppression of polyethylene particle-induced osteolysis by exogenous osteoprotegerin', Journal of Biomedical Materials Research A, vol. 75, no. 2, pp. 288±294. Wahl, D. A. and Czernuszka, J. T. 2006, `Collagen±hydroxyapatite composites for hard tissue repair', European Cells and Materials, vol. 11, pp. 43±56. Wan, Y. Z., Hong, L., Jia, S. R., Huang, Y., Zhu, Y., Wang, Y. L., and Jiang, H. J. 2006, `Synthesis and characterization of hydroxyapatite±bacterial cellulose nanocomposites', Composites Science and Technology, vol. 66, no. 11±12, pp. 1825±1832. Wang, J. and Stevens, R. 1989, `Zirconia-toughened alumina (ZTA) ceramics', Journal of Materials Science, vol. 24, no. 10, pp. 3421±3440. Wang, J. S., Franzen, H., Toksvig-Larsen, S., and Lidgren, L. 1995, `Does vacuum mixing of bone cement affect heat generation? Analysis of four cement brands', Journal of Applied Biomaterials, vol. 6, no. 2, pp. 105±108. Wang, M. and Bonfield, W. 2001, `Chemically coupled hydroxyapatite±polyethylene composites: structure and properties', Biomaterials, vol. 22, no. 11, pp. 1311±1320. Wang, M., Porter, D., and Bonfield, W. 1994, `Processing, characterization, and evaluation of hydroxyapatite reinforced polyethylene composites', British Ceramic Transactions, vol. 93, no. 3, pp. 91±95. Wang, M., Deb, S., Tanner, K. E., and Bonfield, W. 1996, `Hydroxyapatite-polyethylene composites for bone substitution: effects of silanation and polymer grafting', Proceedings of the Seventh European Conference on Composite Materials, vol. 2, pp. 455±460. Wedemeyer, C., Neuerburg, C., Pfeiffer, A., Heckelei, A., von, K. F., Hilken, G., Brankamp, J., Henschke, F., von Knoch, M., Loer, F., and Saxler, G. 2007, `Polyethylene particle-induced bone resorption in substance p-deficient mice', Calcified Tissue International, vol. 80, no. 4, pp. 268±274. Willert, H. G. and Semlitsch, M. 1977, `Reactions of the articular capsule to wear products of artificial joint prostheses', Journal of Biomedical Materials Research, vol. 11, no. 2, pp. 157±164. Woo, S. L. Y., Akeson, W. H., and Levenetz, B. 1974, `Potential application of graphite fiber and methyl methacrylate resin composites as internal fixation plates', Journal of Biomedical Materials Research, vol. 8, no. 5, pp. 321±338. Yoshida, A., Miyazaki, T., Ashizuka, M., and Ishida, E. 2006, `Bioactivity and mechanical properties of cellulose/carbonate hydroxyapatite composites prepared in situ through mechanochemical reaction', Journal of Biomaterials Applications, vol. 21, no. 2, pp. 179±194. Yoshimura, M., Noma, T., Kawabata, K., and Somiya, S. 1987, `Role of H2O on the degradation process of Y-TZP', Journal of Materials Science Letters, vol. 6, no. 4, pp. 465±467. © 2008, Woodhead Publishing Limited

104

Joint replacement technology

Yue, W. and Roeder, R. K. 2006, `Micromechanical model for hydroxyapatite whisker reinforced polymer biocomposites', Journal of Materials Research, vol. 21, no. 8, pp. 2136±2145. Zdrahala, R. J. 1996, `Small caliber vascular grafts. Part II: Polyurethanes revisited', Journal of Biomaterials Applications, vol. 11, no. 1, pp. 37±61. Zhang, H., Yan, Y., Wang, Y., and Li, S. 2002, `Thermal stability of hydroxyapatite whiskers prepared by homogeneous precipitation', Advanced Engineering Materials, vol. 4, no. 12, pp. 916±919. Zhang, L. J., Feng, X.S., Liu, H.G., Qian, D. J., Zhang, L., Yu, X.L., and Cui, F. Z., 2004, `Hydroxyapatite/collagen composite materials formation in simulated body fluid environment', Materials Letters, vol. 58, no. 5, pp. 719±722. Zhang, L., Li, Y., Zhou, G., Wu, L., Mu, Y., and Yang, Z. 2007, `Preparation and characterization of chitosan/nanohydroxyapatite composite used as bone substitute material', High Technology Letters, vol. 13, no. 1, pp. 31±35. Zhang, M. and James, S. P. 2005, `Synthesis and properties of melt-processable hyaluronan esters', Journal of Materials Science, vol. 40, no. 11, pp. 2937±2943. Zhao, F., Yin, Y. Lu, W. W., Leong, J. C. Zhang, W., Zhang, J., Zhang M., and Yao, K., 2002, `Preparation and histological evaluation of biomimetic three-dimensional hydroxyapatite/chitosan-gelatin network composite scaffolds', Biomaterials, vol. 23, no. 15, pp. 3227±3234. Zhitomirsky, I. and Pang, X. 2006, `Chitosan±hydroxyapatite nanocomposite coatings for biomedical applications', TMS Annual Meeting, pp. 31±39.

© 2008, Woodhead Publishing Limited

5

Regulatory issues affecting joint replacement: the case of the UK

E D A M I E N , MHRA, UK, B P A U L , Kyiv Medical Academy, Ukraine and S M D A M I E N and C S D A M I E N , QMUL, UK

5.1

Introduction and background

This chapter is intended as a brief general introduction into the regulatory process of joint replacement devices in the United Kingdom. Joint replacement implants refer to articulating surfaces rather than parts or components of the implant system. This should not be regarded as an authoritative statement of the law.

5.2

The regulatory process

5.2.1

Medical device

According to the Medical Device Directive: Medical device means any instrument, apparatus, appliance, material or other article, whether used alone or in combination, including the software necessary for its proper application intended by the manufacturer to be used for human beings for the purpose of: · diagnosis, prevention, monitoring, treatment or alleviation of disease, · diagnosis, monitoring, treatment, alleviation of or compensation for an injury or handicap, · investigation, replacement or modification of the anatomy or of a physiological process, · control of conception, and which does not achieve its principal intended action in or on the human body by pharmacological, immunological or metabolic means, but which may be assisted in its function by such means.1

Manufacturers and their agents planning to place joint replacement devices on the European Community market are advised to consult the Medical Device Directives and Medical Device Regulations to establish that their devices fall within the scope of the definition of a medical device.

© 2008, Woodhead Publishing Limited

106

5.2.2

Joint replacement technology

Commission Directives

Joint replacement devices are regulated via the Medical Device Directive (MDD), Council Directive 93/42/EEC concerning medical devices, which was adopted by the European Council of Ministers on 14 June 1993 (Official Journal of the European Communities 12 July 1993 ref L169).1 Directive 93/42/EEC was implemented in the UK Medical Device Regulations 2002 (SI No. 618) which consolidates all medical device regulations into a single piece of legislation, which came into force on 13 June 2002.2 Changes to the Directive and the new legislation are documented in the Commission Directive 2005/50/EC of 11 August 2005 on the reclassification of hip, knee and shoulder joint replacements in the framework of Council Directive 93/42/EEC concerning medical devices.1,3 The main purpose of the MDD is to allow free movement of devices throughout the European Union and to ensure optimum device performance and safety. The Directive specifies a device to which it applies. It also specifies the essential requirements to be met prior to placing the device on the market. The essential requirements, set out in the MDD, ensure that the joint replacement devices do not compromise the clinical indications or safety of the end users and the professional users. They also ensure that the implants achieve the intended purpose as claimed by the manufacturer. Any risks associated with the use of particular joint replacement prosthesis should be acceptable when weighed against the benefits to the end user. The essential requirements introduce controls for safety, performance, specification, design, manufacture, sterilisation, labelling and packaging in addition to specifying the actions to be taken by the Competent Authority and the manufacturer following an adverse incident associated with a device. Essential requirements specify the requirements for preclinical assessment of notifications for clinical investigations. The Directive requires the Competent Authorities in each of the Member States of the European Union to designate notified bodies to check that devices conform with the essential requirements by performing conformity assessments to ensure the safety and performance of devices to avoid the introduction of implants that might compromise the health and safety of its users.

5.2.3

Medical Device Regulations

The directives are transposed into UK legislation as regulations, and are published as a statutory instrument. The Medical Device Regulations 2002 (SI No. 618) came into force on 13 June 2002 and implement the provisions of the MDD 93/42/EEC.2 The Medicines and Healthcare Products Regulatory Agency (MHRA) is the UK Competent Authority for the regulations.3 The Regulations implement the MDD into UK law and place obligations on © 2008, Woodhead Publishing Limited

Regulatory issues affecting joint replacement

107

manufacturers to ensure that their devices are safe and fit for their intended purpose before they are CE marked and placed on the market in any European Union Member State. There are currently two sets of Medical Device Regulations implementing all of the MDDs and amendments to date: Statutory Instruments 2002 No. 618 (consolidated legislation) and Statutory Instruments 2003 No. 1697 (Amendments).3±5

5.2.4

Reclassification Directive (2005/50/EC)

The devices covered by the Directive are grouped into four classes. Class I is regarded as low-risk devices. Class IIa and IIb devices are generally regarded as having medium risk where as class III includes high-risk devices. The classes differ according to the choice of the conformity assessment procedures followed. It is initially for the manufacturer to determine the classification of products, select a Notified Body and to perform appropriate conformity assessment procedure. The Notified Body will confirm the classification prior to carrying out the process of conformity assessment. If the Notified Body and the manufacturer disagree on the classification, either party can refer the matter to the Competent Authority for a decision. Some of the major reasons and the rationale for the reclassification originally proposed by the United Kingdom and France are given below. Hip, knee and shoulder replacement devices are associated with increased risk of device failure due to the complexity of the device itself compared with other total joint replacements. The risk of revision surgery is greater than that for other replacement joints. Moreover, hip, knee and shoulder replacement procedures are performed on young and active patients. Therefore, the need for these implants to function properly over the life expectancy of the patients to reduce revision and replacement has been increased. Furthermore, specific clinical data, including long-term performance data are not always available for joint replacement devices before they are placed on the market or put into clinical use. Consequently, conclusions on clinical data collected by the manufacturer of these devices should be verified in order to assess the appropriateness of the available clinical data. Minor design changes to a previously safe device can lead to serious clinical problems and early failure due to unintended and unpredicted consequences. In order to minimise design-related safety issues and to achieve the optimum safety level, the design dossier of hip, knee and shoulder joint replacement devices including the clinical data used to support the performance of these devices and the post-marketing design changes, if any, should be assessed by the Notified Body prior to these devices being commissioned for general clinical use.

© 2008, Woodhead Publishing Limited

108

5.2.5

Joint replacement technology

Hip, knee and shoulder replacement implants

Hip, knee and shoulder replacement implants are reclassified as Class III according to the Reclassification Directive 2005/50/EC. On the basis of the classification rules set out in Annex IX to Directive 93/42/EEC, total joint replacements such as hip, knee and shoulder replacements are Class IIb medical devices. In the Reclassification Directive (2005/50/EC), total joint replacement devices are reclassified as Class III medical devices by way of derogation from the provisions of Annex IX, to ensure an appropriate conformity assessment of these devices before placing them on the market. The Reclassification Directive states that: In order to achieve the optimal level of safety and health protection and to reduce the design related problems to the lowest level, the design dossier of hip, knee and shoulder replacements, including the clinical data used by the manufacturer to support the claimed performance and the subsequent post marketing design and manufacturing changes should be examined in detail by the notified body before these devices are introduced in general clinical use.6

Total implantable joint replacement systems usually comprise two sets of components: · The total joint implant itself which consists of multiple implantable component parts including implantable load-bearing parts. · Ancillary implantable components (screws, wedges, etc.) and the devices and accessories needed to perform the implantation (plates, instruments, etc.). The objective of Directive 2005/50/EC is to reclassify the implantable loadbearing articulating components that function in a similar way to the natural joint as Class III. Ancillary implantable devices and other supplied devices and accessories are not subjected to reclassification. Reclassified component parts are typically placed on the market.

5.3

Planning for the regulatory approval of a product

Manufacturers who want to place their products on the European Union market come within the scope of the Regulations.

5.3.1

Compliance with the regulatory requirements

Compliance with the regulatory requirements means the following, with regards to an orthopaedic prosthetic device: · The device can legally be sold within the European Union (EU) and the European Free Trade Area (EFTA). © 2008, Woodhead Publishing Limited

Regulatory issues affecting joint replacement

109

· The prosthesis can move freely throughout the European Single Market (ESM). · CE marking can be affixed on the device by the Notified Body (NB) (see Section 5.3.3) which is a legal requirement prior to placing a device on the European market. · CE marking will lead to enhanced sales and user satisfaction since the manufacturer can claim that the prosthetic implant meets the essential requirements specified in the MDD. · Ensures that the device meets designated safety standards and quality. · The device promotes public health and safety.

5.3.2

The Medicines and Healthcare Products Regulatory Agency

The MHRA of the Department of Health, the UK Competent Authority (CA), is the government executive agency that is responsible for ensuring that joint replacement devices work, and are acceptably safe to use in order to safeguard public health. However, no device is risk free and the MHRA ensures that the benefits to patients and the public justify the potential risks associated with them. The MHRA will take any necessary action to protect the public as it has the power to withdraw an orthopaedic device from the market if it is necessary. The MHRA can also prosecute a manufacturer or a distributor if the law has been broken, as it is fully accountable to both the government and the public. Medical device regulations establish systems under which a manufacturer must submit to the MHRA information about clinical investigations of joint replacement devices to be performed in the United Kingdom. The MHRA publishes guidance for manufacturers on clinical investigations to be carried out in the United Kingdom and outlines the legal requirements as set out in the regulations in addition to providing background and guidance on how to apply for preclinical assessment of a proposed clinical investigation in humans. The MHRA conducts preclinical assessment of applications for a proposed clinical investigation of joint replacement implants submitted by manufacturers to be carried out in part or in whole in the United Kingdom. Information and assistance on individual cases can be sought from the MHRA.

5.3.3

The Notified Body

A Notified Body (NB) is responsible for carrying out regulatory inspections of orthopaedic device manufacturers and their joint replacement devices under the MDD. Notified Bodies are independent organisations that have been nominated by a Member State and notified by the Commission to perform the conformity assessments according to the conditions set out in the MDD.3,7 NBs assess the manufacturer's conformity to the essential requirements listed © 2008, Woodhead Publishing Limited

110

Joint replacement technology

in the Directive by manufacturer's factory inspection, quality assurance, type examination, or design dossier examination or a combination of these and affix the letters `CE' on the product. The `CE' mark on a device is the manufacturer's claim that a particular device meets the requirements of the European Directive. NB under the full quality assurance system, conduct an examination of the design dossier and of the changes to the previously approved design in accordance with Annex II, point 4 to the Council Directive 93/42/EEC. Designation of Notified Bodies under the MDD in the United Kingdom is by the Secretary of State acting through the MHRA. The MHRA, being the UK CA, monitors Notified Bodies through regular surveillance assessment audits against the requirements to establish the NB's compliance with the requirements. The CA has a duty to ensure that the designated Notified Bodies are formally accredited against the EN 45000 series of standards and to establish basic competence under the regulations. The MHRA has a duty to withdraw designation should the NB fail to meet the requirements set out in Annex XI of the MDD and the associated responsibilities described in Annexes II, III, IV, V, and VI of the MDD.1

5.3.4

Basic steps to CE marking for a joint replacement device

· Establish that the device falls within the scope of the MDD. · Identify and ensure that the device meets the applicable entry requirements. · Identify and ensure that the device meets the applicable harmonised standards. · Classify the device. · Choose the conformity assessment procedure. · Select a Notified Body for conformity assessment. · Declaration of conformity. · Affix the CE marking.

5.3.5

Harmonised standards for joint replacement devices

There are three levels of harmonised standards to follow for joint replacement implants. · Level 1 = The European Standard EN ISO 14630:2005 Nonactive Surgical Implants General Requirements. · Level 2 = The European Standard EN 12010:1998 Nonactive Surgical Implants Joint Replacement Particular Requirements. · Level 3 = The European Standard EN 12563:1999 Nonactive Surgical Implants Joint Replacement Implants Specific Requirements for Hip Joint Replacement Implants. · Level 3 = The European Standard EN 12564:1999 Nonactive Surgical

© 2008, Woodhead Publishing Limited

Regulatory issues affecting joint replacement

111

Implants Joint Replacement Implants Specific Requirements for Knee Joint Replacement Implants.

5.3.6

Centre for Evidence based Purchasing (CEP)

The Centre for Evidence based Purchasing (CEP) collates and gathers evidence to support and assist in the adoption of innovative technologies in healthcare. CEP underpins purchasing decisions by providing evidence to assist the introduction of useful, safe and innovative products in healthcare covering all medical devices including joint replacements on the market and near the market.8

5.4

Summary

Guidance notes on the European Commission MDD are available on the EC website and are intended for manufacturers and users of medical devices.5 Further detailed information can be obtained from other relevant publications produced by the MHRA from the MHRA website.3 Annex VII of the Directive requires the report of adverse incidents associated with CE marked joint replacement devices including custom made devices to the UK Competent Authority, which is the MHRA. The MHRA will investigate the adverse incidents and may issue safety notices including device alerts as necessary to safeguard public health.

5.5

References and useful websites for further information

1 Council Directive 93/42/EEC concerning medical devices (Official Journal No. L169, published 12 July 1993). 2 Medical Device Regulations 2002 SI 2002 No. 618. 3 www.mhra.gov.uk 4 Medical Devices (Amendments) Regulations 2007 SI 2007 No. 400 http://www.opsi.gov.uk/si/si2007/20070400.htm 5 http://www.opsi.gov.uk/index.htm 6 MEDDEV guidance documents on European Commission website. http://ec.europa.eu/enterprise/medical_devices/meddev/index.htm 7 BSI Notified Body link http://www.bsiamericas.com/CEMarking/WhatisCEMarking/ CEProcess.xalter 8 CEP link ± www.pasa.doh.gov.uk/evaluation/propose_project/

© 2008, Woodhead Publishing Limited

Part II

Material and mechanical issues

© 2008, Woodhead Publishing Limited

6

Metals for joint replacement Y T K O N T T I N E N , Helsinki University Central Hospital, Finland, I M I L O SÏ E V , JosÏef Stefan Institute, Slovenia, R T R E B SÏ E , Orthopaedic Hospital Valdoltra, Slovenia, P R A N T A N E N and R L I N D E N , National Agency of Medicines, Finland, V - M T I A I N E N , Orton Orthopaedic Hospital of the Invalid Foundation, Finland and S V I R T A N E N , University of Erlangen-Nuremberg, Germany

6.1

Introduction

6.1.1

Summary

This short overview provides first three different classifications of biomaterials in general, based on their composition, surface reactivity and production, which will help to put metals in context with everyday life. Not all metals can be used as biomaterials since their use in the human body is tightly regulated; the general principles and specific EU standards well reflect the global status and trends in the regulatory field. Metal ions are held together by metallic bonds, which consist of relatively loosely bound valence electrons. The positive metal ions are located in crystal lattice points, which are surrounded by electrons. This provides metals with special properties, such as good thermal and electrical conductivity, metallic lustre, but also their ductile and malleable properties, i.e. the ability to undergo plastic deformation without breaking. Unfortunately, most of the technically important metallic materials are thermodynamically not stable in the metallic state in air or in aqueous solutions. Instead, surface oxidation will take place (during which metal loses electrons). The oxidised metal ions will either dissolve (corrosion), or form an oxide film on the metal surface (passivity). Surgical use of steel, titanium, cobalt-based alloys and tantalum is overviewed. Finally, some future trends such as mini-invasive surgery, resurfacing implants, isoelastic implants, implant coating and osseo-integrating implants are discussed.

6.1.2

Classification

Classification can be based on different premises. Biomaterials can be divided according to the chemical composition as follows: · metals · polymers © 2008, Woodhead Publishing Limited

116

Joint replacement technology

· ceramics · composites; and · materials of biological origin. This chapter describes the use of metals in connection with joint replacements. Biomaterials can also be divided based on their surface reactivity as follows: 1. 2. 3. 4.

Almost inert, with smooth surface. Almost inert, with porous surface. Chemically reactive surfaces. Bioresorbable (bioabsorbable) materials.

Metallic biomaterials in clinical use belong to group 1 or 2 according to their surface reactivity. Often the metal, e.g. stainless steel or cobalt±chrome, evokes a host response so that the implant is surrounded by fibrous tissue or capsule. If the biomaterial, however, is able to endure the effect of its biological surrounding, it is known as biotolerant. Some metals have an oxide layer on their surface, e.g. titanium, and can be in direct contact with the surrounding bone without causing any effects or reactions with its biological surrounding. They are bioinert materials. The group 3 surface reactive materials provoke a tissue response, which can lead to direct bonding to osteoid or bone. This has been demonstrated, for example, for calcium phosphate in, for example, hydroxyapatite, bioglass and glass ceramics. These materials are called bioactive and can be used for surface coating. Group 4 bioabsorbable materials are degraded and replaced by regenerating tissues either fully (for example, polyglycolides) or partially (calcium phosphate). It is also possible on a historical basis to separate three different generations of biomaterials: 1. First generation: minimal tissue±biomaterial interactions, replacement of body parts with materials with an adequate functional performance, e.g. loadbearing biomaterials. 2. Second generation: control or induction of favourable host reactions, e.g. bioresorbable materials, which allow in-growth of host tissues. 3. Third generation: emphasis on regeneration of impaired tissue, tissue engineering products, living cells used in combination with artificial materials. Metals or metal-based materials can belong to any of these, with whole metal implants belonging to the first generation (Fig. 6.1), bioactive-coated metal implants to the second generation and metallic scaffolds to the third generation.

6.2

General requirements for biomaterials

The properties of materials to be used inside the body must meet the requirements concerning biofunctionality as well as biocompatibility. If the purpose of the material is to replace diseased tissue that, for instance, has had specific load© 2008, Woodhead Publishing Limited

Metals for joint replacement

117

bearing or optical properties, then the biomaterial also has to meet such requirements in order to be biofunctional. The main difference between biomaterials and other engineering materials is that the biomaterials and their products additionally need to be biocompatible and safe for the host.

6.1 (a) Hybrid total hip replacement implant with Charnley Muller stainless steel femoral stem articulating with a polyethylene cup. The surface of the stem, in contact with bone cement (poly(methylmethacrylate) or PMMA used for fixation) in the implant bed in the femoral bone below the collar part, is matt (approximately 100 nm rough). The aim of the collar is to minimise migration of polyethylene wear debris produced in the ball-to-cup gliding pair to implant±bone interface, where it might provoke osteolysis and weaken the fixation of the implant (or actually of the cement mantle surrounding the implant stem) to the bone. The ball of the stem has been polished to a mirror surface (approximately 10 nm rough). It snap fits to the socket in the cup produced from ultra-high molecular weight polyethylene (UHMWPE). This particular cup was intended for a cementless application and the holes were made to secure primary fixation of the socket with screws. (b) A somewhat similar total hip replacement prosthesis implanted as seen in a plain radiograph. The polyethylene liner is marked with a metallic ring. Both components have been cemented, although the cement, which contains radio-opaque barium sulphate, is not clearly seen. This is a loose implant after 32 years of function. The plain radiograph shows osteolytic lesions both around the cup, the neck of the stem and the distal part of it. © 2008, Woodhead Publishing Limited

118

Joint replacement technology

The properties and production of biomaterials in the EU economic area are regulated by the European Union. The products have to be suitable for their purpose and they must attain the planned functionality and functional properties. The appropriate use of the products should not endanger the health of the patient, the user or any third person. The national regulations are based on EU Directives about active implants for medical use (90/385/EEC) and medical devices (93/42/EEC). The risks caused by the use of biomaterials must be acceptable and compatible with the purpose of use from the point of view of health, security and high standards of the healthcare system. The manufacturer must recognise the risks associated with the biomaterials and must eliminate or minimise them. The user must be informed of the residual risk after this minimising procedure. All adverse effects have to be acceptable from the point of view of the planned use and properties of the biomaterial. Planning of implants and properties of biomaterials are subject to regulation. The most important properties relate to chemical, physical and biological properties and infections or microbiological contamination. Special attention has to be paid to the selection of biomaterials and the compatibility of the biomaterials with tissues, cells and tissue fluids. The planned life in service of the implant has to be taken into account. Sterilising is subjected to its own requirements. The origin of biomaterials of biological origin is strictly regulated. The source material has to be traceable, but at the same time anonymity is required. Biomaterials of such origin have to meet particular requirements concerning modification, storage, testing and other handling, which all must be performed under optimal circumstances. Viruses and other infectious agents should to be removed or inactivated during processing. Biomaterials of human origin are developing rapidly. The central standards regulating biomaterials have been published in the ISO 10993 series, which includes many standards. ISO refers to International Organization for Standardization. Because the abbreviation of this organisation would have been different in different languages, it was decided that a word derived from the Greek word `isos', meaning `equal', is used. Standards are referred to below by their original international ISO codes. The code EN refers to European standards created by CEN, CENELEC or ETSI.

6.3

Examples of currently valid European Union standards

6.3.1

Biomaterial standards

· ISO 10993-1 provides guidelines for the selection of tests required for the biological evaluation of devices and equipment used in health care. · ISO 10993-2 covers animal welfare requirements to be followed during biological testing of medical devices. © 2008, Woodhead Publishing Limited

Metals for joint replacement

119

· ISO 10993-3 covers evaluation of genotoxicity, carcinogenicity and reproductive toxicity. Genotoxic factors are able to cause mutations, chromosomal aberration and other permanent DNA effects. Because genes contain instructions for cellular behaviour, such changes can have disastrous effects, such as cancer or reproductive toxicity. · ISO 10993-4 regulates the selection of tests to determine haemocompatibility. Blood enables transportation of oxygen and carbon dioxide, defence against infectious pathogens and repair of injured tissues. These functions are of vital importance. Therefore, biomaterials coming into contact with blood have to be haemocompatible. Haemolysis or degradation of red blood cells can be caused by material itself or by mechanical injury caused by the use of the implant. Intravascular haemolysis leads to haemoglobinuria and kidney damage. Deleterious effects on white blood cells can impair defence against infectious agents. In deep granulocytopenia, blood neutrophils are less than 0:5  109/l and risk of infection in, for example, skin or mucosa and even sepsis is high. In deep lymphocytopenia various intracellular viruses and tuberculous and atypical mycobacteria threaten. Blood contains many cascade systems of vital importance, such as complement, coagulation and fibrinolytic systems. The tests that are required for individual devices depend on the classification of that device and differ for blood vessel prosthesis, artificial heart valve, cardiac pacemaker, blood cannulas, stents and heart± lung machine, for example. · ISO 10993-5 defines the test to be used for the determination of cytotoxicity in vitro. This is often done as the first test for planned medical devices to test their biocompatibility. This is because such tests can be rapidly performed, are well standardised, sensitive and cheap. Cytotoxicity testing correlates well with short-term implant fate. Cytotoxicity testing is so important that it has to be performed for all medical devices. Various cell lines, e.g. human colon adenocarcinoma cell line caco-2, can be used for cytotoxicity testing. · ISO 10993-6 provides guidelines on how to analyse local effects after implantation. This standard describes experimental animal tests, tissues, follow-up times, implantation methods and estimation of the biological host responses. Usually the implant is removed before the tissue samples are taken. Particular attention is paid to the interface between the implant and host. · ISO 10993-7 deals with residues produced as a result of ethylene oxide sterilisation. · ISO 10993-8 deals with the selection and qualification of reference materials for biological tests. · ISO 10993-9 provides framework for identification and quantification of potential degradation products. · ISO 10993-10 defines irritation and hypersensitivity tests. Biomaterial (and chemicals released from it) can cause irritation of skin, mucosa or eyes. © 2008, Woodhead Publishing Limited

120

·

·

·

·

Joint replacement technology

Usually irritation leads to inflammation with redness, swelling, tenderness, etc., which is roughly proportional to the concentration of the irritant. Numerous chemicals of biomaterial origin cause either immediate or delayed irritation. Some of the irritants are additives, which are used to facilitate production, whereas some are harmful contaminants. For example, devices, which have been sterilised using ethylene oxide gas, can contain residues, which lead to irritation if their concentration is high enough. In spite of numerous attempts to find in vitro tests, irritation is still mainly tested using animal experiments. Hypersensitivity can arise, if the host is exposed repeatedly or for a long time to biomaterials (or products thereof). Most of these hypersensitivities are delayed hypersensitivity reactions medicated by T lymphocytes. Immediate hypersensitivities mediated by IgE-sensitised mast cells are rarer. Therefore, these tests are usually performed using animal skin testing and epicutaneous or patch tests. Hypersensitivity leads to redness and swelling of the skin. In immediate hypersensitivity, an already hypersensitive individual can mount rapid responses when exposed to allergen. This can be tested using measurements of serum IgE antibodies or intracutaneous prick tests, more rarely using exposure (under controlled conditions). Hypersensitivity reaction is not proportional to the allergen, because even small concentrations of allergen can cause dramatic responses in hypersensitive individuals. ISO 10993-11 deals with toxicity tests. Material released from the implant can cause systemic toxicity, which can affect the function of vital organs. In particular, liver, heart, kidneys and brain are considered important from this point of view. Toxicity tests are based on classical toxicological evaluation used for drugs and chemicals, which must be modified so that they can be applied to solid medical devices. This is particularly pertinent to corrosion products as the body has a limited capacity to secrete various metal ions released from metal implants, which may lead to an accumulation in vital organs. End stage kidney disease can form a contra-indication for metallic implants. ISO 10993-12 contains instructions for the preparation of samples and comparative materials. Test types, appropriate dissolvents and conditions and control materials are described. The standard recommends the use of medical device in its final composition, because the biological testing must cover all substances used for production of the device. ISO 10993-13 provides help for the identification and quantification of degradation products produced from polymers, which in particular in totally replaced joints may cause a problem as a result of debris formation and foreign body reactions. ISO 10993-14 deals with the characterisation of materials. Attention has to be paid to a) materials used in the production,

© 2008, Woodhead Publishing Limited

Metals for joint replacement

121

b) additives, contaminants and residues, c) leachable substances, d) degradation products, e) other components and their interactions in the final product, and f) properties and characteristics of the final products. · ISO 10993-15 deals with the identification and quantification of degradation products from metals and alloys. · ISO 10993-16 provides toxicokinetic study designs for degradation products and leachables. · ISO 10993-17 established allowable limits for leachable substance. Material released from the implant can cause systemic toxic effects, which can adversely affect the function of vital organs. It has seven main components: 1. Materials must be defined so that their composition and character, impurities and debris derived from them can be described and used as a basis for requirements for specifications. 2. Chemicals and degradation products derived from the materials have to be taken into consideration in the assessment of toxicity. 3. The tests planned used have to take into consideration exposure to materials produced from the implant, as a result of degradation or production by other means. 4. Testing has to follow good laboratory practice (GLP) principles and has to be performed by informed experts. 5. All test results have to be available for official government organisations. 6. If the material composition, production or purpose of use is changed, toxicity has to be re-evaluated. 7. All information available from non-clinical sources, clinical studies and postmarketing surveillance has to be taken into consideration in safety assessment.

6.3.2

Corrosion standards

· ISO 17475:2005 `Corrosion of metals and alloys ± Electrochemical test methods ± guidelines for conducting potentiostatic and potentiodynamic polarisation measurements' applies to corrosion of metals and alloys, and describes the procedure for conducting potentiostatic and potentiodynamic polarisation measurements. The test can be used to characterise the electrochemical kinetics of anodic and cathodic reactions, the onset of localised corrosion and the repassivation behaviour of a metal. · ISO 11463: 1995 `Corrosion of metal and alloys ± Evaluation of pitting corrosion' gives guidelines on the selection of procedures that can be used in the identification and examination of pits and in the evaluation of pitting corrosion. © 2008, Woodhead Publishing Limited

122

Joint replacement technology

· ISO 11845: 1995 `Corrosion of metals and alloys ± General principles for corrosion testing' contains the most important general guidelines for carrying out corrosion test under conditions of constant immersion. · ISO 12732:2006 `Corrosion of metals and alloys ± Electrochemical potentiokinetic reactivation measurement using the double loop method (based on Cihal's method)' specifies the method for measuring the degree of sensitisation (DOS) in stainless steel and nickel-based alloys using the double loop electrochemical potentiokinetic reactivation (DL-EPR) test (based on Cihal's method). · ASTM standard G150-99(2004) `Standard test method for electrochemical critical pitting temperature testing of stainless steels' covers a procedure for the evaluation of the resistance of stainless steel and related alloys to pitting corrosion based on the concept of the determination of a potential independent critical pitting temperature (CPT).

6.3.3

Hip standard (as an example of a joint standard)

· SFS EN ISO 14242-1 covers the principle for the simulator testing of artificial hip joints, the specifications for the reagents and materials and apparatus used in the testing and describes the simulator testing procedures and testing reports. · SFS EN ISO 14242-2 covers the measurement methods for the wear testing of artificial hip joints. The two methods described are the gravimetric method and the dimensional change method. The standard gives the principles for the testing methods, specifications for the reagents, materials and apparatus used. Descriptions of the preparation of test specimen, testing procedures and test reports are given.

6.4

Overview of metals

The most widely used metals in medicine are gold and other precious metals, surgical stainless steels, cobalt±chrome alloys, titanium and its alloys, tantalum and mercury-based alloys. Metals are either light or heavy metals with 5 g/cm3 as the cut-off. Aluminium and titanium compounds are light metals, whereas other metals mentioned above are heavy metals. Usually metals have a high chemical reactivity, except for the precious (noble) metals, which as a result usually occur as pure elements in nature. Many metals also contain various alloying elements and/or impurities, which are important for their physical and chemical properties, as well as for their biocompatibility. Biocompatibility refers to the ability of a material to appropriately interact (including inert behaviour) with the host in a specific location, e.g. in blood or bone. Most metals have metallic lustre if not oxidised. Metals form positive ions in solutions and occur in metallic compounds only as positive ions. Metal oxides © 2008, Woodhead Publishing Limited

Metals for joint replacement

123

6.2 Some materials, such as silicon dioxide (SiO 2 ), can exist either in amorphous (a) or in crystalline (b) form. Amorphous material lacks the longterm order of the crystalline material. Further, regularly ordered, repeated crystalline pattern can occur in a monocrystalline or polycrystalline form. Polycrystalline material (c) consists of multiple monocrystalline grains.

form hydroxides rather than acids with water. Owing to their structure and chemistry, metals are subjected to a special form of degradation, known as corrosion. However, when used in its broader sense, the word corrosion can refer to any environmental degradation of polymers, ceramics or polymers. If the atoms of a material are organised into definite repeating pattern the material is called crystalline (in contrast to amorphous materials, Fig. 6.2). The smallest repeating unit of a crystal is called unit cell. The localisation of the atoms in the unit cell defines the crystal structure. There are seven crystal systems defined by the geometry of the unit cell, i.e. by the lattice parameters. These systems can be combined with six different lattice centrings, i.e. adding symmetrical lattice points: no extra points, at the centre, centred on all faces or centred on two opposite faces (three different options). Thus, in theory, there are 42 different combinations but on closer inspection some of them are found to be identical with each other resulting in 14 different geometric arrangements (or Bravais lattices) into which all crystalline solids fit. The crystal structure is usually not perfect, but contains different defects, such as point, line and plane defects. If the repeating pattern extends through the entire piece of material it is called monocrystalline, whereas a material consisting of multiple (mono)crystals is called polycrystalline (Fig. 6.2). The average size of the crystals in a polycrystalline material is called the grain size. If a material has no longterm structure, it is called amorphous. Metals are usually polycrystalline materials and the most common crystal structures found in metals are body centred cubic (bcc), face centred cubic (fcc) and hexagonal close packed (hcp) with atomic packing factors of 0.68, 0.74 and 0.74, respectively (Fig. 6.3). Packing factor refers here to the volume of space taken up by the metal atom spheres in a unit cell. The densest possible packing of equal-sized spheres is achieved with fcc and hcp structures, where the spheres occupy 74.05% of the space. Metals can be allotropic, which means that they can exist in different crystal structures; e.g. iron has bcc (alpha-iron) structure at room temperature but exists © 2008, Woodhead Publishing Limited

124

Joint replacement technology

6.3 The most common unit cell crystal lattice structures found in metals are body centred cubic (bcc), face centred cubic (fcc) and hexagonal closed packed (hcp) with atomic packing factors of 0.68, 0.74 and 0.74, respectively. For example, at room temperature the bcc structure can be found in raw iron, fcc in AISI316L and hcp in titanium and TiAl6V4.

in fcc (gamma-iron) between 900 and 1400 ëC. Amorphous metals or glassy metals are a relatively new innovation and their full commercial potential has yet to be explored. These materials possess interesting physical properties such as high strength and they are used on, for example, surgical blades (Liquidmetal Technologies). There are not enough electrons for the metal atoms to be covalently bonded to each other. In metallic bonds the valence electrons are relatively free, delocalised and only loosely held to the positive metal atom ion cores, which makes the bonds non-directional. There is a strong electrical attraction between the immobile positive metal ions and the mobile negative electrons, which makes the metallic bond. Electrons can readily move in the crystal, so that metals conduct electricity which subjects them to galvanic corrosion. Because of the free electrons in metals, the thermal conductivity of the metals is usually higher than that of ceramics (although aluminium oxide has a high thermal conductivity) and of polymers, which are bonded by ionic and covalent bonds, respectively. For the same reason, metals feel cold. The melting point of the metals is because of the strength of the metallic bond, usually quite high (e.g. Al 660 ëC, Fe 1538 ëC and Ta 3017 ëC). However, mercury (MP ÿ39 ëC), caesium (MP 28 ëC) and gallium (MP 30 ëC) are liquid at or near room temperature and tin has a melting point of 232 ëC. Owing to the non-directional metallic bonds, neighbouring crystallographic planes can move relatively easily in relation to each other without causing a catastrophic failure of the material. This makes metals malleable and ductile (capable of undergoing plastic deformation). However, cold working (or strain hardening, e.g. forging) deforms individual crystals and introduces dislocations, rendering the further formation of dislocations more difficult and the metal stronger. © 2008, Woodhead Publishing Limited

Metals for joint replacement

125

Metals can also be strengthened by introducing impurities (or adding them intentionally to alloy), grain size diminution and precipitation. When adding alloying elements, substitutional replacement means replacing lattice atoms, whereas interstitial replacement means placing atoms between lattice atoms. Heat treatments, such as annealing and quenching, can be used to alter the grain size or to generate precipitates of a harder phase. In the latter case the harder phase acts as the disperse phase in composites, e.g. cementite (Fe3C) in ferrite () iron matrix. An alloy is a combination of two or more metals, or a metal and a non-metal with characteristics of a metal. Alloys are usually prepared by mixing the molten components and then cooling the mixture. If an alloy contains high percentage of iron then it is called ferrous alloy (compared with non-ferrous metals/alloys that do not contain iron or contain it in relatively small amounts). Solid solution alloys are homogeneous mixtures of substitutional (e.g. TiAl6V4) or interstitial (e.g. C in steel) type. Substitutional alloys are made of two components with similar atomic radii (15%) and bonding characteristics. In such alloys, one atom can substitute the other so that the solute atoms can take the positions of the solvent and occupy regular lattice sites. In interstitial alloys the smaller (usually a nonmetal) of the two atoms has a radius of only approximately half the larger one. Therefore, the smaller atoms fit into the spaces or interstices between the larger atoms and the solute occupies interstitial sites in the metallic lattice. The alloy produced is stronger than the pure metal. Steel is an interstitial alloy of iron and carbon, which contains up to 1.7% carbon. In contrast, heterogeneous alloys are non-homogeneous dispersions containing at least two different phases. An intermetallic alloy is, instead of being a solid solution, a compound formed of two different metals and has a definite chemical composition (e.g. CuAl2). The chemical formula dictates the ratio of its components and its chemical properties and crystal structure are different from the parent metals. Naturally, an alloy can be a combination of the options mentioned above.

6.5

Biomechanical properties

From the materials science point of view human implants are a very demanding but not a unique challenge and thus normal material testing methods yield useful information for their development. For example, most physical properties (Table 6.1) of a material are obtained simply by measuring its strain under stress (Fig. 6.4). The stress is usually tension or traction, but shear, torsion and compression are also used. The latter is used especially with brittle ceramic materials. When the deformation is elastic, or `recoverable', the stress and strain are proportional, stress = E  strain. The coefficient E defines Young's modulus or the modulus of elasticity and its unit is pascal (Pa) ± the same as for the stress because in the formula the strain is relative. If the proportionality is not linear the coefficient has to be determined using, for example, the tangent or secant © 2008, Woodhead Publishing Limited

126

Joint replacement technology

Table 6.1 Mechanical properties of selected biomaterials Density (g/cm3) AISI316L CoCrMo (cast) TiAl6V4 Ti (ASTM F 67) Ta (bulk) UHMWPE

7.9 7.8 4.43 4.5 16.6 0.93±0.944

Hardness HV (MPa)

Yield strength (MPa)

1500±3100 205±310 3000 455 3000±3400 830 2400±2700 483 900 165 60±66 (HS) 21.4±27.6

Tensile strength (MPa)

Elong. (min.%)

515±620 655 930 550 205 38.6±48.3

12 10 5±8 13±15 2 230±350

modulus of the stress±strain diagram. The modulus of elasticity describes the material's ability to resist elastic deformation and can therefore be referred as the stiffness of the material. Resilience describes the material's maximum ability to store and release energy when loaded and unloaded to the yielding point. Consequently, its unit is J/m3 which can easily be derived to pascal (Pa). When the stress is so high that the material can no longer assume its original shape the material starts to yield and the deformation is plastic. For most metals the transition from elastic to plastic is gradual and therefore the exact starting point of yield is difficult to define. To overcome this problem a strain offset line parallel to the elastic part of the stress±strain curve is drawn so that the two curves intercept. Normally the offset shift is 0.002 towards higher strain. This level of stress ± or more likely the material's ability to resist it ± is called the yield strength. For materials with non-linear elastic stress±strain behaviour a certain value of strain (e.g. 0.005) is used to define the stress. If the stress is further increased it reaches the point above which the structure cannot resist a rupture under continuous load. This point describes the tensile strength (ultimate strength) of a material. Ductility indicates the amount of plastic deformation at the point of fracture and is expressed either in terms of percentage elongation (%EL) or percentage area reduction (%AR). The first defines the percentage of strain whereas the latter defines the percentage reduction of the cross-section due to the elongation. The opposite of ductile is brittle whereas anelasticity refers to time-dependent recovery of elastic deformation. Toughness describes a material's ability to absorb energy before it breaks. To be tough a material should have both strength and ductility and thus ductile materials are often tougher than brittle ones. Hardness is perhaps the most common way to obtain information on a material mainly because it is fast and easy to measure and because it gives valuable information for the testing of the tribological properties. Because hardness is another measure of a material's ability to resist plastic deformation it gives an indication of other mechanical properties. For example, for most steels the tensile strength is roughly 3.5 times the Brinell hardness. © 2008, Woodhead Publishing Limited

Metals for joint replacement

127

6.4 Schematic representation of a stress±strain curve. The small figure inserts depict the general shape change of the specimen during a stress±strain measurement.

The earliest way of determining the hardness, used mainly by mineralogists, was to compare which material could scratch others. Of such systems the best known is the ten-step Moh's scale in which the hardest material that no other materials can scratch, diamond, is represented by number ten and the softest, talc, by number one. Since then numerous hardness scales based on the principle of making indentations have been developed. A probe of a known geometry is pressed with a constant force on the material under inspection and the size of the indentation defines the hardness. The most common hardness scales based on indentation are Shore, Brinell, Knoop, Rockwell and Vickers. A more sophisticated method used especially with thin films and hardest materials is the nano-indentation, in which a hard, very sharp tip is pressed into the substrate, the shape and the hysteresis of the load±displacement curve are recorded and the hardness is calculated. Sometimes the sets of apparatus are combined to perform nanoscratch and nanowear measurements. Fatigue testing gives information on a material's ability to resist cyclic loads. Such a load, preferably similar to real conditions, is applied on a test piece until © 2008, Woodhead Publishing Limited

128

Joint replacement technology

it breaks. By repeating this experiment with different stress amplitudes a `stress versus number of cycles to failure' (S±N) curve is obtained. The stress level below which fatigue failures do not exist is called the fatigue or endurance limit. For example, for most steels this limit varies between 35 and 65% of the tensile strength. However, there is usually a considerable amount of statistical variation involved in this kind of data. Also, some materials, such as aluminium and most other non-ferrous alloys, do not have a fatigue limit. Therefore, statistical methods are often used to define a suitably low failure probability (Callister, 2000). Good and versatile basic tools for evaluating, testing and screening of the tribological properties of new materials and their combinations are pin-on-disk and pin-on-flat testers. During testing a pin of known dimensions slides on a circular or reciprocating track, chafing a flat specimen. Usually the pin is made of a harder material and its wear can be neglected. A typical implant-related test arrangement would be a CoCrMo pin sliding on a polyethylene slab. The load pressing the pin down, the sliding speed and the environmental parameters, such as temperature, composition of surrounding liquid or gas, and humidity of the surrounding gas, can be adjusted accurately. The number of cycles and the force needed to prevent the pin assembly from moving are recorded simultaneously during the test. The wear of the materials can be measured using several methods. The simplest and most straightforward method is to measure the physical dimensions of the wear track for example with a profilometer. The wear can also be measured using methods more common in simulators, e.g. weighing the test pieces, filtering and analysing the surrounding liquid or, if the materials are hard and the wear is minute, measuring the profile of a marker scratch (Anttila et al., 1999). The data recorded before, during and after the measurement are used to calculate the contact pressure, friction, sliding distance, wear volume and wear factor. The wear factor, k, is obtained by dividing the wear volume with the load and the sliding distance and is normally expressed in terms of mm3/N m. It gives a relative number that can be used to compare the wear resistance of materials. However, when comparing materials a good testing practice is to use similar conditions for all the materials to avoid scale errors. For example, the contact surface area may change significantly if the wear is excessive, thus biasing the results on the favour of less wear-resistant materials. The most important testing method for biomechanical components, apart from in vivo testing, is the use of implant simulators such as hip and knee simulators. The simulators mimic the movements of human joints and the loads associated with them providing tribological and endurance information simultaneously. The human walking gait cycle was first described by J. P. Paul in 1967 (Paul, 1967). It is still used as the basis of the ISO standard, which defines the cyclic loading used in hip implant testing. The Paul gait curve and the simplified ISO © 2008, Woodhead Publishing Limited

Metals for joint replacement

129

standard are compared in Fig. 6.5, which relates them to different phases of the gait cycle. With knee implants the loads are somewhat more complicated. The most important standards for knee and hip implant testing are ISO 14242 and ISO 14243.

6.5 The Paul gait curve for regular approximately 1 Hz walking. The maximum and minimum estimates of the loads in the vertical y axis are given in broken and continuous line, respectively. It can be seen that the maximum load the hip is subjected to during walking reaches approximately four times the body weight (see the scale to the left). The gait cycle starts at heal strike, when the heel of the forward foot first touches the ground. This leads very rapidly to a load peak in the hip, followed by a small notch and then a second peak when finally the big toe of the other foot leaves the ground. The apex of this second part of the first notched major load peak occurs approximately when 13% of the gait cycle time (= a) has passed. This notched first major load peak is followed by a valley, which occurs when 32% of the gait cycle time has passed (= b). It is followed by a second major load peak which occurs when 51% of the length of the gait cycle has passed since the heel strike (= c). When 62% of the gait cycle has passed the stance phase finishes and the swing phase starts when the big toe of the forward foot leaves the ground. The gait cycle finishes when the heal touches (strikes) the ground again. In the simplified ISO standard (thick black line) the first peak occurs after 12% of the cycle has passed, the valley after 32% of the cycle has passed and the second peak after 50% of the cycle has passed. In the ISO curve both load peaks reach 3 kN (see the scale to the right). In this example the ISO curve reaches the value 4 on the left-hand J/W scale, which corresponds to a person weighing 75 kg. © 2008, Woodhead Publishing Limited

130

6.6

Joint replacement technology

Corrosion

Corrosion, the gradual degradation of materials due to electrochemical attack, occurs in the electrolytic environment of the human body. Several forms of corrosion are recognised: · All metals in electrolytic solutions are subjected to certain amount of uniform attack or overall corrosion. However, the metallic materials typically employed for use in the human body show a high resistance against active, uniform dissolution, as they spontaneously form thin but highly protective oxide layers (so-called passive films) on the metal surface. Such passive metals can be susceptible to special types of localised corrosion. · Crevice corrosion begins in narrow crevices containing fluid, e.g. between a screw and a plate. Local depletion of oxygen accelerates corrosion by impairing the passivating surface oxide layer (leading to depassivation of the surface). Changes in the local electrolyte composition and pH also contribute to propagation of crevice corrosion. · Similar mechanisms are also active in pitting corrosion; however, in this case the pit initiation is a first step (which is not required in crevice corrosion, as the crevice already can be considered as a pit site). Pit initiation typically takes place at surface heterogeneities, such as inclusions, intermetallic particles or precipitates. · Galvanic corrosion occurs between two different metals as electrochemical corrosion, owing to a difference between their electrochemical potentials. · The same mechanism is also active in intergranular corrosion due to precipitations at the grain boundary. This leads to formation of internal galvanic couples between the bulk matrix and the surface in the vicinity of the grain boundaries, as depletion of alloying elements present in the grain boundary precipitates takes place. · Leaching is a form of selective corrosion, which occurs, not at the grain boundaries, but within the grains themselves. · Fretting corrosion refers to corrosion at contact areas between materials under load subjected to vibration and slip. · Tension corrosion (or stress corrosion) refers to corrosion of metal subjected to stress, e.g. bending. This will create electrochemical differences between the surfaces subjected to tensile vs. compressive stress. Corrosion is also accelerated if the tensile stress leads to a rupture of the passivation layer.

6.7

Corrosion testing

Since corrosion is an electrochemical process, it is usually studied by electrochemical methods. The most common and relatively simple electrochemical experiment used in the study of orthopaedic alloys is the potentiodynamic polarisation curve. Potentiodynamic curves record the current related to the © 2008, Woodhead Publishing Limited

Metals for joint replacement

131

6.6 (a) A potentiometric set-up demonstrates a computer screen (1) with recorded potentiodynamic curves (shown in detail in Fig. 6.8), recorded by a potentiostat (or galvanostat; 2). Values have been obtained using three electrodes (3, shown in close up in panel b) mounted in a glass cell (container, 4) with a thermostatic jacket coupled to a water bath (5) regulated by a thermostat (6). (b) The three electrodes in this set up are a working electrode (7), which is a metal specimen of interest, a reference electrode (8) against which the potential is measured and a counter electrode (9) against which the current is measured.

electrochemical reaction as a function of electrode potential impressed to the electrode. Potentiodynamic means that the electrode potential is changing linearly with time, the so-called sweep rate dE/dt, within the cathodic and anodic potential limit. Polarisation curves are usually presented as current density (j) as a function of potential applied (E). To perform potentiodynamic curves one needs a potentiostat instrument and an electrochemical cell (Fig. 6.6). The behaviour of a metal in a certain solution depends on the thermodynamics and kinetics of both metal dissolution (anodic or oxidation reaction; M!Mn+ ‡ neÿ , where M is metal in its elemental or zero valence state) and the balancing process (cathodic or reduction reaction; Xnÿ ‡ neÿ ! X) (Bockris and Reddy, 2000). A typical anodic potentiodynamic curve is schematically presented in Fig. 6.7. Anodic potentiodynamic curves recorded for three most common orthopaedic alloys, titanium-based alloys, stainless steel and cobalt-based alloys in Hank balanced physiological solution are presented in Fig. 6.8 (MilosÏev et al., 2000a; MilosÏev and Strehblow, 2000, 2003; Hodgson et al., 2004). Potentiodynamic curves give information about the susceptibility of a certain metal to corrosion and passivation, on the span of the passive region, transpassive oxidation, etc. Quantitative data on the corrosion potential and current density are obtained, although it is more convenient to get these data from the linear polarisation measurement and Tafel plot. The latter are performed in the vicinity of the corrosion potential and are less destructive. Other electrochemical © 2008, Woodhead Publishing Limited

132

Joint replacement technology

6.7 A typical anodic potentiodynamic curve presented as current density (j, in A/cm2) as a function of potential applied (E, in V). The potential sweep (dE/dt) commences at a potential where the metal is immune to corrosion. In this part the cathodic reaction takes place, which is usually the reduction of oxygen. At some potential in a particular medium, the metal will begin to oxidise. Oxidation corresponds to the anodic reaction. The corrosion potential, Ecorr, is the potential at which no net current flows because the anodic and cathodic potentials are equal. The corresponding value of corrosion current, jcorr, can be used for the calculation of corrosion rate. Throughout the active region the current increases exponentially with potential (so-called Tafel region). At a certain critical potential, Ecrit, however, the current drops often by several orders of magnitude and then remains almost constant within relatively broad potential range. The reason for the decrease in current is the formation of an oxide layer, so-called passive layer. This layer acts as a protective barrier between the metal and environment and thus ceases the metal dissolution. The thickness of the passive layer is usually only a few nanometres but its ionic and electronic properties are responsible for its high protection ability. A sudden increase of current within the passive range may occur and is related to the localised breakdown of the passive film, usually to pitting corrosion attack (A). At a certain high positive potential (B), a new anodic process commences related to either oxygen evolution or to so-called transpassive oxidation related to the oxidation of oxide species to a higher oxidation state, e.g. Cr3+ to Cr6+, and Co2+ to Co4+.

methods include potentiostatic (current measured as a function of time at constant potential), galvanostatic method (potential measured as a function of time at constant current) and electrochemical impedance spectroscopy. Electrochemical impedance is usually measured by applying an AC potential to an electrochemical cell and measuring the current through the cell. The response to © 2008, Woodhead Publishing Limited

Metals for joint replacement

133

6.8 Comparison of titanium-based alloys (o), stainless steel (±) and cobaltbased alloys (n) in Hank balanced physiological solution. Contrary to the previous example, where a transition from active to passive state is observed at Ecrit, no such transition is observed for these alloys since they spontaneously form the passive oxide layer and no active dissolution can be observed (1). They differ in the corrosion current density and the span of the passive region. Whereas TiAl6V4 alloy is stable up to 3.5 V, stainless steel and CoCr28Mo6 alloy is stable up to much lower potentials, approximately 0.2 and 0.4 V vs. standard calomel electrode or SCE (2 and 3). Furthermore, stainless steel is subjected to pitting corrosion, as evidenced by a sudden current increase and current oscillations. CoCr28Mo6 is prone to transpassive oxidation, as evidenced by a peak at 0.5 V related to the oxidation of Co2+ and Cr3+ (4). The passive layer formed on the titanium-based alloy is mostly TiO2, whereas the passive films formed on stainless steel and CoCr28Mo6 are duplex layers. The inner layer is in both cases Cr2O3, whereas the outer layer consists of Fe2O3 and CoO, respectively.

this potential is an AC current signal, containing the excitation frequency and its harmonics. This current signal can be analysed as a sum of sinusoidal functions (a Fourier series) and provides parameters like corrosion rate, capacitance of the interface, etc. The advantage of the impedance technique is that is does not strongly change the electrochemical equilibrium, as only a small sinusoidal potential disturbance is applied to the sample. Therefore, this technique allows us to monitor the corrosion of the sample as a function of time. Several standards are available for performing the electrochemical measurements, for example: ISO 17475:2005 `Corrosion of metals and alloys ± Electrochemical test methods ± Guidelines for conducting potentiostatic and potentiodynamic polarisation measurements', ISO 11463: 1995 `Corrosion of metal and alloys ± Evaluation of pitting corrosion'. © 2008, Woodhead Publishing Limited

134

Joint replacement technology

Corrosion testing can also be performed by non-electrochemical methods, i.e. weight-loss method, where the sample is suspended in a solution and weight of the sample is measured at regular intervals over a longer period of time (Pletcher and Walsh, 1990). Assuming that the change in weight represents only a loss of metal to the solution, it can be converted to mol/cm2 (ˆ rate) or a corrosion current in A/cm2 can be calculated from the relation jcorr/nF ˆ rate in mol/cm2 (Pletcher and Walsh, 1990). Corrosion engineers usually express it in microinches/year. The data obtained from such a testing are rather limited and do not describe the corrosion mechanisms. Moreover, in the case of high corrosionresistant passive materials the overall mass loss is quite small (as typically only localised corrosion will take place). This further limits the usefulness of weightloss measurements.

6.8

Metals used in joint replacements

6.8.1

Surgical stainless steel

Pure iron (containing a maximum of 0.006% carbon at room temperature), wrought iron (<0.15% carbon) and cast (pig) iron (containing 2.1±4% carbon) have at room temperature the bcc (-iron, ferrite) crystal structure and they are (ferro)magnetic. They have poor mechanical properties and easily become rusty and corroded. Steel, an alloy of iron and carbon, contains a maximum of 1.7% carbon, which at such concentrations increases strength. Carbon steel can be further processed to stainless steel to diminish corrosion. Stainless steel is an alloy of carbon steel and chromium as a major alloying element; stainless steel contains typically at least 12% chromium. As a result of addition of the chromium, the surface of steel produces a thin and relatively durable passivating oxide layer, which protects against corrosion (rust). The passive film is highly enriched with Cr-oxide. Corrosion properties can be further improved by the addition of molybdenum and nickel. An addition of 2±6% of Mo efficiently increases the resistance against pitting corrosion in NaCl-containing solutions. Nickel stabilises the austenitic ( -iron), fcc phase microstructure of the steel (Fe-Cr steels are ferritic or -iron). Austenitic steel is non-magnetic, which eliminates movement of, for example, vascular stents and heating during magnetic resonance imaging (MRI), although metal-induced artefacts remain. The properties of steels can also, in addition to alloying, be modified by different type of heat treatments leading to microstructural changes. Hard martensite steel, e.g. for piano wire, is formed by rapid cooling (quenching) of austenite steel so it has the same chemical composition but the atom spheres are arranged in a different tetragonal crystalline structure. Owing to this versatility, there are thousands of different brands of steel. At least 50 different types of steel are commercially available and approximately 20 of them are used as biomaterials. However, only some brands, for example © 2008, Woodhead Publishing Limited

Metals for joint replacement

135

American Iron and Steel Institute (AISI) austenitic stainless steel (AISI 316, 0.08% carbon) and its low carbon derivative AISI 316L (<0.03% carbon), are widely used. Steel is relatively cheap compared with other metals. The iron and lowcarbon alloyed chromium (17±20%) containing stainless steel used in medicine also usually contains 2±4% molybdenum and 12±14% nickel, and in addition small amounts of other elements. Corrosion has been minimised to reduce the release of these components as they could lead to toxic, allergic and various other symptoms. For example, nickel can cause toxic and allergic reactions. Iron release from steel can contribute to bacterial infections by acting as an iron source for bacteria. Iron is a reactive transition metal able to catalyse production of hydroxyl radicals in the Haber±Weiss reaction. Wear debris formation increases the effective biomaterial surface, which increases corrosion. This can raise the iron burden of the body. Normally the body contains approximately 3± 4 grams of iron. Clinical experience and accumulated knowledge suggest that human body tolerates leachables from surgical steel implants relatively well. Still, steel is mostly used in applications that are temporary so that the implants are removed after use, or as coated implants. Typical applications are plates, medullary nails, screws, pins, sutures and steel threads and networks used in fixation of fractures. Use of steel in joint replacements has diminished since new cobalt- and titanium-based materials have been taken into use. The production method and the microstructure of metal affect its mechanical properties. Regular AISI 316 has a relatively good yield point in traction, approximately 200±250 MPa. If a higher yield point is required, cold processing is used. Production defects and improper design increase the risk of fatigue fracture. Fatigue refers to a failure of an implant under repeated cyclic stress below the ultimate stress level. Metal implants usually fracture based on fatigue rather than mechanical overloading. For joint replacement of the large weightcarrying joints the fatigue resistance during 107 cycle testing should be at least 400 MPa. Surgical steel has a relatively good fatigue resistance, approximately 350±400 MPa. Walking exposes a hip implant to approximately 106 walking cycles per year. The metallic stem of the total hip would easily last such loading. In reality, total joints are not subjected to such a high load. 400 MPa corresponds to 4000 kg/cm2. Thus, in practice metallic stems should last indefinitely. Improper design, material defects, wear and corrosion can, however, diminish the fatigue resistance. AISI 316L stainless steel has a relatively good corrosion resistance, but compared with cobalt- and titanium-based alloys is sensitive to crevice and pit corrosion. Therefore, steel implants should not have porous surfaces. Large surface area, i.e. porous surface or debris, increases leaching of implant components and additives. AISI 316L steel is relatively heavy, its density being approximately 8 kg/dm3. The coefficient of friction of steel against polyethylene is approximately 0.10, but © 2008, Woodhead Publishing Limited

136

Joint replacement technology

under in vivo circumstances only 0.02. In a natural mixed mode lubricated joint the coefficient of friction has been estimated to be approximately 0.001±0.025 (0.001± 0.01 for pressure film lubrication and approximately 0.1 for contact point lubrication). A high coefficient of friction increases the formation of debris. In a gliding pair consisting of metal±UHMWPE (ultra-high molecular weight polyethylene) polyethylene wears out more easily than the hard metal. Polyethylene wear can be diminished by the use of highly cross-linked polyethylene (HXPE). Sir John Charnley started to use steel stems fixed with poly(methylmethacrylate) (PMMA) and cups produced from polyethylene. This configuration is still used as the golden standard in total joint replacements. Titanium- and cobalt-based alloys have better corrosion resistance. Therefore, titanium- and cobalt-based alloys for stem implants have largely replaced the AISI 316L stems. Steel is approximately ten times stiffer than the cortical bone. Most of the steels currently used have an elastic modulus of approximately 200 GPa. When such stiff steel is used, load is carried by steel implant and the bone is no longer subjected to normal loading. Use of bone cement as an interface decreases this stress shielding effect, because the elastic modulus of bone cement is much lower than that of the steel and as the cement mantle forms an interface. Fixation of stiff steel implants is therefore done with bone cement. Use of steel implants can lead to periprosthetic osteoporosis and to pathological fractures. Stiffness of metal implants is also affected by their design. Above the elastic limit, metals can, after the yield point, undergo extensive plastic deformation under stress before (ductile) failure. Ductility is thus the maximum strain that a material can withstand before undergoing ductile failure. The non-discriminate nature of the metal atoms for neighbours makes it possible for them to change their relative position under load, especially when dislocations are present. Therefore, they are ductile, not brittle like ceramics. The corrosion resistance of steel is much improved by the passivation layer, which is 1±5 nm thick and has a low ion conduction capacity. This thin layer consists of hydrated oxides and contains more chromium compared to the composition of the bulk alloy. Corrosion and biological body fluids affect the composition of this passivation layer. The protective effect of the passivation layer is decreased by heterogeneities in the microstructure, for example at the site of chromium carbides and MnS inclusions. Defects in the passivation layer can lead to localised corrosion.

6.8.2

Titanium and its alloys

Titanium is named after the `Titans', who were sons of the Earth Goddess Gaia in Greek mythology. TiAl6V4, which is a hcp±bcc alloy, is now widely used in orthopaedics. Also commercially pure hcp titanium CPTi is used, particularly for dental implants. TiAl6V4 has a tensile strength of approximately 900±1000 MPa and its fatigue resistance at 107 cycles is 600±700 MPa. © 2008, Woodhead Publishing Limited

Metals for joint replacement

137

The elastic modulus of titanium is 110 GPa, which is only half that of surgical stainless steel or cobalt-based alloys and five times that of cortical bone. This leads to more physiologically sound stress distribution in the peri-implant bone. Therefore, cement as an intermediate layer used for stress distribution in steel and cobalt-based implants would not seem to be so important when titanium implants are used. To guarantee cementless fixation by bone in-growth and micromechanical interlocking instead, porous-coated implants are often used (Fig. 6.9c). Pore size (e.g. 50±400 m), pore interconnectivity (e.g. 75±150 m), particle interconnectivity, volume fraction porosity (e.g. 30±40% for spherical beads), and the area coated (proximal part) are important for such a fixation. The other possibility is hydroxyapatite coating to allow direct chemical bonding. Lately, a new titanium alloy, which also contains zirconium and niobium, has been developed (for example TiZr13Nb13). These alloys share a high strength and low elastic modulus (65±80 GPa). Nickel±titanium shape-memory alloy (Nitinol) implants can be deformed but return to their original shape upon heating. Titanium-based biomaterials have better osteoconductive properties than cobalt-based materials. A titanium-based product (ReGenerexÕ) similar to tantalum Trabecular MetalÕ will enter the market in the near future. Titanium is relatively light, its density being only 4.51 kg/dm3. This is only approximately half that of other biomaterial metals. Wear resistance of titanium is, however, not so good. This can be improved by eloxation (Eloxalverfahren, oxidation treatment), ion implantation and titanium nitride coating. TiAl6V4 alloy has better corrosion resistance than cobalt-based materials and surgical steel. This is because a thin oxide (TiO2 or titania) layer spontaneously formed on the surface of titanium-based implants protects it against corrosion. The TiO2 passive film is very stable, and Ti and its alloys show a high resistance against pitting corrosion in NaCl-containing solutions. Wear may damage this protective oxide layer, but this loss is rapidly replaced with so-called repassivation. This process produces so much oxide, that the peri-implant tissues gradually become dark black. This metallosis can be quite dramatic as seen in revision operation, but its biological effects are usually harmless, although it can induce necrosis in periprosthetic tissue. Another possible impact of the use of TiAl6V4 alloy is the fact that it contains aluminium. Aluminium is known as a cause of osteomalacia and has also been tentatively associated with dementia. However, aluminium and vanadium atoms are located randomly in titanium and do not form any particular aluminium-rich surface layer. Because corrosion resistance of titanium-based alloys is good and the bonding forces strong, no significant amount of aluminium is released. No reports of osteomalacia or dementia have been described upon use of titaniumbased implants. The main disadvantage of CPTi and to some extent also of TiAl6V4 compared with F75 cobalt-based alloys is that it is relatively soft. The Vickers Hardness number for F75 cobalt-based alloy is 300±400 compared with 120±200 for CPTi © 2008, Woodhead Publishing Limited

6.9 (a) A double incision, minimally invasive approach is here used for total hip arthroplasty. The upper incision lies in the interneural plane between tensor fasciae late and sartorius muscles and is used for the placement of the socket. The lower incision lies in the abductor muscles and is used to insert the femoral stem. The aim of the mini-invasive total hip replacement surgery is to use short, soft tissue-saving cuts with the intention of creating better cosmetic results and enable faster recovery than is possible with a conventional approach. (b) A minimally invasive TiAl6V4femoral stem implant before implantation in the femoral bone bed. This is a cementless implant and, therefore, the intercalcar area and the proximal part of the stem is porous coated to facilitate micromechanical locking to the surrounding bone tissue and in-growth of peri-implant bone into the pores of the coating. In contrast, the intermediate part of the stem has a matt surface. The tip of the stem has a mirror surface, which allows some movement during loading of the hip. This particular implant is a modular implant so that a metallic or ceramic head will be mounted on the top of it. This enables selection of a proper size ball in the primary operation and replacement of it in an eventual revision operation without having to remove the wellfixed stem. (c) A cementless acetabular shell (Trilogy, Zimmer) produced of titanium alloy and coated with commercially pure titanium fibres is brought into place as part of a mini-invasive total hip replacement surgery. Later, a polyethylene liner is locked to this metallic shell. © 2008, Woodhead Publishing Limited

WPTF3007

Metals for joint replacement

139

and 310 for TiAl6V4. The corresponding figure for the AISI 316L steel is 130± 180.

6.8.3

Cobalt-based alloys

Cobalt comes from the German `kobald' meaning `goblin' or evil spirit. Cobaltbased alloys usually contain 30±60% cobalt and approximately 20±30% chromium. Chromium is added to improve corrosion resistance and molybdenum to increase strength. Cobalt-based alloys can also contain other elements, such as wolfram, iron, manganese and silicon. The main types are cobalt±chrome± molybdenum alloy, which is produced by casting, and cobalt±nickel± chromium±molybdenum, which is usually forged. In casting, molten metal is poured or pressured into a mould, usually with little pressure because the material is in a molten state. Following solidification, the article is removed by a process known as demoulding. The casting method has several different modifications, which can be named according to the mould (e.g., sand, investment/lost wax, shell, plaster, ceramic) or the force used to move the molten metal or perhaps some other feature of the casting method (gravity, pressure, rotatory, vacuum, squeeze, slip, rheocasting/thixocasting, which processes semisolid thixotropic metal, etc.). Forging utilises ductility of metals, which makes it possible to apply heat and compressive force to metal to obtain plastic deformation to a desired shape. The most widely used cast cobalt-based CoCr29Mo5 (F75 and F76) alloys have yield points (yield strengths) of approximately 450±800 MPa and the nickel containing CoNi35Cr20Mo10 (F-562) alloy 950 MPa. The fatigue resistances are approximately 200±950 MPa and 350±550 MPa for 107 cycles, respectively. As for other metals, production methods and micro-granularity modify their strength. The elastic modulus of cobalt-based alloys is usually 200±300 GPa. Therefore, PMMA bone cement has to be used for implantation of cobalt-based devices. The coefficient of friction between CoCr29Mo5±polyethylene gliding pair is in vitro with the use of synovial fluid 0.16 and in vivo 0.04. Cobalt-based alloys have a good corrosion resistance, although some corrosion occurs (Table 6.2). Accordingly, measurable cobalt and chromium ion concentrations have been observed in blood tests. The International Agency for Research on Cancer (ARC) has classified cobalt to group 2B or possibly carcinogenic to humans. In contrast, trivalent chromium is an essential trace element (a daily uptake of 60 g is necessary) and, therefore, chromium (III) and metallic chromium belong to group 3 and are not classified as carcinogens, although chromium (VI) compounds belong to group 1 and are carcinogenic to humans. Cobalt-based alloys are not sensitive for galvanic corrosion and titanium and cobalt, for example, can be used together. However, surgical steel should not be used in contact with cobalt-based implants, because the relatively poor corrosion © 2008, Woodhead Publishing Limited

140

Joint replacement technology

Table 6.2 Element concentrations in serum in patients with joint replacement implants Element Co Cr Ti

CM/P

M/P

C/P

M/M

0.25  0.23 0.17  0.12 1.11  0.00

0.87  1.33 0.92  0.89 2.57  1.83

0.20  0.10 0.08  0.10 3.15  2.31

2.35  2.39 3.48  2.67 1.87  1.71

Values are given in ng/ml and represent mean  standard deviation. CM/P, cemented metal-to-polyethylene. M/P, cementless metal-to-polyethylene. C/P, ceramic-to-polyetheylene. M/M, metal-to-metal.

resistance of steel leads to rapid galvanic corrosion. Titanium and titaniumbased alloys have the best crevice corrosion properties and are in this respect better than cobalt-based alloys, which again are better than surgical steel. Cobalt-based implants have very good tension corrosion properties (a combination of tension and corrosion) in spite of the fact that the body fluids contain quite a lot of chloride, which facilitates tension corrosion. In general, the corrosion resistance ranking order from the most resistant to the weakest is TiAl6V4, CoCr29Mo5 and AISI 316L (Table 6.3).

6.8.4

Tantalum

Tantalum (Ta) is a chemically very resistant metal. At low temperatures (<150 ëC) tantalum is almost completely immune to chemical attack. Only hydrofluoric acid, acidic solutions containing fluoride ions and free sulphur trioxide have a significant effect on it. Hence, tantalum is completely immune to body fluids and it is also a non-irritating metal (Lide, 2001). The biocompatibility and excellent osseointegration properties have augmented the interest on the use of tantalum on the bone-implant interface of cementless implants. Owing to its high chemical Table 6.3 Corrosion resistance of three commonly used implant metals Alloy

Corrosion potentiala (mV vs. SCE)

Passive current density (A/cm2)

Breakdown potential (mV vs. SCE)

ÿ480 ÿ140 ÿ150

4.85 3.06 2.95

>1500 420 120

TiAl6V4 CoCr29Mo5 AISI 316L

The parameters denoted are from our own experiments produced using the set-up shown in Fig. 6.6 and detailed in the form of a graphical presentation in Fig. 6.8 a The lower the corrosion resistance, the lower is this Ecorr value. © 2008, Woodhead Publishing Limited

Metals for joint replacement

141

resistance it can also serve as a supplementary corrosion barrier when new protective coatings are developed (Kiuru et al., 2002). Bulk tantalum exists normally in bcc -phase and is ductile and relatively soft (HV 0.9 GPa) material that can, for example, be easily drawn into fine heating filaments. However, some deposition conditions of tantalum coatings can produce tetragonal -phase that is thermally unstable and brittle and may cause coating failures (Matson et al., 2000). Globally the main uses of tantalum are in electrolytic capacitors, vacuum furnace parts, nuclear reactors and military industry (Lide, 2001). In the medical industry the high melting point (2996 ëC) and somewhat elevated manufacturing costs of tantalum have restricted the use of bulk tantalum and led to alternative approaches. The most common tantalum-based material used in human implants goes under trade name Trabecular metalÕ. In its manufacturing process carbon foam is tantalum coated with chemical vapour deposition (CVD). The resulting 80% porous trabecular scaffold consists of 99% tantalum and 1% vitrous carbon. The modulus of elasticity is 3 GPa and the pore size 550 m allowing the bone growth into the material (Levine et al., 2006). According to the manufacturer it approximates the physical and mechanical properties of bone more closely than any other prosthetic material (Bobyn et al., 1999).

6.9

Particle disease

Some of the particles formed from metal are so small that they are phagocytosed. Monocyte/macrophages try to digest the metal (or polymer) particles, but without success. This can lead to recruitment of more haematogenous monocytes to the site of inflammation, their maturation to macrophages and multinuclear giant cells and organisation to foreign body granulomas. This socalled foreign body reaction is associated with local production of proinflammatory cytokines, such as tumour necrosis factor- and interleukin-1 . Various proteinases, including matrix metalloproteinases and cathepsin K, are produced. Finally, growth and differentiation factors are produced. These include macrophage-colony stimulating factor and receptor activator of nuclear factor kappa B ligand (RANKL), which further enhance formation of both foreign body giant cells and osteoclasts. Osteoclasts mediate periprosthetic osteolysis and loosening in the long term. Foreign body reaction or `particle disease' is considered to play a central role in aseptic loosening of total joint implants. Therefore, it has been suggested, that the toxicity of cobalt±chrome could be an advantage. Phagocytosis of cobalt±chrome particles can lead to apoptosis and necrosis of macrophages and fibroblast. This could have a moderating effect on the foreign body reaction.

© 2008, Woodhead Publishing Limited

142

6.10

Joint replacement technology

Clinical success of metals used in joint replacement surgery

Aseptic loosening is the most common mode of failure of a total joint arthroplasty (Malchau et al., 1993). Therefore, success will in this section be considered only regarding this particular end point. Metals are predominantly used for bodies of artificial joints that are intended to be incorporated to the adjacent bone either by biological fixation or by means of PMMA (bone cement) anchoring. Contemporary articulating surfaces consist of a metal part articulating with a plastic polymer (UHMWPE in most cases), but other types of bearings such as ceramic on polyethylene, metal-on-metal, and ceramic-on-ceramic are also widely used. The last two types are only used in total hip arthroplasties. Titanium alloys are also used for articulating surfaces, but more rarely. Their use for this purpose is not recommended owing to low wear resistance (Salvati, 1993; Massoud et al., 1997). It is convenient to consider clinical performance of the metals in the prosthetic bodies separately from the metals for the articulating part as they have different relationships to implant failures. All constituents of a total joint arthroplasty may be involved in the pathways leading to clinical failure and can be considered as links in a chain, the weakest defining the overall level of performance. The bearing couple usually presents the weakest link due to release of harmful particles and metal ions that constitute the first step in the cascade of macrophage- or lymphocyte-mediated periprosthetic osteolysis and final failure.

6.10.1 Statistical problems with outcome studies Although the clinical success of various prostheses has been reported in the literature for over 40 years, which type of implant is best for which patient is still controversial. A valid comparison between two prosthesis designs can be made only if the patient populations are similar, if the outcomes are the same and are measured in a similar fashion using similar follow-ups and analyzed using adequate statistical techniques. The confusion originates particularly from the shortage of head-to-head comparisons in long-term randomised clinical studies of new designs with the already established prosthesis. Further uncertainty stems from the large and continuously increasing variability of prostheses in use, the incorrect or incomplete data presentation together with the lack of independent assessment or conclusions from inadequate follow-up. Early studies usually lacked adequate methods for survivorship analysis and subsequent studies have often used surrogate variables such as `radiographic loosening' rather than revision surgery as the primary outcome measurements due to relative shortage of early revisions. `Loose' has, however, been defined in various ways that have added to the confusion. As only a few authors have used © 2008, Woodhead Publishing Limited

Metals for joint replacement

143

confidence intervals, the overall impression of the published results may be too optimistic.

6.10.2 Stainless steel In November 1962 Sir John Charnley originated the modern era of artificial joints with the introduction of an HMWPE cup paired to an already welldocumented stainless steel stem, both fixed with cement, for total hip arthroplasty. No prosthesis seems to have surpassed the cemented stainless steel Charnley in terms of length of service and universality of its use. Thus, the cemented Charnley design represents the gold standard to which all other prostheses should be compared due to its widespread use and consistent reports of its outcomes. The same applies for the material it was made of, namely stainless steel. Initially, `flat back' polished stem made of EN 58J stainless steel was replaced with high nitrogen content stainless steel (0.35±0.50%, ORTRON stem; De Puy International, Leeds, UK) due to high incidence of stem fatigue fractures (Wroblewski, 1982). Despite the diversity of institutions and authors reporting, an amazing consistency is seen in that in many studies, as long as 15 years after the index surgery, about 90% of the prostheses among the surviving patients were still in situ. Some reports with 20 years or more of follow-up indicate that at 20 years, between 80 and 90% of the prostheses will be surviving (Ahnfelt et al., 1990; Hozak et al., 1990; Joshi et al., 1993; Berry and Hamsen, 1998; Wroblewski et al., 1998).

6.10.3 Cobalt±chrome alloy Many cemented prostheses made of cobalt±chrome alloy have come close to or, in certain studies, even surpassed the cemented Charnley. One of them was certainly the T(trapezoidal)28 (Zimmer, Warsaw, IN). In patients who were 60 years or older, survival of the implant was 95%, 15 years after implantation. However, if `radiographic loosening' was considered as failures, over 35% of older patients failed (Amstutz et al., 1998). The Saint George Mark I and Mark II have reached similar longevity with less than 30% of failure rate after more than 20 years of follow-up (Engelbrecht and Kluber, 1998). Lubinus SP II Link has also shown adequate performance according to Scandinavian hip registries and head-to-head comparisons (Jacobsson et al., 1995; Havelin et al., 2000). Cobalt±chrome alloy has also been used in cementless fashion with variable success. It seems that surface finish determines the overall osseointegration potential and thus outcome (Sotereanos et al., 1995). Porous coating (pore size 50±400 m) results in highest in-growth potential. Cumulative probability of survival reached 96% at 12 years for patients more than 60 years of age. Osteolysis was, however, seen in 40% of cases (including patients younger than © 2008, Woodhead Publishing Limited

144

Joint replacement technology

60) at 10 years' follow-up (Engh, 1998). Some systems such as Tri-lock femoral stems (DePuy, Warsaw, IN) and PCA (Howmedica, Rutherford, NJ) have reached nearly 100% and 95% survivorship, respectively, after 12±15 years of follow-up (Pellegrini et al., 1998; Kawamura et al., 2001).

6.10.4 Titanium alloy (TiAl6V4 or NiAlNb) Titanium (Ti) alloys are widely used for the manufacture of orthopaedic implants in total hip arthroplasty both for cemented and cementless prostheses. Three main types of Ti alloys are in use: commercially pure titanium (CPTi), TiAl6V4 and TiAlNb. The former are mostly used for low-profile cementless cups, i.e. BiconÕ (Plus Orthopedics), AllofitÕ (Zimmer) and TOP acetabular cup systemÕ (Waldemar Link). The properties of Ti alloys include good biocompatibility, low modulus of elasticity and their resistance to fatigue and corrosion (Kohn and Ducheyne, 1992; Willert et al., 1996). Cementless stems are more resistant to osteolysis and mechanical failure when compared with similar cemented stems (Head et al., 1995; Emerson et al., 2002; Laupacis et al., 2002).

6.10.5 Cementless fixation Based on their ability to allow bone in-growth, Ti alloys are the material of choice for cementless fixation. There are innumerable types of cementless Tialloy devices uniformly showing excellent performance. In Europe one of the most popular is the Zweymuller type (VariallÕ, Zimmer, SL PlusÕ, Plus Orthopedics) showing excellent survivorship (Table 6.4). In some recent publications a titanium alloy stem (CorailÕ, De Puy, Warsaw, IN) and Anatomic Hip (Zimmer, Warsaw, IN) showed no cases of aseptic loosening after an average 11.5 and 10 years of follow-up, respectively, with no distal osteolysis and no thigh pain in the former (Archibeck et al., 2001; Froimson, 2007).

6.10.6 Cemented fixation In contrast to the good results obtained for cementless Ti alloy stems, the results for cemented Ti alloy stems are inconclusive and need a more detailed discussion. Cemented Ti alloy monoblock stems were introduced in the 1970s (Sarmiento and Gruen, 1985; McKellop and Clarke, 1988) and the initial results were satisfying. The mid-term results for CoCr headed stems, such as BicontactÕ (Eingartner et al., 2001) and UltimaÕ (Bowditch and Villar, 2001), were also promising, showing a survival rate of 97% at 10 and 7.5 years, respectively. Straight Muller stems made of TiAlNb with a ceramic head showed the revision rate of 2.5% at average 6 year's follow-up (Rader et al., 2000). Intermediate results depending on the cementation technique and head © 2008, Woodhead Publishing Limited

Table 6.4 Some recent clinical experiences with titanium alloy total hip replacements Reference

Hips (n)

Stem/cup type

GrÏbl et al. (2002)

208

Zweymuller/ alumina head/CSF cup

Mallory (1998)

103

Mallory/head system

Archibeck et al. (2001)

74

Anatomic hip/HG II cup (commercially pure Ti proximally)

Traulsen et al. (2001)

113

Zweymuller stem/ various cups

Siebold et al. (2001)

227

CLS Spotorno stem/ HG and Morsher press fit cup

Froimson et al. (2007)

147

Corail stem

Mean follow-up (year)

Mean age (year)

10

61

99% for stem; 93% for cup; 92% for any revision at 10 years

7

51

96% for stem; patients younger than 60 years

10

52

100% for stem, 96% for cup; 92% for any revision at 10 years

9

56

96% for stem at 10 years

12

55

94% for stem at 12 years

11.5

63

100% for stem

© 2008, Woodhead Publishing Limited

WPTF3007

Probability of survival

146

Joint replacement technology

material were reported for the long term (Kovac et al., 2006). Good long-term results with 87.3% survival at 20 years were reported with Ceraver±OstealÕ polished and titanium oxide coated stems which, however, were combined with alumina on alumina coupling (Hamadouche et al., 2002). On the other hand, Maurer et al. (2001) reported an increased loosening of cemented straight Muller stems made of TiAl6Nb7 alloy compared with CoNiCr alloy stems at a median of follow-up of 7.7 years. Unacceptable failure rates of 4.5% (Tompkins et al., 1994), 9% (Jacobson et al., 1995) and 11.5% (Jergesen and Karlen, 2002) at 4.8, 5 and 5.5 years, respectively, were reported for Ti alloy cemented stems. The results for CapitalÕ modular Ti alloy stems with either CoCr or TiN coated titanium heads were poor, with 16% loose femoral components at a mean follow-up of 26 months (Massoud et al., 1997). The surface finish of the Ti alloy stem and its elasticity undoubtedly has an important effect on the performance of the prosthesis, with more polished surfaces and larger (less elastic) designs performing better than rough and more elastic models (Agins et al., 1988; Sedel et al., 1990; Jacobson et al., 1995; Massoud et al., 1997; KaÈrrholm et al., 1998; Le Mouel et al., 1998; Sarmiento et al., 1998; Sporer, 1999; Maurer et al., 2001; Collis and Mohler, 2002; Emerson et al., 2002; Jergesen and Karlen, 2002; Janssen et al., 2005). Increased cement± stem interface motion and, accordingly, fretting wear at the interface depending also on their surface finish and elasticity (McKellop et al., 1990; Jacobson et al., 1995) resulted in increased generation of wear debris leading to aseptic loosening (Agins et al., 1988; Witt and Swann, 1991; Salvati et al., 1993).

6.10.7 Metals used for articulating surfaces Metal-on-metal (often abbreviated to M/M) articulating surfaces were introduced for total hip replacements at the same time as metal-on-polyethylene (M/P). However, M/M implants were largely abandoned because of some early failures and excellent early results of Charnley's low-friction arthroplasty at about the same time. Long-term problems with M/P articulating hips that developed serious osteolyses and consequent loosening due to polyethylene particles induced wear (Willert et al., 1978), together with favourable long-term results with selected patients with first generation M/M coupling that showed virtually no osteolysis after more than 20 years of performance (KreuschBrinker et al., 1998) resulted in development of second generation of M/M bearing couples. In the contemporary M/M bearings, most of the critical design parameters that seem to ensure clinical success, i.e. high carbon materials, proper clearance between components, surface finish and roundness, can be achieved with modern manufacturing techniques. Metals differ by the manufacturer; castings or forgings are used for processing in various fashions, like-on-like material combinations are used in the vast majority. Second generation of metal-on-metal couples made of CoCrMo alloy originated © 2008, Woodhead Publishing Limited

Metals for joint replacement

147

in 1988 with the introduction of the Metasul bearing couples (Weber et al., 1992) with the intention to achieve better long-term clinical results than conventional total hip replacements by minimising and possibly eliminating polyethylene wear as a cause of osteolysis and aseptic loosening (Weber, 1992). All contemporary metal-on-metal couples are considered second-generation implants. Manufacturers use CoCr28Mo6 alloys (ASTM F799 and ASTM 1537, ISO 5832-12) to produce M/M total hip replacements. They differ in their content of carbon which might be clinically important. High-carbon alloys contain >0.2 wt% carbon and low-carbon alloys <0.07 wt%. MetasulÕ for instance, is a high-carbon CoCrMo alloy, Sikomet SM 21Õ is a low-carbon alloy, whereas UltimaÕ is a combination of low-carbon head and high-carbon cup (MilosÏev et al., 2005). Low-carbon and low-carbon-on-high-carbon combinations express higher wear rate (Firkins et al., 2001; St. John et al., 2004) and are related to higher number of particles released in the periprosthetic tissue being thus implicated in the development of osteolysis. Use of M/M bearing in total hip surgery results in several advantages such as: 50±100 decreased wear rate over conventional polyethylene in the laboratory tests (McKellop, 2001), 40 decreased linear wear and 200 decreased volumetric wear over conventional polyethylene in vivo (Rieker and KoÈttig, 2002; Rieker et al., 2004), and possibly as a consequence of this, rare appearance of osteolysis (Park et al., 2005; MilosÏev et al., 2006). M/M bearing allows the use of large femoral heads with their improved stability, range of motion, and superior lubrication, concomitantly avoiding the risk of bearing fracture. With some exceptions (BoÈsch and Legenstein, 2004; MilosÏev et al., 2005; Park et al., 2005), all involve low-carbon and low-carbon-on-high-carbon combinations, demonstrating excellent clinical results (MacDonald, 2004). There are disadvantages, too, including: biologic/carcinogenic concerns of metal ions elevations especially in patients with renal failure and in women in childbearing age (Milosev et al., 2005; Ziaee et al., 2007). Those have never been confirmed but do not seem to exist with other bearings, and unclear but increasingly reported `deep' and partly local hypersensitivity to metals (Hallab et al., 2001; MilosÏev et al., 2005; Willert et al., 2005). A summary of clinical results is presented in Table 6.5, which has been updated from our earlier paper (MilosÏev et al., 2006). It is reasonable to conclude that low-carbon M/M bearings produce inferior clinical results, probably because of increased wear in comparison to high-carbon MM bearings.

6.11

Future trends

6.11.1 Resurfacing implants and mini-invasive surgery Hip resurfacing arthroplasty, considered an old, unsuccessful orthopaedic concept, has undergone a resurgence of interest in the past decade with the advent of the © 2008, Woodhead Publishing Limited

Table 6.5 Literature analysis on clinical results obtained on metal-on-metal bearing hip implants (modified from Milosíev et al., 2006) Reference

Bearing

Acetabular component

No. of hips

Mean age (year)

Mean follow-up (range) (year)

Revisions (failures)

Weber (1992)

Metasul

Weber cup

100

59

3.5 (2±7)

5

Hilton et al. (1996)

Metasul

Weber cup

74

71

2.2 (0.5±4)

1

Dorr et al. (2000)

Metasul

Weber cup

56

70

5.2 (4±6.8)

3 (1)

Wagner and Wagner (2000)

Metasul

47 conical, 28 all-metal cups

75

49

5 (3.6±7.3)

3

Jessen et al. (2004)

Metasul

Metasul cups

88

62

7

2

Long et al. (2004)

Metasul

47 APR, 114 IntraOp cups

161

55

6.5 (2±9)

5

Migaud et al. (2004)

Metasul

Hemisph. cup

39

39

5.7 (5.1±6.6)

0

© 2008, Woodhead Publishing Limited

WPTF3007

Survivorship

98.2% at 7 years for loosening of cup; 94.1% at 7 years for any reason

97.5% at 8 years

100% at 5 years

Delaunay (2004)

Metasul

CSF cup

100

60

6 (1.4±10.5)

5

Lombardi et al. (2001)

Biomet

M2a cup

78

49

3.29 (1.64±5.40)

10

Lombardi et al. (2004)

Biomet

M2a cup

53

50

5.7 (5±8)

10

100% at 5 years

BÎsch et al. (2004)

Biomet

PPF system

148

NA

8.2

34 (16)

80% at 8.2 years

Jacobs et al. (2004)

Ultima

Ultima cup

95

53

3.9 (3.0±5.7)

1

Korovessis et al. (2003)

Sikomet

Bicon±Plus

350

55

4.3 (3.1±7.7)

8

99.4% for cup; 96.8% for stem at 7.6 years

MacDonald (2004)

Biomet

M2a cup

22

61

3.2 (2.2±3.9)

0

100% survival

Randle and Gordiev (1997)

Metasul

57

63

± (5 months± 2.6 years)

0

100% survival

Milosíev et al. (2006)

Sikomet

640

57

7.1 (2.3±10.5)

34

91% both components for any reason at 10 years

Bicon±Plus

© 2008, Woodhead Publishing Limited

WPTF3007

98.7% (81%) at 8 years for any reason for hip without (with) sleeve

150

Joint replacement technology

second generation of M/M bearings in the 1990s. It is the most anatomical way of hip replacement, where the femoral head is not removed but instead modelled, to allow the coverage with a low-profile resurfacing prosthesis (Fig. 6.10). It provides superior stability for the hip joint and improved range of motion defined by individual head/neck ratio, normalised biomechanics while avoiding the problems of leg length inequality (as the sizes of the `ball' and the length of the collum are anatomical). It conserves and preserves bone for a potential revision surgery. The disadvantages, besides those inherent to M/M bearings or metallosis (MilosÏev et al., 2000b, 2005; Hallab et al., 2001; MacDonald, 2004) include demanding surgical technique not suitable for low-volume surgeon, increased exposure ± more soft tissue releases, relatively small fixation area for the femoral component especially with tiny femoral heads, new potential modes of failure as avascular necrosis of the remaining bone and neck fracture (Shimmin

© 2008, Woodhead Publishing Limited

Metals for joint replacement

151

et al., 2005). The early and mid-term results seem promising at lest for a selected patient population (Amstutz et al., 1998, Australian registry, http:// www.dmac.adelaide.edu.au/aoanjrr/publications.jsp). In the years after 2000, several articles emerged in the literature regarding minimally invasive hip and knee surgery (MIS). It can be performed from short incisions (Fig. 6.9a), via which the femoral stem (Fig. 6.9b) and a modular head (Fig. 6.9c) are implanted. The issue became a major topic in contemporary orthopaedics with substantial media coverage that increased patient and surgeon interest. For all of the standard approaches a less invasive variant come into sight, each of them having its own proponent. Some were definitely supported and propagated by industry. In hip arthroplasty surgery mini-anterior, minianterolateral, mini-direct lateral, mini-posterior and a novel mini two-incision (Berger, 2003) dominated the literature. In the knee a mini subvastus, a mini mid-vastus and a quad sparing approach were its counterparts. The main advantages were claimed to be less or even no damage to soft tissues, a better cosmetic result, less blood loss and a quicker recovery after the surgery. The disadvantages included a potential of misplacement of the components and more tissue damage due to reduced vision and difficult extensibility of the approach when needed. A special attention and extensive promotion was given to a novel twoincision approach (Berger, 2003, 2004; Berry et al., 2003). The authors claimed no muscle or tendon damage, rapid rehabilitation and even the potential for 6.10 (opposite) (a) A metallic CrCoMo femoral component of the resurfacing hip replacement implant. (b) A metal-to-metal resurfacing total hip replacement implant with the femoral component mounted in the acetabular cup. The cup is made of cobalt±chrome alloy with sintered beads on the outer surface to enable cementless fixation. The femoral component is made of the same material but fixed with cement, although there are also cemenless femoral components on the market. They are increasingly used for young patients whose life expectancy exceeds that of the implant survival (so that they are likely to need a revision operation). The potential advantage of the bone-saving primary operation is mainly easier revision. The eventual revision operation is similar to the primary total hip replacement. It also usually avoids risks of leg lengthening and offers an anatomical offset. Patients may also better retain proprioception (osseoperception), stress shielding in the femoral shaft is prevented and the natural size of the ball reduces the incidence of dislocations. (c) Plain radiograph of a post-traumatic osteoarthritic hip in a 41-year-old patient before a resurfacing operation. The joint space is very narrow, the joint margins contain osteophytes and several loose bodies surround the joint. This hip is very suitable for a resurfacing implant because of good bone quality, large femoral head, long neck, lack of cystic changes and valgus anatomy. (d) A metal-to-metal resurfacing total hip cup is implanted at an angle of about 45ë and the femoral part should preferably be about 140ë to femoral anatomic axis. Saving of the femoral neck offers some advantages, but also introduces some new modes of failure such as aseptic necrosis of the remaining femoral head and collum fractures. © 2008, Woodhead Publishing Limited

152

Joint replacement technology

outpatient surgery. Soon after the early promotion, heavy criticisms were published showing that no scientific data supported the notion that the twoincision approach is functionally better than any other total hip arthroplasty approach (Archibeck and White, 2004). Cadaver work studies dispelled the belief that the two-incision approach can be done without damaging muscle or tendon (Mardones et al., 2005). The actual incidence of perioperative complications was dangerously high (Bal et al., 2005; Pagnano et al., 2005). In a direct comparison patients preferred other approaches when applied on the contralateral hip (Pagnano et al., 2006). Between conventional THR and resurfacing THR, there is a particular series of implants that require resection of the femoral head for the implantation but not the neck (Fig. 6.11). Actually they are intended to be biologically fixed or in

6.11 The orthopaedic community is always in search of an ideal hip implant. The figure depicts a bone-sparing titanium alloy thrust plate hip endoprosthesis (TPP). It is inserted into the femoral neck and secured just below the greater trochanter with a thrust plate that is fixed with two screws in the metaphyseal bone for primary fixation. The device relies on proximal femoral metaphyseal bone ingrowth for fixation. This has the theoretical advantage of leaving diaphyseal bone intact for easier conversion to a stemmed prosthesis. It is an alternative to the resurfacing implants. The ball has a mirror surface and below it there is a collar for a more physiological load transmission to avoid stress transfer and thus thigh pain, which can occur when stiff intramedullary devices are used. The potential disadvantage is its reduced surface area in contact with bone. Already discrete proximal osteolysis that would yet not be detrimental for a stemmed prosthesis could lead to loosening of such a short prosthesis. It is thus advisable to couple it with a hard-on-hard bearing system to avoid wear debris formation and development of osteolysis. © 2008, Woodhead Publishing Limited

Metals for joint replacement

153

other words osseo-integrated into the femoral neck bone. One of the first implants of this particular type was the Huggler±Jacob prosthesis in the 1980s (Huggler and Jakob, 1980). Today, there is a resurgence of this bone-sparing design. Among the most interesting are Gothenburg hip inspired by dental implants (Carlsson et al., 2006a,b), DSP hip (Zimmer), the contemporary version of the Huggler implant (Jacob et al., 2007) and Proxima and Silent hip (Johnson&Johnson). These short implants are bone friendly and very easy to implant even using the least invasive approaches (Santori et al., 2005, 2006). On the other hand, they probably need a certain unloading period to allow integration. Their survival is very dependent on the proximal femoral bone quality. Even very limited osteolysis can be detrimental for these types of implant. It is thus advisable to avoid polyethylene bearings with this particular design. There is some data regarding survivorship of these hips that show promising results that, however, cannot compare to the best conventional hips.

6.11.2 Isoelasticity Studies have shown that by increasing stiffness of the stem, either by stiffer material or by increasing cross-sectional dimensions, the amount of load carried by proximal femur decreases, resulting in reduced stress transfer and bone resorption in the proximal femur thus compromising the fixation of the proximal stem (Crowninshield et al., 1980; Huiskes, 1980; Lewis et al., 1984). To overcome the mismatch between a stiff stem and the more elastic bone, the concept of isoelasticity was introduced in 1970s (Fig. 6.12). This concept was based on the assumption that the implant and the bone should deform as one unit to avoid stress shielding. Studies confirmed almost no loss of bone stock from the proximal femur or the acetabulum, even in cases where the implants were clearly loose, with use of isoelastic implant (Horne et al., 1987; NiinimaÈki and Jalovaara, 1995). However, decreasing the stiffness excessively resulted in higher implant±bone or implant± cement motion, leading to early debonding and failure (TrebsÏe et al., 2005). Computer-simulated models have proven high proximal stem/bone interface stresses, which may cause interface debonding and relative motions, possibly affecting implant loosening (Burke et al., 1991; Huiskes et al., 1992). Thigh pain that occurs in different percentages in patients with bipolar total hip replacement is believed to be a result of stiffness mismatch between stem and bone. In one particular design (Anatomic Porous Replacement; Sultzer Orthopedics, Austin, TX) the clinical incidence of thigh pain was significantly reduced after the stem was hollowed to decrease the stiffness below that of the bone (Dorr and Wan, 1996). Therefore, when introducing the concept of isoelasticity, one should aim for an optimal stem flexibility, to diminish stress shielding but to keep interface stresses low enough. It is probable that the optimal elasticity of an implant is © 2008, Woodhead Publishing Limited

154

Joint replacement technology

6.12 Isoelastic Robert Mathys cementless total hip replacement implant. Isoelastic implants were developed to overcome the mismatch between a stiff stem and the more elastic bone. The stem is produced of polyacetal resin around a metal core that extends proximally forming the neck of the prosthesis. The elasticity of the device comes close to that of the host bone. This part of the implant is fixed into the pre-shaped femoral canal bone. On the neck of the prosthesis a modular ball is inserted that articulates with a screw-in titanium plasma sprayed polyethylene cup, which is additionally fixed with two slightly eccentric pegs and with shallow circular grooves which allow bone in-growth. © 2008, Woodhead Publishing Limited

Metals for joint replacement

155

different for every patient concerning bone quality, shape and dimensions; a task difficult to achieve even with a customised implant.

6.11.3 Coating of implants Metal implants interact with their surrounding via their surface. The properties of the bulk can be fine tuned using specialised and purpose-designed coatings. Versatility and functionality have and will be extensively introduced to implants, which may have different functional domains, e.g. against bone or bone cement, and in an articular or modular gliding pair. One very promising approach is high-quality amorphous diamond, rich in diamond (sp3) bonds, produced from industrial graphite using plasma acceleration in a pulse arc discharge setting. It basically eliminated the formation of wear debris from the gliding surfaces and of corrosion products from the implant surface in contact with body fluids (reviewed in Santavirta, 2003).

6.12

Acknowledgements

This study was supported by evo funds, Sigrid JuseÂlius Foundation, Finska LaÈkaresaÈllskapet, COST 537 `Core laboratories for the improvement of Medical Devices in Clinical Practice from the Failure of the Explanted Prostheses Analyses FEPA' and COST 533 `Materials for Improved Wear Resistance of Total Artificial Joints'.

6.13

Sources of further information and advice: useful websites

http://www.dmac.adelaide.edu.au/aoanjrr/publications.jsp http://mrsec.wisc.edu/Edetc/IPSE/educators/amMetal.html http://www.liquidmetal.com/applications/dsp.medical.asp http://en.wikipedia.org/wiki/Precipitation_hardening http://www.dmac.adelaide.edu.au/aoanjrr/publications.jss

6.14

References

Agins H J, Alcock N W, Bansal M, Salvati E A (1988), `Metallic wear in failed titaniumalloy total hip replacements. A histological and quantitative analysis', J Bone Joint Surg [Am], 70-A, 347±56. Ahnfelt L, Herberts P, Malchau H, Anderson G B J (1990), `Prognosis of total hip replacement. A Swedish multicenter study of 4,664 revisions', Acta Orthop Scand, suppl. 238, 61, 1±25. Amstutz H, Dorey F, Finerman G A M (1998), `The cemented T28/TR28 prosthesis', in Finerman G A M, et al., Eds: Total Hip Arthroplasty Outcomes, Kidlington, Churchill Livingstone, 55±63. © 2008, Woodhead Publishing Limited

156

Joint replacement technology

Anttila A, Lappalainen R, Heinonen H, Santavirta S, Konttinen Y T (1999), `Superiority of diamondlike carbon coating on articulating surfaces of artificial hip joints', New Diamond Frontier Carbon Technology, 9, 283±8. Archibeck M J, White R E (2004), `Learning curve for the two-incision total hip replacement', Clin Orthop, 429, 232±8. Archibeck M J, Berger R A, Jacobs J J, Quigley L R, Gitelis S, Rosenberg A G, Galante J O (2001), `Second-generation cementless total hip arthroplasty. Eight to elevenyear results', J Bone Joint Surg [Am] , 83-A, 1666±73. Bal B S, Haltom D, Aleto T, Barrett M (2005), `Early complications of primary total hip replacement performed with two-incision minimally invasive technique', J Bone Joint Surg [Am] , 87-A, 2432±8. Berger R A (2003), `THA using the minimally invasive two incision approach', Clin Orthop, 417, 232±41. Berger R A (2004), `Rapid rehabilitation and recovery with minimally invasive total hip arthroplasty', Clin Orthop, 429, 239±47. Berry D J, Berger R A, Callaghan J J, Dorr L D, Duwelius P J, Hartzband M A, Lieberman J R, Mears D C (2003), `Minimally invasive THA: development early results, and critical analysis. Presented at Annual Meeting of the AOA, Charleston SC June 14, 2003', J Bone Joint Surg [Am] , 85-A, 2235±46. Berry D J, Hamsen W S, (1998), `The Charnley The Mayo Clinic. Total hip artroplasty outcomes', in Finerman G A M, et al., eds: Total Hip Arthroplasty Outcomes, Kidlington, Churchill Livingstone, 31±40. Bobyn J D, Stackpool G J, Hacking S A, Tanzer M, Krygier J J (1999), `Characteristics of bone ingrowth and interface mechanics of a new porous tantalum biomaterial', J Bone Joint Surg [Br], 81-B, 907±14. Bockris J O'M, Reddy A K N, Eds. (2000), Modern Electrochemistry, Vol. 2B, 2nd edn, New York, Kluwer Academics/Plenum Publishers. BoÈsch P, Legenstein R (2004), `Plus and Minus der Metall-Metall-Paarung' [German], OrthopaÈdie, 5, 16±19. Bowditch M, Villar R (2001), `Is titanium so bad? Medium-term outcome of cemented titanium stems', J Bone Joint Surg [Br] , 83-B, 680±5. Burke D W, O'Connor D O, Zalenski E B, Jasty M, Harris W M (1991), `Micromotion of cemented and uncemented femoral components', J Bone Joint Surg [Br], 73, 33±7. Callister Jr W D (2000), Materials Science and Engineering ± An Introduction, 5th edn, New York, John Wiley & Sons Inc. Carlsson L V, Albrektsson B E, Albrektsson B G, Albrektsson T O, Jacobsson C M, Macdonald W, Regner L, Rostlund T, Weidenhielm L R (2006a), `Stepwise introduction of a bone-conserving osseointegrated hip arthroplasty using RSA and a randomized study: I. Preliminary investigation ± 52 patients followed for 3 years', Acta Orthop, 77, 549±58. Carlsson L V, Albrektsson T, Albrektsson B E, Jacobsson C M, Macdonal W, Regner L, Weidenhielm L R (2006b), `Stepwise introduction of a bone-conserving osseointegrated hip arthroplasty using RSA and a randomized study. II. Clinical proof of concent ± 40 patients followed for 2 years', Acta Orthop, 77, 559±66. Collis D K, Mohler C G (2002), `Comparison of clinical outcomes in total hip arthroplasty using rough and polished cemented stems with essentially the same geometry', J Bone Joint Surg Am , 84-A, 586±92. Crowninshield R D, Brand R A, Johnston R C, Milroy J C (1980), `An analysis of femoral component stem design in total hip arthroplasty', J Bone Joint Surg Am, 62-A, 68± 78. © 2008, Woodhead Publishing Limited

Metals for joint replacement

157

Delaunay C P (2004), `Metal-on-metal bearings in cementless primary total hip arthroplasty', J Arthroplasty, 19(8 Suppl 3), 35±40. Dorr L D, Wan Z (1996), `Comparative results of a distal modular sleeve, circumferential coating, and stiffness relief using the APR-II', J Arthroplasty, 11, 419±28. Dorr L D, Wan Z, Longjohn D B, Dubois B, Murken R (2000), `Total hip arthroplasty withuse of the Metasul metal-on-metal articulation. Four to seven-year results', J Bone Joint Surg [Am], 82, 789±98. Eingartner C, Volkmann R, Winter E, Maurer F, Ihm A, Weller S, Weise K (2001), `Results of cemented titanium alloy straight femoral shaft prosthesis after 10 years of follow-up', Int Orthop, 25, 81±4. Emerson R H, Head W C, Emerson C B, Rosenfeldt W, Higgins L L (2002), `A comparison of cemented and cementless titanium femoral components used for primary total hip arthroplasty', J Arthroplasty, 17, 584±91. Engelbrecht E, Kluber S D (1998), `The model St. George/Mark I/Mark II prosthesis', in Finerman G A M, et al., eds: Total Hip Arthroplasty Outcomes, Kidlington, Churchill Livingstone, 65±83. Engh C A (1998), `The anatomic medullary locking prosthesis', in Finerman G A M, et al., eds: Total Hip Arthroplasty Outcomes, Kidlington, Churchill Livingstone, 117±39. Firkins P J, Tipper J L, Saadatzadeh M R, Ingham E, Stone M H, Farrar R, Fisher J (2001), `Quantitative analysis of wear and wear debris from metal-on-metal hip prostheses tested in a physiological hip joint simulator', Bio-Medical Materials and Engineering, 11, 143±57. Froimson M I, Garino J, Machenaud A (2007), `Minimum 10-year results of tapered, titanium, hydroxyapatite-coated hip stem. Independent review', J Arthroplasty, 22, 1±7. GruÈbl A, Chiari C, Gruber M, Kaider A, Gottsauner-Wolf F (2002), `Cementless total hip arthroplasty with a tapered, rectangular titanium stem and a threaded cup: a minimum ten-year follow-up', J Bone Joint Surg [Am], 84-A, 425±31. Hallab N, Merritt K, Jacobs J J (2001), `Metal sensitivity in patients with orthopaedic implants', J Bone Joint Surg [Am] , 83-A, 428±36. Hamadouche M, Boutin P, Daussange J, Bolander M E, Sedel L (2002), `Alumina-onalumina total hip arthroplasty: a minimum of 18.5-year follow-up study', J Bone Joint Surg [Br] , 84-B, 69±77. Havelin L I, Engesaeter L B, Espehaug B, Furnes O, Lie S A, Vollset S E (2000), `The Norwegian Arthroplasty Register: 11 years and 73,000 arthroplasties', Acta Orthop Scand, 71, 337±53. Head W C, Bauk D J, Emerson R H (1995), `Titanium as the material of choice for cementless femoral components in total hip arthroplasty', Clin Orthop Relat Res, 311, 85±90. Hilton K R, Dorr L D, Wan Z, McPherson E J (1996), `Contemporary total hip replacement with metal on metal articulation', Clin Orthop Relat Res, 329 Suppl, S99±105. Hodgson A W E, Kurz S, Virtanen S, Fervel V, Olsson C-O A, Mischler S (2004), `Passive and transpassive behavior of CoCrMo in simulated biological solutions', Electrochimica Acta, 49, 2167±78. Horne G, Berry N, Collis D (1987), `Isoelastic uncemented hip arthroplasty ± early experience', Aust N Z J Surg, 57, 461±6. Hozack W J, Rothman R J, Booth R E, Balderston R A, Cohn J C, Pickens G T (1990), `Survivorship analysis of 1041 Charnley total hip arthroplasties', J Arthroplasty, 5, 41±7. © 2008, Woodhead Publishing Limited

158

Joint replacement technology

Huggler AC, Jacob HA (1980), `A new approach towards hip-prosthesis design', Arch Orthop Trauma Surg, 97, 141±4. Huiskes R (1980), `Some fundamental aspects of human joint replacement. Analyses of stresses and heat conduction in bone±prosthesis structures', Acta Orthop Scand, Suppl 185, 1±208. Huiskes R, Weinans H, van Rietbergen B (1992), `The relationship between stress shielding and bone resorption around total hip stems and the effects of flexible materials', Clin Orthop Relat Res, 274, 124±34. Jacob H A, Bereiter H H, Buergi M L (2007), `Design aspects and clinical performance of the thrust plate hip prosthesis', Proc Inst Mech Eng [H], 221, 29±37. Jacobs M, Gorab R, Mattingly D, Trick L, Southworth C (2004), `Three- to six-year results with the Ultima metal-on-metal hip articulation for primary total hip arthroplasty', J Arthroplasty, 19(Suppl 2), 48±53. Jacobsson S A, Ivarsson I, Djerf K, Wahlstrom O (1995), `Stem loosening more common with ITH than Lubinus prosthesis', Acta Orthop Scand, 65, 425±31. Janssen D, Aquarius R, Stolk J, Verdonschot N (2005), `Finite-element analysis of failure of the Capital Hip designs', J Bone Joint Surg [Br], 87-B, 1561±7. Jergesen H E, Karlen J W (2002), `Clinical outcome in total hip arthroplasty using a cemented titanium femoral prosthesis', J Arthroplasty, 17, 592±9. Jessen N, Nickel A, Schikora K, BuÈttner-Janz K (2004), `Metal/metal ± a new (old) hip bearing system in clinical evaluation. Prospective 7-year follow-up study', OrthopaÈde, 33, 594±602 [German], Erratum in OrthopaÈde, 33, 602. Joshi A B, Porter M L, Trail I A Hunt L P, Murphy J C, Hardinge K (1993), `Long-term results of Charnley low friction arthroplasty in young patients', J Bone Joint Surg [Br], 75-B, 616±23. KaÈrrholm J, Frech W, Nivbrant B, Malchau H, Snorrason F, Herberts P (1998), `Fixation and metal release from the Tifit femoral stem prosthesis', Acta Orthop Scand, 69, 369±78. Kawamura H, Dunbar M J, Murray P, Bourne R B, Rorabeck C H (2001), `The porous coated anatomic total hip replacement. A ten to fourteen year follow-up study of a cementless total hip arthroplasty', J Bone Joint Surg [Am], 83-A, 1333±8. Kiuru M, Alakoski E, Tiainen V-M, Lappalainen R, Anttila A (2002), `Tantalum as a buffer layer in diamond-like carbon coated artificial hip joints', J Biomed Mater Res Part B: Appl Biomater, 66B, 425±8. Kohn D H, Ducheyne P (1992), `Materials for bone and joint replacement' in Williams D F, ed., Materials Science and Technology: Medical and Dental Materials, Vol. 14, Weinheim, VCH, 31±109. Korovessis P, Petsinis G, Repanti M (2003), `Zweymueller with metal-on-metal articulation: clinical, radiological and histological analysis of short-term results', Arch Orthop Trauma Surg, 123, 5±11. Kovac S, TrebsÏe R, MilosÏev I, PavlovcÏicÏ V, PisÏot V (2006), `Long-term survival of the straight titanium stem', J Bone Joint Surg [Br] , 88-B, 1567±73. Kreusch-Brinker R, Schwetlick G, Sparmann M, Hoppe S (1998), `The McKee±Farrar prosthesis' in Finerman G A M, et al., eds: Total Hip Arthroplasty Outcomes, Kidlington, Churchill Livingstone, 249±74. Laupacis A, Bourne R, Roarbeck C, Feeny D, Tugwell P, Wong C (2002), `Comparison of total hip arthroplasty performed with and without cement. A randomized trial', J Bone Joint Surg [Am], 84-A, 1823±8. Le Mouel S, Allain J, Goutallier D (1998), `Ten-year survival analysis of 156 alumina polyethylene total hip arthroplasties', Rev Chir Orthop, 84, 338±45. © 2008, Woodhead Publishing Limited

Metals for joint replacement

159

Levine B R, Sporer S, Poggie R A, Della Valle C J, Jacobs J J (2006), `Experimental and clinical performance of porous tantalum in orthopedic surgery', Biomaterials, 27, 4671±81. Lewis J L, Askew M J, Wixson R L, Kramer G M, Tarr R R (1984), `The influence of prosthetic stem stiffness and of a calcar collar on stresses in the proximal end of the femur with a cemented femoral component', J Bone Joint Surg [Am], 66-A, 280±6. Lide D R, ed. (2001), CRC Handbook of Chemistry and Physics, 82nd edn, New York, CRC Press. Lombardi A V Jr, Mallory T H, Alexiades M M, Cuckler J M, Faris P M, Jaffe K A, Keating E M, Nelson C L Jr, Ranawat C S, Williams J, Wixson R, Hartman J F, Capps S, Kefauver C A (2001), `Short-term results of the M2a taper metal-on-metal articulation', J Arthroplasty, 16(Suppl 1), 122±8. Lombardi A V Jr, Mallory T H, Cuckler J M, Williams J, Berend K R, Smith T M (2004), `Mid-term results of a polyethylene-free metal-on-metal articulation', J Arthroplasty, 19(7 Suppl 2), 42±7. Long W T, Dorr L D, Gendelman V (2004), `An American experience with metal-onmetal total hip arthroplasties: a 7-year follow-up study', J Arthroplasty, 19(8 Suppl 3), 29±34. MacDonald S J (2004), `Metal-on-metal total hip arthroplasty: the concerns', Clin Orthop Relat Res, 429, 86±93. Malchau H, Herberts P, Ahnfelt L (1993), `Prognosis of total hip replacement in Sweden', Acta Orthop Scand, 64, 497±506. Mallory T H (1998), `Measurement of polyethylene wear in acetabular components inserted with and without cement. A randomized trial', J Bone Joint Surg [Am], 80A, 766. Mardones R, Pagnano M W, Nemanich J P, Trousdale R T (2005), `The Hip Society Frank Stinchfield Award: muscle damage after total hip arthropalsty done with twoincision and mini-posterior techniques', Clin Orthop, 441, 63±7. Massoud S N, Hunter J B, Holdsworth B J, Wallace W A, Juliusson R (1997), `Early femoral loosening in one design of cemented hip replacement', J Bone Joint Surg [Br], 79-B, 603±8. Matson D W, McClanahan U E D, Rice J P, Lee S L, Windover D (2000), `Effect of sputtering parameters on Ta coatings for gun bore applications', Surface Coatings Technol, 133±134, 411±16. Maurer T B, Ochsner P E, Schwarzer G, Sumacher M (2001), `Increased loosening of cemented straight stem prostheses made from titanium alloys. An analysis and comparison with prostheses made of cobalt±chromium alloy', Int Orthop, 25, 77±80. McKellop H A (2001), `Bearing surfaces in total hip replacement: state of the art and future developments', Instr Course Lecture, 20, 165±79. McKellop H A, Clarke I (1988), `Wear of artificial joint materials in laboratory tests', Acta Orthop Scand, 59, 342±57. McKellop H A, Sarmiento A, Schwinn C P, Ebramzadeh E (1990), `In vivo wear of titanium alloy hip prostheses', J Bone Joint Surg [Am], 72-A, 512±17. Migaud H, Jobin A, Chantelot C, Giraud F, Laffargue P, Duquennoy A (2004), `Cementless metal-on-metal hip arthroplasty in patients less than 50 years of age: comparison with a matched control group using ceramic-on-polyethylene after a minimum 5-year follow-up', J Arthroplasty, 19(8 Suppl 3), 23±8. MilosÏev I, Strehblow H-H (2000), `The behavior of stainless steels in physiological solution containing complexing agent studied by X-ray photoelectron © 2008, Woodhead Publishing Limited

160

Joint replacement technology

spectroscopy', J Biomed Mater Res, 52, 404±12. MilosÏev I, Strehblow H-H (2003), `The composition of the surface passive film formed on CoCrMo alloy in simulated physiological solution', Electrochimica Acta, 48, 2767±74. MilosÏev I, MetikosÏ-Hukovic M, Strehblow H-H (2000a), `Passive film on orthopaedic TiAlV alloy formed in physiological solution investigated by X-ray photoelectron spectroscopy', Biomaterials, 21, 2103±13. MilosÏev I, AntolicÏ V, MinovicÏ A, CoÈr A, Herman S, PavlovcÏicÏ V, Campbell P (2000b), `Extensive metallosis and necrosis in failed prostheses with cemented titaniumalloy stems and ceramic heads', J Bone Joint Surg [Br], 82-B, 352±7. MilosÏev I, PisÏot V, Campbell P (2005), `Serum levels of cobalt and chromium in patients with Sikomet metal±metal total hip replacements', J Orthop Res, 23, 526±35. MilosÏev I, TrebsÏe, KovacÏ S, Cor A, PisÏot V (2006), `Survivorship and retrieval analysis of Sikomet metal-on-metal total hip replacements at a mean of seven years', J Bone Joint Surg [Am], 88, 1173±82. NiinimaÈki T, Jalovaara P (1995), `Bone loss from the proximal femur after arthroplasty with an isoelastic femoral stem. BMD measurements in 25 patients after 9 years', Acta Orthop Scand, 66, 347±51. Pagnano M W, Leone J, Lewallen D G, Hanssen A D (2005), `Two incision THA had modest outcomes and some substantial complications', Clin Orthop Relat Res, 441, 86±90. Pagnano M W, Trousdale R T, Meneghini R M, Hanssen A D (2006), `Patients preferred a mini-posterior THA to a contralateral two-incision THA', Clin Orthop Relat Res, 453, 156±9. Park Y-S, Moon Y-W, Lim S-J, Yang J-M, Ahn G, Choi A-l (2005), `Early osteolysis following second-generation metal-on-metal hip replacement', J Bone Joint Surg [Am], 87-A, 1515±21. Paul J P (1967), `Forces transmitted by joints in the human body', Proc Inst Mech Eng, 181, 3J, 8±15. Pellegrini V D, Olcott C W, McCollister Evarts C (1998), `The Tri-lock femoral systems' in Finerman G A M, et al., eds: Total Hip Arthroplasty Outcomes, Kidlington, Churchill Livingstone, 181±93. Pletcher D, Walsh F C (1990), Industrial Electrochemistry, 2nd edn, London, New York, Chapman and Hall. Rader C P, Hendrich C, LoÈw S, Walther M, Eulert J (2000), `5-bis 8-Jahres-Ergebnisse nach HuÈfttotalendoprothese mit der MuÈller-Geradshaft-prothese (zementierter TiAlNb-Shaft)', Unfallchirurg, 103, 846±52. Randle R, Gordiev K (1997), `Metal-on-metal articulation in total hip arthroplasty: preliminary results in 57 cases', Aust N Z J Surg, 67, 634±6. Rieker C B, KoÈttig P (2002), `In vivo tribological performance of 231 metal-on-metal hip articulations', Hip International, 12, 73±6. Rieker C B, SchoÈn R, KoÈttig P (2004), `Development and validation of second-generation metal-on-metal bearings', J Arthroplasty, 19(Suppl 3), 5±11. Salvati E A, Betts F, Doty S B (1993), `Particulate metallic debris in cemented total hip arthroplasty', Clin Orthop, 293, 160±73. Santavirta S (2003), `Compatibility of the totally replaced hip. Reduction of wear by amorphous diamond coating', Acta Orthop Scand, Suppl, 74, 119. Santori F, Albanese C, Rendine M, Duffy G, Learmonth I D (2005), `Bone preservation with a conservative methaphyseal loading implant'. Presented at the EFORT Meeting, Lisbone, 4±7 June 2005. © 2008, Woodhead Publishing Limited

Metals for joint replacement

161

Santori F, Manili M, Fredella N, Tonci Ottieri M, Santori N (2006), `Ultra-short stems with proximal load transfer: Clinical and radiographic results at five-year followup', Hip International, 16, Suppl 3, S31. Sarmiento A, Gruen T A (1985), `Radiographic analysis of a low-modulus titanium-alloy femoral total hip component. Two to six year follow-up', J Bone Joint Surg [Am], 67-A, 48±56. Sarmiento A, Ebramzadeh E, Normand P, Llinas A, McKellop H A (1998), `The stainless-steel and titanium alloy femoral prosthesis' in Finerman G A M, et al., eds: Total Hip Arthroplasty Outcomes, Kidlington, Churchill Livingstone, 41±53. Sedel L, Kerboull L, Christel P, Meunier A, Witvoet J (1990), `Alumina-on-alumina hip replacement: results and survivorship in young patients', J Bone Joint Surg [Br], 72-B, 658±63. Shimmin A J, Bare J, Back D L (2005), `Complications associated with hip resurfacing arthroplasty', Orthop Clin North Am, 36, 187±93. Siebold R, Scheller G, Schreiner U, Jani L (2001), `Long-term results with the cementfree Spotorna CLS shaft' (German), OrthopaÈde, 30, 317±22. Sotereanos N, Engh C A, Glassman A H et al. (1995), `Cementless femoral components should be made of cobalt chrome', Clin Orthop Relat Res, 313, 146±53. Sporer S M, Callaghan J J, Olejniczak J P, Goetz D D, Johnston R C (1999), `The effects of surface roughness and polymethylmethacrylate precoating on the radiographic and clinical results of the Iowa hip prosthesis. A study of patients less than fifty year old', J Bone Joint Surg Am, 81-A, 481±92. St. John K R, Zardiackas L D, Poggie R A (2004), `Wear evaluation of cobalt-chromium alloy for use in a metal-on-metal hip prostheses', J Biomed Mater Res B Appl Biomater, 68, 1±14. Tompkins G S, Lachiewicz P F, DeMasi R (1994), `A prospective study of titanium femoral component for noncemented total hip arthroplasty', J Arthroplasty, 9, 623± 30. Traulsen F C, Hassenpflug J, Hahne H J (2001), `Long-term results with cement-free total hip prosthesis (Zweymuller)' (German), Z Orthop Ihre Grenzgeb, 139, 206±11. TrebsÏe R, MilosÏev I, KovacÏ S, Mikek M, PisÏot V (2005), `Poor results from the isoleastic total hip replacement: 14±17-year follow-up of 149 cementless prostheses', Acta Orthop, 76, 169±76. Wagner M, Wagner H (2000), `Medium-term results of a modern metal-on-metal system in total hip replacement', Clin Orthop Relat Res, 379, 123±33. Weber B G (1992), `Metal±metal THR: back to the future', [German] Z OrthopaÈdie, 130, 306±9. Willert H G, Semlitsch M, Buchhorn G, Kriete U (1978), `Materialverschleiss und Gewebereaktion bei kunstlichen Gelenken', OrthopaÈde, 7, 62±7. Willert H G, Broback L G, Buchhorn G H, Jensen P H, KoÈster G, Lang I, Ochsner P, Schenk R (1996), `Crevice corrosion of cemented titanium alloy stems in total hip replacements', Clin Orthop Relat Res, 333, 51±75. Willert H G, Buchhorn G H, Fayyazi A, Flury R, Windler M, KoÈster G, Lohmann C H (2005), `Metal-on-metal bearings and hypersensitivity in patients with artificial hip joints. A clinical and histomorphological study', J Bone Joint Surg [Am], 87-A, 28± 36. Witt J D, Swann M (1991), `Metal wear and tissue response in failed titanium alloy total hip replacements', J Bone Joint Surg [Br], 73-B, 559±63. Wroblewski B M (1982), `Fractured stem in total hip replacement. A clinical review of 120 cases', Acta Orthop Scand, 53, 279±84. © 2008, Woodhead Publishing Limited

162

Joint replacement technology

Wroblewski B M, Siney P D, Fleming P A, (1998) The Charnley LFA `The Wrightington hospital', in Finerman G A M, et al., eds, Total Hip Arthroplasty Outcomes, Kidlington, Churchill Livingstone, 15±29. Ziaee H, Daniel J, Datta A K, Blunt S, McMinn D J W (2007), `Transplacental transfer of cobalt and chromium in patients with metal-on-metal hip arthroplasty', J Bone Joint Surg [Br], 89-B, 301±5.

© 2008, Woodhead Publishing Limited

7

Ceramics for joint replacement D K L U E S S , W M I T T E L M E I E R and R B A D E R , University of Rostock, Germany

7.1

Introduction

Ceramics were initially used in total hip arthroplasty (THA) and in total knee arthroplasty (TKA) more than 30 years ago. The high wear rates of early metal-on-metal and metal-on-polyethylene bearings, together with the particleinduced osteolysis, necessitated advanced materials in joint replacement. Owing to its corrosion resistance based on the high level of oxidation, as well as its excellent bearing performance, alumina was the first ceramic material being applied in THA (1970) (Boutin, 1972) and in partial knee replacement (1972) (Langer, 2002). The extremely low wear rate of ceramic-on-ceramic bearings, the biocompatibility and allergological benefits compared with metal components made ceramics the most favourable choice in articulating joint replacement materials (Wang et al., 2003). A striking disadvantage of alumina ceramic is its brittleness and lower tensile strength manifested in relatively low fracture toughness. A number of femoral head component fractures and ceramic cup breakages demonstrated this disadvantage compared with ductile metallic components for total hip replacement. Moreover, complications encountered with ceramics were often connected to the ceramic±bone interface and the insufficient osseous integration of ceramic materials (Hannouche et al., 2005). The first attempt to face the low fracture toughness of alumina was the introduction of zirconia in 1985 (Clarke et al., 2003). Alumina and zirconia are both oxide ceramics, but the biphasic structure of zirconia causes a flexural strength almost twice as high as alumina. However, owing to the instability of the biphasic structure, zirconia undergoes an undesired ageing process. Further developments of ceramics in joint replacement have addressed the hydrothermal degradation of zirconia. The most recent development is a composite ceramic material which combines the stableness of alumina with the high fracture toughness of zirconia (Rack and Pfaff, 2000; Merkert, 2003). For instance, the BioloxÕ delta ceramic, introduced in 2000, consists of an alumina matrix with zirconia grains and strontium platelets. The high performance of composite © 2008, Woodhead Publishing Limited

164

Joint replacement technology

oxide ceramics advances them for further applications in total hip, knee and shoulder arthroplasty.

7.2

Material and mechanical properties of ceramics

Ceramics can generally be divided into non-oxide and oxide ceramics. The ceramics predominantly applied in joint replacement (alumina, zirconia, composite) belong to oxide ceramics. The mechanical properties of ceramics (Table 7.1) are mainly determined by the fabrication process and the internal structure of the material (Hannouche et al., 2005; Lee and Ahn, 2007). Ceramics are usually manufactured from a fine powder, which is mixed with water and an organic binder. The mixture is then pressed into a mould to obtain a shape that is close to the desired geometry. The subsequent steps are the hot isostatic pressure (HIP) and the sintering process, i.e. a high-temperature treatment of the shaped ceramic material. During sintering, the density of the material increases while the volume shrinks. Grain boundaries are formed and grain growth causes the reduction of pores. After sintering, the ceramic body is machined into the final desired geometry (Burger, 2000; Hannouche et al., 2005).

7.2.1

Alumina

The ceramic material initially introduced into the field of orthopaedic implants is aluminium oxide (Al2O3), so-called alumina. Alumina consists of a polycrystalline monophasic structure. It exhibits the highest state of oxidation, hence allowing hydrodynamic stability (no ageing), chemical inertness and resistance to corrosion. The chemical inertness and corrosion resistance are the basis for the excellent biocompatibility of ceramic materials. The high hardness of alumina makes the material resistant to scratches and wear; however the flexural strength and the fracture toughness are limited. Table 7.1 Material properties of selected alumina, zirconia and composite ceramics (Clarke and Willmann, 1994; Cales and Stefani, 1995; Burger, 2000; Merkert, 2003)

Density (g/cm3) Grain size (m) Flexural strength (MPa) Hardness Poisson ratio Young's modulus (GPa)

© 2008, Woodhead Publishing Limited

Alumina (BioloxÕ forte)

Zirconia (Y-TZP)

Composite BioloxÕ delta

3.98 <2 580 2300 HV0.5 0.23 380

6.08 < 0.5 1050 1250 HV0.5 0.3 210

> 4.36 < 1.5 1150 1975 HV1 0.22 350

Ceramics for joint replacement

7.2.2

165

Zirconia

Based on its higher flexural strength compared with alumina, zirconia (ZrO2) was established in joint replacement in the 1980s. Zirconium oxide (zirconia) consists of a polycrystalline biphasic (tetragonal and monoclinic) structure. The tetragonal phase of pure zirconia exists in an unstable state, which is why stabilisation, e.g. with yttrium, is performed (Y-TZP: yttrium stabilised zirconia). The tetragonal phase of Y-TZP is stable at room temperature, but unstable at higher temperatures, e.g. during exposure to frictional heat or in the autoclave. The phase transformation from the metastable tetragonal phase into the monoclinic phase results in a volume increase of approx. 3±4%, hence ageing of zirconia may result in surface roughening. Zirconia has lower hardness, stiffness and grain size and higher density and flexural strength than alumina, resulting in a lower risk of breakage in orthopaedic applications (Clarke and Willmann, 1994; Cales and Stefani, 1995; Blaise et al., 2001; Corfield et al., 2007).

7.2.3

Composite ceramic

The development of composite ceramics is a result of the disadvantageous ageing of zirconia and fracture toughness of alumina. The common composite ceramics in the field of joint replacement are zirconia toughened alumina (ZTA) and zirconia and platelet reinforced alumina (ZPTA). For instance, the BioloxÕ delta ceramic is a composite consisting of alumina matrix (AMC), in which zirconia grains (approx. 25%) and strontium platelets plus chrome oxide are added. The percentage of zirconia still contains a metastable tetragonal phase, but it was shown that surface roughening due to ageing did not occur (Corfield et al., 2007). The stiffness and hardness of BioloxÕ delta ceramic are in the range of alumina, while its flexural strength is almost twice as high.

7.3

Ceramics in total hip replacement

Alumina ceramics were firstly implanted in THA by Boutin (France, 1970) (Boutin, 1972). Further developments in collaboration with Sedel resulted in the introduction of a cemented ceramic socket in combination with a ceramic taperlocked femoral head design in 1977 (Sedel et al., 1990). In Germany, Mittelmeier (Mittelmeier, 1978; Mittelmeier and Harms, 1979) introduced a ceramicon-ceramic THA with a non-cemented monoblock alumina cup combined with a taper-locked alumina mushroom-shaped head in 1974, which gained US Food and Drug Administration (FDA) approval in 1982. The first cups in ceramic-onceramic bearings exhibited insufficient osseous integration owing to the chemical inertness, resulting in implant loosening and migration. Since the mid-1980s modular metal-backed sockets with ceramic liners have been © 2008, Woodhead Publishing Limited

166

Joint replacement technology

7.1 Different ceramic components (ball heads and liners based on alumina and composite ceramics) for total hip replacement.

established to achieve biological implant fixation. Besides metal-backed sockets that are directly connected to the ceramic liner, sandwich constructions with a polyethylene layer between the metal-back and the ceramic liner are still available on the market. Today, ceramic materials are limited only to the articulating function in total hip replacement (Fig. 7.1), but recent developments are striving for the fixation of ceramic implants by direct bone contact.

7.3.1

Ceramic-on-ceramic and ceramic-on-polyethylene bearings

The excellent wear behaviour of ceramic-on-ceramic bearings has been proven in numerous in vitro and in vivo studies (Garino, 2000; D'Antonio et al., 2002; Kaddick and Pfaff, 2002; Fisher et al., 2006). The question arises whether the wear behaviour of ceramic-on-polyethylene is superior to the wear behaviour of metal-on-polyethylene. The wear performance of total hip replacement bearings is tested in vitro under standardised conditions (ISO/FDIS 14242-1) using different hip joint simulators. Generally, the femoral and the acetabular component are each mounted in a test station and exposed to rotational movements under definite load characteristics to simulate physiological loading during walking. The tests are performed in bovine or calf serum diluted to 15% (Kaddick and Wimmer, 2001), with a loaded soak control tested in parallel to enable objective measurement of weight loss as a result of wear. The measured wear rates of various combinations of ceramic, polyethylene and metal couples (Kaddick and Pfaff, 2002) are shown in Table 7.2. The total wear rates of the cup and the head of ceramic-on-ceramic couples are about 35-fold lower than those of metal-on-metal. The head wear rate alone is about 100 to 300-fold © 2008, Woodhead Publishing Limited

Ceramics for joint replacement

167

Table 7.2 Gravimetric wear rates of ceramic-on-ceramic, metal-on-metal, metal-onpolyethylene and ceramic-on-polyethylene wear couples in hip simulator studies (Kaddick and Pfaff, 2002) Ball head

Cup liner

Biolox forte Biolox delta Metal Metal Biolox forte Biolox delta

Biolox forte Biolox delta Metal UHMWPEa UHMWPEa UHMWPEa

a

Head Head wear diameter rate (mm) (mg/million cycles) 28 22.2 28 28 28 28

Cup wear rate (mg/million cycles)

Total wear rate (mg/million cycles)

0.084 0.072 1.571

0.089 0.086 3.141

0.005 0.014 1.570 ± ± ±

70.62 26.57 16.08

UHMWPE: ultra-high molecular weight polyethylene

Table 7.3 Volumetric wear rates of ceramic-on-crosslinked polyethylene and metalon-crosslinked polyethylene wear couples in hip simulator studies over the period from 2 to 7 million cycles (Fisher et al., 2006) Ball head Biolox forte Cobalt±chromium

Cup liner

Head diameter (mm)

Total wear rate (mm3/million cycles)

XL-PEa XL-PEa

36 36

4.3 9.5

a

XL-PE: crosslinked polyethylene.

lower. The hip joint simulator results for the comparison of metal-onpolyethylene and ceramic-on-polyethylene couples show a significant reduction of wear with ceramic ball heads in combination with ultra-high molecular weight polyethylene (UHMWPE) cups. With regard to newly developed crosslinked polyethylene (XL-PE) cups, the wear rate of BioloxÕ forte ceramic heads turned out to be less than half of cobalt±chromium heads (Table 7.3) (Fisher et al., 2006). Moreover, the wear performance of XL-PE provides an overall significant wear reduction compared to UHMWPE.

7.3.2

Large ceramic heads in total hip replacement

In recent experimental (Bader et al., 2004a,b), computational (Bader et al., 2002) and finite element (FE) analyses (Kluess et al., 2005, 2007, 2008) the clear advantages of large femoral heads regarding higher range of motion and enhanced dislocation stability have been demonstrated. Dislocation-prone manoeuvres were simulated using a previously validated FE model, and the © 2008, Woodhead Publishing Limited

168

Joint replacement technology

7.2 Course of the resisting moments during internal rotation at 90ë flexion of head sizes from 28 to 40 mm. Cup position is 45ë inclination and 15ë anteversion. Stem antetorsion is 0ë and neck thickness is 14 mm for all head sizes (Kluess et al., 2007).

resisting moments were calculated during impingement and subluxation. The progression of the resisting moments shows the range of motion (ROM) until impingement, the maximum resisting moment during impingement and the ROM until dislocation. The results of the simulation of internal rotation at 90 flexion with four different head sizes are given in Fig. 7.2. It is shown that large heads provide a greater ROM until impingement, higher resisting moments during impingement and subluxation, as well as higher ROM until total dislocation. Stress analysis with the same FE models also revealed a better stress distribution during subluxation due to a greater area of articulating surfaces. Although large head diameters are historically uncommon in total hip joint replacement, today several attempts have been undertaken to establish larger ball heads in THA. A crucial question in introducing bigger ceramic heads is to clarify if the wear performance is adversely affected by larger head diameters. A series of studies showed that with large diameters, wear of ceramic heads is distinctly reduced compared to metallic heads. Alumina-on-alumina bearings showed no tendency of higher wear with larger heads. Furthermore, large ceramic bearings are less sensitive to manufacturing tolerances in terms of sphericity and clearance (Pandorf, 2007). The introduction of larger heads enables a reconsideration of liner design owing to the enhanced dislocation stability. In hip resurfacing, where large metal heads have been in clinical application for decades, the cups are designed with reduced head coverage compared with cups in standard total hip replacement systems (Kluess et al., 2008). We recently carried out an FE analysis to identify the influence of head coverage of ceramic liners in combination with large © 2008, Woodhead Publishing Limited

Ceramics for joint replacement

169

ceramic ball heads on the impingement and dislocation behaviour. Dislocationprone manoeuvres of a 40 mm diameter ceramic-on-ceramic bearing with head coverage angles of 180ë, 160ë and 140ë were simulated and the resisting moments during subluxation were calculated. With reduced head coverage, ROM until impingement was increased by up to 30ë, but very low head coverage angles tended to cause very little resisting moments during subluxation. With highly abducted cups in combination with low head coverage, dislocation without impingement was predicted. In conclusion, it was shown that reduction of head coverage with large ceramic heads yields a large increase in ROM; however, attention must be paid to safe implant positioning with regard to the lateral abduction (inclination) and anteversion angles. The use of ceramic heads in revision hip arthroplasty is limited since the application of new ceramic heads on used tapers raises the risk of subsequent head fracture, because scratches and material elevations on the used taper, caused by removal of the primary head, can induce high stress peaks if a new ceramic head is applied. A solution to bypass these complications in the metal-ceramic taper interface is provided by the BioloxÕ option ceramic heads (Fig. 7.3). A

7.3 Ceramic heads (BioloxÕ option) for revision hip arthroplasty, providing different diameters and neck lengths. © 2008, Woodhead Publishing Limited

170

Joint replacement technology

metallic sleeve included in the set establishes a safe connection between the used taper and the ceramic head. The scratched surface of the used taper is covered by a ductile metal cone, while the undamaged outer surface of the metallic sleeve allows a safe connection to the ceramic head. This technique expands the application of ceramic heads to the field of revision hip arthroplasty.

7.4

Ceramics in total knee replacement

Ceramic knee endoprostheses were initially implanted in the form of noncemented tibial plateaux by Langer (2002) in 1972 (Fig. 7.4). The extension of alumina ceramic to a total knee replacement system was realised in 1982 by Oonishi (Bal and Oonishi, 2003), named KOM-1 after the accordant Kokuritsu Osaka Minami Hospital in Japan. Further developments (KOM-2, 1982 and KOM-3, 1990) included the implementation of a polyethylene tibial inlay for

7.4 X-ray of a non-cemented ceramic tibial plateau performed by Langer in 1972 (Langer, 2002). © 2008, Woodhead Publishing Limited

Ceramics for joint replacement

171

improved articulation as well as advanced cement fixation with ceramic beads on the implant surface (Bal and Oonishi, 2003). The promising results of BioloxÕ delta composite ceramic in THA justified the re-evaluation of ceramic components for TKA. The objective of the development of a ceramic femoral component was the avoidance of potential allergic materials (in particular cobalt±chromium) and a further reduction in the release of polyethylene particles and their contribution to implant loosening. An international prospective multicentre study started in November 2006 at nine centres in Germany, Italy and Spain to control the postoperative performance obtaining clinical and radiological data during follow-up. The patients received an unconstrained Lima Multigen Plus knee (Fig. 7.5) with a BioloxÕ delta femoral component, polyethylene inlay and a cobalt-chromium tibial component. Clinical and functional outcome as well as quality of life is evaluated using standard scoring systems to enable comparison of the results to those with standard total knee replacements. Clinical performance is evaluated using the Ranawat and Shine (HSS) score. Patient satisfaction and quality of life is evaluated using the WOMAC-Score and the Short Form 36-score. The position

7.5 Multigen Plus knee endoprosthesis with a BioloxÕ delta ceramic femoral component. © 2008, Woodhead Publishing Limited

172

Joint replacement technology

7.6 Left: finite-element model of the ceramic femoral component, polyethylene tibial inlay, bone cement and the bone stock. Right: maximum principal stresses (S. Max. Principal) developed in the ceramic femoral component during gait. Regions of higher stresses are detected at the anterior radii (2a, 2b) and close to the pegs (1a, 1b).

of the femoral component, the tibial plate and patella are radiologically evaluated in the follow-up. In the investigations three months postoperatively no adverse effects related to the ceramic material were observed. However, with regard to the brittleness of ceramics and the complicated geometry of the femoral component, the risk of failure as a result of stress peaks has to be considered. Stress peaks can be minimised through sufficient implant design, proper material composition and optimum force transmission between implant and bone stock. Using non-linear FE models (Schultze et al., 2007), we analysed the force transmission from the distal femur through the bone cement into the femoral component with different bone cement layer thicknesses. Higher stress regions occurred in the area of the force transmission and in the adjoining transitional radii of the femoral component (Fig. 7.6). The maximum calculated stresses of the implant, however, were below the flexural strength. Stresses were found to be lower with a thin layer of cement.

7.5

Summary

The use of ceramics in total hip replacement is widely accepted with good longterm results and is a standard in modern orthopaedic surgery. Ceramic-onceramic bearings in THA not only generate fewer wear particles than metal-onpolyethylene or metal-on-metal bearings, but the ceramic wear particles are also less damaging in terms of adverse biological effects (Wang et al., 2003). The most common ceramics in THA are the oxide ceramics, i.e. alumina, zirconia © 2008, Woodhead Publishing Limited

Ceramics for joint replacement

173

and newly developed composite ceramics (Rack and Pfaff, 2000; Merkert, 2003; Kaddick and Pfaff, 2002; Hannouche et al., 2005). High-performance bioceramics provide an extension of the applicability of ceramics in joint replacement. New trends are found in the application of large femoral heads for enhanced range of motion and dislocation stability as well as in the application of ceramics in revision THA (Pandorf, 2007). With the substantial improvements in the mechanical properties, ceramic components for total knee replacement have been reintroduced (Mittelmeier et al., 2006). Improved long-term implant survival rates based on the high corrosion resistance and reduced release of polyethylene wear particles can be expected in TKA in the near future.

7.6

References

Bader, R., Steinhauser, E., Gradinger, R., Willmann, G. and Mittelmeier, W. (2002) ComputergestuÈtzte Bewegungssimulation an HuÈftendoprothesen mit KeramikKeramik-Gleitpaaarung. Analyse der Einflussparameter Implantat-Design und Position. Z Orthop Ihre Grenzgeb, 140, 310±16. Bader, R., Scholz, R., Steinhauser, E., Busch, R. and Mittelmeier, W. (2004a) Methode zur Evaluierung von Einfluûfaktoren auf die LuxationsstabilitaÈt von kuÈnstlichen HuÈftgelenken. Biomed Tech, 49, 137±44. Bader, R., Scholz, R., Steinhauser, E., Zimmermann, S., Busch, R. and Mittelmeier, W. (2004b) The influence of head and neck geometry on stability of total hip replacement: a mechanical test study. Acta Orthop Scand, 75, 415±21. Bal, B. S. and Oonishi, H. (2003) Ceramic femoral components in total knee replacement. In Zippel, H. and Dietrich, M. (Eds) Ceramics in Orthopaedics 8th BIOLOXÕ Symposium Proceedings. Darmstadt, Steinkopff. Blaise, L., Villermaux, F. and Cales, B. (2001) Ageing of zirconia: everything you always wanted to know. Key Engineering Materials, 192±195, 553±556. Boutin, P. (1972) [Total arthroplasty of the hip by fritted aluminum prosthesis. Experimental study and 1st clinical applications]. Rev Chir Orthop Reparatrice Appar Mot, 58, 229±46. Burger, W. (2000) Oxidische Hochleistungswerkstoffe und deren endkonturnahe Formgebung. In Kriegesmann, J. (Ed.) Technische keramische Werkstoffe. Cologne, Fachverlag deutscher Wirtschaftsdienst. Cales, B. and Stefani, Y. (1995) Yttrium-stabilized zirconia for improved orthopaedic prostheses. In Wise, D. L. (Ed.) Encyclopedic Handbook of Biomaterials and Bioengineering. New York, Marcel Dekker Inc. Clarke, I. C. and Willmann, G. (1994) Structural ceramics in orthopaedics. In Cameron, H. U. (Ed.) Bone Implant Interface. St. Louis Baltimore Boston, Mosby Year Book. Clarke, I. C., Manaka, M., Green, D. D., Williams, P., Pezzotti, G., Kim, Y. H., Ries, M., Sugano, N., Sedel, L., Delauney, C., Nissan, B. B., Donaldson, T. and Gustafson, G. A. (2003) Current status of zirconia used in total hip implants. J Bone Joint Surg Am, 85-A Suppl 4, 73±84. Corfield, V., Khan, I. and Scott, R. (2007) Hydrothermal stability of ceramic femoral heads. In Chang, J. D. and Billau, K. (Eds) Ceramics in Orthopaedics 12th BIOLOXÕ Symposium Proceedings. Darmstadt, Steinkopff. D'Antonio, J., Capello, W., Manley, M. and Bierbaum, B. (2002) New experience with

© 2008, Woodhead Publishing Limited

174

Joint replacement technology

alumina-on-alumina ceramic bearings for total hip arthroplasty. J Arthroplasty, 17, 390±7. Fisher, J., Jennings, L. M. and Galvin, A. L. (2006) Wear of highly crosslinked polyethylene against cobalt chrome and ceramic femoral heads. In Benazzo, F., Falez, F. and Dietrich, M. (Eds) Ceramics in Orthopaedics 11th BIOLOX Õ Symposium Proceedings. Darmstadt, Steinkopff. Garino, J. P. (2000) Modern ceramic-on-ceramic total hip systems in the United States: early results. Clin Orthop Relat Res, 379, 41±7. Hannouche, D., Hamadouche, M., Nizard, R., Bizot, P., Meunier, A. and Sedel, L. (2005) Ceramics in total hip replacement. Clin Orthop Relat Res, 430, 62±71. Kaddick, C. and Pfaff, H. G. (2002) Results of hip simulator testing with various wear couples. In Garino, P. and Willmann, G. (Eds) Bioceramics in Joint Replacement: Proceedings. Stuttgart, New York, Thieme. Kaddick, C. and Wimmer, M. A. (2001) Hip simulator wear testing according to the newly introduced standard ISO 14242. Proc Inst Mech Eng [H], 215, 429±42. Kluess, D., Martin, H., Bader, R., Mittelmeier, W. and Schmitz, K. P. (2005) Impingement and dislocation of total hip replacement ± Validation of a finite element analysis. Biomed Tech, 50, 1507±1508. Kluess, D., Martin, H., Mittelmeier, W., Schmitz, K. P. and Bader, R. (2007) Influence of femoral head size on impingement, dislocation and stress distribution in total hip replacement. Med Eng Phys, 29, 465±71. Kluess, D., Zietz, C., Lindner, T., Mittelmeier, W., Schmitz, K. P. and Bader, R. (2008) Limited range of motion of hip resurfacing arthroplasty due to unfavorable ratio of prosthetic head size and femoral neck diameter. Acta Orthopaedica, in press. Langer, G. (2002) Ceramic tibial plateau of the 70s ceramics for total knee replacement: status and options. In Garino, J. P. and Willmann, G. (Eds) Bioceramics in Joint Arthroplasty. Stuttgart, New York, Thieme. Lee, M. C. and Ahn, J. W. (2007) Ceramic femoral prosthesis ± present and future. In Chang, J. D. and Billau, K. (Eds) Ceramics in Orthopaedics 12th BIOLOXÕ Symposium Proceedings. Darmstadt, Steinkopff. Merkert, P. (2003) Next generation ceramic bearings. In Zippel, H. and Dietrich, M. (Eds) Ceramics in Orthopaedics 8th BIOLOXÕ Symposium Proceedings. Darmstadt, Steinkopff. Mittelmeier, H. (1978) Early results of hip alloplasty with endoprostheses with ceramic carrying ribs. Z Orthop Ihre Grenzgeb, 116, 594±5. Mittelmeier, H. and Harms, G. (1979) Present-day state of cement-free anchoring of combined ceramics-metal prostheses. Z Orthop Ihre Grenzgeb, 117, 478±81. Mittelmeier, W., Ansorge, S., Kluess, D., Kircher, J. and Bader, R. (2006) Ceramic knee endoprosthesis: reality or future? In Benazzo, F., Falez, F. and Dietrich, M. (Eds) Ceramics in Orthopaedics 11th BIOLOXÕ Symposium Proceedings. Darmstadt, Steinkopff. Pandorf, T. (2007) Wear of large ceramic bearings. In Chang, J. D. and Billau, K. (Eds) Ceramics in Orthopaedics 12th BIOLOXÕ Symposium Proceedings. Darmstadt, Steinkopff. Rack, R. and Pfaff, H. G. (2000) A new ceramic material for orthopaedics. In Willmann, G. and ZweymuÈller, K. (Eds) Bioceramics in Joint Replacement: Proceedings. Stuttgart, New York, Thieme. Schultze, C., Kluess, D., Martin, H., Hingst, V., Mittelmeier, W., Schmitz, K. P. and Bader, R. (2007) Finite element analysis of a cemented ceramic femoral component for the assembly situation in total knee arthroplasty. Biomed Tech, 52, 301±7. © 2008, Woodhead Publishing Limited

Ceramics for joint replacement

175

Sedel, L., Kerboull, L., Christel, P., Meunier, A. and Witvoet, J. (1990) Alumina-onalumina hip replacement. Results and survivorship in young patients. J Bone Joint Surg Br, 72, 658±63. Wang, A., Dumbleton, J. H., Manley, M. T. and Serekian, P. (2003) Role of ceramic components in the era of crosslinked polyethylene for THR. In Zippel, H. and Dietrich, M. (Eds) Ceramics in Orthopaedics 8th BIOLOXÕ Symposium Proceedings. Darmstadt, Steinkopff.

© 2008, Woodhead Publishing Limited

8

Joint bearing surfaces and replacement joint design R L A P P A L A I N E N and M S E L E N I U S , University of Kuopio, Finland

8.1

Introduction

Joint bearing surfaces are critical for the painless movement of limbs. When they are severely damaged, eventually joints must be replaced by either tissue implants or artificial materials. The operation can normally restore basic joint movements and allow sufficient range of motion of the joint. In spite of a good outcome, the performance of artificial materials is poor compared with natural joints, and a large number of wear debris particles are released from the articulating surfaces. They irritate tissues and lead to aseptic loosening of the implant components. This chapter introduces some basics related to joint design, describes different artificial joint bearing materials and compares their performance. Although this discussion focuses on the hip joint and implants, similar concepts can be applied to other joints such as the knee, ankle, shoulder and elbow.

8.2

Articulating surfaces in natural joints

In natural healthy joints, the smoothness and elasticity of the cartilage-covered and bone-supported counterfaces and joint lubricants are essential for the absorption of the impact loads and the low friction movement of bones against each other. The excellent tribological properties of the human and animal joints are supposed to be due to a mixed mode lubrication, which includes pressure film lubrication and in particular contact point lubrication. Intact joint surfaces and high-quality joint lubricants seem to be essential for proper joint function. Normally, joint structure and function are maintained by proper use of joints for healthy living, including walking, bicycling and swimming, to mention a few of our favourite activities. Inactivity leads to atrophy, and overuse to an accelerated degeneration. It is not movement per se that maintains the joint architecture and physiology but it is the changes that these movements lead to in joint structures. These include interstitial fluid flow, which has been well recognised for its importance for joint nutrition. However, owing to water and ion movements this © 2008, Woodhead Publishing Limited

Joint bearing surfaces and replacement joint design

177

also leads to associated electrical and electromagnetic phenomena, i.e. joint mechanics are closely related to electrodynamics. Owing to the extraordinary structure of joint articulating surfaces, joints are able to carry weights that are 8±12 times the body weight and glide with extremely low friction of 0.002±0.005 at 3000 kg/cm2 load (Charnley, 1959). Furthermore, even after stable loading with squeezing out of the liquid film between articulating surfaces, as the patient moves the joint, excellent lubrication is quickly achieved. From the engineering point of view this is an excellent achievement, since one of the best synthetic boundary lubricants, Teflon, can only provide lubrication with a friction coefficient at least 10 times higher than the natural joint, i.e. around 0.1 (Adamson, 1967).

8.3

Demands for the bearing surfaces

It is evident that the performance of natural joints is hard to achieve using synthetic materials. Therefore, conventionally, artificial joint bearing surfaces are designed to achieve two main goals, i.e. low wear rate and sufficiently low friction. Low wear rate is necessary to maintain wear debris generation at low level, i.e. below a threshold to avoid aseptic loosening owing to a large number of foreign particles from the sliding surfaces being released into the joint capsule and surrounding tissues. Low friction at the articulating surfaces guarantees low bending torque values on the fixation surfaces of implants, e.g. on the back surface of a hip joint cup. As demonstrated also in clinical use, the above criteria can be met by different combinations of basic material types, i.e. polymers, metals and ceramics. Friction and wear rate values for different kinds of combination are schematically illustrated in Fig. 8.1.

8.1 Comparison of friction and wear properties of different sliding pairs (modified from Schmidt et al., 1996). © 2008, Woodhead Publishing Limited

178

Joint replacement technology

In principle, additional requirements for articulating surface materials include biocompatibility, good corrosion resistance in synovial fluid and degradation resistance in long-term use. However, these demands have been compromised since CoCr alloys are commonly used in metal components of sliding surfaces in spite of the fact that both Co and Cr are known to be toxic and cause allergic reactions. Ultra-high molecular weight polyethylene (UHMWPE) is also the most common plastic material for implant bearing surfaces, even though there is long-term degradation of UHMWPE components and fractures are common especially in knee implants.

8.4

Different solutions available

Two very relevant factors related to tribological performance of different sliding pairs are wear mechanisms and lubrication. There are five generally recognised wear mechanisms which can be summarised as follows: 1. Abrasive ± the displacement of materials by hard particles. 2. Adhesive ± the transference of material from one surface to another during relative motion by the process of solid-phase welding. 3. Fatigue ± the removal of materials as a result of cyclic stress variations. 4. Erosive ± the loss of material from a solid surface due to relative motion in contact with a fluid that contains solid particles. This is often subdivided into impingement erosion and abrasive erosion. If no solid particles are present, erosion can still take place, such as rain erosion and cavitation. 5. Corrosive ± a process in which chemical or electrochemical reactions with the environment dominates, such as oxidative wear. Consideration of these mechanisms is important to the discussion which follows with respect to UHMWPE as well as the other bearing surfaces. Lubrication refers to adding a lubricant between two bearing surfaces in order to control friction and wear. Change in the lubrication conditions can have a major impact on both friction and wear of total hip replacements. This is especially true in the case of polymers, where the wear rate may vary even by several orders of magnitude in two quite similar lubricants. Furthermore, the geometry, topography and surface chemistry of bearing surfaces determine how the lubricant operates and remains on the sliding surfaces. Three alternatives are presented in Fig. 8.2 of how the lubricant operates on the sliding surface. Furthermore, especially hard third-body wear debris particles can significantly increase scratching and wear of the sliding surfaces, especially in the case of metals.

8.4.1

Polymers on bearing surfaces

UHMWPE is the polymer that has been generally used as an implant material for over four decades. Sir John Charnley reported his clinical experience with total © 2008, Woodhead Publishing Limited

Joint bearing surfaces and replacement joint design

179

8.2 Schematic of lubrication regimes and associated characteristics (fluid film, boundary and mixed lubrication regimes, based on Jin, 2006).

hip replacement in 1961. Charnley's original low-friction arthroplasty was based on a polyethylene acetabular component and a metallic femoral component. Today the number of total hip replacements done worldwide each year is over 800 000, most of which still utilise polyethylene on one of the bearing surfaces. A property of UHMWPE that distinguishes it from other polymers is its structure with extremely long and highly entangled molecular chains, which © 2008, Woodhead Publishing Limited

180

Joint replacement technology

makes it resistant to wear. This is one main reason why UHMWPE has been so successful on implant bearing surfaces. Furthermore, UHMWPE is biocompatible and stable in the body. Several factors affect the wear resistance of the finished polymer implant and some of them can be traced back to its processing. Different processing steps can be used to modify the characteristics which affect performance strongly and the wear of the artificial joint. Molecular weight and degree of crystallinity are among the most significant ones. Other important factors include the state of crosslinking and oxidation index. It has been proved that crosslinked UHMWPE functions better on the sliding surfaces than ordinary UHMWPE. The present manufacturing processes are carried out so that there will be as little oxygen as possible present at any stage. Also the sterilisation is performed by irradiating in inert gas or using ethylene oxide gas, minimising oxidative degradation. It has been possible to improve UHMWPE as a sliding surface but it has not been possible to prevent wear and release of particles. Although the present implants wear out considerably less, even a minor yield of particles can be extremely harmful. There is now considerable evidence implicating UHMWPE wear debris in osteolysis and loosening of prostheses. There is, therefore, much interest in understanding the wear mechanisms, so that the best solutions can be developed. The compressive yield strength of the UHMWPE is about 12 MPa. The maximum nominal contact stresses in the hip joint are in the range 3±6 MPa. It is known that the acetabular component can still deform elastically, although the surface layer may experience large plastic flow. It has been found clinically that three different contact conditions favour the occurrence of surface plastic flow. Firstly, one or both of the contact surfaces may be rough and this often results in contact stresses exceeding the yield strength at the asperities. The second case follows when third-body abrasive particles remain stuck between the sliding surfaces. In many examples, the particles come from the bone cement that is used to fix implants. Bone cement contains hard ceramic particles, such as zirconium oxide or barium sulphate. In the third poorly understood case, both sliding surfaces are extremely smooth and no particles appear between the sliding surfaces. It has been perceived clinically that in some cases there are high wear rates (10±6 mm3/N m) although the sliding surfaces of the implant can be extremely smooth (Atkinson, 1985; Hall, 1996).

8.4.2 Metal±metal bearing surfaces Oil-lubricated metal bearings are common in machinery and they were also adopted in the early phase of the development of the modern THR. These first generation metal-on-metal bearings were based on stainless steel (McKee, 1951; Wiles, 1957). In the 1960s and early 1970s, the metal-on-metal design by © 2008, Woodhead Publishing Limited

Joint bearing surfaces and replacement joint design

181

McKee and Farrar became popular and successfully competed with the Charnley-type metal-on-polyethylene designs. These prostheses were made of CoCrMo alloy and had head diameters of 32±42 mm. Poor fit between head and socket resulted quite often in high friction and wear although significant improvements were achieved with time. It was assumed that high friction values measured for the sliding pairs of the early McKee±Farrar joint caused failure of the bone cement fixation due to high torque on the cup. Later, simulator studies have confirmed that the wear rates are low and the loosening was more likely caused by mechanical factors. The bearing surface qualities and tribological properties of the later McKee prostheses were actually very good. Retrospective studies have shown that in clinical use wear was very low as long as the bearing components were properly matched with 0.15 to 0.20 mm clearance to allow a lubricating fluid interface. The first generation metal-on-metal bearings and prostheses were practically abandoned in the 1970s partly due to their problems and the better overall results achieved by the Charnley-type low friction implant which had bearing solutions based on polyethylene. Weber was one of the first to realise that in fact the low wear rates in metalon-metal THRs could be related to reduced loosening (Weber et al., 1989). His findings indicated that technically well-implanted first generation metal-onmetal prostheses usually had a very good clinical and radiological outcome. Based on these findings, Weber and his industrial partners started the development of the second generation metal-on-metal sliding pairs to optimise the clearance between head and liner, to enhance the low roughness and microstructure of CoCrMo alloy by using a wrought alloy instead of cast alloys, to develop optimal sphericity on the bearing and to use modern quality control. The resulting MetasulTM metal-on-metal THR prosthesis was introduced to clinical use in 1988 and approved by the US Food and Drug Administration (FDA) in 1999. Based on the promising outcome, most of the major implant manufacturers have a metal-on-metal THR system on the market. During the past decade, most hip implant manufacturers have introduced surface replacements having metal-on-metal bearing combinations with large head diameters of about 44±54 mm. Both the metal ball and the socket are manufactured with high precision tolerance. Better CoCrMo alloys and high precision on the bearing surfaces have made it possible to utilise these larger heads without increasing the wear rate significantly in spite of longer sliding distance with increasing head size. An additional benefit is the improved stability of the hip joint and reduced risk for dislocation, by as much as a factor of 10 compared with conventional implants. Furthermore, the reduced risk of impingement and point contact damage in rim contact during separation of the head and cup lead to decreased risk for severe damage to the bearing surfaces. The wear rates of bearing surfaces of metal-on-metal prostheses have been estimated both from retrieved implants and by using hip joint simulators. Generally, during an initial running-in phase, the wear rate is around 20 m/year © 2008, Woodhead Publishing Limited

182

Joint replacement technology

corresponding to a volumetric wear rate of about 2 mm3/year. In the steady state, the wear rate is much lower, i.e. about 0.1±0.2 mm3/year, and remains about the same for forthcoming years. This might be due to the fact that the bearing surfaces are to a certain degree self polishing and tough. The size of metallic wear particles is generally below 100 nm and therefore they are too small to irritate the resident macrophages (Willert et al., 1996). In patients with metalon-metal THRs, an elevation of Co and Cr concentrations in blood serum and urine has been measured (Brodner et al., 1997). These concentrations seem to decrease after the initial one-year run-in phase. It has been thought that patients with chronic renal failure should not have a metal-on-metal THR implanted. Although hypersensitivity to the metals is theoretically possible, there is not much data on its clinical relevance (Willert et al., 2000). Furthermore Visuri and Koskenvuo (1991) showed that over a period of 15 years there was no elevated cancer risk in patients with McKee±Farrar type CoCrMo metal-on-metal THRs. As a conclusion the clinical experience accumulated from current and older metal-on-metal designs is extensive. Such bearings have been developed to be tough, nearly unbreakable and in the long run generate very little wear debris.

8.4.3

Ceramic±ceramic bearing surfaces

Owing to the extreme hardness of ceramics they should provide the lowest wear rate as sliding surfaces unless critical fractures occur. In hard sliding pairs, in addition to smoothness, an accurate fit of articulating surfaces is even more important than with pairs using polyethylene. Current manufacturing techniques can provide tolerances of bearing surface sphericity better than 1 m. These well-matched ball±cup pairs should allow hydrodynamic lubrication with a continuous fluid film. However, clinical surveys have indicated that articulating surfaces are partly in contact and that adhesive and abrasive wear occur (Streicher, 1995). Therefore, in identical sliding pairs, the materials should be as hard as possible to minimise wear. Aluminium oxide (alumina) and partially stabilised ZrO2 (zirconia) are the most used ceramics on bearing surfaces and they have been systematically studied since the early 1970s. The clinical performance of first generation ceramic±ceramic implant systems was poor owing to shortcomings in fixation, design and material properties. Sudden component fractures and catastrophic wear were quite common. Through a systematic development in material properties and proper designs, more consistent properties and wear performance have been achieved. In fact, a modern aluminium oxide ceramic head paired with an acetabular component of the same material has turned out to be reliable in clinical use and has been recommended especially for young and active patients (Sedel et al., 1994). Very low linear wear rates and low concentrations of wear particles in periprosthetic tissues have been confirmed by several clinical studies (e.g. BoÈhler et al., 2000). Aluminum oxide ceramics provide © 2008, Woodhead Publishing Limited

Joint bearing surfaces and replacement joint design

183

superior performance in terms of wear resistance, bioinertness and little local cellular response to wear particles compared with other bearing couples (BoÈhler et al., 2000; Mochida et al., 2001). The wear properties including lubrication, friction and wear in alumina-onalumina bearings are excellent. Clinical data of wear rates for alumina±alumina pairs vary a lot for several reasons such as stability/instability of fixation, possible catastrophic failures and poor material grades and design. For wellfixed and stable implants the wear rates of 1±3 m/year are quite typical (Clarke and Willmann, 1994; Skinner, 1999). As mentioned above, good performance of alumina ceramics on sliding surfaces requires systematic development in material characteristics, especially a fine and uniform grain size to achieve high density, elimination of flaws and greater strength, by a manufacturing method called hot isostatic pressing (HIP). In addition to better fracture resistance, this kind of fine-grained material also wears less. In the case of ceramics, a single critical flaw can lead to complete failure of an implant. Therefore, 100% proof testing during the final inspection of finished ceramic implants has improved reliability since high stresses applied in testing would immediately propagate cracks from critical flaws, for example on the bearing surface, and lead to failure of the implant in testing. The criteria for such testing are based on empirical and theoretical data. Clinical fractures have decreased from around 10% in the 1980s to 0.004% with contemporary alumina implants meeting stringent quality criteria (Willman, 1999). Alumina wear particles as well as bulk alumina are very biocompatible with an average particle size below a micrometre. They are removed from the hard sliding surface and phagocytosed easily by macrophages in tissues, leading to a benign response. Nowadays the excellent tribological properties of alumina ceramic bearings can be utilised with several successful THR systems and designs. Current modular ceramic bearings are also clinically successful with good tribological properties, very low wear rate, and good local and systemic biocompatibility.

8.4.4

Coatings for bearing surfaces

In principle, coatings can be utilised on bearing surfaces of both hard±soft and hard±hard bearing couples. Several coating methods can be effectively used to modify implant bearing surface properties such as wettability (contact angle) or scratch and wear resistance. Although the typical average surface roughness of metal implant gliding surfaces initially is in the range 0.01±0.05 m, the roughness of these surfaces increases with time because of wear, corrosion and third body particles. This roughening can significantly increase the wear rate of metal and especially a polymer counterpart. Attempts to minimise polyethylene wear have been tried using two main approaches, namely, by optimising the properties of crosslinked polyethylene © 2008, Woodhead Publishing Limited

184

Joint replacement technology

and by optimising the material properties of the counterface. In addition to bulk ceramics like alumina or zirconia, coatings such as TiN (Pappas et al., 1995), CrN, CrCN (Fisher et al., 2002), zirconium oxide (OxiniumÕ, Smith&Nephew Orthopaedics, Memphis, TN), alumina (Yen and Hsu, 2001) or amorphous diamond (Lappalainen et al., 1998, 2003; Affatato et al., 2000) have given some promising results. In principle, these coatings can be readily applied to metallic implant materials to provide a high strength ceramic outer layer. This ceramic layer can improve, for example, wettability of the implant, which is known to improve tribological performance of bearing surfaces in the case of sliding pairs with polyethylene. If the ceramic layer is thick enough, it can also protect the metal surface against third body particles (Santavirta et al., 1999). These might be ceramic components from the bone cement used for implant fixation, i.e. typically ZrO2 or BaSO4, which are two agents commonly used to make bone cement radiopaque. Ceramic coating may provide good improvement in corrosion resistance as well, if its adhesion and corrosion resistance on the metallic substrate are good on long-term exposure to body fluids. Unfortunately, these requirements have not been met in all the recently used clinical coated implants, leading to poor outcome in long-term clinical use. Based on the simulator experiments and clinical trials, ceramic surfaces can remain stable and minimise long-term PE wear with typically two to four times lower wear than with Co± Cr±Mo heads (McKellop, 1998; Oonishi et al., 1989, Schuller and Marti, 1990). On the other hand, coatings can give much worse results than uncoated CoCrMo against polyethylene (e.g., Jones et al., 2001). This is because a certain class of materials and coatings prepared using different methods and set-ups may have a large variation in characteristics and behaviour. Ceramic coatings have potential for hard sliding pairs due to good wear resistance. In this combination, in addition to smoothness, an accurate fit of bearing surfaces is even more important than with pairs using PE. However, high tolerances can be achieved using current manufacturing techniques. Theoretically, the well-matched ball±cup pairs should allow hydrodynamic lubrication with a continuous fluid film at the gliding interface. However, the clinical surveys have shown that articulating surfaces are partly in contact and that adhesive and abrasive wear occur (Streicher, 1995). Therefore, the coating material should be smooth and as hard as possible to minimise wear. Even in this case, ceramic coatings could offer several advantages over bulk ceramics. For example, because of extreme hardness and good tribological characteristics of diamond, continuous film lubrication is not needed in the case of amorphous diamond coatings. When the coating is thick enough (>20 m), it can withstand high contact stresses and the wear rate is negligible (less than 10 nm per 15 million cycles in a simulator) (Lappalainen et al., 2003). As shown in Fig. 8.3, ceramic particles of bone cement cannot damage the coating in simulator testing. In contrast, a CoCr head is easily damaged, leading to increasing wear of both the head and the cup. Furthermore, the coefficient of friction is generally fairly © 2008, Woodhead Publishing Limited

Joint bearing surfaces and replacement joint design

185

8.3 Comparison of total wear (cup and ball) in a hip simulator test with bone cement third body particles for 2 million cycles in the middle of 5 million series. (MOM ˆ metal on metal; COC ˆ ceramic on ceramic: AD-on-AD ˆ amorphous diamond coating on amorphous diamond coating.)

low (<0.1) for ceramic-on-ceramic sliding pairs even in the early stages of an implant life cycle. Low friction is accompanied by low bending torque on fixation surfaces of the prostheses. However, the most important advantage of ceramic coatings compared with bulk ceramics is the fact that they are less prone to sudden complete failure, which is a feared, though rare (less than 0.1%), complication of current ceramic-on-ceramic total hip sliding pairs.

8.5

Special concepts and designs for bearing surfaces

Sometimes only one surface of a natural joint is damaged, e.g. the femoral head is damaged due to necrosis after a fracture of the femoral neck. In this case, the diseased femoral head can be replaced with an artificial ball of the same dimensions, typically 45±56 mm in diameter in a surgical procedure called hemiarthroplasty. The artificial head then articulates against the healthy cartilage in the socket. The ball may be rigidly fixed on a stem (monopolar) or a large ball may articulate freely on a smaller ball with a polyethylene liner between them (bipolar design). These designs lead to bearing surfaces consisting of natural cartilage and artificial implant material. In spite of the fact that hemiarthroplasty is a well-established treatment, several animal in vivo studies and human followup studies with metal implants have shown that cartilage was severely worn © 2008, Woodhead Publishing Limited

186

Joint replacement technology

especially in long-term use (e.g. van der Meulen et al., 2002; Parsons et al., 2004). Evidently, better solutions would be welcome for this application, too. Since less invasive procedures are becoming more and more popular, different procedures have been developed, for example to replace part of the joint surface due to defect. These techniques include the use of tissue grafts, osteochondral grafts and cell-based techniques. In the best cases, these implants are well incorporated in the cartilage and functional even several months or years postoperatively.

8.6

Comparison of bearing surface solutions

One piece of evidence that different main bearing solutions, namely, highly cross-linked UHMWPE combined with metal or ceramic, CoCrMo±CoCrMo and alumina±alumina pairs, have their own merits and demerits is the fact that they are still in clinical use (Table 8.1). In spite of significant developments, none of them can provide all the advantages. On the other hand, they all have been clearly improved over the years and can offer THR longevity. Correct counterface design, proper material combinations, surface finish and tolerances are necessary to optimise tribological aspects. Based on long comparative simulator studies as well as implant registers, only minor differences in performance can be assumed on average. Figure 8.4 compares the typical annual penetration rates and friction values measured for different bearing combinations. However, some recommendations can be drawn based on the properties of different materials and clinical experiences. For example, if metal hyperTable 8.1 Comparison of strengths and weaknesses of different bearing materials Joint bearing

Strengths

Weaknesses

Metal-on-UHMWPE

· Many options · Toughness

· Extreme wear

Metal-on-crosslinked UHMWPE

· Many options · Toughness

· Prone to third-body wear

Ceramic-on-crosslinked UHMWPE

· Reduced wear · Abrasion resistance · Low friction

· Fracture risk · No head exchanges · Fewer sizes

Metal-on-metal

· Reduced wear · Head size options · Toughness

· High ion levels · Fewer liner options · Sensitive to abrasion

Ceramic-on-ceramic

· Reduced wear · Abrasion resistance · Low friction

· Fracture risk · Limited options · Revision challenges

Ceramic coatings

· All those listed above

· Delamination risk

© 2008, Woodhead Publishing Limited

Joint bearing surfaces and replacement joint design

187

8.4 Different combinations for bearing surfaces with their annual clinical penetration values and friction coefficients.

sensitivity might be a problem or retrieval (revision surgery) is carried out due to metallosis, UHMWPE combined with a ceramic head or totally ceramic-onceramic pair might be a better choice. On the other hand, if impingement or luxation are to be expected or retrieval (revision surgery) is due to ceramic failure, polyethylene cups or metal±metal combinations are more reliable, since they are not easily fractured by impingement or impacts on the rim section of the cup.

8.7

Future trends

Although several solutions for bearing couples already exist, some further developments can be expected, based on current know-how; these are listed next. One major advantage of better materials for bearing surfaces, for example highly crosslinked polyethylene and wear-resistant CoCrMo alloys by powder metallurgy, is the possibility of using larger head sizes in hip implants without significantly increasing wear debris release and the risk of osteolysis. Larger head sizes reduce the risk for impingement and luxation. Novel polymers such as polycarbonate urethane have shown potential for use as an acetabular bearing material (Khan et al., 2005). Polyurethane elastomers © 2008, Woodhead Publishing Limited

188

Joint replacement technology

have a unique combination of durability, toughness and flexibility in comparison with UHMWPE, and may provide a good alternative after several years of systematic development. In spite of the serious shortcomings of pure quality coatings on bearing surfaces, coatings such as amorphous diamond or carbon nitride have the potential to give good long-term results as hard±hard bearing couples, if the coatings fulfil stringent requirements (Lappalainen and Santavirta, 2005). Perhaps the most important advantage of ceramic coatings compared with bulk ceramics is that they are less prone to sudden complete failure, which is a feared, if rare (less than 0.1%), complication of current ceramic-on-ceramic total hip sliding pairs.

8.8

References

Adamson AW, Physical Chemistry of Surfaces, New York, Wiley, 1967. Affatato S, Frigo M, Toni A, `An in vitro investigation of diamond-like carbon as a femoral head coating', J Biomed Mater Res, 2000 53 221±226. Atkinson JR, `Laboratory wear tests and clinical observations of the penetration of femoral heads into acetabular cups in total replacement hip joints', Wear, 1985 104 225±244. BoÈhler M, Mochida Y, Bauer TW, Salzer M, Plenk Jr H, `Characterization of wear debris from alumina-on-alumina THA', J Bone Joint Surg, 2000 82-B 901±909. Brodner W, Bitzan P, Meisinger V, Kaider A, Gottsauner-Wolf F, Kotz R, `Elevated serum cobalt with metal-on-metal articulating surfaces', J Bone Joint Surg, 1997 79-B 316±321. Charnley J, `Lubrication of animal joints', In Proceedings of I Mech E Conference on Biomechanics, London, 12, 1959. Clarke IC, Willmann G, `Structural ceramics in orthopedics', in Cameron HU, Bone Implant Interface, Mosby, St. Louis, 203±252, 1994. Fisher J, Hu XQ, Tipper JL, et al., `An in vitro study of the reduction in wear of metal-onmetal hip prostheses using surface-engineered femoral heads', Proc Inst Mech Eng H , 2002 216 219±230. Hall RM, `Wear in retrieved Charnley acetabular sockets', J Eng Med, 1996 210 197±207. Jin ZM, `Biotribology', Current Orthopaedics, 2006 20 32±40. Jones VC, Barton DC, Auger DD, Hardaker C, Stone MH, Fisher J, `Simulation of tibial counterface wear in mobile bearing knees with uncoated and ADLC coated surfaces', Biomed Mater Eng, 2001 11 105±115. Khan I, Smith N, Jones E et al., `Analysis and ealuation of a biomedical polycarbonate urethane teted in an in vitro study and an ovine arthroplasty model', Biomaterials, 2005 26 633±643. Lappalainen R, Santavirta S, `Potential of coatings in total hip replacements', Clin Orthop, 2005 430 72±79. Lappalainen R, Anttila A, Heinonen H, `Diamond coated total hip replacements', Clin Orthop, 1998 352 118±127. Lappalainen R, Selenius M, Anttila A et al., `Reduction of wear in total hip replacement prostheses by amorphous diamond coatings', J Biomed Mater Res, 2003 66 410± 413.

© 2008, Woodhead Publishing Limited

Joint bearing surfaces and replacement joint design

189

McKee GK, `Artificial hip joint', Proceedings of the East Anglian Orthopaedic Club, J Bone Joint Surg, 1951 33-B 465. McKellop H, `Assessment of wear of materials for artificial joints', in Callaghan J, Rosenberg A, Rubash H, The Adult Hip, New York, Lippincott-Raven, 231±246, 1998. Mochida Y, BoÈhler M, Salzer M, Bauer TW, `Debris from failed ceramic-on-ceramic and ceramic-on-polyethylene hip prostheses', Clin Orthop, 2001 389 113±125. Oonishi H, Igaki H, Takayama Y, `Comparison of wear of UHMWPE sliding against metal and alumina in total hip prostheses', Bioceramics, 1989 1 272±277. Pappas MJ, Makris G, Buechel FF, `Titanium nitride ceramic film against polyethylene. A 48 million cycle wear test', Clin Orthop 1995 317 64±70. Parsons IM, Millett PJ, Warner JJ, `Glenoid wear after shoulder hemiarthroplasty: quantitative radiagraphic analysis', Clin Orthop Relat Res 2004 421 120±125. Santavirta S, Lappalainen R, Pekko P, Antila A, Konttinen YT, `The counterface, surface smoothness, tolerances, and coatings in total joint prostheses', Clin Orthop, 1999 369 92±102. Schmidt M, Weber H, SchoÈn R, `Cobalt chromium molybdenum metal combination for modular hip prostheses', Clin Orthop, 1996 329 35±47. Schuller H, Marti R, `Ten-year socket wear in 66 hip arthroplasties. Ceramic versus metal heads', Acta Orthop Scand, 1990 61 240±243. Sedel L, Nizard RS, Kerboull L, Witvoet J, `Alumina±alumina hip replacement in patients younger than 50 years old', Clin Orthop,1994 298 175±183. Skinner HB, `Ceramic bearing surfaces', Clin Orthop Rel Res 1999 369 83±91. Streicher RM, `Tribology of artificial joints', in Morscher EW, Endoprosthetics, Berlin, Springer, 34±48, 1995. van der Meulen MCH, Beaupre GS, Smith RL et al., `Factors influencing changes in articular cartilage following hemiarthroplasty in sheep', J Orthop Res, 2002 20(4) 669±675. Visuri T, Koskenvuo M, `Cancer risk after McKee total hip replacement', Orthopedics 1991 14 137±142. Wiles P, `The surgery of the osteoarthritic hip', Br J Surg 1957 45 488±497. Willert HG, Buchhorn GH, GoÈbel D, KoÈster G, Schaffner S, Schenk R, Semlitsch M, `Wear behaviour and histopathology of classic cemented metal on metal hip prostheses', Clin Orthop 1996 329 160±186. Willert HG, Buchhorn GH, Fayyazi A, Lohmann CH, `Histopathologische VeraÈnderungen bei Metall/Metall Gelenken geben Hinweis auf eine zellvermittelte È berempfindlichkeit: vorlaÈufige Untersuchungsergebnisse von 14 FaÈllen', U Osteologie 2000 9 2±16. Willman G, `Ceramic ball head retrieval data', in Sedel L, Willmann G, Reliability and Long-Term Results of Ceramics in Orthopaedics, 4th International CeramTec Symposium, Stuttgart, Georg Thieme Verlag, 1999. Weber HG, Fiechter T, `PolyaÈthylen-Verschleiss und SpaÈtlockerung der Totalprothese des HuÈftgelenkes ± Neue Perspektiven fuÈr die Metall/Metall Paarung fuÈr Pfanne und Kugel', OrthopaÈde 1989 18 370±376. Yen SK, Hsu SW, `Electrolytic Al2O3 coating on Co-Cr-Mo implant alloys of hip prosthesis', J Biomed Mater Res 2001 54 412±418.

© 2008, Woodhead Publishing Limited

9

Cementless fixation techniques in joint replacement

M J C R O S S and J S P Y C H E R , The Australian Institute of Musculoskeletal Research, Australia

9.1

Introduction

Modern joint replacement without the use of cement demonstrates outstanding results and compels us to question the routine of using cement to improve the quality of patient care. Survivorship has been reported to be 96% at 13 years1 for cementless total knee arthroplasty and 97.5% for femoral stems in total hip arthroplasty after 20 years.2 These excellent results are possible by applying the main principles of cementless fixation in bone. These features include implant mechanics and design, rigorous surgical technique, knowledge of surface structure properties, bone ingrowth behaviour, and biological and pharmacological enhancement methods. Without doubt, the discovery and use of poly(methylmethacrylate) (PMMA) revolutionised the history of joint replacement and implant fixation in bone when it was introduced in the late 1950s. The credit for discovering the use of cement in orthopaedic surgery belongs to Sir John Charnley, who began his pioneering work at Wrightington in the United Kingdom.3±5 After some initial failures with materials (Teflon instead of polyethylene) the combination of bone cement and high-density polyethylene rapidly became the gold standard in hip replacement. It took a number of years before the US Food and Drug Administration (FDA) allowed the use of bone cement and once the procedure was set in motion, other joints beyond the hip were considered. In 1971 Gunston introduced the first cemented arthroplasty of the knee joint.6 The first shoulder, ankle and elbow prostheses were introduced and slowly the age of cement took a firm grip on the orthopaedic community and bestowed it with unprecedented success in arthroplasty. It was not until the early 1980s that numerous reports appeared in the literature documenting the drawbacks of cement.7±11 The main reasons for prosthesis failure were osteolysis and loosening of the implants. Cement disease became apparent in revision surgery. The two biggest problems were bone loss, and cement removal, particularly in femoral stems. The combination of cement © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

191

debris and polyethylene particles was an accelerating factor in osteolysis. Recently it has been shown that polyethylene particles are the major factors in osteolysis.8 These failures prompted renewed interest in cementless fixation that focused on how to improve the qualities of implants to allow permanent stability in bone, without the use of an additional interface. Extensive research in this field merits the successful long-term outcomes of large clinical studies published in current literature. This in turn will continue to produce innovative discoveries in material technology and biological modulation. This chapter advances our knowledge of uncemented prostheses, showing that excellent long-term results can be achieved by understanding and applying the basic principles of cementless fixation.

9.2

Cementless fixation

The primary goal of cementless fixation is to improve the longevity of the implant. To achieve this objective certain fundamental principles must be applied and any neglect can lead to the implant's early failure. The three main principles that guarantee the best outcome of cementless implant fixation are: 1. sound initial stability; 2. osseous integration; 3. mechanical properties of the implant. Failure to follow these principles caused many of the early uncemented prostheses to fail. A good example is the early development of cementless total knee replacements. Early designs were insufficient in both mechanical and anatomical terms. An important factor in the design of tibial prostheses is peripheral cortical support and this requires a sufficient number of differing sizes to fit the varying anatomical configurations. The porous coated anatomic (PCA) knee replacement had an insufficient range of tibial sizes, allowing sinkage into the cancellous bone of the tibial metaphysis. Other issues associated with the failure of cementless designs had nothing to do with fixation methods. Rapid osteolysis due to particulate debris of early polyethylenes, insufficient polyethylene-locking mechanisms in hips and knees as well as flat articulating surfaces in knees occurred in both cemented and uncemented implants. These were failures related to design features and particularly the materials used. A classical example of inappropriate material in total knee prostheses was the Miller Galante I that had a titanium femoral component. As a bearing surface, titanium proved to be too soft. This caused premature wear and severe synovitic metallosis as well as rapid production of early polyethylene particulate debris. This in turn led to massive osteolysis and bone loss, making revision difficult.

© 2008, Woodhead Publishing Limited

192

9.3

Joint replacement technology

Initial stability

Rigid initial stability of a prosthesis is achieved when the forces required to dislodge the components are equal to or less than the physiological forces to which the implant is exposed upon postoperative loading. This involves an immediate tight fit in or on the bone, either by a three-dimensional press-fit configuration or by additional fixation using screws. The specific loading mechanics are different for each joint and depend on their anatomy and function. When a joint is subjected to compressive forces it is more amenable to cementless fixation. The most successful cementless implants in joint replacement have been the acetabular cup and the femoral stem (Fig. 9.1), followed by the distal femur. For primary total hip arthroplasty, the uncemented press fit acetabular cup has emerged as the most common standard of fixation. The compressive forces directed from the centre of rotation to the cup±bone interface result in an ideal mechanical equilibrium where shear and rotational forces are practically negligible. In this case the cup is press fitted (the reamed diameter is slightly smaller than the actual diameter of the cup). The proximal femur, however, is exposed to more complicated patterns of rotational, shear, tensile and compressive forces that place higher demands on implant design. Bourne et al.12 evaluated the stability of fixation, stress shielding and thigh pain of cementless stem designs. They concluded that the tapered stem showed better results than anatomical or cylindrical stems. A proximal femoral prosthesis must resist initial `sinking' (taper) and rotational forces (cross-section, in-built anteversion). In the knee it is more difficult to achieve stable initial fixation of the tibial component than of the femoral component. Resurfacing of the distal femur allows for a component, designed to be fixed in a press-fit fashion with the anterior and posterior bevels at an angle of 1±3ë. This creates a minimal wedge for the femoral epiphysis to fit into distributing forces over the whole component and avoids the problem of too much loading of the distal surface. The pattern of compressive forces along the tibial component during normal gait is complex. Due to rolling, sliding, toggling and rotation of the femur on the

9.1 Uncemented acetabular cup and femoral stem. © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

193

tibia, application of an unsupported flat tray on the proximal tibial surface does not provide sufficient resistance to compressive failure of cancellous bone, lift off and micromotion. Miura et al.13 showed that the addition of a stem or screws are mandatory to achieve stable initial fixation. Surface roughness of the implant interface is an additional determinant to initial stability.14±16 The resistance to frictional motions is substantially increased by the presence of a rough surface. A minimum roughness of 3.5 m has proven favourable for the promotion of sufficient primary fixation and for growth stimulating micro-movements. After establishing the optimal design and instrumentation of a given prosthesis, the next step in gaining initial rigid fixation is careful surgical technique. Preoperative planning is key. Special attention should be given to deformities, bone loss and correcting malalignments. Instrumentation of the prosthesis must yield precise bone cuts, and bone moulding must accurately accommodate the implant. Undersizing the mould of an acetabular cup or femoral metaphysis in short-stem prosthesis significantly increases the stability of implant fixation.16±20 Special care must be given not to apply deforming forces on jigs or instruments that can jeopardise accuracy. For implants that rely on cortical support, e.g. the proximal tibia, efforts must be made to choose the ideal implant size that maximises cortical contact area without over sizing the component. Therefore, tibial trays need to have at least 10 sizes (personal experience of over 7000 cases). Various studies have shown that the compaction of cancellous bone increases the initial fixation strength of the implant up until four weeks postoperatively.21±23 This simple technical detail must be considered in instrument design and should be applied wherever possible, especially in proximal femora and tibiae. It is also important to note that if the periphery of the implant succeeds in producing complete bone ingrowth and in creating a watertight fit, then the transport of particulate debris to the bone±implant interface is rendered virtually impossible.24 After having performed the bone cuts, no attempt should be made to clean the surfaces of blood and its osteogenic marrow components until firm implantation of the prosthesis. The quality of bone influences initial stability.25 In cases of gross osteoporosity a finger pressure test can be carried out. A positive result indicates a possible contra-indication of an uncemented design. Please refer to Section 9.6 on `Why do you still use cement?' Only if rigid initial stability is realised can the crucial secondary phase of successful osseo-integration begin, which in the long run will determine the longevity of the fixation.26,27

9.4

Osseous integration of cementless implants

The direct intimate contact of bone tissue to the surface of an implant is unique to the technology of cementless fixation and is key to preventing the effects of © 2008, Woodhead Publishing Limited

194

Joint replacement technology

implant loosening. The preference for cementless fixation is that osseointegration occurs rapidly; thereby minimising the risk that early forces will act on the implant causing displacement. The main aspects affecting the quality and rate of osseo-integration are: · · · · ·

implant stability, micromotion and interface gaps; surface geometry characteristics; biocompatibility of materials; bioactive surface coatings; physiology of osseo-integration.

9.4.1

Implant stability, micromotion and interface gaps

Excessive micromotion of an implant in bone renders bone ingrowth impossible. The amount of permitted minimal movement within an interface has been reported to be 28±150 m, and repetitive higher displacements allow only ingrowth of fibrous tissue.28,29 Micromotion is a function not just of primary implant stability but also of the differences in the elastic modulus of bone and in the implant material. In tibial trays this mismatch can produce motions of up to 150 m. Retrieved implants have shown best bone ingrowth near fixation pegs and inconsistent ingrowth patterns at the periphery.30 Even for the most talented surgeons, it remains impossible to produce an absolute conforming host site that shows no gaps over the entire surface area. Those areas that lack direct contact have a negative effect on osseo-integration. Numerous models with analysis of controlled gaps in stable implants reveal that bone ingrowth is reduced up to sixfold in the presence of a 2 mm gap as compared with direct bone contact, These results demonstrate the supreme importance of meticulous surgical technique.20,31 In our study of hydroxyapatite coated total knee replacements we proved that gaps up to 2 mm usually filled in after two years, but only if the implant was absolutely stable.32

9.4.2

Surface geometry characteristics

Numerous modifications of surface geometry were studied to determine optimal pore sizes and roughness. Porous surfaces are usually applied as a coating on the implant through a variety of techniques. The most common techniques involve the application of vacuum-sputter or plasma-sprayed coatings, flame spraying techniques or, alternatively, roughening the surfaces through grit blasting (Fig. 9.2).33 Friedman et al. have demonstrated higher shear strength for arc-deposited titanium (flame spraying technique), than for grit-blasted surfaces or sprayed beads.34 The ideal pore size that allows bone ingrowth has been extensively investigated, mostly in canine models.35±38 These studies indicate that the © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

195

9.2 Beaded coating on tibial tray.

optimal range of pore sizes is between 100 and 400 m. Although most prostheses are coated with a porosity size within this range, retrieval analysis has shown that in tibial plateaux bone ingrowth is not routinely achieved. Further research led to cast mesh being applied to the surface of implants, which demonstrated superior fixation strength when compared to porous coatings.39,40

9.4.3

Biological compatibility of materials

Modern bone implant materials can be classified according to their biocompatibility into biotolerant, bioinert and bioactive materials.41 Biotolerant materials such as PMMA are characterised by a thin fibrous tissue interface, whereas bioinert materials such as titanium and aluminium oxide typically integrate well into bone. Bioactive materials, such as calcium phosphate ceramics and glass, demonstrate direct chemical bonding of the implant with bone. Wilke et al.42 proved that cell proliferation as a marker for osseo-integration was highest on hydroxyapatite surfaces, followed by titanium and chrome±cobalt±molybdenum alloy. Under transmission electron microscopy the implant±bone ultrastructure of commercially available pure titanium, Ti6Al4V, and cobalt±chromium alloy are comparable. In recent years, a new implant material has produced substantially higher volumetric bone ingrowth into implants. It possesses a much higher porosity ± up to 80%, compared with previous porosity of only 35±50% ± without sacrificing the structural integrity of the implant. This new implant consists of commercially pure tantalum that is deposited on a vitreous carbon skeleton by chemical vapour deposition/infiltration. This trabecular-like metal can be made © 2008, Woodhead Publishing Limited

196

Joint replacement technology

into complex shapes and used either as a bulk implant or as a surface coating. It shows a remarkably high resistance to corrosion and excellent biocompatibility. Its high frictional properties make it conducive to biological fixation and its low modulus of elasticity allows for more physiological load transfer and relative preservation of bone stock. Most clinical publications to date report very promising results with this novel material.43±47 A radiographic analysis of 86 acetabular cups showed that all gaps up to 5 mm were filled after 24 weeks of implantation indicating the strong osteoconductive, possibly even osteoinductive properties of trabecular metal.48 An experimental study conducted by Bobyn et al.49 in a canine model proved bone ingrowth into pores averaging 430 m of up to 80% with histological evidence of Haversian remodelling at 52 weeks.

9.4.4

Bioactive surface coatings

The quest for improving the rate and amount of osseo-integration of an implant in bone led to the introduction of osteo-conductive surface chemistry and osteoinductive biomodulators on or in implant surfaces. The application of calcium phosphates, especially hydroxyapatite (HA), as an osteo-conductive mediator has significantly improved the quality of implant fixation. This has been demonstrated in numerous experimental and clinical studies. Cooley et al. investigated ingrowth of HA-coated titanium compared with pure titanium at three time periods. Mechanical evaluation showed significantly greater interface bond strength, and histological analysis revealed nearly twice the percentage of direct bone contact for coated implants.50 Soballe and coworkers51±53 showed that HA coatings in dogs significantly enhanced gap healing and, when exposed to polyethylene particles in unstable implants, provided a superior sealing effect at the periphery. The same study showed shorter healing time, higher tolerance to micromotion and improved loading anchorage in bone. Many later clinical studies have confirmed the superior results of HA-coated implants.54±59 In our review of 1000 consecutively performed total knee replacement surgeries that employed HA-coated stemless implants, we demonstrated a survivorship of 99% at 10 years with excellent clinical and functional outcome.60 In our collective of patients over 75 years of age, as well as younger very active patients, we observed a low rate of revision and infection.61,62 Especially in the younger age group, where previous studies have shown unsatisfactory mid- to long-term results for cemented knee replacement, we strongly advocate the use of a HA-coated, uncemented knee replacement (Fig. 9.3). We believe the main advantage of HA coatings on implants lies in accelerated and increased bone ingrowth, earlier stability, reduced migration rates and less loosening, all of which allows earlier return to normal function. This incorporation seals the bone±prosthesis interface blocking synovial penetration which causes osteolysis.

© 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

197

9.3 Hydroxyapatite-coated total knee replacement.

9.4.5

Physiology of osseo-integration

Bone ingrowth refers to the formation of bone within an irregular surface of an implant, which improves the implant's integration into bone. The presence of a porous-coated implant evokes a cellular and physiological response that resembles the healing cascade of cancellous defects. In porous implants, the void spaces are filled with newly formed bone tissue in a stable situation.63 This process is similar to that of primary fracture healing in stable osteosynthesis, in which haematoma develops into mesenchymal tissue that is then replaced by woven bone and eventually undergoes remodelling to lamellar bone without ever passing through an intermediate stage of fibrocartilagenous tissue. Various studies have proven that this process is enhanced in an optimal mechanical environment of suitable loading (Figs 9.4(a) and (b)).64±66

9.5

Mechanical properties of the implant

One of the key challenges in the development of implant materials is the mechanical mimicry of biologically active bone. Bone is a composite material © 2008, Woodhead Publishing Limited

198

Joint replacement technology

9.4 (a) and (b) Bone ingrowth into beaded HA-coated surface on a retrieval implant.

made of a collagenous fibre matrix stiffened by HA crystals, which account for 69% of bone mass. It is constantly in dynamic interaction with its physiological and mechanical environment. One of the most important terms in implant technology is stiffness. Stiffness is an extensive material property that is defined by the resistance of an elastic body to deflection or deformation by an applied force. This is measured by Young's modulus of elasticity, E, which is a ratio for the rate of change of stress (force per area) with strain (deformation) in gigapascal (GPa). Cancellous bone has an E of 0.04±1.0 GPa, pure titanium 105 GPa, titanium alloys 55±110 GPa and cobalt chromium alloys 200±250 GPa. The regenerating and remodelling process in bone is directly related to loading. As Wolff's law states: bone subjected to loading or stress will regenerate and bone not subjected to stress will atrophy. An implant that is much stiffer than bone demonstrates an unphysiological redistribution of force transmission at the interface, which is referred to as stress shielding. Locations of high load transmission cause surrounding bone to generate (hypertrophy) and locations of reduced load transmission cause surrounding bone to degenerate. Numerous studies have shown that the degree of stress shielding is directly related to the difference in stiffness of implant and bone.67±70 In hip replacement, this adverse effect results in proximal/metaphyseal bone loss and potentially compromises the long-term stability of an implant that may have shown ideal mid-term osseo-integration. This was particularly observed in proximal femoral stems made of cobalt±chromium alloys. Titanium, which has a much lower stiffness than cobalt chromium, was used as a femoral stem in cemented implants. The results were unfavourable due to early loosening, the titanium showed excessive flexibility in a stiff cement mantle. Only when titanium was applied in cementless fashion did it gain popularity in hip replacement. © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

199

One trend in recent years to minimise stress shielding has been toward components with only proximal/metaphyseal fixation in longer implants. One target for this trend has been the proximal femur, where elasticity mismatch has a higher effect. The new implants either restrict coating in stems of standard length to the metayphyseal section or produce short stem implants. Both techniques have shown an improvement of bone loss due to less stress shielding.71±73 Extensive research into material technology is being undertaken to minimise the mismatch of bone±implant stiffness. Advances in metallurgy have recently yielded new titanium alloys that undergo heat treatment and result in altered microstructures, i.e. metastable , that have reduced stiffness by 50% to 55 GPa. Tantalum, which was mentioned in Section 9.4.3, not only shows low stiffness and excellent biocompatibility, but should demonstrate a structural loading pattern similar to that of bone because the empty space of 80% volumetric porosity gets filled up by bone. An additional issue closely related to material stiffness and cementless implantation is micromotion. A study conducted by Simon et al.74 focused on the influence of implant material stiffness on stress distribution and micromotion at the interface. They found that the low-stiffness implant showed more homogeneous stress distribution with fewer peak loads than the high stiffness implants, but, contrary to their hypothesis, the low-stiffness implant showed more micromotion. The elasticity of a structural implant must be such that it resists deformation by physiological loading. This is especially true for thin, flat implants such as the tibial component in total knee replacements or for long implants such as the proximal femur in total hip replacement. Under such circumstances, low stiffness would allow for repetitive deformation, excessive micromotion at the bone±implant interface, and therefore early breakdown.

9.6

Why do you still use cement?

No prosthesis or implantation technique can fully create a biological replica of the tissues lost. This chapter attempts to shed some light on the advantages and disadvantages of uncemented fixation techniques, especially in joint replacement.

9.6.1

Uncemented implants and revision surgery

Even though the vast majority of implants survive for longer than 15 years, it is imperative that doctors consider the implications for revision surgery when inserting these devices. This is especially important in young patients. A recently published randomised, controlled trial by Khaw et al.75 of cemented versus uncemented total knee replacement showed equivalent survival rates after 10 years. Wood,76 in a recently published review of the most recent data on cemented versus uncemented total knee replacement, concluded that uncemented fixation has the potential to improve longevity of knee replacement © 2008, Woodhead Publishing Limited

200

Joint replacement technology

9.5 Aseptic loosening of tibial tray before revision, demonstrating minimal bone loss.

and delay the expenses and risks of revision surgery. Another study by Nilsson et al.77 comparing three different fixation techniques for the tibial component in total knee replacement concluded that uncemented fixation of HA-coated implants seemed to be the best solution, especially for younger patients, because their migration in radiostereometric analysis came to a halt after three months, whereas in cemented implants migration continued for up to two years. Regarding this last study, it is likely that this migration is due to micro-movements taking place in the thin fibrous layer between biotolerant cement and bone. Migration creates more play in the host site, leading to more motion and eventually more bone loss. We think that this may be an additional reason why we observe less bone loss at revision for aseptic loosening in an uncemented implant than we do in cemented implants (Fig. 9.5). An advantage of revising an uncemented prosthesis is a significant reduction in operating time and, in the case of knee revisions, in tourniquet time. Cemented implants often show irregular debonding of the interfaces between the implant and the cement and between the cement and the bone. Removal of cement from bone can be both arduous and time consuming, especially in the proximal femur. A possible, albeit rare, disadvantage in revision of an uncemented prosthesis involves a solid implant that has to be removed. This is seldom the case, since a stable implant does not often need to be removed even in the case of infection. On occasion, however, retrieval is required, for example when the revision component is not compatible with the stable implant. In such cases the removal of the stable implant can cause extensive bony destruction, such as the conical Wagner and threaded stems of the proximal femur.

9.6.2

Osteoporosis ± do you really need cement?

Severe osteoporosis can be a contra-indication to uncemented fixation, especially in implants that rely on a competent cancellous structure in the presence of © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

201

marginal cortical rim support, such as the tibial plateau. If osteoporosis is so severe that sufficient primary support is not given, then cementing is warranted. Age itself is not a reliable predictor of the degree of osteoporosis. Our experience in cementless total knee replacement shows that elderly patients over 75 years of age do just as well as patients in the younger age group.61 An interesting study conducted by Therbo et al.78 analysed the influence of preoperative bone mineral content of the proximal tibia in uncemented TKA on the revision rate. They found that low trabecular bone quality was not a predictor for later revision surgery. Takashi et al. also demonstrated that the type of osteoblastic response to osteoarthritis according to Bombelli79 as atrophic, normotrophic and hypertrophic had no effect on the predictive outcome of cementless cup fixation in total hip replacement after seven years. These results are confirmed by numerous other studies80±84 and relate to all forms of osteoporosis without regards to the etiology. Bone grafting in large defects, as seen in rheumatoid arthritis, haemophilia and pigmented villonodular synovitis, can with expert judgement be filled with local cuttings from the surgery, with excellent success. This will induce osteogenesis and is preferable to filling these defects with large amounts of cement (Figs 9.6 and 9.7).

9.6 Tibial bone grafting in post-traumatic osteoarthritis using an uncemented prosthesis.

9.7 Filling of a giant cyst in the distal femur with autologous bone graft. © 2008, Woodhead Publishing Limited

202

Joint replacement technology

Cementless fixation can thus reliably be accomplished in all but the most severe cases of osteoporosis. However, initial stability must be guaranteed to allow permanent secure fixation.

9.6.3

Infected implant in the presence of cement

Infection is a potentially disastrous complication in joint replacement. Surprisingly few publications evaluate the influence of cement on the risk for infection. It seems that the presence of an additional material and interface should at least minimally increase this risk. There is, however, one study from the Norwegian Arthroplasty Register that compared the risk of infection for cemented versus cement + gentamycin versus uncemented fixation out of 56 275 primary total hip replacements.85 Their results showed that the risk for infection was higher in the cement-only group, whereas no increased risk could be detected for both uncemented and antibiotic-loaded cemented groups. There is sufficient evidence in the literature to conclude that the presence of cement reduces resistance to infection.86±91 Once a prosthetic joint is infected, the retention potential is higher in noncemented arthroplasties. This has been demonstrated by Freeman et al.92 and correlates with our treatment of infected total knee replacements.93 A reason for this may be that the avascular cement±prosthesis interface could be a site where bacteria are protected from debridement and antibiotic penetration. In an infected prosthesis that is stable, has no periprosthetic radiolucent lines and no signs of osteomyelitis, there is a 75% chance of eradication of the infection by arthroscopic synovectomy while retaining the prosthesis.93

9.6.4

Saving cement saves time, costs and complications

Implantation of an uncemented prosthesis saves the time needed for cement preparation and polymerisation. This process takes up to 15 minutes and can take twice as long when proceeding with staged cementing. In knee replacement, cement preparation extends the tourniquet significantly, which in turn increases the possibility of deep venous thrombosis and pulmonary emboli. Hirota et al.94 used transoesophageal echocardiography to relate the number of emboli to tourniquet time. Reducing operative time puts patients, especially elderly ones, at less risk for general complications.

9.6.5

Blood loss

One disadvantage of cementless fixation is increased blood loss due to nonwatertight fit of the implant on a relatively large surface of cancellous bone. Yet this downside can be overcome. We were able to demonstrate in a prospective © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

203

randomised study that this blood loss could be reduced by 30% with the use of low vacuum drains and still ensure sufficient drainage of haematoma.95

9.6.6

Costs: modern uncemented implants are cheaper

One of the biggest arguments against cementless knee prostheses rests on the implant's higher price tag. Numerous studies have recommended cemented prosthesis on the basis of lower cost. We believe this argument does not properly consider the market. As the supply of cementless implants has increased, their price has decreased. In comparing the actual prices of a number of cemented total knee replacements and their uncemented counterparts on the market today, we notice a price increase of 0±10%. The most commonly used components are often the same price. This is even more the case in total hip replacements. An increased cost of 10% for materials is more than saved by the shorter use of the operating theatre and by obviating the cost of cement. Future expenses are also cut by reduced revision rates, higher retention rates in the case of infection and prevention of treatment costs that may arise from increased general complication rates due to longer surgery.

9.7

Future trends

Joint replacement has made major strides in recent years and now provides excellent pain relief and return of function for patients suffering from destructive joint disease. Whereas only two decades ago patients were told that their new hip joints might survive 10 years, some of these same implants are now demonstrating survival rates of over 90% after 20 years. The success rate of knee joints has recorded similar growth. Despite these promising developments, however, there is a continuing increased revision surgery. Kurtz et al.96 used the National Inpatient Sample and US Census data to quantify historical trends and to make future projections regarding primary and revision joint arthroplasty. They anticipated that the number of knee replacements would double by 2015 and hip replacements would double by 2026. The number of revision knee replacements was projected to increase 500% from 2005 to 2030. The chief objective of modern research is to minimise the revision load by developing products and techniques that maximise the longevity of primary implants. Ideally, joint replacement should be a biological replication of lost tissue. Although this goal remains distant, the most promising developments in technology and design are based on the principle of biological mimicry. This, combined with therapeutic alteration of a host's reaction to an implant, are the main concepts in current research. The pursuit of new treatment options must not deter us from advancing the even more beneficial principle of prevention. New measures preventing or slowing down the progression of joint disease must be introduced in the form of patient education and modification of diet and exercise; the promotion of © 2008, Woodhead Publishing Limited

204

Joint replacement technology

protective gear and rule changes in sports will reap great benefits. Additionally, supportive funding for research into conservative treatment methods of early joint disease must be prioritised. In regard to cementless fixation the two main areas of current research are the development of new surface materials and the alteration of existing surfaces to promote better bonding. Literature today in multiple fields of research is replete with the mention of nanotechnology as a scientific breakthrough promising to be the instrument of modern invention. A double-stranded DNA helix is roughly 2 nm in diameter. Nanotechnology has many promising theoretical applications in bone fixation but it also poses some potential dangers to human health. It can be seen as an extension of existing sciences and technologies into the nanoscale, or a recasting of existing sciences into a more modern term.97 The ultrastructure of bone can also be called a nanostructure of collagen fibres and hydroxyapatite. Current implementations of nanotechnology in implant development range from construction of bulky bone-like structures to composition of `osteophilic' surfaces on implants. To our knowledge, there are no commercial uses of products consisting of nanoparticles in implant technology to date. There are numerous research groups simulating the composition of biological structures or biological processes by the use of nanomaterials such as nanorods, nanotubes and so-called fullerenes. Experiments have been able to demonstrate enhancement of osteogenesis in animal studies98 on nano-sized and nano-structured titanium surfaces. There are concerns, however, that nanoparticles could have harmful effects on human health. There is growing evidence indicating a potential toxicity of nanoparticles, especially due to their higher chemical reactivity and biological activity. Nanoparticles are capable of crossing all biological barriers in cells, tissues and organs, and because of their miniscule size are much more readily taken up by the human body. This field of research remains in its earliest stages and extensive in vitro and in vivo studies are warranted before clinical testing can be undertaken. Bisphosphonates administered locally as well as systemically have demonstrated an enhancement of the osteoconductive effect in HA-coated implants.99±103 To our knowledge, there are no published results available yet of its use in clinical trials. In animal studies the local application increases biomechanical fixation strength two-fold compared with non-treated HA coated implants. Clearly, more work remains to be done. Currently the most common experiments in osteoinduction are the application of bone growth factors on or preferably in HA coatings to prevent burst release of growth factors upon implantation.104 Because of their physiological high expression during bone ingrowth, the most studied growth factors are tumour growth factor 1 (TGF 1) and insulin-like growth factor 1 (IGF1). Both proteins show an improvement of mechanical fixation and osseointegration of titanium implants without production of fibrous tissue in the interface when delivered in a biodegradable coating.105±107 © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

205

Another modification of existing coatings is the addition of antibiotics to HA coatings. Alt et al.108 showed significant improvement of infection prophylaxis in gentamycin HA-coated implants infected with Staphylococcus aureus in rabbits. Chemical treatment of titanium mesh surfaces without HA coatings by a hydrogen peroxide solution containing tantalum chloride has also been shown to enhance bone ingrowth.109 Other research groups are carrying out extensive investigations into the use of such biopolymers as chitosan, a biologically produced polymer closely related to chitin, which is the second most abundant form of polymerised carbon in nature.110,111 Seung Yun Shin et al.112 were able to confirm the biocompatibility of chitosan nanofibre membranes in rats with evidence of enhanced bone regeneration and no evidence of an inflammatory reaction. Tissue engineering is a biological approach attempting to obtain osteoinductive coatings by in vitro cultivation of patient cells to form bone tissue. Bruijn et al. conducted experiments by coating implants with cultured osteogenic bone marrow cells from rats, goats and humans. Calcium phosphatecoated implants were seeded with bone marrow cells cultured for a week to facilitate osteogenic proliferation and extracellular modification. These hybrid implants were then subcutaneously implanted in nude mice and retrieved after four weeks. Histological examination revealed newly formed bone in both porous implants and on flat metallic surfaces.104 Gene therapy is another vast field of research in medicine that shows promising applications in orthopaedics. The principle of providing genetic information via a viral or non-viral delivery vehicle into a living cell is one of the most elegant ways to improve host site physiology. In cementless fixation this information could contain the codes for producing growth factors in continuous fashion by periprosthetic cells. Gene therapy could also provide breakthrough in regeneration of cartilage, which remains the crux of joint disease therapy to this day. In summary cementless fixation: · · · · · · ·

works; saves bone; reduces operative exposure and infection rates; decreases the number of interfaces; saves time; simplifies revision; with hydroxyapatite, seals the bone interface and prevents osteolysis.

9.8

References

1 Mohan, A.R. and M. Gross, Cementless total knee replacement ± a prospective 12 to 15 year follow-up study. J Bone Joint Surg Br, 2004. 86-B(Supp III): p. 318.

© 2008, Woodhead Publishing Limited

206

Joint replacement technology

2 Engh, C.A. Sr, Outcome of total hip arthroplasty using extensively porous-coated components at 20-year follow-up. In Annual meeting, AAOS 2006. 2006. 3 Charnley, J., Anchorage of the femoral head prosthesis to the shaft of the femur. J Bone Joint Surg Br, 1960. 42-B: pp. 28±30. 4 Charnley, J., Arthroplasty of the hip. A new operation. Lancet, 1961. 1(7187): pp. 1129±32. 5 Charnley, J., [Adherence of Prostheses to Living Bone.]. Acta Orthop Belg, 1964. 30: pp. 663±72. 6 Gunston, F.H., Polycentric knee arthroplasty. Prosthetic simulation of normal knee movement. J Bone Joint Surg Br, 1971. 53(2): pp. 272±7. 7 Mirra, J.M., et al., The pathology of the joint tissues and its clinical relevance in prosthesis failure. Clin Orthop Relat Res, 1976(117): pp. 221±40. 8 Schmalzreid, T.P. and J.J. Callaghan, Current concepts review ± wear in total hip and knee replacements. J Bone Joint Surg Am, 1999. 81(1): pp. 115±36. 9 Horowitz, S.M., et al., Studies of the mechanism by which the mechanical failure of polymethylmethacrylate leads to bone resorption. J Bone Joint Surg Am, 1993. 75(6): pp. 802±13. 10 Jones, L.C. and D.S. Hungerford, Cement disease. Clin Orthop Relat Res, 1987(225): pp. 192±206. 11 Jasty, M.J., et al., Localized osteolysis in stable, non-septic total hip replacement. J Bone Joint Surg Am, 1986. 68(6): p. 912±19. 12 Bourne, R.B., Rorabeck, G.H., Inman, K.J., Tozakoglou, E. Cementless total hip replacement: anatomic, cylindrical or tapered. In 67th Annual Meeting of the American Academy of Orthopaedic Surgeons. 2000. Orlando, FL. 13 Miura, H., et al., Effects of screws and a sleeve on initial fixation in uncemented total knee tibial components. Clin Orthop Relat Res, 1990(259): pp. 160±8. 14 Hayakawa, T., et al., Trabecular bone response to surface roughened and calcium phosphate (Ca-P) coated titanium implants. Biomaterials, 2002. 23(4): pp. 1025±31. 15 Shalabi, M.M., et al., Implant surface roughness and bone healing: a systematic review. J Dent Res, 2006. 85(6): pp. 496±500. 16 Shalabi, M.M., J.G. Wolke, and J.A. Jansen, The effects of implant surface roughness and surgical technique on implant fixation in an in vitro model. Clin Oral Implants Res, 2006. 17(2): pp. 172±8. 17 Kold, S., et al., Bone compaction enhances fixation of weightbearing titanium implants. Clin Orthop Relat Res, 2005(431): pp. 138±44. 18 Kold, S., et al., Bone compaction enhances fixation of weight-bearing hydroxyapatite-coated implants. J Arthroplasty, 2006. 21(2): pp. 263±70. 19 Kold, S., et al., Bone compaction enhances fixation of hydroxyapatite-coated implants in a canine gap model. J Biomed Mater Res B Appl Biomater, 2005. 75(1): pp. 49±55. 20 Dalton, J.E., et al., The effect of operative fit and hydroxyapatite coating on the mechanical and biological response to porous implants. J Bone Joint Surg Am, 1995. 77(1): pp. 97±110. 21 Green, J.R., et al., The effect of bone compaction on early fixation of porous-coated implants. J Arthroplasty, 1999. 14(1): pp. 91±7. 22 Channer, M.A., et al., Use of bone compaction in total knee arthroplasty. J Arthroplasty, 1996. 11(6): pp. 743±9. 23 Chareancholvanich, K., et al., In vitro stability of cemented and cementless femoral stems with compaction. Clin Orthop Relat Res, 2002(394): pp. 290±302. 24 W. Norman Scott, Insall & Scott's Surgery of the Knee e-dition. 4th edition 2006, © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

207

vol. I: Churchill Livingstone. 25 Au S. Wong, G. Isaac, A.M.R. New and M. Taylor. Influence of bone quality on the initial stability of cementless hip stem in total hip arthroplasty. In 2003 Summer Bioengineering Conference. 2003. Sonesta Beach Resort in Key Biscayne, Florida. 26 Ryd, L., et al., Roentgen stereophotogrammetric analysis as a predictor of mechanical loosening of knee prostheses. J Bone Joint Surg Br, 1995. 77(3): pp. 377±83. 27 Karrholm, J., et al., Radiostereometry of hip prostheses. Review of methodology and clinical results. Clin Orthop Relat Res, 1997(344): pp. 94±110. 28 Burke, D.W. et al., Dynamic measurement of interface mechanics in vivo and the effect of micromotion on bone ingrowth into a porous surface device under controlled loads in vivo. Trans Orthop Res Soc, 1991 16, 103. 29 Pilliar, R.M., J.M. Lee, and C. Maniatopoulos, Observations on the effect of movement on bone ingrowth into porous-surfaced implants. Clin Orthop Relat Res, 1986(208): pp. 108±13. 30 Sumner, D.R., et al., Bone ingrowth and wear debris in well-fixed cementless porous-coated tibial components removed from patients. J Arthroplasty, 1995. 10(2): pp. 157±67. 31 Sandborn, P.M., et al., Tissue response to porous-coated implants lacking initial bone apposition. J Arthroplasty, 1988. 3(4): pp. 337±46. 32 Cross, M.J., et al., Enhanced fixation of uncemented knee replacement with hydroxyapatite. Key Engineering Materials, 2003, 240±242: pp. 773±776. 33 Goldberg, V.M., et al., Biology of grit-blasted titanium alloy implants. Clin Orthop Relat Res, 1995(319): pp. 122±9. 34 Friedman, R.J., et al., Influence of biomaterial surface texture on bone ingrowth in the rabbit femur. J Orthop Res, 1996. 14(3): pp. 455±64. 35 Bobyn, J.D., et al., The optimum pore size for the fixation of porous-surfaced metal implants by the ingrowth of bone. Clin Orthop Relat Res, 1980(150): pp. 263±70. 36 Cook, S.D., K.A. Walsh, and R.J. Haddad, Jr, Interface mechanics and bone growth into porous Co-Cr-Mo alloy implants. Clin Orthop Relat Res, 1985(193): pp. 271± 80. 37 Clemow, A.J., et al., Interface mechanics of porous titanium implants. J Biomed Mater Res, 1981. 15(1): pp. 73±82. 38 Jasty, M., et al., Bone ingrowth into porous coated canine total hip replacements. Quantification by backscattered scanning electron microscopy and image analysis. Scanning Microsc, 1989. 3(4): pp. 1051±6; discussion 1056±7. 39 Dammak, M., et al., Friction properties at the bone±metal interface: comparison of four different porous metal surfaces. J Biomed Mater Res, 1997. 35(3): pp. 329±36. 40 Bellemans, J., Osseointegration in porous coated knee arthroplasty. The influence of component coating type in sheep. Acta Orthop Scand Suppl, 1999. 288: pp. 1±35. 41 Osborn, J.F., [Biomaterials and their application to implantation]. SSO Schweiz Monatsschr Zahnheilkd, 1979. 89(11): pp. 1138±9. 42 Wilke, A., et al., Biocompatibility analysis of different biomaterials in human bone marrow cell cultures. J Biomed Mater Res, 1998. 40(2): pp. 301±6. 43 Levine, B., C.J. Della Valle, and J.J. Jacobs, Applications of porous tantalum in total hip arthroplasty. J Am Acad Orthop Surg, 2006. 14(12): pp. 646±55. 44 Shuler, M.S., M.D. Rooks, and J.R. Roberson, Porous tantalum implant in early osteonecrosis of the hip: preliminary report on operative, survival, and outcomes results. J Arthroplasty, 2007. 22(1): pp. 26±31. 45 Itala, A., J. Reach, F.H. Sim, K.N. An and D.G. Lewallen, Patellar tendon © 2008, Woodhead Publishing Limited

208

46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64

Joint replacement technology attachment and healing to porous tantalum: an experimental canine study. J Arthroplasty, 2006. 21(2): p. 303. Nasser, S. and R.A. Poggie, Revision and salvage patellar arthroplasty using a porous tantalum implant. J Arthroplasty, 2004. 19(5): pp. 562±72. Nehme, A., D.G. Lewallen, and A.D. Hanssen, Modular porous metal augments for treatment of severe acetabular bone loss during revision hip arthroplasty. Clin Orthop Relat Res, 2004(429): pp. 201±8. Macheras, G.A., et al., Radiological evaluation of the metal±bone interface of a porous tantalum monoblock acetabular component. J Bone Joint Surg Br, 2006. 88(3): pp. 304±9. Bobyn, J.D., et al., Characteristics of bone ingrowth and interface mechanics of a new porous tantalum biomaterial. J Bone Joint Surg Br, 1999. 81(5): pp. 907±14. Cooley, D.R., et al., The advantages of coated titanium implants prepared by radiofrequency sputtering from hydroxyapatite. J Prosthet Dent, 1992. 67(1): pp. 93±100. Soballe, K., et al., Gap healing enhanced by hydroxyapatite coating in dogs. Clin Orthop Relat Res, 1991(272): pp. 300±7. Mouzin, O., K. Soballe, and J.E. Bechtold, Loading improves anchorage of hydroxyapatite implants more than titanium implants. J Biomed Mater Res, 2001. 58(1): pp. 61±8. Rahbek, O., et al., Superior sealing effect of hydroxyapatite in porous-coated implants: experimental studies on the migration of polyethylene particles around stable and unstable implants in dogs. Acta Orthop, 2005. 76(3): pp. 375±85. Epinette, J.A. and M.T. Manley, Hydroxyapatite-coated total knee replacement: clinical experience at 10 to 15 years. J Bone Joint Surg Br, 2007. 89-B(1): pp. 34± 38. Munro, J., et al., Total hip replacement hip taper-design stem and hydroxyapatite coating. Clinical outcome and quantitative CT-assisted osteodensitometry. J Bone Joint Surg Br, 2005. 87-B(Supp I): p. 29. Shetty, A.A., et al., Results of a hydroxyapatite-coated (Furlong) total hip replacement: a 13- to 15-year follow-up. J Bone Joint Surg Br, 2005. 87-B(8): pp. 1050±4. Landor, I., et al., Hydroxyapatite porous coating and the osteointegration of the total hip replacement. Arch Orthop Trauma Surg, 2007. 127(2): pp. 81±9. Geesink, R.G. and N.H. Hoefnagels, Six-year results of hydroxyapatite-coated total hip replacement. J Bone Joint Surg Br, 1995. 77(4): pp. 534±47. Rasquinha, V.J., C.S. Ranawat, and A.J. Mauriello, Jr, Hydroxyapatite: catalyst or conjuror? J Arthroplasty, 2002. 17(4 Suppl 1): pp. 113±17. Cross, M.J. and E.N. Parish, A hydroxyapatite-coated total knee replacement: prospective analysis of 1000 patients. J Bone Joint Surg Br, 2005. 87(8): pp. 1073±6. Dixon, P., et al., Hydroxyapatite-coated, cementless total knee replacement in patients aged 75 years and over. J Bone Joint Surg Br, 2004. 86(2): pp. 200±4. Tai, C.C. and M.J. Cross, Five- to 12-year follow-up of a hydroxyapatite-coated, cementless total knee replacement in young, active patients. J Bone Joint Surg Br, 2006. 88(9): pp. 1158±63. Kienapfel, H., et al., Implant fixation by bone ingrowth. J Arthroplasty, 1999. 14(3): pp. 355±68. Frost, H.M., A brief review for orthopedic surgeons: fatigue damage (microdamage) in bone (its determinants and clinical implications). J Orthop Sci, 1998. 3(5): pp. 272±81.

© 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

209

65 Huiskes, R., et al., Effects of mechanical forces on maintenance and adaptation of form in trabecular bone. Nature, 2000. 405(6787): pp. 704±6. 66 Chicurel, M.E., et al., Integrin binding and mechanical tension induce movement of mRNA and ribosomes to focal adhesions. Nature, 1998. 392(6677): pp. 730±3. 67 Sychterz, C.J., et al., Effect of femoral stiffness on bone remodeling after uncemented arthroplasty. Clin Orthop Relat Res, 2001(389): pp. 218±27. 68 P. Christel, A.Meunier, and A.J.C. Lee, Biological and biomechanical performance of biomaterials. In Advances in Biomaterials, vol. 6, 1985, Fifth European Conference on Biomaterials, Paris: Elsevier. 69 Huiskes, R. and E.Y. Chao, A survey of finite element analysis in orthopedic biomechanics: the first decade. J Biomech, 1983. 16(6): pp. 385±409. 70 Long, M. and H.J. Rack, Titanium alloys in total joint replacement ± a materials science perspective. Biomaterials, 1998. 19(18): pp. 1621±39. 71 Glassman, A.H., J.D. Bobyn, and M. Tanzer, New femoral designs: do they influence stress shielding? Clin Orthop Relat Res, 2006. 453: pp. 64±74. 72 Francesco P., L. Malfetta, and M. Grandizio, Preservation of the femoral neck in hip arthroplasty: results of a 13- to 17-year follow-up. J Orthop Traumatology, 2000. 1(1): pp. 31±9. 73 Sumner, D.R. and J.O. Galante, Determinants of stress shielding: design versus materials versus interface. Clin Orthop Relat Res, 1992(274): pp. 202±12. 74 Simon, U., et al., Influence of the stiffness of bone defect implants on the mechanical conditions at the interface ± a finite element analysis with contact. J Biomech, 2003. 36(8): pp. 1079±86. 75 Khaw, F.M., et al., A randomised, controlled trial of cemented versus cementless press-fit condylar total knee replacement. Ten-year survival analysis. J Bone Joint Surg Br, 2002. 84(5): pp. 658±66. 76 Wood, J.E.J., Uncemented fixation in total knee replacement: a promising future. Curr Opin Orthop, 18(1): pp. 61±5. 77 Nilsson, K.G., et al., Uncemented HA-coated implant is the optimum fixation for TKA in the young patient. Clin Orthop Relat Res, 2006. 448: pp. 129±39. 78 Therbo, M., et al., Influence of pre-operative bone mineral content of the proximal tibia on revision rate after uncemented knee arthroplasty. J Bone Joint Surg Br, 2003. 85(7): pp. 975±9. 79 Bombelli, R., Osteoarthritis of the Hip: Classification and Pathogenesis: The Role of Osteotomy as a Consequent Therapy. 2nd edn. 1983. Berlin: Springer-Verlag. 80 Panagiotis, Z., et al., Biological fixation of cementless stems in patients over 75 years with diagnosed osteoporosis. J Bone Joint Surg Br, 2004. 86-B(Supp III): p. 320. 81 Healy, W.L., Hip implant selection for total hip arthroplasty in elderly patients. Clin Orthop Relat Res, 2002(405): pp. 54±64. 82 Ekelund, A., N. Rydell, and O.S. Nilsson, Total hip arthroplasty in patients 80 years of age and older. Clin Orthop Relat Res, 1992(281): pp. 101±6. 83 Effenberger, H., et al., Successful hip arthroplasty using cementless titanium implants in rheumatoid arthritis. Arch Orthop Trauma Surg, 2002. 122(2): pp. 80±7. 84 Jana, A.K., et al., Total hip arthroplasty using porous-coated femoral components in patients with rheumatoid arthritis. J Bone Joint Surg Br, 2001. 83-B(5): pp. 686±90. 85 Engesaeter, L.B., et al., Does cement increase the risk of infection in primary total hip arthroplasty? Revision rates in 56,275 cemented and uncemented primary THAs followed for 0±16 years in the Norwegian Arthroplasty Register. Acta Orthop, 2006. 77(3): pp. 351±8. © 2008, Woodhead Publishing Limited

210

Joint replacement technology

86 Petty, W., The effect of methylmethacrylate on chemotaxis of polymorphonuclear leukocytes. J Bone Joint Surg Am, 1978. 60(4): pp. 492±8. 87 Petty, W., The effect of methylmethacrylate on bacterial phagocytosis and killing by human polymorphonuclear leukocytes. J Bone Joint Surg Am, 1978. 60(6): pp. 752±7. 88 Petty, W., The effect of methylmethacrylate on the bacterial inhibiting properties of normal human serum. Clin Orthop Relat Res, 1978(132): pp. 266±78. 89 Petty, R., Influence of skeletal implant material on infection. Trans ORS, 1983. 8: pp. 137. 90 Samuelson, K.M., D. Au, G.L. Rasmussen, and R.L. Ellingson, Evaluation of cemented versus cementless fixation and infections in caninetotal joint arthroplasties. Trans ORS, 1983. 8: p. 230. 91 Gristina, A.G. and J. Kolkin, Current concepts review. Total joint replacement and sepsis. J Bone Joint Surg Am, 1983. 65(1): pp. 128±34. 92 Freeman, M.A., et al., The management of infected total knee replacements. J Bone Joint Surg Br, 1985. 67(5): pp. 764±8. 93 Dixon, P., E.N. Parish, and M.J. Cross, Arthroscopic debridement in the treatment of the infected total knee replacement. J Bone Joint Surg Br, 2004. 86(1): pp. 39±42. 94 Hirota, K., et al., The relationship between pneumatic tourniquet time and the amount of pulmonary emboli in patients undergoing knee arthroscopic surgeries. Anesth Analg, 2001. 93(3): pp. 776±80. 95 Morgan-Jones, R.L., M.M. Perko, and M.J. Cross, Uncemented total knee replacement ± the favourable influence of low over high pressure drainage. Knee, 2000. 7(3): pp. 149±50. 96 Kurtz, S., et al., Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am, 2007. 89(4): pp. 780±5. 97 http://en.wikipedia.org/wiki/Nanotechnology. 98 Meirelles, L., et al., Increased bone formation to unstable nano rough titanium implants. Clin Oral Implants Res, 2007. 18: pp. 326±32. 99 Stadelmann, V. and D.P. Pioletti, Using Orthopedic Implants for Local Bisphosphonate Delivery in Osteoporotic Animals: A Theoretical Model, Biomechanical Orthopedics Laboratory, Swiss Federal Institute of Technology: Lausanne, Switzerland. 100 Jakobsen, T., et al., Local alendronate increases fixation of implants inserted with bone compaction: 12-week canine study. J Orthop Res, 2007. 25(4): pp. 432±41. 101 Jensen, T.B., et al., Systemic alendronate treatment improves fixation of press-fit implants: a canine study using nonloaded implants. J Orthop Res, 2007. 25(6): pp. 772±8. 102 Eberhardt, C., et al., The bisphosphonate ibandronate accelerates osseointegration of hydroxyapatite-coated cementless implants in an animal model. J Orthop Sci, 2007. 12(1): pp. 61±6. 103 Eberhardt, C., et al., Osseointegration of cementless implants with different bisphosphonate regimens. Clin Orthop Relat Res, 2006. 447: pp. 195±200. 104 de Bruijn, J.D., et al., Bone induction by implants coated with cultured osteogenic bone marrow cells. Adv Dent Res, 1999. 13: pp. 74±81. 105 Lamberg, A., et al., Locally delivered TGF-beta1 and IGF-1 enhance the fixation of titanium implants: a study in dogs. Acta Orthop, 2006. 77(5): pp. 799±805. 106 Aspenberg, P., T. Albrektsson, and K.G. Thorngren, Local application of growthfactor IGF-1 to healing bone. Experiments with a titanium chamber in rabbits. Acta Orthop Scand, 1989. 60(5): pp. 607±10. © 2008, Woodhead Publishing Limited

Cementless fixation techniques in joint replacement

211

107 Lind, M., et al., Transforming growth factor-beta stimulates bone ongrowth. Hydroxyapatite-coated implants studied in dogs. Acta Orthop Scand, 1996. 67(6): pp. 611±16. 108 Alt, V., et al., The effects of combined gentamicin-hydroxyapatite coating for cementless joint prostheses on the reduction of infection rates in a rabbit infection prophylaxis model. Biomaterials, 2006. 27(26): pp. 4627±34. 109 Taeseong Kim, M.S., C. Ohtsuki, K. Masuda, H. Tamai, E. Watanabe, A. Osaka, and H. Moriya, Enhancement of bone growth in titanium fiber mesh by surface modification with hydrogen peroxide solution containing tantalum chloride. J Biomed Mater Res Part B: Appl Biomater, 2003. 64B(Issue 1): pp. pp. 19±26. 110 Bumgardner, J.D., et al., The integration of chitosan-coated titanium in bone: an in vivo study in rabbits. Implant Dent, 2007. 16(1): pp. 66±79. 111 Jarry, C., et al., Effects of steam sterilization on thermogelling chitosan-based gels. J Biomed Mater Res, 2001. 58(1): pp. 127±35. 112 Shin, S.Y., et al., Biological evaluation of chitosan nanofiber membrane for guided bone regeneration. J Periodontol, 2005. 76(10): pp. 1778±84.

© 2008, Woodhead Publishing Limited

10

Bone cement fixation: acrylic cements J - S W A N G , Lund University, Sweden and N D U N N E , Queen's University of Belfast, UK

10.1

Introduction

Bone cements based on poly(methylmethacrylate) (PMMA) are important products in joint replacement surgery. Originally developed for dental applications, they have been used effectively in joint replacement surgery for almost 50 years. They are widely perceived to be simple cold curing polymer powder/ liquid monomer systems; however there are many ways in which bone cements vary, leading to considerably different properties.

10.2

Acrylic bone cements ± history and evolution

PMMA was formulated by Otto Rohm in 1902, and is a solid, glass-like material with good biocompatibility. In 1936, Kulzer demonstrated that a dough material could be produced by mixing PMMA powder and a liquid monomer, which cured when benzoyl peroxide (BPO) was added and the blend was heated to 100 ëC. The first clinical application of PMMA was to repair cranial defects in the monkeys in 1938. As greater understanding of the PMMA system was achieved, surgeons moved towards using these materials in reconstructive surgery on humans. Subsequently, the heat-curing polymer Paladon 65Õ (Heraeus Kulzer, Hanau, Germany) was used for repairing cranial defects in humans by manufacturing plates in the laboratory and later trimming the cured material in situ (Kleinschmitt, 1941). In 1943, polymer chemists discovered that polymerisation of methylmethacrylate (MMA) would take place by itself under ambient conditions if a coinitiator was added, and the companies Degussa and Kulzer established a procedure for the chemical manufacturing of PMMA bone cements by utilising tertiary aromatic amines. These protocols still remain the cornerstones for PMMA bone cement production today. In 1949, the Judet brothers were the first to use PMMA to manufacture a hip prosthesis (Judet and Judet, 1956). Before long, however, it became obvious that the PMMA prosthesis used by them could neither be successfully incorporated © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

213

nor articulated in the body for biological and mechanical reasons. In 1958, Sir John Charnley successfully fixed an intramedullary stem prosthesis using selfpolymerising PMMA bone cement (Charnley, 1960). Charnley named the material acrylic-based bone cement. In 1970, Charnley first described a novel surgical technique for total hip joint replacement surgery (Charnley, 1970). At the beginning of the 1970s in an effort to alleviate periprosthetic infection, the most feared complication after total joint replacement, Buchholz et al. (1981) advocated the addition of antibiotics to bone cement. Their idea was to add antibiotics to the cement in order to reduce the incidence of infection, which was high at that time. Gentamicin powder mixed with PMMA cement was found to be stable and offered a suitable spectrum of antibiotic activity. Since the introduction of PMMA bone cement by Charnley, millions of people have had their lives dramatically and remarkably improved by his innovation. After nearly 50 years of clinical application, PMMA bone cement is still the most frequently used material for the fixation of joint prostheses.

10.3

Clinical application and function

PMMA bone cements are primarily used for the fixation of joint prostheses. In the fixation of joint replacement the self-curing cement fills the free space between the prosthesis and the bone, and constitutes a very important interface (Fig. 10.1). Owing to their optimum elasticity, the bone cements can evenly cushion the forces acting against the bone. The close association between the cement and the bone leads to optimal distribution of the stresses and interface strain energy.

10.1 Schematic diagram of prosthesis and PMMA bone cement in an acetabular socket and femur. © 2008, Woodhead Publishing Limited

214

Joint replacement technology

The transfer of the force from bone-to-implant and implant-to-bone is the primary function of bone cement. The ability to do this reliably for a long time is crucial for long-term survival of the implant. Adequate cement interdigitation and reinforcement of the cancellous bone are of utmost importance. If the extreme stresses generated exceed the capability of the bone cement to transfer and absorb forces, a fatigue fracture is possible. Antibiotic loaded bone cements are also used as drug-delivery systems. Artificial implants are more susceptible to bacterial colonisation on their surfaces because the germs can then hide from the natural protection via the body and cause periprosthetic infection. When loaded with antibiotics, bone cement functions as carrier matrix.

10.4

Composition

PMMA bone cements are two-component systems, comprising a polymer powder and a liquid monomer (Figs 10.2 and 10.3). The polymer powder

10.2 Chemical composition and properties of MMA.

10.3 Chemical composition and properties of PMMA. © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

215

Table 10.1 Components of polymer powder of PMMA bone cement Polymer

Poly(methylmethacrylate) and/or copolymers of methylmethacrylate

Initiator

Benzoyl peroxide

Radiopacifier

Barium sulphate Zirconium dioxide

Antibiotics

Gentamicin sulphate Clindamycin hydrochloride Tobramycin Erythromycin±glucoheptonate Colistin±methane±sulphonate±sodium

Additives

Dye (chlorophyll) Plasticiser (di-cyclo-hexyl phthalate)

Table 10.2 Components of liquid monomer of PMMA bone cement Monomer

Methylmethacrylate Butylmethacrylate

Activator

DmpT (N,N-dimethyl-p-toluidine) DmapE (2-[4-(dimethylamino)phenyl] ethanol)

Inhibitor

Hydroquinone

Additives

Dye (chlorophyll)

component is composed of PMMA and/or methacrylate copolymers (Table 10.1). Furthermore, the polymer powder contains BPO, which acts as an initiator of the free radical polymerisation reaction. The BPO can form part of the polymer micro-beads or simply be incorporated into the powder. The powder also contains an X-ray contrast agent and possibly an antibiotic. In the liquid phase MMA is the main constituent but sometimes other methacrylates such as butylmethacrylate are also present (Table 10.2). In order for the MMA to be used for bone cements it must be polymerisable, therefore it must contain a carbon double bond which can be broken. As an activator of the formation of radicals the liquid contains an aromatic amine, such as N,N-dimethyl-p-toluidine (DmpT). Additionally, it contains an inhibitor (hydroquinone) to avoid premature polymerisation during storage and optionally a colouring agent such as chlorophyll.

10.5

Polymer powder/liquid monomer ratio

Most of the commercial bone cement formulations have a polymer powder/ liquid monomer ratio of 2/1 (weight, g/volume, ml). When this ratio increases © 2008, Woodhead Publishing Limited

216

Joint replacement technology

(by increasing the amount of PMMA polymer powder or by decreasing the amount of MMA liquid monomer), the viscosity of the bone cement dough increases. This change causes difficulty in workability and creates problems in applying the dough cement into the surgical cavity. It also results in a shorter growth time for the free radicals as well as shorter time for the occurrence of the peak temperature with fast polymerisation. Conversely, lower polymer powder/ liquid monomer ratios give lower peak concentrations of free radicals, larger peak temperatures and a high monomer release to the surrounding tissue.

10.6

Polymerisation reaction

Acrylic bone cements are polymeric materials produced by radical polymerisation of MMA (Fig. 10.4). The process starts when the polymer powder and liquid monomer are mixed, resulting in a reaction between the BPO initiator and the DmpT activator, forming free radicals at room temperature. Consequently, the DmpT liquid causes a breakdown of the BPO powder in a reduction/ oxidation process by transfer of electrons, resulting in the formation of benzoyl free radicals (Fig. 10.5). These unpaired electrons are energetically unstable and need to be paired and stabilised (Fig. 10.6). When they find any electrons to pair with, they do so. The carbon±carbon double bond in a vinyl monomer, such as in MMA, has a pair of electrons which is attacked by the free radical to form a new chemical bond between the initiator fragment and one of the double carbon bonds of the monomer molecule. The other electron of the double bond stays on the carbon atom that is not bonded to the initiator fragment, creating a new free radical. This unpaired electron is now capable of attacking the double bond of a new monomeric unit. The whole process, the breakdown of the initiator molecule to form radicals, followed by the radicals' reaction with a monomer molecule is called the initiation step of the polymerisation (Fig. 10.7). This new radical reacts with another MMA molecule in the same way as the

10.4 Polymerisation reaction of PMMA. © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

10.5 Reduction±oxidation process of BPO caused by DmpT.

10.6 Initial radical formation by decomposition of BPO. © 2008, Woodhead Publishing Limited

217

218

Joint replacement technology

10.7 Initiation step of the polymerisation.

initiator fragment did. Another radical is always formed when this reaction takes place over and over again. This process of adding more monomer molecules to the growing chain is called propagation (Fig. 10.8). As long as the radical and the monomer are present, more and more MMA molecules are added, and they build a long chain containing n monomeric units. As the polymerisation continues, the viscosity of the whole mass increases, reducing the diffusion of growing chains and the combination of chain ends. Therefore, the rate of termination decreases. At this point, the polymerisation rate and the temperature increase. By regrouping of the two radical chains the system becomes depleted of free radicals and the polymerisation ceases (Fig. 10.9). During the formation of the polymer matrix, PMMA particles from the polymer powder component of the bone cement become entrapped in the newly formed polymer matrix. Varying the dimensions of the polymer powder, the concentration of the initiator and activator will influence the wetting of the powder which directly affects the viscosity, dough time and setting time.

10.8 Chain propagation reaction. © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

219

10.9 Termination of the polymerisation reaction by recombination.

10.7

Polymerisation heat

The free radical polymerisation of PMMA bone cement is a highly exothermic chemical reaction. As a consequence of the proceeding polymerisation and the growing dough viscosity, the temperature of the cement increases, so that a polymerisation heat of 57 kJ (13.8 kcal) is produced per mole of MMA. The peak temperatures demonstrated during the reaction are only exhibited for a short period. However, it has been reported that these peak temperatures have the potential to cause thermal necrosis, which is allegedly one of the mitigating factors for aseptic loosening of an implant fixed with bone cement. In particular connective tissue reactions near loose implants have been regarded as occurring as a result of primary damage to the bone bed. Nonetheless, the peak temperatures recorded in vitro do not correlate with those achieved in vivo (Fig. 10.10). Clinical studies have shown significantly lower peak temperatures (40± 46 ëC) at the bone±cement interface being recorded in situ during the joint replacement surgery (Toksvig-Larsen et al., 1991). The upper temperature limit is understood to be reached only in cement mantles of at least 3 mm thickness without cement penetration into the cancellous bone (Breusch, 2001). With acceptable surgical technique and preservation of the cancellous bone it seems highly improbable that the protein coagulation temperature is exceeded, © 2008, Woodhead Publishing Limited

220

Joint replacement technology

10.10 Exothermic temperatures generated for PMMA bone cement under in vitro (ISO 5833) and in vivo conditions.

particularly because of the dissipation of heat from the system via the implants and the local blood circulation. The maximum temperature exhibited during polymerisation is influenced by the liquid monomer composition, the polymer powder/liquid monomer ratio and radiopacifier content. Variation of these factors can result in a significant difference in management/delivery of the cement and its performance under mechanical load.

10.8

Polymerisation shrinkage

Acrylic bone cement cannot be produced from MMA monomer on its own, as the duration of the polymerisation reaction would be too long. Furthermore the level of shrinkage due to the polymerisation reaction would be too high, and the heat occurring during the polymerisation of the MMA monomer could not be adequately controlled. During the polymerisation reaction of bone cement, many MMA monomer molecules combine to form long polymer molecules and it is unavoidable that volume shrinkage will occur. Pure MMA will shrink by approximately 21%. By using a pre-polymerised polymer powder component, the portion of the MMA in the system is reduced to a third. Therefore, the theoretical shrinkage is 6±7%. Charnley (1970) found that cement has 3±5% volumetric shrinkage after curing. If prepared under atmospheric conditions, PMMA cement tends to shrink to less than bone cement mixed under the application of a vacuum (Dunne and Orr, 2002). In vivo a major part of the volume shrinkage is counteracted by © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

221

water uptake of the cement. The water uptake for commercial PMMA bone cements is approximately 1±2% for pure bone cements and slightly higher for antibiotic-loaded cements (Ege et al., 1998). Concerns over shrinkage have focused primarily on the stability of the implant. With vacuum mixing, the volumetric shrinkage may be increased from 3±5% to 5±7% in different cements (Muller et al., 2002). Studies have, however, been unable to find differences in diametrical shrinkage or gap between cement and prosthesis when a reduced porosity cement is used (Davies and Harris, 1995; Wang et al., 1999). Shrinkage within the cancellous bone can be regarded as beneficial, as some interface gaps allow for re-vascularisation (Draenert et al., 1999). Vacuum mixing of cement has not been found to negatively affect interface strength between cement and prosthesis. Roentgen stereophotogrammetric analyses (RSA) have shown a stable cemented implants in early and medium-term studies of vacuum mixed cemented implants (Nivbrant et al., 2001; Adalberth et al., 2002).

10.9

Molecular weight and sterilisation

The molecular weight is the conventional parameter used to describe the length of the polymer chains incorporated in the powder component or resulting from the polymerisation of the MMA. The molecular weight of the polymer powder primarily influences the swelling properties and viscosity, and thereby the working time, as well as the fatigue properties of the hardened cement. The molecular weight of the powder is strongly influenced by the sterilisation method. Sterilisation by gamma ( ) or beta ( ) radiation significantly lowers the molecular weight, while sterilisation with ethylene oxide does not affect the molecular weight of the polymer (Fig. 10.11). The advantage of radiation is the high penetration depth facilitating the cement to be sterilised as the final product. Conversely, it is known that irradiation causes variations in the properties of plastic materials. Sterilisation using ethylene oxide (EtO) is the preferred technique for PMMA bone cements, because of the low risk of alterations in the material properties. However, sterilisation with EtO is more expensive than sterilisation with radiation or radiation because it is a more complex and time-consuming technique. The residual EtO also has to be desorbed from the powder.

10.10 Residual monomer and monomer release Free radical polymerisation of the MMA in bone cement generally does not proceed to completion, because the mobility of the remaining MMA monomer is hindered at high conversion rates; therefore, there is residual monomer trapped within the polymer. The content of residual monomer directly after curing is approximately 2±6% (Ege et al., 1998). In the first 2±3 weeks this content © 2008, Woodhead Publishing Limited

222

Joint replacement technology

10.11 Molecular weight of various commercial PMMA bone cements and the effect of sterilisation method (Kuehn, 2000).

decreases to approximately 0.5% by slow post-polymerisation and the ratio will be kept in the material in the body for many years (Kirschner, 1978). The main part (approximately 80%) of the total residual is slowly polymerised (Rudigier et al., 1981). A small portion of the residual monomer is released and metabolised to carbon dioxide and water in the citric acid cycle (Wenda et al., 1988).

10.11 Viscosity and handling properties The viscosity of bone cements at the dough stage is determined mainly by the chemical composition of the polymer powder and liquid components and the powder/liquid ratio (Dunne and Orr, 1998). These characteristics should never be altered in the operating theatre in an attempt to modify the viscosity of the cement. The acceptable method to adjust the viscosity of the cement is through prechilling of the polymer powder and liquid components prior to mixing. The velocity of the reaction, and ultimately the viscosity, depends on the temperature. Prechilling of the polymer powder and liquid components, especially of high-viscosity cements, has been used with the introduction of vacuum mixing systems to make mixing of the cement more convenient and improve the quality of the cement, especially with respect to porosity (Lidgren et al., 1987). Bone cements can be divided into two main categories: high- and lowviscosity cements. High-viscosity bone cements have shorter mixing phases and lose their stickiness quickly. During the working phase the viscosity remains constant and slowly increases toward the end of this phase. Generally, the © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

223

working phase is long. Low-viscosity bone cements have a long-lasting liquid to low-viscosity mixing phase. The cement remains sticky for three minutes or longer. In its working phase the viscosity quickly increases. During the working phase there is already heat formation caused by polymerisation and therefore the temperature of the cement dough increases. The waiting phase of the lowviscosity cements lasts a few minutes. High-viscosity cements are injectable almost directly after the mixing phase and they have a longer application phase. The high viscosity during mixing could be a disadvantage because it allows air to become entrapped. The handling of bone cements can be described by four different phases (Fig. 10.12) with their corresponding viscosities: mixing phase, waiting phase, application or working phase and setting or hardening phase (Dunne and Orr, 2002). The mixing phase (up to 1 minute) is the time for thorough homogenisation of the polymer powder and the liquid components. Significant differences are observed in the mixing phase for different bone cements. Some bone cements mix more readily than others because of higher initial viscosity. Therefore, care must be taken not to introduce air bubbles into the dough at this early stage, which could lead to high porosity within the cement, resulting in premature failure when subjected to mechanical loading. The homogeneity of the dough is influenced by a number of factors such as the design of the mixing

10.12 The four stages of polymerisation for PMMA bone cements can be described by four different phases as a function of ambient temperatures: I mixing phase, II waiting phase, III application phase, IV setting phase. © 2008, Woodhead Publishing Limited

224

Joint replacement technology

vessel and the spatula, the mixing speed, number of strokes or revolutions per minute, and so on. The longer and more vigorous the cement is mixed the more porous it becomes. The waiting phase (up to several minutes, depending of bone cement type and the handling temperature) is the period for the cement to achieve a non-sticky consistency and therefore be ready for use. The application or working phase (2±4 minutes, depending on bone cement type and handling temperature) is the period when the cement in injected into the bone prior to implantation of the prosthesis. At this stage of the polymerisation, the cement dough is of moderate viscosity. To achieve a proper interdigitation of interface at the cement mantle and cancellous bone it is important for the surgeon to know the working time in order to correctly pressurise the cement and insert the prosthesis at an optimal viscosity. The viscosity at beginning of the application stage must not be too low, otherwise the injected dough will not withstand the bleeding pressure (13 kPa) in the bone. Blood infiltration into the bone cement will lead to laminations, ultimately reducing its mechanical strength. This is the main problem when using low-viscosity cements with their short working time. High-viscosity cements in this regard are more user-friendly and forgiving, therefore resulting in better long-term performance (Breusch, 2001). Late application of bone cement of too high viscosity can also result in problems such as poor interdigitation of the cement into the cancellous bone. The setting or hardening phase (1±2 minutes) is the period of the final setting process and the development of the polymerisation heat. The temperature of the cement prior to mixing and of the surroundings has a significant influence on the progress of the bone cement through the different phases (Figs 10.12 and 10.13). The information given by the temperature vs. elapsed time from start of mixing is not always comparable, because manufacturers

10.13 Four stages of polymerisation for PMMA bone cement that has been stored under different temperatures. © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

225

use different measuring techniques resulting in different lengths of the working phases. This disagreement is caused primarily by the lack of a universal test method. Wide experience and knowledge by the surgeon are therefore beneficial to determine the optimal time to inject the cement and insert the prosthesis. For the surgeon the viscosity is the most important handling characteristic and determines the working properties of the bone cement. The timing for injection and pressurisation of the PMMA bone cement is critical to the success of the surgical procedure.

10.12 Antibiotics in poly(methylmethacrylate) bone cement Bone cement may perform as a matrix for the local application of antibiotics. Owing to the high local concentration of an antibiotic surrounding the prosthesis, the use of antibiotic-loaded bone cements has great advantages compared with systemic antibiotic therapy. An implanted prosthesis is particularly sensitive to bacterial growth on its surface, because the microorganisms may multiply out of reach for the immune system of the body. As the bacteria reproduce they form a protective biofilm, which has low sensitivity to antibiotics. Local antibiotic treatment is therefore essential (Gristina, 1987). The kinetics of antibiotic release from bone cement is of clinical significance. However, not all antibiotics are appropriate for application in acrylic bone cements. To reduce the incidence of a bacterial strain developing resistance to an antibiotic loaded in bone cement, an initial high level of antibiotic release for the first few days or weeks is essential. Many bacteriological, physical and chemical factors should be considered in the selection of an antibiotic (Table 10.3) (Breusch and Kuehn, 2003). It has been reported that the presence of small doses of antibiotics (less than 1 g per 40 g pouch of polymer powder) in bone cement has no detrimental influence on the compressive or diametral tensile strength of the cured cement, and does not Table 10.3 Bacteriological, physical and chemical factors that should be taken into account when selecting an antibiotic (Breusch and Kuehn, 2003) Broad antibacterial spectra Good bactericidal effect at low concentrations Low frequency of primary resistant bacteria Low progress of resistances Low protein bonding Low probability of allergic reaction Insignificant effect on cement performance Chemically and thermally stable Good solubility in water Good rate of antibiotic release from bone cement

© 2008, Woodhead Publishing Limited

226

Joint replacement technology

10.14 Percentage content of gentamicin in polymer powder of various commercial PMMA acrylic bone cements and in vitro release of gentamicin from polymerised bone cement after 7 days implantation (Kuehn, 2000).

change the thermal or viscosity characteristics (Mark et al., 1976; Murray, 1984), but Lautenschlager et al. (1976) and Klekamp et al. (1999) showed that large doses of antibiotics (greater than 1 g per 40 g pouch of polymer powder) decreased the compressive and tensile strengths of bone cements. If surgeons hand-mix antibiotics into bone cement at the time of surgery, the quality of the antibiotic loaded cement may be affected. The disadvantages of this approach are poor mechanical properties and a deleterious effect of elution kinetics. Based on the prerequisites outlined in Table 10.3 and release measurements (Fig. 10.14), gentamicin has become the antibiotic of choice for bone cements since the 1970s. The release mechanism for antibiotic from bone cement is surface and diffusion processes (Frommelt, 2001). Release studies show a high rate of release within the first six hours after implantation, followed by a continuous decrease during the next few days. It has been reported that even after five years there is still a low rate of antibiotic release (Wahlig and Dingeldein, 1980).

10.13 Radiopacifier in poly(methylmethacrylate) bone cement PMMA bone cement is not a radiopaque material, therefore it is impossible to detect bone cement by ordinary X-ray imaging techniques. Until 1972, PMMA cement did not contain any radiopaque materials and was, therefore, radiolucent. It is important for the orthopaedic surgeon to easily monitor and evaluate the healing and loosening processes and to analyse changes in both bone and bone © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

227

10.15 Scanning electron micrographs of PMMA bone cement containing radiopacifier: (a) Palacos, R, (1) polymer bead, (2) zirconium dioxide; (b) Simplex B, (1) polymer bead, (3) barium sulphate.

cement after a joint replacement, so well-defined opacity is required. This is the reason that radiopaque substances are added to PMMA bone cements. Barium sulphate (BaSO4) or zirconium dioxide (ZrO2) are now used as radiopacifiers in all commercially available bone cements (Fig. 10.15). These radiopacifiers are not a part of the polymer chain. They are dispersed uniformly in the polymer powder and thus in the resulting hardened bone cement. All PMMA bone cements contain 8±15% X-ray contrasting agent within the polymer powder (Kuehn, 2000). Compared with bone cements with BaSO4, those containing ZrO2 have a significantly higher opacity. Bone cements with 15% ZrO2 in the powder component have the highest opacity. The addition of radiopacifiers to bone cement has many disadvantages. Animal experiments with different cell cultures have shown greater differences in bone resorption around the bone cement application area by using BaSO4 in comparison to ZrO2 (Sabokbar et al., 2001). In spite of the low solubility of BaSO4, toxic barium ions can be released. In comparison, the abrasive properties of ZrO2 appear to be their primary disadvantage. The disadvantage is apparent only in cases of loosening of the prosthesis, or when free cement particles migrate into the joint articulation. The radiopaque materials remain in the body for decades as components of PMMA bone cement. When the static and dynamic mechanical properties of PMMA bone cement that included BaSO4 and ZrO2 were examined, it was shown that the addition of ZrO2 significantly increased the tensile strength, the fracture toughness and the fatigue crack propagation resistance. In contrast, the addition of BaSO4 produces a decrease in tensile strength, but does not affect the fracture toughness and improves crack propagation resistance (Ginebra et al., 2002).

10.14 Mechanical properties PMMA bone cement has mainly a mechanical function, by evenly distributing the contact stresses and transferring them from the implant to the bone. This is © 2008, Woodhead Publishing Limited

228

Joint replacement technology

the main reason why the mechanical properties of acrylic bone cements have been measured and reported by many research groups. Unfortunately, the majority of the literature relating to the mechanical properties of bone cements cannot be compared because of a lack of information about the preparation and storage of the test specimens and the test methods used. The mechanical properties of bone cement can be divided into short term and long term. The important short-term properties are: tensile strength, compressive strength, bending strength and modulus of elasticity. Some of these are covered by the ISO 5833, ISO 527 and ASTM F 451 standards (Fig. 10.16). There are differences in data from the three static test methods described. The compressive strength of PMMA bone cement is superior to the bending strength, which is higher than the tensile strength. This order is observed for all polymeric materials. This means that tensile loading may be a greater risk factor for failure than compressive loading. In vivo, simple tensile loading, however, does not play a major role; complex combinations of different modes of loading are more appropriate. From a physical point of view, bending combines tensile and compressive loading; therefore, the bending test is the most informative test. Another static test applied to bone cements is the shear strength test according to ASTM D 732. This mechanical parameter is important because debonding of the implant±cement interface has been known to initiate the failure of cement femoral prostheses. The interfacial static shear strength is influenced by surface roughness, cement type and porosity (Wang et al., 2003).

10.16 Typical test methods to determine the static properties for PMMA bone cement. © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

229

Furthermore, there are test methods to determine the fracture properties, such as fracture toughness (ASTM E 399 and ISO 13, 586) and impact strength (ISO 179/ISO 180) (Sih and Bernam, 1980). There is a strong relationship between impact strength and fracture toughness (Lewis and Mladsi, 2000). The long-term viscoelastic properties are creep, stress relaxation and fatigue. The test methods used to determine creep and stress relaxation are described in ASTM D 2990. Bone cements demonstrate plastic properties. Therefore it is possible that they intrude slowly into cavities and close them after polymerisation. This important characteristic allows cements to adapt to changes in bone shape with time. The creep depends on several factors, such as composition, temperature, load size and duration. Equally important are the fatigue properties of the cement, which define the ability of the cement to withstand repeated load cycles without failing. Many studies have focused on the fatigue properties of PMMA bone cement (Krause and Mathis, 1988; Lewis, 1997). Three different testing techniques are used to characterise the fatigue behaviour; ISO 5833, ISO 527 and ASTM F 2118 (Fig. 10.17). Typically, fatigue testing of PMMA bone cement is conducted by bending as the required equipment is relatively simple. Additionally, the preparation of specimens for tension±compression and tension±tension is more complex than the preparation technique for the bending tests. Such studies are very time consuming, requiring between 1 and 10 million cycles until failure is

10.17 Typical test methods to determine the dynamic properties for PMMA bone cement. © 2008, Woodhead Publishing Limited

230

Joint replacement technology

Table 10.4 Physical and mechanical properties of PMMA bone cement (Dunne and Orr, 1998, 2001, 2002; Dunne et al., 2003) PalacosÕ R bone cement

Test method Hand mixed Vacuum mixed @ ÿ2 kPa

Density (kg/m3) Porosity (%) Shrinkage (%) Ultimate tensile strength (MPa) Ultimate compressive strength (MPa) Bending strength (MPa) Bending modulus (MPa) Number of cycles to failure

Displacement 1.18±1.26 Displacement 16.40 Dilatometry 3±5 ISO 527 25 ISO 5833 66 ISO 5833 55 ISO 5833 2110 ISO 527 24,163

1.23±1.30 3.17 7±8 54 80 70 2900 53,528

reached. The results from fatigue testing have been shown to correlate well with the clinical performance of the cement. Mechanical properties of bone cements are affected by various factors, and it is difficult to report strength characteristics (Table 10.4). Some of the factors that influence mechanical properties are the composition of the cement, molecular weight of the polymer powder component (Lewis, 2000), the addition of radiopacifying agents and antibiotics (Gruenert and Ritter, 1974; De Wijn et al., 1975; Kuehn 2000), the porosity (Jasty et al., 1990; Lewis and Mladsi, 1998; Dunne et al., 2003), the sterilisation method used for the polymer powder and the liquid monomer components (Lewis and Mladsi, 1998), the mixing methods (Davies et al., 1987; Lidgren et al., 1987; Wixson et al., 1987; Lewis, 2000; Dunne and Orr, 2001) and the environmental test conditions (Freitag and Cannon, 1977; Johnson et al., 1989).

10.15 Mixing methods When bone cement was first used in arthroplasty, it was hand mixed in a bowl in the operating room and then inserted by hand or transferred and injected into the desired location. Because PMMA comes as a powder composed of pre-polymerised particles to be mixed with the liquid monomer, monomer fumes are released into the air. Furthermore with hand mixing, a certain amount of porosity in the final material is unavoidable owing to air entrapment, even in lower-viscosity cements. During the 1980s different techniques were introduced in the hope of improving mixing and thereby bone cement properties (Burke et al., 1984; LindeÂn, 1991). The results, however, were not convincing. Lidgren et al. (1984) introduced vacuum mixing of bone cement. The quality of the bone cement was improved. Today, vacuum mixing is widely accepted as the method of choice for achieving homogeneous cement, reducing porosity and increasing cement strength, which is why it is an integral part of the modern cementing technique (Malchau and Herberts, 1996). © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

231

Vacuum mixing systems reduce the monomer exposure to the operating theatre staff by 50±70% (Schlegel et al., 2004) and eliminate contact with bone cement during delivery (Darre et al., 1988; Buchhorn et al., 1992; Bettencourt et al., 2001; Eveleigh, 2002). The working environment for the theatre staff is improved, and the risk of fume-induced headaches, respiratory irritation and allergic reactions are minimal. Conventional mixing of bone cement produces a porosity of 5±16%. Vacuum mixing produces porosity of 0.1±1% (LindeÂn and Gillquist, 1989; Wang and Kjellson, 2001). Porosity has been found to be the major cause of decreased mechanical performance of bone cement. To ensure its in vivo survival, the cement must be able to withstand the varying loads it endures. Thus fatigue property, which is directly affected by porosity, is as important in determining the long-term survival of a joint replacement as static strength. Fatigue failure occurs when cement cracks are initiated from defects in the cement mantle. It is known that vacuum mixing of cement improves mechanical properties (Lidgren et al., 1984; Alkire et al., 1987; Wixson e et al., 1987; Schreurs et al., 1988; LindeÂn and Gillquist, 1989; Askew et al., 1990; Davies and Harris, 1990; Mau et al., 2004) largely as a result of minimising micro- and macropores (Wang et al., 1993, 1996). Numerous studies have confirmed that vacuum mixing enhances the fatigue life of the bone cement (Fig. 10.18; Lewis, 2000; Harper and Bonfield, 2000; Wilkinson et al., 2000; Dunne and Orr, 2001; Schelling and Breusch, 2001; Yau et al., 2001; Murphy and Prendergast, 2002).

10.18 Summary of the 50%-probability-of-fracture life (cycle) estimates. Vacuum mixed cement significantly increased fatigue strength (Lewis, 2000). © 2008, Woodhead Publishing Limited

232

Joint replacement technology

10.19 Void on a fracture surface. Many partially unpolymerised PMMA particles and zirconium dioxide particles are seen in the voids (Wang et al., 1994).

Incomplete mixing of the monomer and polymer may lead to partially united and, in some cases, free unbonded cement particles (Fig. 10.19). Vacuum mixing of bone cement not only decreases the number of voids, but also improves the microscopic homogeneity of bone cement (Wang et al., 1994). When cement fracture occurs, inhomogeneous cement may release PMMA and contrast media particles to the bone±cement interface. These particles may evoke a foreign body response or stimulate osteoclast activity (Sabokbar et al., 1997, 2001; Wimhurst et al., 2001), resulting in osteolysis of the surrounding bone. Extensive porosity at the cement±stem interface has been found in retrieved cement mantles and in laboratory-prepared specimens (James et al., 1993; Bishop et al., 1996). This interface porosity is caused by entrapment of air at the stem surface during stem insertion and by residual porosity in the cement. When cement is mixed under vacuum, cement porosity is significantly reduced, thus producing less porosity at the cement±prosthesis interface (Bishop et al., 1996; Wang et al., 1998) (Fig. 10.20). Various studies have shown that interface porosity weakens the resistance of the cement to torsional load (Davies et al., 1995) and decreases fatigue life of the cement±metal interface (Iesaka et al., 2003). Interface porosity has also been linked to the initiation of cement cracks (Jasty et al., 1991; James et al., 1993; Verdonschot, 1995). The evidence is convincing that reduction of interface porosity improves the strength of the interface, thereby increasing the survival of cemented implants. The variation of cement porosity from different mixing systems is still © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

233

10.20 Samples from a cemented implant. The cement was mixed at atmospheric pressure (left), and under vacuum (right) (M = metal; BC = bone cement).

considerable (Dunne et al., 2004; Mau et al., 2004; Wang, 2005). Various studies indicate that macropores increase the risk of fatigue failure, and the current opinion is that efforts should be made to minimise the number and size of macropores. The development and use of a pre-packed bone cement mixing and delivery system to further minimise PMMA exposure, reduce porosity and make handling easier seems warranted.

10.16 Joint replacement cementing technique The key to successful cemented implants is to establish a durable interface between cement and cancellous bone and a tight interface between cement and prosthesis, by means of an even cement mantle. To get a well-cemented implant, careful preparation of the bone bed and effective cement pressurisation is of the utmost importance. Details of the recommendations for cementation in the hip and knee are provided below.

10.16.1 Cemented hip replacement The cementing technique was introduced by Charnley (1970): `The cement is forced down the track of the medullary canal as a stiff dough and the insertion of the point of the tapered stem of the prosthesis expands the stiff dough and injects it into the cancellous lining of the marrow space'. The further development of cementing technique includes saving of high-quality trabecular bone in the implant bed, cleansing of the bone bed by jet-lavage, the use of a distal intramedullary plug and proximal seal for pressurisation. All of these techniques will reduce the risk for revision in hip arthroplasty (Malchau and Herberts, 1998). Currently, this technique is used routinely at almost all clinics in Scandinavia and at many other clinics in the world. © 2008, Woodhead Publishing Limited

234

Joint replacement technology

Acetabular cementing The improvement of long-term results with cemented acetabular components compared with those of Charnley's first series has been attributed to the use of modern cementing technique creating a better cement mantle (Noble and Swarts, 1983; Ranawat et al., 1997; Joshi et al., 1998; Malchau and Herberts, 1998; Ritter et al., 1999). The appearance and anatomy of the acetabulum raises particular difficulties for cement pressurisation and penetration, quite different from the femur where high pressures are achieved more easily. The subchondral bone plate, which is a thin but dense structure, plays an important role in stress distribution in the acetabulum even after the insertion of the prosthesis. However, no cement penetration can occur through the subchondral bone plate and all cement interdigitation takes place in the anchorage holes. In cemented joint replacement, a major prerequisite for lasting implant fixation is an initially stable and secure fixation at the cement±bone interface (Karrholm et al., 1994; MjoÈberg, 1994; Ryd et al., 1995; Stocks et al., 1995). Incomplete cement±bone interdigitation can cause micromotion between cement and bone and, by influence from hydrostatic pressure from the joint access of fluid to the interface with the possibility of migration of wear debris along the interface with later implant loosening (Eftekhar and Nercessian, 1988; Schmalzried et al., 1992; Aspenberg and Van der Vis, 1998). A recent clinical study by Flivik (2005) indicates that careful removal of the subchondral bone plate with six to eight anchorage holes results in a superior cement±bone interface with less development of radiolucent lines and less early micromotion. It is also recommended to partially preserve the subchondral bone plate of the acetabular roof, but to open the cancellous spaces for cement interdigitation with a combination of reaming, multiple drill holes and copious pulsatile lavage (Parsch and Breusch, 2005). A thorough pulse lavage is the basis for cement penetration. Keeping a dry bone bed is also essential. Cement pressurisation is the key for reaching optimal cement±bone integration. The cement can be applied in a rather early phase for high-viscosity cement, or in a middle viscosity stage, when combined with a seal or a pressuriser. Keeping the pressure for 1±3 minutes until the cement reaches a high viscosity followed by cup insertion and keeping the pressure until the cement sets is recommended. A 2±3 mm cement mantle should be aimed for in all areas of acetabulum. The cement should not be applied too early as low viscosity makes it more difficult to pressurise, and there will be more cement leakage. In an animal study it has been shown that low-viscosity cement gives less penetration than high-viscosity cement (Breusch et al., 2002). Pressure above 4 kPa can prevent bleeding, thus avoiding blood lamination at the interface (Benjamin et al., 1987; Shelley and Wroblewski, 1988). Experimentally, pressures of 35±50 kPa for 30±60 seconds have been found to produce near-optimal cement penetration into cleaned cancellous bone (Noble and © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

235

Swarts, 1983). The maximum pressure for the acetabular cup is often achieved during insertion of the implant, when this is done at a time when the cement has reached a high-viscosity stage. However, experimentally, the best cement penetration occurs during pressurisation rather than during cup insertion (Abdulghani et al., 2007). Cement penetration between 2 and 5 mm has been considered optimal to give the cement±bone interface a high tensile and shear strength (Krause et al., 1982; Askew et al., 1984; MacDonald et al., 1993; Majkowski et al., 1993; Mann et al., 1997). Protection of the bearing surfaces from cement contamination is important in order to avoid cement particles from causing accelerated wear (Kesteris et al., 2001). Femoral cementing To achieve adequate cement interdigitation and a viable interlock, preservation of cancellous bone stock is essential. Charnley (1970) advocated preservation of cancellous bone for cemented anchorage. He believed that it was an advantage to have a layer of cancellous bone interposed between the cement surface and cortical bone. Based on his further clinical experience, Charnley (1979) later recommended preservation of 2±3 mm of strong cancellous bone adjacent to the endosteal surface. Therefore, the reaming and broaching processes should aim at preservation of a healthy layer of cancellous bone, with the blood supply minimally disrupted. A stable distal intramedullary cement restrictor allows for cement containment and better pressurisation, which results in improved cement penetration (Markolf and Amstutz, 1976; Indong et al., 1978) and better clinical outcome (Harris et al., 1982; Harris and McGann, 1986; Mulroy and Harris, 1990; Karrholm et al., 2005). A distal plug is always used in Sweden and is considered mandatory in cemented total hip arthroplasty to ensure adequate cement pressurisation. Using pulse lavage prior to plug insertion may prevent fat embolism (Breusch, 2005). Brushing of the femoral cavity will be beneficial as if cleans out debris and opens the bone trabecular spaces (Lidgren and Robertsson, 2005). The use of bone lavage prior to cementation has improved cement penetration and increased interface shear strength (Bannister and Miles, 1988). In further studies, Breusch et al. (2000a, 2001) investigated the effect of the technique and volume of lavage on cement penetration in human cadaver and in sheep bone. Their results showed that the use of 1 litre pulse lavage yielded significantly improved rates of cement penetration compared with syringe or bladder-syringe lavage (Fig. 10.21). The use of pulse lavage is considered mandatory for cleansing the bone bed in cemented hip replacement. It not only improves cement penetration, but also significantly reduces the risk of embolic complications during cement pressurisation. A dry bone bed is essential for optimal interdigitation of cement. © 2008, Woodhead Publishing Limited

236

Joint replacement technology

10.21 Microradiographs of cemented femora. Cement penetration in different levels for jet lavage (left) and syringe lavage section (right) (Breusch, 2005).

The key to the cementing technique is to drive cement into the trabecular structure of cancellous bone, and create micro-interlocks. Two issues should be considered during cement penetration into the trabeculae. One is that a bleeding bone surface will prevent the flow of cement. The pressure generated by bleeding has been measured at 3.53 kPa (Heys-Moore and Ling, 1982) which could be sufficient to displace cement from trabeculae or cause laminations, especially with cement of low viscosity. Secondly, bone cement changes viscosity with time and the change varies not only with formulation but also with other factors such as the ambient temperature and humidity. This complex relationship remains the technical challenge of cementing during surgery. First of all, 80 g (at least 60 g) bone cement should be used in order to achieve better filling and subsequent pressurisation. With regard to pressure on cement penetration, Panjabi et al. (1983) used a canine model to analyse the role of insertion pressure on cement penetration. They concluded that 520 kPa was sufficiently high to achieve adequate penetration of cement, but sufficiently low to avoid complications. The shear strength at the cement±bone interface increased significantly with pressure until the pressure reached 410 kPa (Bean et al., 1988). Continuous pressure measurement throughout cementation has demonstrated that stem insertion achieves the highest pressures (Song et al., 1994). Greater pressures during stem insertion were generated distally (359±758 kPa) when compared with proximal pressures (131±200 kPa) (Bourne et al., 1984). The timing of prosthesis insertion can affect the pressure generated. Late stage femoral stem insertion had significantly less radiolucency and an increased cement±bone interface contact compared with early stem insertion (Churchill et al., 2001; Dayton et al., 2002), suggesting © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

237

that late stem insertion associated with high cement viscosity generates higher intramedullary pressure leading to improve cement penetration into the bone. In general, pressurisation has two stages: firstly, pressure will be generated by an entirely sealed pressuriser combined with a gun and a medium viscosity phase. This phase is regarded as the most crucial and ideally full cement interlock should be achieved at this stage before prosthesis insertion. Secondly, (distal) pressure can be generated by the prosthesis during insertion; this utilises a relatively higher cement viscosity. The quality of pressurisation has been demonstrated to be related to a good clinic outcome (Fig. 10.22). The prosthesis should not be hammered at the end of process of insertion. Finally it is important to maintain the prosthesis position until final polymerisation.

10.22 Poisson analyses showing that the use of a proximal plug in femoral sealing reduces the long-term risk of aseptic loosening (all diagnoses and reasons, 1992±2005; Swedish Hip Register, 2005). ß Swedish Hip Arthroplasty Register. © 2008, Woodhead Publishing Limited

238

Joint replacement technology

Optimum cement mantle thickness is still not clearly defined. There is little doubt that a deficient cement mantle may be detrimental with regard to longterm implant survival. Thin layers of cement have less potential for energy absorption and may crack and fail (Huiskes, 1980; Jasty et al., 1986), in particular in the proximal and distal portions of the cement mantle (Kawate et al., 1998). Furthermore, a deficient cement mantle may create a pathway for particulate wear debris to migrate along the stem±cement interface down to the cement±bone interface, thus initiating or accelerating particle-induced osteolysis and loosening (Jasty et al., 1986; Howie et al., 1988). In contrast complete cement mantles with a minimum thickness of 2±3 mm have been reported to be associated with better long-term radiographic outcome (Joshi et al., 1993, 1998; Ebramzadeh et al., 1994). Using a distal stem centraliser it is possible to obtain a homogeneous cement mantle at the tip of the stem and avoid cement mantle deficiencies, but using such a device does not completely control thin cement mantles in the middle and proximal zones. The benefit of centralisers with regard to long-term outcome remains subject to debate (Tolo et al., 1998). A good cement mantle should cover the entire femoral stem and achieve good contact to bone without defects.

10.16.2 Cemented knee replacement Total knee arthroplasty (TKA) surgery started in the 1960s and early 1970s. A combination of metal and plastics were inserted using bone cement. Many studies from the late 1970s and early 1980s indicated that the cementing techniques are important to stabilise the implants in TKA (Cooke et al., 1978; Tremblay et al., 1979; Convery and Malcom, 1980; Askew et al., 1984; Walker et al., 1984). Proper preparation of the bone surface and correct application of cement have been shown to improve the survival of TKA (Ritter et al., 1994; Hofmann et al., 2006). During the period from the mid-1980s to the mid-1990s the use of uncemented implants was relatively common, but the risk of revision was 1.4 times higher than with cemented implants (Robertsson et al., 2001). In recent years bone cement has been used in the majority of knee joint replacements. More than 95% of TKAs are done with cement in various countries, such as the United Kingdom, Norway, Australia (Phillips et al., 1996; Lutz and Halliday, 2002; Norwegian Arthroplasty Register, 2006), and in Sweden reaching even 99% (Lidgren and Robertsson, 2006; Fig. 10.23). Fundamental to cemented implant longevity are meticulous technique, bone preparation and handling of the cement. There are at the present time limited published scientific studies showing that certain techniques for the cementation are clinically superior. Established methods include bone impaction in the tibia and the femoral canal to prevent the cement from penetrating too far into the medullary canals, and the uses of high-pressure irrigation and pressurisation to get better cement penetration and interdigitation. © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

239

10.23 The yearly distribution for cemented, uncemented and hybrid fixation of components (ß 2006 Swedish Knee Register).

Prior to cement application, hard sclerotic bone should be drilled or abraded to allow the cement to grasp the bone surface. The bone surface requires lavage and the most efficient way is by using pulse lavage. Three litre pulse lavage before implantation is effective in removing bone debris and organics (Maistrelli et al., 1995; Helmers et al., 1999; Norton and Eyres, 2000; Weiss et al., 2003; Niki et al., 2007). After lavage, the bone surface should be dried and is even better cleaned by suction (Norton and Eyres, 2000; Stannage et al., 2003). Cleaning of debris from the trabecular bone allows penetration of the cement and improves fixation strength to both shear and tensile force (Askew et al., 1984; Walker et al., 1984; Ritter et al., 1994; Maistrelli et al., 1995; Weiss et al., 2003). Devices for delivering the bone cement into the trabecular bone together with sealing membranes or a pressuriser to achieve pressure in the proximal tibia are extensively used. The techniques have improved cement penetration into the bone and increased the tensile strength of the cement±bone interface (Ritter et al., 1994; Mann et al., 1997; Norton and Eyres, 2000; Bauze et al., 2004). The ideal cement penetration into the bone is 1±2 mm. With soft rheumatoid bone, deeper cement penetration may occur. A maximal penetration depth of 10 mm has been suggested as acceptable before the large volume of cement may cause © 2008, Woodhead Publishing Limited

240

Joint replacement technology

thermal necrosis (Huiskes and Sloof, 1981). The desirable cement penetration should be approximately 1 mm. All excess cement must be removed from around the components to prevent cement particles from breaking loose. The presence of entrapped cement between the articular surfaces leads to third-body wear and damage to the polyethylene. Pulse lavage of 1±2 litre after cementation will be effective for removing the remaining bone debris and cement particles during cementing of TKA (Helmers et al., 1999; Niki et al., 2007).

10.17 Problems with acrylic cements 10.17.1 Infection Since the early 1970s, the use of antibiotics in bone cement has been widespread. Antibiotic-loaded bone cement is used prophylactically in both primary and aseptic revision joint replacement surgery (Buchholz et al., 1984; Espehaug et al., 1997). A survey including a total of 25 controlled randomised trials published between 1966 and 1998 showed the overall rate of infection in THR surgery to be 1% (2.1% including the TKR patients). In the trials Staphylococcus aureus and Staphylococcus epidermidis were the most frequently isolated pathogens (Glenny and Song, 1999). The largest European multi-centre study with 8000 joint replacements was done by Lidwell et al. (1982, 1984). They found a clear effect of combining local and systemic antibiotics in addition to the use of laminar airflow. Norwegian and Swedish long-term register studies show that bone cement containing antibiotic alone is less effective than systemic antibiotics, whereas the combination of both is still better (Malchau and Herberts, 1998; Walenkamp and Murray, 2001). In a randomised study with 340 primary knee arthroplasties, a reduced rate of deep infection was found when antibiotic-impregnated cement was combined with systemic antibiotic prophylaxis compared with bone cement without antibiotics (Chiu et al., 2002). The synergistic effect of using short-term systemic antibiotic together with a cement containing antibiotic seems promising. The antibiotics are released from the bone cement into the tissues surrounding the implant. This local concentration of antibiotics is sufficient to kill the antibiotic-sensitive bacteria left in the wound (Walenkamp and Murray, 2001; Hendriks, 2003; Breusch and Malchau, 2005). Gentamicin is used in many brands of cements worldwide. A concern could be an increased risk of bacterial resistance to gentamicin in patients after joint replacement (SanzeÂn and Walder, 1988). However, this has not been proven but there is a need for continuous monitoring of bacteria and resistance patterns. The most common bacterial species cultured today in joint prosthetic infection in Scandinavia are coagulase negative staphylococci, but other so-called `lowvirulence anaerobes' are also causing infections (Lidgren, 2001). Apart from gentamicin, other antibiotics have been used in bone cement, such as © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

241

tobramycin, vancomycin, clindamycin and fusidic acid. Combinations of these antibiotics, for example clindamycin and gentamicin in bone cement, have been used for revision surgery (Penner et al., 1996; Konig et al., 2001; Koo et al., 2001; Walenkamp and Murray, 2001; Breusch and Malchau, 2005).

10.17.2 Risk of fat embolism This is a clinical complication that is still debated. A transient fall of blood pressure 1±2 minutes after introducing bone cement into the medullary canal of the femur is often seen in clinical practice. Intra-operative mortality during cemented hip replacement has been evaluated by Parvizi et al. (1999) in a large retrospective study involving 29 431 patients, and was found to occur in 0.08%. Two separate causes have been considered. One is the chemical factor, i.e. whether the MMA monomer causes the problem. Intra-operative measurements in patients have showed in vivo concentrations ranging from 0.3 to 5.9 mg/ 100 ml. Death occurred in an animal model after injection of doses corresponding to hundredfold MMA concentrations (125 mg/100 ml) (Charnley 1970; Breusch and Malchau, 2005). Clinically correlation between measured MMA concentrations and pressure drop could not be found by Wenda et al. (1988). It is highly unlikely that MMA, under clinical conditions, is a cause of death during surgery (Kuehn, 2000). The second possible cause is fat embolism. A summary of 21 publications from 1970 to 2001 considering the intra-operative mortality during cemented hip replacement showed that fat and marrow embolism occurred in 32 of 37 cases (85.6%) at autopsy (Clarius et al., 2005). Pressurisation, femoral stem implantation and reposition of the hip were found to be the most embolism-prone operative steps during the operation (Breusch and Malchau, 2005). Most intra-operative mortality is associated with fat and marrow embolism. A thorough pulsatile lavage preparation has been proved to be able to clean fat and marrow from the bone bed (Byrick et al., 1989; Breusch et al., 2000a,b). Therefore, if strict pulsatile lavage is implemented before cementation, the risk of fat and marrow embolism will be minimised.

10.17.3 Wear particles One of the major long-term problems with the use of bone cement is loosening that originates from periprosthetic osteolysis secondary to cement fragmentation and the inflammatory and foreign body reaction to wear particles. PMMA-wear particles are often found in tissue surrounding an implant together with polyethylene and metal particles. The size of the particles ranges from 0.5 to 50 m (Walenkamp and Murray, 2001). Such particles activate macrophages from the periprosthetic tissues expressing the pro-inflammatory cytokines interleukin-1, interleukin-6 and tumor necrosis factor alpha. These cytokines play a major role in the process of periprosthetic osteolysis. (These aspects are © 2008, Woodhead Publishing Limited

242

Joint replacement technology

dealt with in detail in Chapter 15 of this book.) Particles combined with micromotion or fluid pressure in vivo cause foreign body and chronic inflammatory reaction which may accelerate bone resorption (Goodman, 1994; Aspenberg and Van der Vis, 1998). Loose particles can migrate along a cement fracture line to the bone bed or along the gap between cement and prosthesis up to the joint space, causing wear problems. Strengthening the bone cement, achieving proper cement interdigitation and close cement contact to the prosthesis are factors for avoiding wear particle generation and access.

10.18 Summary Since Sir John Charnley introduced bone cement for joint replacement surgery nearly 50 years ago, acrylic cement has been used widely throughout the world. With thousands of studies our understanding of the properties and use of bone cement has increased. The users of cemented joint replacement need to know the chemical and physical properties of bone cement, which change by even slight variations in the chemical composition. The final bone cement is produced by nurses and surgeons in the operating theatre. These users have enormous influence on the quality of the final cement. Control of handling procedures is of the utmost importance in producing a well-cemented implant. The use of modern cementing techniques has demonstrated increased long-term survival rate of cemented implants. Further education and training on bone cement and cementation techniques will reduce the complications of bone cement use and lead to yet longer implant survival.

10.19 References Abdulghani S, Wang J-S, McCarthy I, Flivik G (2007), `The influence of initial pressurization and cup introduction time on the cement penetration depth in an acetabular model', Acta Orthop, 78, 333±339. Adalberth G, Nilsson KG, KaÈrrholm J, Hassander H (2002), `Fixation of the tibial component using CMW-1 or Palacos bone cement with gentamicin: similar outcome in a randomized radiostereometric study of 51 total knee arthroplasties', Acta Orthop Scand, 73, 531±538. Alkire MJ, Dabezies EJ, Hastings PR (1987), `High vacuum as a method of reducing porosity of polymethylmethacrylate', Orthopaedics, 10, 1533±1539. Askew MJ, Steege JW, Lewis JL, Ranieri JR, Wixson RL (1984), `Effect of cement pressure and bone strength on polymethylmethacrylate fixation', J Orthop Res, 1, 412±420. Askew MJ, Kufel MF, Fleissner PR, Gradisar, IA, Salstrom SJ, Tan J (1990), `Effect of vacuum mixing on the mechanical properties of antibiotic-impregnated polymethylmethacrylate bone cement', J Biomed Mater Res, 24, 573±580. Aspenberg P, Van der Vis H (1998), `Migration, particles, and fluid pressure. A discussion of causes of prosthetic loosening', Clin Orthop, 352, 75±80. Bannister GC, Miles AW (1988), `The influence of cementing technique and blood on the

© 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

243

strength of the bone±cement interface', Eng Med, 17, 131±133. Bauze AJ, Costi JJ, Stavrou P, Rankin WA, Hearn TC, Krishnan J, Slavotine JP (2004), `Cement penetration and stiffness of the cement-bone composite in the proximal tibia in a porcine model', J Orthop Surg (Hong Kong), 12, 194±198. Bean DJ, Hollis, JM, Woo SLY, Convery FR (1988), `Sustained pressurization of polymethylmethacrylate: a comparison of low- and moderate-viscosity bone cements', J Orthop Res, 6, 580±584. Benjamin JB, Gie GA, Lee AJ, Ling RS, Volz RG (1987), `Cementing technique and the effects of bleeding', J Bone Joint Surg, 69-B, 620±624. Bettencourt A, Calado A, Amaral J, Vale FM, Rico JM, Monteiro J, Castro M (2001), `The influence of vacuum mixing on methylmethacrylate liberation from acrylic cement powder', Int J Pharm, 219, 89±93. Bishop NE, Ferguson S, Tepic S (1996), `Porosity reduction in bone cement at the cement±stem interface', J Bone Joint Surg, 78-B, 349±356. Bourne RB, Oh I, Harris WH (1984), `Femoral cement pressurization during total hip replacement. The role of different femoral stems with reference to stem size and shape', Clin Orthop, 183, 12±16. Breusch S (2001), `Cementing technique in THR: factors influencing survival of femoral components', in: Walenkamp GHIM, Murray DW (Eds) Bone cements and bone cement techniques, Berlin, Heidelberg, Springer. Breusch SJ (2005), `Bone preparation: femur', in Breusch S and Malchau H, The Wellcemented Total Hip Arthroplasty. Theory and Practice, Berlin Heidelberg, Springer, 125±140. Breusch S, Kuehn K-D (2003), `Bone cements based on polymethylmethacrylate', Orthopaedics, 32, 41±50. Breusch S, Malchau H (2005), The Well-cemented Total Hip Arthroplasty. Theory and Practice, Berlin, Heidelberg, Springer-Verlag. Breusch S, Norman TL, Schneider U, Reitzel T, Blaha JD, Lukoschek M (2000a), `Lavage technique in THA: Jet-lavage produces better cement penetration than syringe-lavage in the proximal femur', J Arthroplasty, 15, 7921±7927. Breusch S, Reitzel T, Schneider U, Volkmann M, Ewerbeck V, Lukoschek M (2000b), `Cemented hip prosthesis implantation-decreasing the rate of fat embolism with pulsed pressure lavage', Orthopaed, 29, 578±586. Breusch S, Schneider U, Reitzel T, Kreutzer J, Ewerbeck V, Lukoschek M (2001), `Significance of jet lavage for in vitro and in vivo cement penetration', Z Orthop Ihre Grenzgeb, 139, 52±63. Breusch S, Heisel C, Muller J, Borchers T, Mau H (2002), `Influence of cement viscosity on cement interdigitation and venous fat content under in vivo conditions: a bilateral study of 13 sheep', Acta Orthop Scand, 73, 409±415. Buchholz H, Elson R, Engelbrecht E, LodenkaÈmper H, RoÈttger J, Siegel A (1981), `Management of deep infection of total hip replacement', J Bone Joint Surg, 63-B, 342±353. Buchholz HW, Elson RA, Heinert K (1984), `Antibiotic-loaded acrylic cement: current concepts', Clin Orthop, 190, 96±108. Buchhorn G, Streicher R, Willert (1992), `Exposure of surgical/orthopadic operating room personnel to monomer vapors during the use of bone cements ± review of the literature and report of experiences', Biomed Tech (Berl), 37, 293±302. Burke D, Gates E, Harris WH (1984), `Centrifugation as a method of improving tensile and fatigue properties of acrylic bone cement', J Bone Joint Surg, 66-A, 1265± 1273. © 2008, Woodhead Publishing Limited

244

Joint replacement technology

Byrick RJ, Bell RS, Kay JC, Waddell JP, Mullen JB (1989), `High-volume, high-pressure pulsatile lavage during cemented arthroplasty', J Bone Joint Surg, 71-A, 1331± 1336. Charnley J (1960), `Anchorage of the femoral head prostheses of the shaft of the femur', J Bone Joint Surg, 42B, 28±30. Charnley J (1970), Acrylic Cement in Orthopedic Surgery, Edinburgh, London, E&S Livingstone. Charnley J (1979), Low Friction Arthroplasty of the Hip: Theory and Practice, Berlin, Heidelberg, Springer. Chiu F-Y, Chen C-M, Lin C-FJ, Lo W-H (2002), `Cefuroxime impregnated cement in primary total knee arthroplasty. A prospective, randomized study of three hundred and forty knees', J Bone Joint Surgery, 84-A, 759±762. Churchill DL, Incavo SJ, Uroskie JA, Beynnon BD (2001), `Femoral stem insertion generates high bone cement pressurization', Clin Orthop, 393, 335±344. Clarius M, Heisel C, Breusch SJ (2005), `Pulmonary embolism in cemented total hip arthroplasty', in: Breusch S and Malchau H, The Well-cemented Total Hip Arthroplasty. Theory and Practice, Berlin, Heidelberg, Springer, 320±331. Convery FR, Malcom LL (1980), `Prosthetic fixation wih controlled pressurized polymerization of polymethylmethacrylate', Proceedings of the 26th Orthopedic Research Society, February, 5±7, 77. Cooke FW, Cipolletti GB, Lunceford EM, Sauer BW (1978), `The influence of surgical technique on the strength of cement fixation', Proceedings of the 24th Orthopedic Research Society, February, 21±23, Dallas, Texas, 89. Darre E, Gottlieb J, Nielsen P M, Jensen J S (1988), `A method to determine methylmethacrylate in air', Acta Orthop Scand, 59, 270±271. Davies J, O'Connor D, Greer J, Harris W (1987), `Comparison of mechanical properties of Simplex P Zimmer Regular and LVC bone cements', J Biomed Mater Res, 21, 719±730. Davies JP, Harris WH (1990), `Optimization and comparison of three vacuum mixing systems for porosity reduction of Simplex P Cement', Clin Orthop, 254, 261±269. Davies JP, Harris WH (1995), `Comparison of diametral shrinkage of centrifuged and uncentrifuged Simplex P bone cement', J Appl Biomater, 6, 209±211. Davies JP, Kawate K, Harris WH (1995), `Effect of interfacial porosity on the torsional strength of the cement±metal interface', 41st Annual Meeting Orthopedic Research Society, February, 13±16, Orlando, Florida, 713. Dayton MR, Incavo SJ, Churchill DL, Uroskie JA, Beynnon BD (2002), `Effects of early and late stage cement intrusion into cancellous bone', Clin Orthop, 405, 39±45. De Wijn J, Sloof T, Driessens F (1975), `Characterization of bone cements', Acta Orthop Scand, 46, 38±51. Draenert K, Draenert Y, Garde U, Ulrich C (1999), Manual of Cementing Technique, Berlin, Heidelberg, Springer. Dunne N, Orr J (1998), `Flow characteristics of curing polymethyl methacrylate bone cement', Proc Inst Mech Engrs, J Eng Med, 212, 199±207. Dunne N, Orr J (2001), `Influence of mixing techniques on the physical properties of acrylic bone cement', Biomaterials, 22, 1819±1826. Dunne N, Orr J (2002), `Thermal characteristics of curing acrylic bone cement', J Mater Sci: Mater Med, 13, 17±22. Dunne N, Orr J, Mushipe M, Eveleigh R (2003), `The relationship between porosity and fatigue characteristics of bone cements', Biomaterials, 24, 239±245. Dunne NJ, Carey G, Orr J, Beverland D (2004), `Current affairs: bone cement mixing', J © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

245

Adv Perioperative Care, 2, 11±18. Ebramzadeh E, Samiento A, Mckellop HA, LLinas A, Gogan W (1994), `The cement mantle in total hip arthroplasty. Analysis of long-term radiographic results', J Bone Joint Surg, 76-A, 77±87. Eftekhar NS, Nercessian O (1988), `Incidence and mechanism of failure of cemented acetabular component in total hip arthroplasty', Orthop Clin North Am, 19, 557± 566. Ege W, Kuehn K, Tuchscherer C, Maurer H (1998), `Physical and chemical properties of bone cements', in: Walenkamp GHIM (ed) Biomaterials in Surgery, Stuttgart, Georg Thieme. Espehaug B, Engesaeter L, Vollset S, Havelin L, Langeland N (1997), `Antibiotic prophylaxis in total hip arthroplasty', J Bone Joint Surg, 79-B, 590±595. Eveleigh R (2002), `Fume levels during bone cement mixing', Br J Perioper Nurs, 12, 145±7, 149±150. Flivik G (2005), `Fixation of the cemented acetabular component in hip arthroplasty', PhD Thesis, Lund University. Freitag T, Cannon S (1977), `Fracture characteristics of acrylic bone cements. II. Fatigue', Biomed Mater Eng, 11, 609±624. Frommelt L (2001) `Gentamicin release from PMMA bone cement: mechanism and action on bacteria', in Walenkamp GHIM and Murray DW (Eds), Bone Cement and Cementing Technique, Berlin, Heidelberg, Springer. Ginebra M, Albuixech L, FernaÂndez-BarragaÂn E, Aparicio C, Gil F, San R, VaÂzquez B, Planell J (2002), `Mechanical performance of acrylic bone cements containing different radiopacifying agents', Biomaterials, 23, 1872±1882. Glenny AM, Song F (1999), `Antimicrobial prophylaxis in total hip replacement: a systematic review', Health Technol Assessment, 3, 21. Goodman SB (1994), `The effects of micromotion and particulate materials on tissue differentiation. Bone chamber studies in rabbits', Acta Orthop Scand Suppl, 258, 1± 43. Gristina A (1987), `Biomaterial-centered infection: microbial adhesion versus tissue integration', Science, 237, 1588±1595. Gruenert A, Ritter G (1974), `Alterations of the physical properties of so-called bone cements after admixing foreign ingredients', Arch Orthop Unfallchir, 78, 336±342. Harper EJ, Bonfield W (2000), `Tensile characteristics of ten commercial acrylic bone cements', J Biomed Mater Res (Appl Biomater), 53, 605±616. Harris WH, McGann WA (1986), `Loosening of the femoral component after the use of the medullary-plug cementing technique. Follow-up note with a minimum five-year follow-up', J Bone Joint Surg, 68-A, 1064±1066. Harris WH, McCarthy JC, O'Neill DA (1982), `Femoral component loosening using contemporary techniques of femoral cement fixation', J Bone Joint Surg, 64-A, 1063±1067. Helmers LS, Sharkey PF, McGuigan FX (1999), `Efficacy of irrigation for removal of particulate debris after cemented total knee arthroplasty', J Arthroplasty, 14, 549± 552. Hendriks H (2003), `Antibiotic release from bone cement under simulated physiological conditions', PhD thesis. Heys-Moore GH, Ling RSM (1982), `Current cementing techniques', in: Marti R (ed.) Progress in Cemented Total Hip Surgery and Revision, Proceedings of a Symposium held in Amsterdam, Amsterdam, Geneva, Hong Kong, Princeton, Tokyo: Excerpta Medica, 71. © 2008, Woodhead Publishing Limited

246

Joint replacement technology

Hofmann AA, Goldberg TD, Tanner AM, Cook TM (2006), `Surface cementation of stemmed tibial components in primary total knee arthroplasty: minimum 5 year follow-up', J Arthroplasty, 21, 353±357. Howie DW, Vernon-Roberts B, Oakshott R, Manthey B (1988), `A rat model of resorption of bone at the cement±bone interface in the presence of polyethylene wear particles', J Bone Joint Surg, 70-A, 257±263. Huiskes R (1980), `Some fundamental aspects of human joint replacement. Analyses of stress and heat conduction in bone±prosthesis structures', Acta Orthop Scand, Suppl, 185, 109±200. Huiskes R, Sloof TJ (1981), `Thermal injury of cancellous bone following pressurized penetration of acrylic cement', Proceedings of the 27th Orthopedic Research Society, February, 24±26, Las Vegas, Nevada, 134. Iesaka K, Jaffe WL, Kummer FJ (2003), `Effects of preheating of hip prostheses on the stem±cement interface', J Bone Joint Surg, 85-A, 421±427. Indong Oh, Carlson CE, Tomford WW, Harris WH (1978), `Improved fixation of the femoral component after total hip replacement using a methacrylate intramedullary plug', J Bone Joint Surg, 60-A, 608±613. James SP, Schmalzried TP, McGarry FJ, Harris WH (1993), `Extensive porosity at the cement-femoral prosthesis interface: a preliminary study', J Biomed Mater Res, 27, 71±78. Jasty MJ, Floyd WE3rd, Schiller AL, Goldring SR, Harris WH (1986), `Localized osteolysis in stable, non-septic total hip replacement', J Bone Joint Surg, 68-A, 912±919. Jasty M, Davies J, O'Connor D, Burke D, Harrigan T, Harris WH (1990), `Porosity of various preparations of bone cements', Clin Orthop, 259, 122±129. Jasty M, Maloney WJ, Bragdon CR, O'Connor DO, Haire T, Harris WH (1991), `The initiation of failure in cemented femoral components of hip arthroplasties', J Bone Joint Surg, 73-B, 551±558. Johnson J, Provan J, Krygier J, Chan K, Miller J (1989), `Fatigue of acrylic bone cementeffect of frequency and environment', Biomed Mater Eng, 23, 819±831. Joshi AB, Porter ML, Trail A, Hunt LP, Murphy JC, Hardinge K (1993), `Long-term results of Charnley low friction arthroplasty in young patients', J Bone Joint Surg, 75-B, 616±623. Joshi RP, Eftekhar NS, McMahon DJ, Nercessian OA (1998), `Osteolysis after Charnley primary low-friction arthroplasty. A comparison of two matched paired groups', J Bone Joint Surg, 80-B, 585±590. Judet J, Judet R (1956), `The use of an artificial femoral head for arthroplasty of the hip joint', J Bone Joint Surg, 32-B, 166. Karrholm J, Borssen B, Lowenhielm G, Snorrrason F (1994), `Does early micromotion of femoral stem prostheses matter? 4±7 year stereoradiographic follow-up of 84 cemented prostheses', J Bone Joint Surg, 76-B, 912±917. Karrholm J, Garellick G, Herberts P (2005), `The Swedish Hip Arthroplasty Register. Annual Report 2005', http://www.jru.orthop.gu.se/. Kawate K, Maloney WJ, Bragdon CR, Biggs SA, Jasty MJ, Harris WH (1998), `Importance of a thin cement mantle. Autopsy studies of eight hips', Clin Orthop, 355, 70±76. Kesteris U, Carlsson L, Haraldsson C, Lausmaa J, Lidgren L, OnnerfaÈlt R, Wingstrand H (2001), `Contamination of polyethylene cups with polymethyl methacrylate particles: an experimental study', J Arthroplasty, 16, 905±908. Kirschner P (1978), `Experimentelle Untersuchungen mechanischer und chemischer © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

247

Eigenschaften von Knochenzementen nach Langzeitimplantation in menschlichen KoÈrper', Mainz, Germany: Professorial dissertation. Kleinschmitt O (1941), `Plexiglas zur Deckung von SchaÈdelluÈcken', Chirurg, 13, 273. Klekamp J, Dawson J, Haas D, DeBoer D, Christie M (1999), `The use of Vancomycin and Tobramycin in acrylic bone cement: biomechanical effects and elution kinetics for use in joint arthroplasty', J Arthroplasty, 14, 339±346. Konig DP, Schierholz JM, Hilgers RD, Bertram C, Perdreau-Remington F, Rutt J (2001), `In vitro adherence and accumulation of Staphylococcus epidermidis RP 62 A and Staphylococcus epidermidis M7 on four different bone cements', Langenbecks Arch Surg, 386, 328±332. Koo KH, Yang JW, Cho SH, Song HR, Park HB, Ha YC, Chang JD, Kim SY, Kim YH (2001), `Impregnation of Vancomycin, Gentamicin, and Cefotaxime in a cement spacer for two-stage cementless reconstruction in infected total hip arthroplasty', J Arthroplasty, 16, 882±892. Krause W, Mathis R (1988). `Fatigue properties of acrylic bone cements: review of the literature', J Biomed Mater Res, 22, 155±182. Krause WR, Krug W, Miller J (1982), `Strength of the cement±bone interface', Clin Orthop, 163, 290±299. Kuehn K-D (2000), Bone Cements: Up-to-date Comparison of Physical and Chemical Properties of Commercial Materials, Berlin, Heidelberg, Springer. Lautenschlager E, Jacobs J, Marshall G, Meyer P (1976), `Mechanical properties of bone cements containing large doses of antibiotic powders', J Biomed Mater Res, 10, 929±938. Lewis G (1997), `Properties of acrylic bone cement: state of art review', J Biomed Mater Res (Appl Biomater), 38, 155±182. Lewis G (2000), `Relative roles of cement molecular weight and mixing method on the fatigue performance of acrylic bone cements Simplex P versus Osteopal', J Biomed Mater Res (Appl Biomater), 53, 119±130. Lewis G, Mladsi S (1998), `Effect of sterilization method on properties of Palacos R acrylic bone cement', Biomaterials, 19, 117±124. Lewis G, Mladsi S (2000), `Correlation between impact strength and fracture toughness of PMMA-based bone cements', Biomaterials, 21, 775±781. Lidgren L (2001), `Joint prosthetic infections: a success story', Acta Orthop Scand, 72, 553±556. Lidgren L, Robertsson O (2005), `Acrylic bone cements: clinical developments and current status: Scandinavia', Orthop Clin N Am, 36, 55±61. Lidgren L, Robertsson O (2006), Swedish Knee Register, (http://www.knee.nko.se). Lidgren L, Drar H, Moller J (1984), `Strength of polymethylmethacrylate increased by vacuum mixing', Acta Orthop Scand, 55, 536±541. Lidgren L, Bodelind B, Moller J (1987). `Bone cement improved by vacuum mixing and chilling', Acta Orthop Scand, 58, 27±32. Lidwell OM, Lowbury EJU, Whyte W, Blowers R, Stanley SJ, Lowe D (1982), `Effect of ultraclean air in operating rooms on deep sepsis in the joint after total hip or knee replacement: a randomized study', BMJ, 285, 10±14. Lidwell OM, Lowbury EJU, Whyte W, Blowers R, Stanley SJ, Lowe D (1984), `Infection and sepsis after operations for total hip of knee joint replacement: influence of ultraclean air, prophylactic antibiotics and other factors', J Hyg Camb, 93, 504±529. LindeÂn U (1991), `Mechanical properties of bone cement. Importance of the mixing technique', Clin Orthop, 272, 274±278. LindeÂn U, Gillquist J (1989), `Air inclusion in bone cement. Importance of the mixing © 2008, Woodhead Publishing Limited

248

Joint replacement technology

technique', Clin Orthop, 247, 148±151. Lutz MJ, Halliday BR (2002), `Survey of current cementing techniques in total knee replacement', Anz J Surg, 72, 437±439. MacDonald W, Swarts E, Beaver R (1993), `Penetration and shear strength of cement± bone interfaces in vivo', Clin Orthop, 286, 283±288. Maistrelli GL, Antonelli L, Fornasier V, Mahomed N (1995), `Cement penetration with pulsed lavage versus syringe irrigation in total knee arthroplasty', Clin Orthop, 312, 261±265. Majkowski RS Miles AW, Bannister GC, Perkins J, Taylor GJ (1993), `Bone surface preparation in cemented joint replacement', J Bone Joint Surg, 75-B, 459±463. Malchau H, Herberts P (1996), `Prognosis of total hip replacement; surgical and cementing technique in THR: a revision-risk study of 134 056 primary operations', 63rd Annual Meeting of the American Academy of Orthopedic Surgeons, February, 22, Atlanta, Georgia, 22±26. Malchau H, Herberts P (1998), `Prognosis of total hip replacement in Sweden: Revision and re-revision rate in THR', The 65th Annual Meeting of the American Academy of Orthopedic Surgeons, March, 19, New Orleans, Louisiana. Mann KA, Ayers DC, Werner FW, Nicoletta RJ, Fortino MD (1997), `Tensile strength of the cement±bone interface depends on the amount of bone interdigitated with PMMA cement', J Biomech, 30, 339±346. Mark K, Nelson C, Lautenschlager E (1976), `Antibiotic-impregnated acrylic bone cement', J Bone Joint Surg, 58-A, 358±364. Markolf KL, Amstutz HC (1976), `In vitro measurement of bone-acrylic interface pressure during femoral component insertion', Clin Orthop, 121, 60±66. Mau H, Schelling K, Heisel C, Wang JS, Breusch SJ (2004), `Comparison of different vacuum mixing systems and bone cements with respect to reliability, porosity and bending strength', Acta Orthop Scand, 75, 160±172. MjoÈberg B (1994), `Theories of wear and loosening in hip prostheses. Wear-induced loosening vs. loosening-induced wear ± a review', Acta Orthop Scand, 65, 361±371. Muller SD, Green SM, McCaskie AW (2002), `The dynamic volume changes of polymerising polymethyl methacrylate bone cement', Acta Orthop Scand, 73, 684± 687. Mulroy RD, Harris WH (1990), `The effect of improved cementing techniques on component loosening in total hip replacement', an 11-year radiographic review', J Bone Joint Surg, 72-B, 757±760. Murphy BP, Prendergast PJ (2002), `The relationship between stress, porosity, and nonlinear damage accumulation in acrylic bone cement', J Biomed Mater Res, 59(4), 646±654. Murray W (1984), `Use of antibiotic-containing bone cement', Clin Orthop, 190, 89±95. Niki Y, Matsumoto H, Otani T, Tomatsu T, Toyama Y (2007), `How much sterile saline should be used for efficient lavage during total knee arthroplasty? Effects of pulse lavage irrigation on removal of bone and cement debris', J Arthroplasty, 22, 95±99. Nivbrant B, Karrholm J, Rohrl S, Hassander H, Wesslen B (2001), `Bone cement with reduced proportion of monomer in total hip arthroplasty: preclinical evaluation and randomized study of 47 cases with 5 years' follow-up', Acta Orthop Scand, 72, 572±584. Noble PC, Swarts E (1983), `Penetration of acrylic bone cements into cancellous bone', Acta Orthop Scand, 54, 566. Norwegian Arthroplasty Register (2006), http://www.haukeland.no/nrl/. Norton MR, Eyres KS (2000), `Irrigation and suction technique to ensure reliable cement © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

249

penetration for total knee arthroplasty', J Arthroplasty, 15, 468±474. Panjabi MM, Goel VK, Drinker H, Wong J, Kamire G, Walter SD (1983), `Effect of pressurization on methylmethacrylate±bone interdigitation: an in vitro study of canine femora', J Biomech, 16, 473±480. Parsch D, Bruesch SJ (2005), `Bone preparation: acetabular', in Bruesch S and Malchau H, The Well-cemented Total Hip Arthroplasty. Theory and Practice, Berlin, Heidelberg, Springer, 141±144. Parvizi J, Holiday AD, Ereth MH, Lewallen DG (1999), `The Frank Stinchfield Award. Sudden death during primary hip arthropalsty', Clin Orthop, 369, 39±48. Penner MJ, Marsri BA, Duncan CP (1996), `Elution characteristics of Vancomycin and Tobramycin combined in acrylic bone±cement', J Arthroplasty, 11, 939±944. Phillips AM, Goddard NJ, Tomlinson JE (1996), `Current techniques in total knee replacement: results of a national survey', Ann Coll Surg Engl, 78, 515±520. Ranawat CS, Peters LE, Umlas ME (1997), `Fixation of the acetabular component', Clin Orthop, 344, 207±215. Ritter MA, Herbs SA, Keating M, Faris PM (1994), `Radiolucency at the bone-cement interface in total knee replacement', J Bone Joint Surg, 76-A, 60±65. Ritter MA, Zhou H, Keating CM, Keating EM, Faris PM, Meding JB, Berend ME (1999), `Radiological factors influencing femoral and acetabular failure in cemented Charnley total hip arthroplasty', J Bone Joint Surg, 81-B, 982±986. Robertsson O, Knutson K, Lewold S, Lidgren L (2001), `The Swedish Knee Arthroplasty Register 1975±1997. An update with special emphasis on 41,223 knees operated on in 1988±1997', Acta Orthop Scand, 72, 503±513. Rudigier J, Scheuermann H, Kotterbach B, Ritter G (1981), `Release and diffusion of methyl methacrylic monomers after the implantation of self curing bone cements. Study on laboratory specimens and animal experiments', Unfallchirurgie, 7, 132± 137. Ryd L, Albrektsson BE, Carlsson L, Dansgard F, Herberts P, Lindstrand A, Regner L, Toksvig-Larsen S (1995), `Roentgen stereophotogrammetric analysis as a predictor of mechanical loosening of knee prostheses', J Bone Joint Surg, 77-B, 377±383. Sabokbar A, Fujikawa Y, Murray DW, Athanasou NA (1997), `Radio-opaque agents in bone cement increase bone resorption', J Bone Joint Surg, 79-B, 129±134. Sabokbar A, Athanasou N, Murray D (2001), `Osteolysis induced by radiopaque agents', in: Walenkamp GHIM, Murray DW (eds), Bone Cements and Bone Cement Techniques, Berlin, Heidelberg, Springer, 149±161. SanzeÂn L, Walder M (1988), `Antibiotic resistance of coagulase-negative staphylococci in an orthopaedic department', J Hosp Inf, 12, 103±108. Schelling K, Breusch SJ (2001), `Efficacy of a new prepacked vacuum mixing system with Palamed G bone cement', in Walenkamp GHIM, Murray DW (Eds), Bone Cement and Cementing Techniques, Berlin, Heidelberg, Springer, 97±107. Schlegel UJ, Sturm M, Ewerbeck V, Breusch S (2004), `Efficacy of vacuum mixing systems in reducing methylmethacrylate fume exposure. Comparison of 7 different vacuum mixing devices and open bowl mixing', Acta Orthop Scand, 75, 559±566. Schmalzried TP, Kwong LM, Jasty M, Sedlacek RC, Haire TC, O'Connor DO, Bragdon CR, Kabo JM, Malcolm AJ, Harris WH (1992), `The mechanism of loosening of cemented acetabular components in total hip arthroplasty, Analysis of specimens retrieved at autopsy', Clin Orthop, 274, 60±78. Schreurs BW, Spierings PT, Huiskes R, Slooff TJ (1988), `Effects of preparation techniques on the porosity of acrylic cements', Acta Orthop Scand, 59, 403±409. Shelley P, Wroblewski BM (1988), `Socket design and cement pressurization in the © 2008, Woodhead Publishing Limited

250

Joint replacement technology

Charnley low-friction arthroplasty', J Bone Joint Surg, 70-B, 358±363. Sih G, Bernam A (1980), `Fracture toughness concept applied to methyl methacrylate', J Biomed Mater Res, 14, 311±324. Song Y, Goodman S, Jaffe R (1994), `An in-vitro study of femoral intramedullary pressures during hip replacement using modern cement techniques', Clin Orthop, 302, 297±304. Stannage K, Shakespeare D, Bulsara M (2003), `Suction technique to improve cement penetration under the tibial component in total knee arthroplasty', Knee, 10, 67±73. Stocks GW, Freeman MA, Evans SJ (1995), `Acetabular cup migration. Prediction of aseptic loosening', J Bone Joint Surg, 77-B, 853±861. Swedish Hip Arthroplasty Register (2005), Annual Report, 2005, (http:// www.jru.orthop.gu.se). Toksvig-Larsen S, Franzen H, Ryd L (1991), `Cement interface temperature in hip arthroplasty,' Acta Orthop Scand, 62, 102±105. Tolo ET, Wright JM, Bostrom MP-G, Pellicci P, Salvati EA (1998), `The effect of two different types of distal centralizers on the cement mantle thickness and stem alignment in total hip arthroplasty', AAOS Annual Meeting, SE053, March, 19, New Orleans, Louisiana. Tremblay GR, Miller JE, Burke DL, Ahmed A, Krause W, Keleby LC (1979), `Improved fixation of acrylic cement to cancellous bone by pressure injection: an in vivo experimental study', Proceedings of the 25th Orthopedic Research Society, February, 20±22, San Francisco, 67. Verdonschot N (1995), `Biomechanical failure scenarios for cemented total hip replacement', PhD Thesis, Katholieke University Nijmegen. Wahlig H, Dingeldein E (1980), `Antibiotics and bone cements: experimental and clinical observations', Acta Orthop Scand, 51, 49±56. Walenkamp GHIM, Murray DW (2001), Bone Cement and Cementing Techniques, Berlin, Heidelberg, Springer. Walker PS, Soudry M, Ewald FC, McVickar H (1984), `Control of cement penetration in total knee arthroplasty', Clin Orthop, 185, 155±164. Wang JS (2005), `Mixing: The benefit of vacuum mixing', In Breusch S, Malchau H (Eds), The Well-cemented Total Hip Arthroplasty. Theory and Practice, Berlin, Heidelberg, Springer-Verlag, 107±112. Wang JS, Kjellson F (2001), `Bone cement porosity in vacuum mixing system', In Walenkamp GHIM, Murray DW (Eds), Bone Cements and Cementing Technique, Berlin, Heidelberg, Springer, 81±95. Wang JS, FranzeÂn H, Jonsson E, Lidgren L (1993), `Porosity of bone cement reduced by mixing and collecting under vacuum', Acta Orthop Scand, 64, 143±146. Wang JS, Goodman S, FranzeÂn H, Aspenberg P, Lidgren L (1994), `The effects of vacuum mixing on the microscopic homogenicity of bone cement', Eu J Exper Musculoskeletal Res, 2, 159±165. Wang JS, Toksvig-Larsen S, MuÈller-Wille P, FranzeÂn H (1996), `Is there any difference between vacuum mixing systems in reducing bone cement porosity?', J Biomed Mater Res (Appl Biomaterials), 33, 115±119. Wang JS, Aspenberg P, Goodman S, Lidgren L (1998), `Interface porosity in cemented implants in vitro study', 8th European Research Orthopedics Society Meeting, May, 7±10, Amsterdam, The Netherlands, 2. Wang JS, FranzeÂn H, Lidgren L (1999) `Interface gap implantation of a cemented femoral stem in pigs', Acta Orthop Scand, 70, 229±233. Wang JS, Taylor M, Flivik G, Lidgren L (2003), `Factors affecting the static shear © 2008, Woodhead Publishing Limited

Bone cement fixation: acrylic cements

251

strength of the prosthetic stem±bone cement interface', J Mater Sci Med, 53, 55±61. Weiss RJ, Heisel C, Breusch SJ (2003), `Patellar component stability improves with pulsatile lavage in total knee arthroplasty', Int Orthop, 27, 18±21. Wenda K, Scheuermann H, Weitzel E, Rudigier J (1988), `Pharmacokinetics of methylmethacrylate monomer during total hip replacement in man', Arch Ortop Trauma Surg, 107, 316±321. Wilkinson JM, Eveleigh R, Hamer AJ, Milne A, Miles AW, Stockely I (2000), `Effect of mixing technique on the properties of acrylic bone cement', J Arthroplasty, 15, 663±667. Wimhurst J, Brooks R, Rushton N (2001), `The effects of particulate bone cements at the bone±implant interface', J Bone Joint Surg, 83-B, 88±92. Wixson RL, Lautenschlager EP, Novak MA (1987), `Vacuum mixing of acrylic bone cement', J Arthroplasty, 2, 141±149. Yau WP, Ng TP, Chiu KY, Poon KC, Ho WY, Luk DK (2001), `The performance of three vacuum mixing cement guns ± a comparison of the fatigue properties of Simplex P cement', International Orthopaedics, 25, 290±293.

© 2008, Woodhead Publishing Limited

11

Bone±cement fixation: glass±ionomer cements P V H A T T O N , V R K E A R N S and I M B R O O K , University of Sheffield, UK

11.1

Introduction

Glass±ionomer cements (GICs) have been employed extensively in the repair of tooth tissue since the 1970s, and this long history suggests that they are among the most biocompatible dental materials available. Their apparent safety and history of good biocompatibility led scientists and clinicians to consider them for wider surgical uses in the 1980s and 1990s, and much of this early work has been reviewed (Brook and Hatton, 1998; Kenny and Buggy, 2003; Hatton et al., 2006). This chapter in many ways represents the first attempt to connect the data in published studies and reviews to our current technical knowledge of GICs, their setting chemistry and structure±property relationships.

11.2

Structure and properties of glass±ionomer cements

Conventional GICs are formed from the combination of high molecular weight polymeric acids (e.g. polyacrylic acid), a basic fluoroaluminsilicate glass powder and water. Many compositions include tartaric acid to extend the working time. The properties of GICs result from these components and their setting reaction, surface chemistry, physical structure and bulk composition. Set GICs may be described as composites with inorganic glass particles set in a relatively insoluble hydrogel matrix (see Fig. 11.1). Freshly mixed, unset GIC is able to chemically bond to both bone (apatite) and metals (McLean, 1988; Wilson and McLean, 1988). This is advantageous as it means that fixation is not achieved with mechanical interlocking alone. In simple terms, GICs set in stages. First, carboxylic acid residues on the polymeric acid ionise in the presence of water. Protons then react with the surface of the basic glass particles to liberate cations. Specific ions such as Ca2+ and Al3+ are then able to crosslink the ionised carboxylic acid residues, setting the cement. Although at reduced concentrations relative to the parent ionomer glasses (Hatton and Brook, 1992a), many of the component ions remain very mobile long after initial setting, leading to the © 2008, Woodhead Publishing Limited

Bone±cement fixation: glass±ionomer cements

253

11.1 Transmission electron photomicrographs of (a) a G338 glass particle in a set glass±ionomer cement and (b) a glass particle in set Ketac-cem. Field width approximately 5 m.

potential for complex `curing' and leaching of ions. The setting reaction is not exothermic, unlike acrylics. This is highly beneficial as it does not cause thermal damage to tissues at the site of implant, or to any temperature-sensitive drugs that may be incorporated into the matrix phase of the cement. In addition, there is no significant shrinkage of the cement on setting (Wood and Hill, 1991a). The mechanical properties of GICs are generally considered inferior to those of acrylic cements, but are sufficient for low to intermediate load-bearing applications with some potential for improvements with the use of different components. The mechanical properties of the GIC may be controlled to a degree by varying the volume fraction of the glass and hydrogel phases, enabling some properties to be matched with those of the surrounding bone.

11.3

Biological evaluation

11.3.1 Ion release and bioactivity The osteoconductivity exhibited by specific GICs is of particular interest. It has been suggested that this is due to ion exchange with the biological environment (Brook et al., 1991a, Wood and Hill, 1991a, Hatton and Brook, 1992b, Hatton et al., 2006). The ability of the material surface to bind certain biological factors that may recruit and regulate osteogenic cells could also assist the formation of a more stable bone±implant interface and thus improve the potential for clinical success. Immunohistochemical studies of implanted GIC have shown close association of the non-collagenous extracellular matrix proteins of bone (osteopontin, fibronectin and tenascin) with the GIC surface (Carter et al., 1991; Johal et al., 1996). These factors, which are believed to play an important role in ontogenesis and the osseointegration of biomaterials (Weiss and Reddi, 1981; Clark et al., 1982; Mackie et al., 1987; Carter et al., 1991; Bagambisa et al., 1993, 1994) together with the hydrophilic surface of GIC, may explain the osteoconductive properties of implanted GIC (Jonck et al., 1989a,b; Brook et al., 1991a,b, 1992; Doherty, 1991; Nicholson et al., 1991; Hatton and Brook, 1992b; Sasanaluckit et al., 1993; Hill et al., 1995; Johal et al., 1995). © 2008, Woodhead Publishing Limited

254

Joint replacement technology

The bulk composition of GIC acts as a reservoir for ion release (Nicholson et al., 1991; Wood and Hill, 1991b; Sasanaluckit et al., 1993; Hill et al., 1995; Johal et al., 1995). As mentioned previously, certain ions released from the glass particles during the gelation process remain mobile once setting is complete. Studies have reported the presence of such ions in the matrix of the cement (Hatton and Brook, 1992a) and in adjacent bone (Hatton and Brook, 1992b), with exchange of ions taking place with the (aqueous) environment (McLean, 1988; El Mallakh and Sarkar, 1990, Forsten, 1991). Glass composition (Johal et al., 1995) determines ion release, as well as the biochemical environment of the implant bed (Devlin et al., 1994). Fluoride ion release from GIC has been comprehensively reported, although the majority of studies have related to the use of GIC in dental applications (Wilson and McLean, 1988; El Mallakh and Sarkar, 1990; Forsten, 1991). Even though caution must be exercised when interpreting the results, due in the main to poor standardisation and incomplete reporting of methods, it is clear that relatively large quantities of fluoride are released from GICs for periods of up to one year. It was originally proposed that fluoride release was the most significant factor affecting biocompatibility of glass±ionomers. Whereas the absence of fluoride has been reported to result in the least in vitro toxicity (although this material contained no phosphate, complicating interpretation of results) (Brook et al., 1991a), it also produced the lowest osteoconductivity and integration in vivo (Brook et al., 1991b, Johal et al., 1995). The effect of fluoride ions appears to be dose-dependent. Although high fluoride concentrations result in enzyme inhibition in vitro, bone-forming cells exhibit increased proliferation and alkaline phosphatise activity in vivo (Farley et al., 1983; Lundy et al., 1986; Turner et al., 1989; Brook et al., 1991b). Fluoride is also used to treat bone resorption in patients with osteoporosis (Pak et al., 1989; Sùgaard et al., 1995), owing to its ability to increase the density of trabecular bone. Studies have reported a greater volume of bone formation associated with GIC than with more inert ceramic bone substitutes and, more recently, related increased fluoride release with bone formation (Brook et al., 1991b; Johal et al., 1995). It is speculated that this is because fluoride release from GIC during bone formation results in mineralisation containing fluorapatite, which is more resistant to resorption. The release of other ions from GIC is less well reported, although it has been proposed that ion release is a major factor in the bioactivity of different GICs (Brook et al., 1991a, 1992; Doherty, 1991; Nicholson et al., 1991; Hatton and Brook, 1992b; Sasanaluckit et al., 1993). Ion exchange with the tissues of the implant bed has been confirmed to be a significant determinant in the bioactivity of a GIC (Devlin et al., 1994; Johal et al., 1995). Aluminium ion release is particularly important and plays a somewhat controversial role (Nicholson et al., 1991), particularly following reports of four cases of post-otoneurosurgery aluminium encephalopathy and deaths among patients treated with a glass± © 2008, Woodhead Publishing Limited

Bone±cement fixation: glass±ionomer cements

255

ionomer bone cement (Renard et al., 1994; Reusche et al., 2001). Data published to date suggest that aluminium leaching occurs only in the initial period following setting (Crisp et al., 1980; Wilson and McLean, 1988). Increased aluminium ion content and release result in decreased biocompatibility in vitro (Devlin et al., 1994) and may be a more significant factor in biocompatibility than fluoride, although it may have a similarly complex effect. No reports have identified aluminium ions as the sole cause of in vitro cytotoxicity. Low concentrations of aluminium ions have been shown to stimulate the proliferation of osteoblasts in vitro and new bone formation (Quarles et al., 1990; Meyer et al., 1993). Additionally, aluminium particles have been observed inside cultured osteoblasts cells without any detrimental effect on the cells or impairment of bone formation (Blumenthal and Posner, 1984; Hatton and Brook, 1992b; Szulczewski et al., 1993). Aluminium ion release increased the amount of osteoid formation in vivo but interferes with the early stages of bone mineralisation (Blumenthal and Posner, 1984; Quarles, 1991; Meyer et al., 1993), resulting in a reduction of this process. Collagen synthesis is also inhibited (Goodman, 1985; Goodman and O'Connor, 1991). Aluminium also mediates mobilisation of calcium from bone by a cell-independent mechanism. Further research is required into the role of metal ions on osteogenesis. Calcium and phosphate ions may be expected to enhance this process, whereas the effect of other ions is less clear. It is likely that the combination of various factors, particularly the combination and relative concentration of ionic species and their interaction with each other and the biological environment, will result in different outcomes (El Mallakh and Sarkar, 1990; Forsten, 1991; Lau et al., 1991; Nicholson et al., 1991; Devlin et al., 1994; Johal et al., 1995).

11.3.2 In vitro evaluation Glass±ionomers have been tested extensively in vitro and interpretation of these studies demonstrates that these materials should be classed as `bioactive'. Consideration of the results should therefore take into account that bioactive materials often perform less well than more inert materials in vitro (Gross et al., 1987). Furthermore, the in vitro model should attempt to replicate the clinical situation as far as possible. Many studies have evaluated GICs and other bone cements in terms of the migration of osteoblasts onto their surfaces. Appropriate cell source and culture conditions can also be manipulated to induce formation of a bone-like tissue in vitro. Standard cytotoxicity tests involving a variety of cell types (osteoblasts, osteoclasts, fibroblasts, neonate rat calvaria) have been carried out. Positive interactions between certain types of GIC and bone cells have been reported (Brook et al., 1991a, 1992; Doherty, 1991; Meyer et al., 1993; Szulczewski et al., 1993). Cultured cells are particularly sensitive to wet cements; studies in which wet cements were placed in direct contact with neonate rat calvaria resulted in cell death (Brook et al., 1991a). This makes this © 2008, Woodhead Publishing Limited

256

Joint replacement technology

method of study unsuitable for investigating unset cements. Any observed toxicity of set cements has been attributed to the presence of a toxic leachate or the rough surface of the cement. The effect of fluoride, aluminium ion leaching has been discussed in detail above. The low pH of GICs (due to release of protons) while setting and maturing has been proposed as a cause of cyto- and neurotoxicity (Gross et al., 1987; Loescher et al., 1994a; Brook and Hatton, 1998). In summary, the mechanisms responsible for in vitro cytotoxicity are complex and may be unrelated to their in vivo and clinical performance. It might in the future be beneficial to evaluate GICs in vitro following the pre-treatments described for bioactive glasses, but no detailed studies that add substantially to the knowledge reviewed here have been reported.

11.3.3 In vivo evaluation In vivo evaluation allows more meaningful testing of GICs than in vitro methods. Encouraging results have been reported from testing of set cements. In vivo osteoconduction has been widely reported (Jonck et al., 1989a,b; Brook et al., 1991b; Johal et al., 1995). The first preclinical studies into the use of GIC for orthopaedic surgery involved implantation into baboon tibia and gave promising results (Jonck et al., 1989a,b). In one study, GICs with certain compositions were found to support new bone formation six weeks following implantation in rat femora, with the tissue stable over the course of one year. Another important finding of this study was the demonstration that in vitro studies are not always a reliable predictor of the performance of GICs in vivo, as one composition (fluoride- and phosphate-free) that was considered to be biocompatible failed to achieve osseo-integration. Direct bone±cement contact was observed only with fluoride-containing glasses, with the formation of a stable interface (Hatton and Brook, 1992b). The GIC±bone interface morphology varies, ranging from a lamina limitans-like interface shown in Fig. 11.2 (Brook et al., 1991a,b), which mimics the normal boundary of bone with osteocyte lacunae (Van Blitterswijk et al., 1985) to an interdigitation of collagen fibres with the GIC (Brook et al., 1992). The latter of these represents a more bioactive bond, reported on specimens that have not been distorted by decalcification (Jarcho et al., 1977; Kotani et al., 1991). Once the initial post-implantation inflammatory response to a material, which is inevitable, has subsided, the medium-term biological response can be evaluated. GICs demonstrate their osteoconductive potential, unlike poly(methylmethacrylate) (PMMA) cements. Evidence of the bioactive properties of GICs was reported (Fischer, 1986; Carter et al., 1997); no incidence of fibrous encapsulation was observed. Unfortunately, the issue of aluminium ion release appeared to influence the results of in vivo implantation. Two independent studies demonstrated that the new bone tissue that formed rapidly on the surface suffered from defective mineralisation, and that this was associated with the © 2008, Woodhead Publishing Limited

Bone±cement fixation: glass±ionomer cements

257

11.2 Transmission electron photomicrograph of interface (arrowed) between new bone tissue (B) and a set glass±ionomer cement with glass particles (G) and matrix (M). Field width approximately 10 m.

localisation of aluminium in the tissue (Blades et al., 1991, Carter et al., 1991). Despite this concern, a study of the use of particulate GICs as a bone allograft expander concluded that there were no adverse effects on bone remodelling or recovery or the performance of the bone allograft. GIC implanted into the tibia of baboons (Jonck et al., 1989b; Brook et al., 1991a) had previously generated favourable results, leading to the limited clinical application in orthopaedics noted above (Jonck and Grobbelaar, 1990). While numerous reports describe the response of bone to set GICs, very few studies investigating the response to freshly mixed GICs have been reported, and in general the less encouraging tissue responses are observed when GICs are placed `wet' during setting. Studies based on diffusion chambers containing a glass±ionomer and bone marrow, implanted into baboon femora for up to three years demonstrated that the cement promoted osteoblastic activity, with no inhibitory effect on bone tissue. The relevance of these results is limited, however, owing to the separation of host and material by the diffusion chamber. Although freshly mixed glass±ionomers placed on the rat femur suggested initial bone bonding, evidence of a pronounced periosteal reaction and sub-periosteal resorption was found at 6 and 12 weeks. New bone formation was, however, observed at 12 weeks (Brook et al., 1991b; Hatton and Brook, 1992b). Additionally, there was a short-term inflammatory response, likely to be the result of either reduction of tissue pH, causing local tissue necrosis, or release of free glass particles and metal ions from unset cement contaminated by tissue fluid or blood. Water has an adverse effect on the setting reaction and the unfavourable in vivo response to unset cements highlights the importance of avoiding excess moisture during surgery.

11.3.4 Clinical evaluation GICs have been used in various surgical applications. Favourable outcomes have been reported when GICs have been used in granule or cement form in orthopaedic cases where conventional care had failed (Jonck and Grobbelaar, © 2008, Woodhead Publishing Limited

258

Joint replacement technology

1990), but it is the opinion of the authors that these materials are not suitable for situations in which the strength of the cement is critical to the outcome. The use of GIC to reinforce osteoporotic femoral heads has been reported to improve the primary stability of dynamic hip screws (McElveen, 1994), although long-term data or additional information on bone mineral density was not reported. GICs have been particularly successful in otological surgery, being used as a cement or formed into prosthetic implants (Babighian, 1992; Ramsden et al., 1992; Geyer and Helms, 1993; Muller et al., 1993, 1994; Babighian et al., 1994). GIC is the highest-performing material in terms of clinical efficacy in ossicular chain reconstruction, where the cement is used to repair bony ossicles in their normal position, and in cementation of cochlear implants (Babighian, 1992; Ramsden et al., 1992; Muller et al., 1993; Babighian et al., 1994). One study reported 167 patients who had been treated with GIC in middle ear surgery (Geyer and Helms, 1990). Another study reported a 94% four-year success rate for 945 instances of GIC ossicular implant placement (Geyer and Helms, 1993). Furthermore, out of 74 cases of posterior canal wall repair using GIC, only 12 cases required revision surgery (Geyer and Helms, 1993). Oral surgical procedures may also involve GIC, in particular as a bone substitute to prevent bone loss following tooth extraction and as a filler for graft donor sites and cyst cavities (Nordenvall, 1992). GIC are also used as a surgical dressing following exposure of teeth prior to orthodontic alignment (Nordenvall, 1992). The successful outcomes following the use of GIC in various surgical procedures led to their application in neuro-otological and skull base surgery and repair of cerebrospinal fluid (CSF) fistulas and skull defects (Ramsden et al., 1992; Helms and Geyer, 1994). At the same time, evaluation of GIC in terms of their biocompatibility with neural tissue was being carried out. Although seemingly biocompatible, clinical data reported that exposure to viscous GIC resulted in a potentially irreversible block in nerve conduction (Loescher et al., 1994a,b). It was already recommended that unset GIC should not come into contact with soft tissue and that it should be placed in a `dry field', but the results of these studies lead to a further restriction ± that GIC bone cement should not be allowed to come into contact with neural tissue. Sadly, these warnings came too late to prevent four cases of post-otoneurosurgery aluminium encephalopathy, resulting in two deaths (Renard et al., 1994). The issues surrounding these cases are complex, but it is likely that the release of polyacid during the setting reaction and disruption of the setting reaction due to exposure to body fluid/blood, led to the release of large quantities of metal ions and glass particles with disastrous results. It is notable that no deaths resulted from operations where the brain was protected from contact with the cement, and it might be concluded that correct surgical technique is essential when using modern `bioactive' medical materials. Overall, the story of the development of glass±ionomer bone cements is a salient reminder of some of the key points of biomaterials science: biocompatibility is application-specific, and the ultimate © 2008, Woodhead Publishing Limited

Bone±cement fixation: glass±ionomer cements

259

behaviour of a medical device in the body is as much related to the expertise and experience of the surgeon as it is to the properties of the biomaterials used.

11.4

Future trends

The incorporation of antibiotics into acrylic cements is associated with a reduction in the incidence of post-operative infection following orthopaedic surgery. Early studies trying to replicate this success using drug-loaded GIC were less successful (Wijnbergen Buijen Van Weelderen et al., 1983), although the release of biologically active chlorhexidine was reported from two dental cements (Ribeiro and Ericson, 1991). The Sheffield group also reported that GICs provided a favourable matrix for the release of potential biologically active molecules such as proteins (Wittwer et al., 1994). Given the promising properties reported here, it appears advisable to revisit GICs for use as a matrix for drug delivery in bone tissue. GICs have some advantages for orthopaedic applications, including their truly adhesive nature and non-exothermic setting reaction, which does not result in shrinkage and may improve the release of incorporated therapeutic compounds. Their mechanical properties are, however, inferior to those of acrylic bone cements, limiting their load-bearing capacity. It is their interaction with the biological environment that is both encouraging and problematic in equal measure. Biological evaluation has provided some encouraging results, for example the so-called bioactive nature of GICs, by which a composition, site and tissue-dependent ion exchange has encouraged an appropriate host response. More worryingly, the release of aluminium has also been associated with a mineral defect in bone tissue following the use of GICs in orthopaedic applications, and in extreme examples the incorrect placement of the cement has been associated with the death of patients. Concerns regarding aluminium release from glass±ionomers have led recently to the development of aluminium-free GICs, where the Al2O3 was substituted entirely with Fe2O3 (Hurrell-Gillingham et al., 2005, 2006). Further work is ongoing on these promising compositions to establish whether or not they represent a practical alternative to the conventional GIC bone cements already used successfully in ear, nose and throat (ENT) surgery. However, a more radical approach to biomaterial design will be required if glass±ionomer bone cements are to be used successfully for the fixation of medical devices in total joint replacement.

11.5

References

Babighian, G. (1992) Use of a glass ionomer cement in otological surgery: a preliminary report. Journal of Laryngology and Otology, 106, 954±959. Babighian, G., Dominguez, M., Pantano, N. & Tomasi, P. (1994) Multichannel cochlear implant: personal experience. L'impianto cocleare multicanale. Nostra esperienza, Acta Otorhinolaryngol Ital. Mar±Apri 14, 107±125. © 2008, Woodhead Publishing Limited

260

Joint replacement technology

Bagambisa, F. B., Joos, U. & Schilli, W. (1993) Mechanisms and structure of the bond between bone and hydroxyapatite ceramics. Journal of Biomedical Materials Research, 27, 1047±1055. Bagambisa, F. B., Kappert, H. F. & Schilli, W. (1994) Cellular and molecular biological events at the implant interface. Journal of Cranio-Maxillo-Facial Surgery, 22, 12±17. Blades M. C., Moore D. P., Revell P. A. & Hill R. (1991) In vivo skeletal response and biomechanical assessment of two novel polyalkenoate cements following femoral implantation in the female New Zealand white rabbit. Journal of Materials Science: Materials in Medicine, 9, 701±706. Blumenthal, N. C. & Posner, A. S. (1984) In vitro model of aluminium-induced osteomalacia: inhibition of hydroxyapatite formation and growth. Calcified Tissue International, 36, 439±441. Brook, I. M. & Hatton, P. V. (1998) Glass±ionomers: bioactive implant materials. Biomaterials, 19, 565±571. Brook, I. M., Craig, G. T. & Lamb, D. J. (1991a) In vitro interaction between primary bone organ cultures, glass±ionomer cements and hydroxyapatite/tricalcium phosphate ceramics. Biomaterials, 12, 179±186. Brook, I. M., Craig, G. T. & Lamb, D. J. (1991b) Initial in-vivo evaluation of glass± ionomer cements for use as alveolar bone substitutes. Clinical Materials, 7, 295± 300. Brook, I. M., Craig, G. T., Hatton, P. V. & Jonck, L. M. (1992) Bone cell interactions with a granular glass±ionomer bone substitute material: in vivo and in vitro culture models. Biomaterials, 13, 721±725. Carter, D. H., Sloan, P. & Aaron, J. E. (1991) Immunolocalization of collagen Types I and III, tenascin, and fibronectin in intramembranous bone. Journal of Histochemistry and Cytochemistry, 39, 599±606. Carter, D. H., Sloan, P., Brook, I. M. & Hatton, P. V. (1997) Role of aluminium in the integration of ionomeric (glass polyalkenoate) bone substitutes. Biomaterials, 18, 459±466. Clark, R. A. F., Lanigan, J. M., Dellapelle, P., Manseau, E., Dvorak, H. F. & Colvin, R. B. (1982) Fibronectin and fibrin provide a provisional matrix for epidermal cell migration during wound reepithelialization. Journal of Investigative Dermatology, 79, 264±269. Crisp, S., Lewis, B. G. & Wilson, A. D. (1980) Characterization of glass±ionomer cements. 6. A study of erosion and water absorption in both neutral and acidic media. Journal of Dentistry, 8, 68±74. Devlin, A. J., Hatton, P. V., Hill, R., Henn, G., De Barra, E., Craig, G. T. & Brook, I. M. (1994) Initial investigation of ion release from novel polyalkenoate cements related to in vitro biocompatibility. 11th European Conference on Biomaterials. Pisa, Italy, Tipographia Vigo Cursi. Doherty, P. J. (1991) Biocompatibility evaluation of glass ionomer cement using cell culture techniques. Clinical Materials, 7, 335±340. El Mallakh, B. F. & Sarkar, N. K. (1990) Fluoride release from glass±ionomer cements in de-ionized water and artificial saliva. Dental Materials: Official Publication of the Academy of Dental Materials, 6, 118±122. Farley, J. R., Wergedal, J. E. & Baylink, D. J. (1983) Fluoride directly stimulates proliferation and alkaline phosphatase activity of bone-forming cells. Science, 222, 330±332. Fischer, A. A. (1986) Reactions to acrylic bone cement in orthopaedic surgeons and patients. Cutis, 37, 425±426. © 2008, Woodhead Publishing Limited

Bone±cement fixation: glass±ionomer cements

261

Forsten, L. (1991) Fluoride release and uptake by glass ionomers. Scandinavian Journal of Dental Research, 99, 241±245. Geyer, G. & Helms, J. (1990) Reconstructive measures in the middle ear and mastoid using a biocompatible cement ± preliminary clinical experience. In Heimke, E., Soltese, U. & Lee, A. J. C. (Eds) Advances in Biomaterials. Amsterdam, The Netherlands, Elsevier. Geyer, G. & Helms, J. (1993) Ionomer-based bone substitute in otologic surgery. European Archives of Oto-Rhino-Laryngology, 250, 253±256. Goodman, W. (1985) Bone disease and aluminum: pathogenic considerations. American Journal of Kidney Diseases, 6, 330±335. Goodman, W. G. & O'Connor, J. (1991) Aluminum alters calcium influx and efflux from bone in vitro. Kidney International, 39, 602±607. Gross, U., Schmitz, H. J., Kinne, R., Fendler, F. R. & Struntz, V. (1987) Tissue or cell culture versus in vivo testing of surface reactive biomaterials. In Pizzoferrato, A., Ravaglioli, A. & Lee, A. J. C. (Eds) Biomaterials and Clinical Applications. Amsterdam, The Netherlands, Elsevier Science. Hatton, P. V. & Brook, I. M. (1992a) Characterisation of the ultrastructure of glass± ionomer (poly-alkenoate) cement. British Dental Journal, 173, 275±277. Hatton, P. V. & Brook, I. M. (1992b) X-ray microanalysis of bone and implanted bone substitutes. Micron and Microscopica Acta, 23, 363±364. Hatton P.V., Hurrell-Gillingham, K. & Brook I. M. (2006) Biocompatibility of glass± ionomer bone cements. Journal of Dentistry, 34, 598±601. Helms, J. & Geyer, G. (1994) Closure of the petrous apex of the temporal bone with ionomeric cement following translabyrinthine removal of an acoustic neuroma. Journal of Laryngology and Otology, 108, 202±205. Hill, R., Hatton, P. V. & Brook, I. (1995) Factors influencing the biocompatibility of glass polyalkenoate (ionomer) cements with bone tissue. In Ravaglioli, A. (Ed.) Proceedings of the Ceramics Cells and Tissues. Faenza, Italy, Gruppo Editoriale. Hurrell-Gillingham, K., Reaney, I. M., Brook, I. & Hatton, P. V. (2005) Novel Fe2O3containing glass ionomer cements: Glass characterisation. Key Engineering Materials, 284±286, 799±802. Hurrell-Gillingham, K., Reaney, I. M., Brook, I. & Hatton, P. V. (2006) In vitro biocompatibility of a novel Fe2O3 based glass ionomer cement. Journal of Dentistry, 34, 533±538. Jarcho, M., Kay, J. F., Gumaer, K. I., Doremus, R. H. & Drobeck, H. P. (1977) Tissue, cellular and subcellular events at a bone±ceramic hydroxyapatite interface. Journal of Bioengineering, 1, 79±92. Johal, K. K., Craig, G. T., Devlin, A. J., Brook, I. M. & Hill, R. (1995) In vivo response of ionomeric cements: effect of glass composition, increasing soda or calcium fluoride content. Journal of Materials Science: Materials in Medicine, 6, 690±694. Johal, K., Carter, D. H., Hatton, P. V., Sloan, P. & Brook, I. (1996) The interaction of bone extracellular matrix proteins with ionomeric implants increasing in sodium content. Journal of Dental Research, 75, 1152. Jonck, L. M. & Grobbelaar, C. J. (1990) Ionos bone cement (glass±ionomer): an experimental and clinical evaluation in joint replacement. Clinical Materials, 6, 323±359. Jonck, L. M., Grobbelaar, C. J. & Strating, H. (1989a) The biocompatibility of glass± ionomer cement in joint replacement: bulk testing. Clinical Materials, 4, 85±107. Jonck, L. M., Grobbelaar, C. J. & Strating, H. (1989b) Biological evaluation of glass± ionomer cement (Ketac-0) as an interface material in total joint replacement. A © 2008, Woodhead Publishing Limited

262

Joint replacement technology

screening test. Clinical Materials, 4, 201±224. Kenny, S. M. & Buggy, M. (2003) Bone cements and fillers: a review. Journal of Materials Science: Materials in Medicine, 14, 923±938. Kotani, S., Fujita, Y., Kitsugi, T., Nakamura, T., Yamamuro, T., Ohtsuki, C. & Kokubo, T. (1991) Bone bonding mechanism of tricalcium phosphate. Journal of Biomedical Materials Research, 25, 1303±1315. Lau, K. H. W., Yoo, A. & Wang, S. P. (1991) Aluminum stimulates the proliferation and differentiation of osteoblasts in vitro by a mechanism that is different from fluoride. Molecular and Cellular Biochemistry, 105, 93±105. Loescher, A. R., Robinson, P. P. & Brook, I. M. (1994a) The effects of implanted ionomeric and acrylic bone cements on peripheral nerve function. Journal of Materials Science: Materials in Medicine, 5, 108±112. Loescher, A. R., Robinson, P. P. & Brook, I. M. (1994b) The immediate effects of ionomeric and acrylic bone cements on peripheral nerve function. Journal of Materials Science: Materials in Medicine, 5, 551±556. Lundy, M. W., Farley, J. R. & Baylink, D. J. (1986) Characterization of a rapidly responding animal model for fluoride-stimulated bone formation. Bone, 7, 289±293. Mackie, E. J., Thesleff, I. & Chiquet-Ehrismann, R. (1987) Tenascin is associated with chondrogenic and osteogenic differentiation in vivo and promotes chondrogenesis in vitro. Journal of Cell Biology, 105, 2569±2579. McElveen Jr, J. T. (1994) Ossiculoplasty with polymaleinate ionomeric prostheses. Otolaryngologic Clinics of North America, 27, 777±784. McLean, J. W. (1988) Glass±ionomer cements. British Dental Journal, 164, 293±300. Meyer, U., Szulczewski, D. H., Barckhaus, R. H., Atkinson, M. & Jones, D. B. (1993) Biological evaluation of an ionomeric bone cement by osteoblast cell culture methods. Biomaterials, 14, 917±924. Muller, J., Geyer, G. & Helms, J. (1993) Ionomer-based cement in cochlear implant surgery. Laryngo-Rhino-Otologie, 72, 36±38. Muller, J., Geyer, G. & Helms, J. (1994) Good audiological results by reconstruction of defects of the incudo-stapedial joint in the middle ear by reconstructing the ossicles in their normal position. Laryngo-Rhino-Otologie, 73, 160±163. Nicholson, J. W., Braybrook, J. H. & Wasson, E. A. (1991) The biocompatibility of glass±poly(alkenoate) (glass±ionomer) cements: a review. Journal of biomaterials science. Polymer edition, 2, 277±285. Nordenvall, K. J. (1992) Glass ionomer cement used as surgical dressing after radical surgical exposure of impacted teeth. Swedish Dental Journal, 16, 87±92. Pak, C. Y. C., Sakhaee, K., Zerwekh, J. E., Parcel, C., Peterson, R. & Johnson, K. (1989) Safe and effective treatment of osteoporosis with intermittent slow release sodium fluoride: augmentation of vertebral bone mass and inhibition of fractures. Journal of Clinical Endocrinology and Metabolism, 68, 150±159. Quarles, L. D. (1991) Paradoxical toxic and trophic osseous actions of aluminum: potential explanations. Mineral and Electrolyte Metabolism, 17, 233±239. Quarles, L. D., Murphy, G., Vogler, J. B. & Drezner, M. K. (1990) Aluminum-induced neo-osteogenesis: a generalized process affecting trabecular networking in the axial skeleton. Journal of Bone and Mineral Research, 5, 625±635. Ramsden, R. T., Herdman, R. C. D. & Lye, R. H. (1992) Ionomeric bone cement in neuro-otological surgery. Journal of Laryngology and Otology, 106, 949±953. Renard, J. L., Felten, D. & Bequet, D. (1994) Post-otoneurosurgery aluminium encephalopathy [15]. Lancet, 344, 63±64. Reusche, E., Pilz, P., Oberascher, G., Lindner, B., Egensperger, R., Gloeckner, K., © 2008, Woodhead Publishing Limited

Bone±cement fixation: glass±ionomer cements

263

Trinka, E. & Iglseder, B. (2001) Subacute fatal aluminum encephalopathy after reconstructive otoneurosurgery: a case report. Human Pathology, 32, 1136±1140. Ribeiro, J. & Ericson, D. (1991) In vitro antibacterial effect of chlorhexidine added to glass±ionomer cements. Scandinavian Journal of Dental Research, 99, 533±540. Sasanaluckit, P., Albustany, K. R., Doherty, P. J. & Williams, D. F. (1993) Biocompatibility of glass ionomer cements. Biomaterials, 14, 906±916. Sùgaard, C. H., Mosekilde, L., Schwartz, W., Leidig, G., Minne, H. W. & Ziegler, R. (1995) Effects of fluoride on rat vertebral body biomechanical competence and bone mass. Bone, 16, 163±169. Szulczewski, D. H., Meyer, U., Moller, K., Stratmann, U., Doty, S. D. & Jones, D. B. (1993) Characterisation of bovine osteoclasts on an ionomeric cement in vitro. Cells and Materials, 3, 83±92. Turner, R. T., Francis, R., Brown, D., Garand, J., Hannon, K. S. & Bell, N. H. (1989) The effects of fluoride on bone and implant histomorphometry in growing rats. Journal of Bone and Mineral Research, 4, 477±484. Van Blitterswijk, C. A., Grote, J. J. & Kuypers, W. (1985) Bioreactions at the tissue/ hydroxyapatite interface. Biomaterials, 6, 243±251. Weiss, R. E. & Reddi, A. H. (1981) Role of fibronectin in collagenous matrix-induced mesenchymal cell proliferation and differentiation in vivo. Experimental Cell Research, 133, 247±254. Wijnbergen Buijen Van Weelderen, M., Van Mullem, P. J. & De Wijn, J. R. (1983) Release of an anti-inflammatory drug from some dental cements. Biomaterials, 4, 52±54. Wilson, A. D. & Mclean, J. W. (1988) Glass±Ionomer Cement. Chicago, Quintessence Publishing Co. Wittwer, C., Devlin, A. J., Hatton, P. V., Brook, I. M. & Downes, S. (1994). Release of serum proteins and dye from glass±ionomer (polyalkenoate) cements. Journal of Materials Science: Materials in Medicine 5, 108±112. Wood, D. & Hill, R. (1991a) Glass ceramic approach to controlling the properties of a glass±ionomer bone cement. Biomaterials, 12, 164±170. Wood, D. & Hill, R. (1991b) Structure±property relationships in ionomer glasses. Clinical Materials, 7, 301±312.

© 2008, Woodhead Publishing Limited

12

Failure mechanisms in joint replacement

M B U R K E and S G O O D M A N , Stanford University Medical Center, USA

12.1

Introduction

Joint replacement has revolutionized the treatment of arthritic disorders of the hip, knee, shoulder, and other articulations in the body. According to the American Academy of Orthopaedic Surgeons, there were 220 000 primary total hip replacement, 108 000 partial hip replacements, and 418 000 primary total knee replacements performed in the United States in 2003.1 However, during this same time period, there were also 36 000 revision total hip replacements and 33 000 revision total knee replacements. These latter procedures cost $1.66 billion and $1.47 billion in hospital costs respectively. As the general population continues to age, the number of joint replacements will continue to increase. Although the longevity for joint replacements has continued to improve, revision surgeries are increasing in number, and the accompanying burden on the patient, their family, and society is substantial. The most common reasons for revision surgery include implant loosening, wear and periprosthetic osteolysis, infection, recurrent dislocation, malalignment, stiffness, periprosthetic fracture, and implant failure or fracture. In addition, when an implant is placed in bone, there is a redistribution of stresses and subsequent remodeling in the bony bed according to Wolff's law. The rearrangement of the bony architecture in the presence of an implant can have adverse consequences. In this chapter, we will explore some of the mechanisms of failure of joint replacements in current and past usage. We will concentrate on specific etiologies of failure, including implant loosening, wear and the generation of wear debris, dislocation, bony remodeling of the implant bed, and the phenomenon of stress shielding, and failure due to surgical technique. Some of these concepts will also be developed in other related chapters.

12.2

Wear and debris

When materials are in contact with each other and undergo relative motion, wear of the materials occurs. As the reason for performing a joint replacement is to © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

265

obtain pain-free motion and improved function of an articulation, it is not surprising that all implants for total joint replacement undergo wear. The concepts of friction, wear, and lubrication are critical to understanding one of the most important challenges in joint replacement surgery today: the construction of a joint replacement that will last a lifetime while the patient partakes in normal daily activities. The subject of tribology is discussed in Chapter 2, and in other chapters on the different materials for total joint replacement. In this section, the clinical aspects of wear and particle generation will be discussed. Wear often occurs at multiple interfaces of a joint replacement. McKellop has classified wear of joint replacements into four types or modes.2,3 Mode 1 occurs at interfaces that are normally supposed to articulate and undergo wear, for example, the metal ball and polyethylene insert of a hip replacement, or the metal-on-polyethylene articulation of a knee replacement. Mode 2 wear occurs between one normal side of an articulation and another side that should not normally articulate. Mode 2 wear occurs, for example, when a metal ball of a hip replacement burrows through the polyethylene insert to articulate with the metal backing that surrounds the polyethylene. Mode 3 wear occurs when a `third body' particle or other structure, not normally present at that location, interposes itself in a joint articulation. Examples of type 3 wear include wear caused by retained cement, metallic or bone particles, or broken wires that have migrated into a total hip articulation. Mode 4 wear occurs between two surfaces that are not normally meant to undergo wear due to relative motion, for example, socalled `backside wear' of the acetabular polyethylene insert against the metal backing of a modular cementless cup. Another example of mode 4 wear is impingement of the prosthetic femoral neck on the side of the acetabular component. The clinical consequences of wear of joint replacements are threefold.4 First, as wear proceeds, the tolerances between the bearing surfaces become altered. This may lead to changes in the biomechanics, function, and range of motion of the joint (which may be increased or decreased), impingement, subluxation, or dislocation. Second, wear may subsequently alter the physicochemical properties of the bearings, surface coatings, and other treatments. Third, wear of the materials generates particulate debris which may lead to a chronic synovitis, foreign body, and chronic inflammatory reaction, periprosthetic osteolysis, loosening, or pathologic fracture. Prosthetic by-products due to wear may have both local and systemic consequences. With a metal-on-plastic articulation such as a hip joint, progressive wear may compromise the biomechanics of the joint such that sliding occurs in addition to rolling. Patients may complain of the hip suddenly giving way or feeling unstable. Continued wear may lead to impingement of the prosthetic neck on the polyethylene liner, disruption of the locking mechanism of a cementless metalbacked cup and dislodgement of the liner. As the femoral head bores into the cup, the range of motion may become restricted; impingement of the prosthetic © 2008, Woodhead Publishing Limited

266

Joint replacement technology

femoral neck on the side of the cup may cause subluxation or dislocation. With further erosion of material, the head may come to articulate with the metal backing of a cementless shell (mode 2 wear) or pierce the polyethylene completely into the cement mantle of a cemented cup. Although wear may have mechanical consequences, in a metal-onpolyethylene articulation, hundreds of thousands of polyethylene particles around 0.5±5 m in size are generated with every step.3,4 These particles undergo phagocytosis and invoke an adverse foreign body and chronic inflammatory reaction that can have serious local consequences. When the number of wearassociated particles overload local homeostatic mechanisms, a state of disequilibrium occurs.5,6 This leads to upregulation of pro-inflammatory cytokines, chemokines, eicosanoids, the nitric oxide and other metabolic pathways, that stimulate the degradative pathways and inhibit the formative pathways of bone.4,7±13 This tilts the balance in favor of bone destruction, called periprosthetic osteolysis. The cellular processes involved in this reaction will be described in further detail in Chapter 15. Interestingly, in some patients, progressive wear may evoke little or no osteolysis, whereas in others, seemingly minor wear is associated with large osteolytic lesions. There may be a genetic basis for some of these idiosyncratic reactions. In most cases, wear and progressive osteolysis are silent, that is, are asymptomatic until significant wear, synovitis, and loss of bone stock occur.14±16 The eventual symptoms may include those from a chronic synovitis, or due to microfractures or frank breakage of the bone with displacement. Clinically, a chronic synovitis leads to swelling, pain, and warmth of the joint, simulating a joint infection. However, aspiration, microscopic analysis, and culture of the synovial fluid will yield a sterile synovitis containing mostly macrophages, lymphocytes, and wear debris, rather than bacteria and polymorphonuclear leukocytes classically seen in infection. Chronic synovitis may lead to expansion of the joint space, capsular, and ligamentous laxity and complaints of joint instability. This may lead to subluxation or even recurrent dislocation of the joint. Progressive wear and periprosthetic osteolysis undermine the bone stock that forms the foundation of the cementless implant or the surrounding cement mantle. With continued loading, micromotion of the implant within bone results from the lack of support for the prosthesis. This micromotion further compromises the underlying bone, resulting in macromotion and eventually, frank loosening or failure of the implant (Fig. 12.1). Pathologic fractures through areas of particle-induced osteolysis are usually acute painful events, often without prior symptoms. In addition, avulsion of a tendinous insertion may occur through osteolytic bone, such as avulsion of the greater trochanter. The mainstay of treatment of osteolysis is prevention.4,14±16 Careful patient selection, choosing the optimal bearing couple, detailed pre-operative planning © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

267

12.1 These radiographs demonstrate extensive osteolysis and `cement disease' around both the femoral and acetabular components: (a) anteroposterior view; (b) frog lateral view; (c) cross-table lateral view. Both components were revised utilizing impaction grafting of a new press-fit acetabular component and an extensively porous coated long-stemmed revision femoral component. An allograft femoral strut graft was utilized: (d) anteroposterior view and (e) cross-table lateral view.

and meticulous surgical technique are important principles to follow. Periodic clinical and radiographic surveillance is also critical so that early progressive osteolysis can be identified and the patient informed of the different treatment options.4,14±16 The basic principles of treatment include debridement of the debris and synovium, revising the worn articulation and any malaligned components, reconstructing lost bone stock, and stabilizing any fractures as necessary.

12.3

Implant or bone fracture

The incidence of fracture around total joint replacements is increasing. This is due to the increasing prevalence of patients with arthroplasties, expanding indications (to include younger and more active patients), and an increase in revision surgeries. While trauma affects a cross-section of society, arthroplasty patients who sustain fractures have additional confounding variables including osteolysis, diminished bone stock and the presence of surgical implants. Furthermore, medical co-morbidities often influence the decision-making process in this relatively elderly population. © 2008, Woodhead Publishing Limited

268

Joint replacement technology

Patients with an increased risk for periprosthetic fractures include those who have osteoporosis (low bone mass), osteomalacia (pathologic poor bone quality), and those patients who are prone to injury. Therefore, this group includes patients who are elderly, have a history of chronic steroid use, neurologic deficit due to stroke or neuropathy, alcoholism or metabolic bone disease. Fortunately, many fractures can be avoided by meticulous surgical technique and by diligent post-operative care. All total joint patients should be followed closely to evaluate for the presence and progression of periprosthetic osteolytic defects, as these represent a frequent site of fracture. Revision of worn components can minimize these defects and bone stock can be restored. Fractures around total joints can be divided into those that occur intraoperatively vs. postoperatively. Intra-operative fractures are not uncommon and can be influenced by implant design and surgical technique. The incidence is much greater during revision procedures than primary operations, and when using press-fit cementless components rather than cemented ones.

12.3.1 The hip Fractures occur at several key steps during a total hip arthroplasty (THA) including dislocation, broaching, reaming, and impacting press-fit acetabular and femoral components. The anterior bow of the femur or anatomic variability can lead to fracture when the stem and femur do not match well. Fractures can occur around press-fit acetabular shells that are eccentrically or under-reamed. Most surgeons under-ream cementless acetabular components by one or two millimeters based on perceived bone quality. Under-reaming is used to improve the press-fit of a component. This is thought to increase initial stability and bone ingrowth potential. It may be, intuitively, an attractive option for osteoporotic patients; however, unfortunately, this population is prone to fracture. Reaming line-to-line decreases the fracture risk at the potential expense of implant stability. Clearly, under-reaming increases the force necessary to impact an acetabular cup (approx 2000 N for 2 mm under-reaming and 3000 N for 4 mm of under-reaming).17 In a study of cadaveric specimens, Kim et al.17 fractured 18 sockets, and had a clear predominance of fractures with 4 mm of under-reaming. This laboratory data is consistent with Sharkey et al.'s (1999) operative experience in which 13 fractures occurred during seating of the acetabular component.18 In their study, 8 of 13 fractures occurred in hips under-reamed by 2 mm; 3 of which were under-reamed by 3 mm and only 1 of which was underreamed by 1 mm. This effect can be exacerbated during the insertion of elliptical acetabular shells. The monoblock elliptical design has been shown to be an independent risk factor for fracture during impaction.19 Fractures of the femur after total hip replacement are often detected intraoperatively by direct observation, but visualization of acetabular fractures can be obscured by the implant. For similar reasons radiographs clearly show femur © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

269

fractures, while acetabular fractures can be more subtle. Therefore, when intraoperative acetabular fracture is suspected, direct observation should be made by removing the component if necessary, especially if the component is not fully seated or loose. In the post-operative setting, oblique radiographs or a computed tomography (CT) scan with metal suppression can be helpful. Prior healed fractures can leave weakened bone that is prone to re-fracture. Screw holes concentrate stress, thereby predisposing to fractures. For this reason, many surgeons bypass such stress-risers by at least two cortical diameters.

12.3.2 The knee Fractures around a total knee arthroplasty (TKA) can occur in the patella, femur or tibia. Intra-operative fractures are rare, but result from high loads transmitted to the patella, overzealous impaction of the components or from imprecise bone cuts. Fractures occur around stems (as they are points of stress concentration) and are more common in patients who have ipsilateral hip and knee arthroplasties (due to stress concentration). Constrained implants also predispose to fractures via rigid transmission of torsional stresses. Removal of part of the anterior femoral cortex (notching) when performing a TKA may predispose the femur to supracondylar fracture, although this is controversial. This fracture is due to removal of the cortical origin of the trabecular bone in the distal femoral condyles. Therefore, it is important to properly size the femoral component and determine its position in the sagittal plane. The relative risk of fracture, however, has been a matter of debate. In biomechanical studies, the strength of notched femurs decreased in both bending and torsion by 18 and 39% respectively, and when loaded to failure, they resulted in a different fracture pattern from non-notched femurs.20 This effect was exacerbated by osteoporosis (as a function of the polar moment of inertia).21 In several series of patients who sustained supracondylar fractures, a disproportionate number of patients had `notched' femurs (rates of 10±46%).22±26 The relative risk of fracture after femoral notching has recently been evaluated by Ritter et al.27 In their review of 1089 total knee arthroplasties, they found 328 `notched' distal femurs. After an average five-year clinical follow-up, they experienced no fractures within this group. They did, however, have two fractures above non-notched femurs. They did not measure any excess risk of femur fracture after notching.27 Given these studies, we can conclude that the risk of fracture is small after anterior femoral notching, but it remains inadvisable given the ample evidence linking notching with decreased strength, and a preponderance of notched femurs in cohorts of patients who experienced supracondylar fractures. A precise technique is also advisable when cutting the intercondylar notch for a posterior stabilized implant. An imprecise, trapezoidal, or shallow resection © 2008, Woodhead Publishing Limited

270

Joint replacement technology

can cause the square box of the femoral component to act as a wedge when impacted. Excessive force used to seat an ill-fitting component therefore can lead to fracture. Although this fracture is rare, it deserves attention from surgeons and design engineers. Long-stemmed knee components are particularly challenging to implant for several reasons. They are frequently used in revision procedures with patients who have poor bone quality or quantity. In accordance with their design, the stem±cortex interface is a site of stress concentration. Preservation of cortical bone strength is important, hence reaming must be done carefully. Offset stems can improve intramedullary fit in bones with abnormal anatomy, decreasing the potential for fracture. Patellar fractures are the most common periprosthetic fracture around a TKA. These fractures occur due to direct impaction from patient falls or from large tensile forces through porotic bone with prosthesis anchoring holes. Predisposition to fractures has been reported with increased patellar resection, insufficient patellar resection (overstuffing the joint), asymmetric patellar resection or malpositon of the femoral or tibial components Preservation of residual patellar thickness of at least 10 mm as well as protection of its blood supply can help minimize the risk of fracture. Flexion of the femoral component effectively over-stuffs the patello-femoral joint. Internal rotation of either the femoral or tibial component leads to increased risk of pain, mal-tracking, dislocation, and patellar fracture.28 Furthermore, patellar fractures can result from increased joint reactive forces that occur with high knee flexion, or with changes in joint-line position. Patellar fractures have also been attributed to large central pegs rather than three peripheral smaller pegs.28 Compromise of the patellar blood supply can also lead to weakening of the patella. The blood vascular supply to the patella is provided by a peri-patellar anastamosis from the geniculate vessels and the anterior tibial recurrent vessels traversing retrograde through the infrapatellar fat pad. During surgery, the medial contribution is compromised by the medial parapatellar arthrotomy. The anterior±inferior contribution is compromised by resection of the fat pad. Care, therefore must be taken to preserve the remaining lateral blood supply. Commonly a lateral retinacular release is necessary to centralize patellar tracking, which can further compromises the patellar blood supply precipitating the cascade of osteonecrosis, fracture, implant loosening, and failure. Therefore effort should be directed towards optimizing patellar tracking prior to performing a lateral retinacular release. This includes maintaining proper femoral and tibial component external rotation. Some surgeons release the tourniquet prior to performing a lateral release to ensure the extensor mechanism is not entrapped/constrained under the tourniquet. When a lateral release is necessary, the surgeon should attempt to visualize and protect the superior lateral geniculate vessels at the inferior margin of the vastus lateralis.

© 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

271

12.3.3 Classification of periprosthetic fractures There are multiple classification systems for each of the locations where periprosthetic fractures occur. Key principles are reflected in each of the classification systems. They include the position of the fracture relative to the implant's fixation, the remaining bone stock, and the stability of the implant. This has important repercussions on implant retention and fixation options in each of these locations. The Vancouver classification for proximal periprosthetic femur fractures is the most widely utilized system. It classifies the fracture based on location relative to the stem: Above the level of the prosthesis (A), at or just below the tip of the prosthesis (B), or well below the level of the prosthesis (C). Treatment is guided by an evaluation of the remaining bone stock and implant stability. In type A fractures, those of the greater tuberosity are distinguished from those involving the lesser tuberosity. Both tuberosities are sites of insertion of major muscles around the hip. Type B fractures (at the level of the stem) are subdivided based upon implant stability, and bone stock. Subtype B1 is a stable implant with a fracture at or below the level of the stem. Subtype B2 fractures occur with loose stems and adequate bone stock. Subtype B3 fractures occur in association with severe loss of bone stock (either due to osteolysis or comminution). In type C fractures, the fracture occurs well below the level of the prosthesis, rendering the implant unaffected by the fracture. Acetabular fractures around a prosthesis are classified by Peterson and Lewallen based upon implant stability.29 Stable implants within a fractured acetabulum are Type I while Type II fractures render the acetabular shell grossly loose. Fractures about the femoral component of a TKA (supracondylar fractures) have been classified by Lewis and Rorabeck with regard to the degree of fracture displacement and the stability of the femoral component.30 Type I fractures have a stable component with a non-displaced fracture. Type II fractures have a displaced fracture with a stable femoral component. A type III fracture is any pattern that results in an unstable prosthetic component. Tibial fractures are similarly classified based on location (tibial plateau, adjacent to the stem, distal to the prosthesis or at the tibial tubercle), the implant (stable vs. unstable), and timing (intra-operative vs. post-operative). Clearly, each factor has treatment implications. Patellar fractures are integrally related to extensor mechanism function and the stability of the patellar component (in the case of resurfaced patellae). Hozack et al. classified patella fractures based on displacement, extensor mechanism function, distal pole displacement, and failure of prior (non-operative treatment).31

12.3.4 Treatment options Periprosthetic fractures are occasionally treated with activity modification, restricted weight-bearing, immobilization, and close radiographic and clinical © 2008, Woodhead Publishing Limited

272

Joint replacement technology

follow-up. More often, however, they are treated with osteosynthesis and bone grafting to optimize anatomic alignment, and provide sufficient bony stability to allow joint motion and to help restore lost bone stock. Periacetabular fractures often go unnoticed intra-operatively. The fracture site is obscured from the surgeon's vision by the implanted cup and surrounding soft tissues. The only indication that a fracture has occurred may be sudden seating of a tight-fitting component, or the subtle change in pitch heard during impaction. If an intra-operative fracture is suspected, the cup should be removed to inspect the underlying bone. Sharkey et al. demonstrated that initial stability of the cup is critical to overall outcome. Unstable shells fail to gain stability leading eventually to revision surgery.18 Treatment is therefore directed toward obtaining a stable construct in the face of fracture. In their series, three of the four fractures diagnosed on a delayed basis migrated and/or failed clinically. Of the fractures that were initially diagnosed intra-operatively, cup fixation and the fractures were reenforced with acetabular screws. These patients had 6±8 weeks of restricted weight-bearing. The combination of activity restriction and improved cup stability (by placing screws) resulted in improved outcomes (though compromised compared with patients who do not experience fractures). Initial stability must be achieved to optimize patient outcome. Haidukewych et al. reiterated this point with a larger cohort of patients.19 Their incidence of acetabular fracture was 0.4% (21 of 7121 hip arthroplasties). In their cohort of patients, they obtained immediate fixation with 17 components by rim fixation, despite the fracture, but had to exchange four components for multi-holed shells with screws to gain stability. In each of their patients, the fractures healed and the acetabular components performed well. Cup designs were evaluated, and elliptical designs were associated with increased fracture risk. Intra-operative fractures are usually non-displaced, but traumatic postoperative fractures can result from high-energy trauma and may be associated with areas of osteolysis and bearing wear. In this situation stability should be achieved through supplemental acetabular shell screws and, where necessary, plate and screw fixation if major portions of the supporting walls and columns are involved. The application of plates often requires increased soft tissue dissection, blood loss, and operative risk to neurovascular structures. Bony defects can be grafted with structural or morselized bone graft and fragments stabilized with well-described techniques of acetabular column and wall buttress plating. Small wall fractures can be ignored if they do not impact the stability of the component. Large wall fractures should be stabilized with internal fixation with or without bone graft. Column fractures should be plated. Medial wall fractures should be bone grafted and large defects may necessitate the use of an anti-protrusion cage or ring. Periprosthetic fractures around the femoral component are usually treated operatively, because they compromise the stability of the implant. Nevertheless, © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

273

non-operative treatment with activity and weight-bearing modification or with traction and close follow-up for signs of progression or loosening is occasionally appropriate. Operative treatment of a periprosthetic fracture is usually the use of cerclage cables or wires, plates or allograft struts. Severe cases with compromised bone can be treated with proximal femoral replacement or with tumor or customized prostheses. Fractures of the greater trochanter influence hip abductor strength, which is a critical component of gait, function, and hip joint stability. Therefore, fractures that remain non-displaced can be treated with a period of toe-touch weightbearing with close radiographic follow-up. However, if the fracture occurs intraoperatively or if the fracture displaces, rendering the abductor mechanism compromised, the trochanter should be stabilized with wires or a trochanteric plate. If structurally significant osteolysis is present, it should be grafted and the bearing surfaces exchanged. Treatment of femoral shaft fractures around a femoral hip stem depends upon component stability, fracture comminution and remaining bone stock. Vancouver type B1 fractures can be reliably treated with osteosynthesis plates stabilized by screws, cerclage wires/cables. Cortical strut allograft bone can be used to augment bone stock on the anterior and/or lateral surfaces. These fractures reliably heal, but can increase the risk of infection and/or hip instability. Vancouver B2 fractures (fractures around a loose femoral stem), have adequate bone stock for revision surgery. The loose stem must be removed from the proximal fracture fragment (and may require a proximal trochanteric osteotomy). The proximal femur is reconstructed around a new femoral stem that must achieve 5 cm of distal fixation (in the intact distal femoral shaft). The proximal fracture fragments can be stabilized with plates and screws and/or allograft augmentation. These fractures are at increased risk for non-union, malunion and infection (Fig. 12.2). Vancouver B3 fractures occur in femurs with severe osteolysis and/or comminution. The reconstruction requires revision of the femoral stem with allograft struts or a proximal femoral structural allograft. This can jeopardize the attachment of the hip abductors, and hence the stability of the hip. Another alternative is a tumor prosthesis. Owing to the high risk of instability with these constructs, securing the component with a constrained acetabular liner may be advisable. Similar concepts govern the treatment of fractures around a TKA. Function relies upon the restoration of a stable, properly aligned implant. The collateral ligaments, like the hip-abductor mechanism, must be structurally competent. If the fracture is non-displaced and stable, with adequate bone stock, it is reasonable to attempt to treat the fracture in a long-leg cast or functional brace with a 6±12 weeks of protected weight-bearing. If a fracture above a TKA is displaced, rigid internal fixation, enough to allow postoperative knee motion must be obtained with either a blade-plate, condylar screw with side-plate, peri-articular locking plate, or intramedullary © 2008, Woodhead Publishing Limited

274

Joint replacement technology

12.2 These radiographs are of a patient who mis-stepped 4 weeks after right THA for osteoarthritis of the hip. The patient sustained a Vancouver B2 periprosthetic femur fracture around the proximal end, and the prosthesis subsided within the bone (a). The old femoral component was excised, the fracture reduced and stabilized with cerclage wires and an extensively porous coated stem was placed (b).

nail. Intramedullary fixation is attractive because it can be performed with minimal dissection and can be used reliably for posterior cruciate retaining prostheses (as a nail will easily fit through the area between the femoral condyles). If, on the other hand, a posterior-stabilized implant is present, the surgeon must be aware of the variable presence of a pre-drilled hole in the central box created for this contingency. If a hole is not present, one can be created with a carbide drill, but other methods of fixation may be simpler. Fractures that compromise knee stability via avulsion of the collateral ligaments require fixation or revision to stemmed constrained implants. If the origin of the collateral ligaments is disrupted, the fracture requires bone grafting, or if there is soft-tissue interposition, the fracture site should be visualized directly, and rigid internal fixation should be obtained. A loose prosthesis should be removed, and replaced with stemmed components over a fracture that is stabilized and bone grafted as necessary. In the rare case where there is significant comminution, with minimal remaining bone stock, distal femoral allograft around a stemmed distal femoral component or a custom implant can be considered. Occasionally, factors such as the patient's medical condition, the presence of chronic infection or multiple injuries necessitate long-term traction treatment or amputation. Treatment of periprosthetic tibial fractures is dependent upon implant stability and the location of the fracture. Fractures of the tibial plateau with a stable component (type I) are treated with screw(s) and buttress plate as necessary. If these fractures occur intra-operatively, the fracture site and tibial plateau should be offloaded and bypassed by use of a stemmed component. If the fracture results in a loose component or a type II fracture (one that occurs around the stem of the tibial component) the implant should be revised utilizing bone grafting and stemmed components. Type III fractures (those that occur distal to the tibial component) usually do not affect the stability or function of the implant and should be treated non-operatively with casting and/or bracing. Type IV fractures (avulsions of the tibial tubercle) affect the extensor mechanism. © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

275

Small non-displaced avulsions can be treated with extension bracing and activity modification. Fractures that displace or render the extensor mechanism incompetent require internal fixation to re-establish extensor continuity. The treatment of patellar fractures depends upon the stability of the polyethylene component, the function of the extensor mechanism and the remaining bone stock. A stable component may be retained. An extensor mechanism that is non-functional due to a displaced patellar fracture requires internal fixation with a tension band and a new component if it is loose. However, these operations are frequently unsuccessful due to non-union. Partial or complete patellectomy remains a viable option when insufficient bone stock or comminution precludes fixation and re-implantation. In patients who demonstrate failure of the above treatments, the final option is transplantation of an allograft comprising the quadriceps tendon, patella, patellar tendon and tibial tuberosity. Non-operative treatment consisting of six weeks in an extension brace with minimal weightbearing is appropriate only for non-displaced transverse or vertical patella fractures in which the patellar component remains well fixed. Some of these factors cannot be influenced by the patient and/or physician, however, early treatment of osteolysis and proper implant selection can minimize the risk.

12.3.5 Implant fracture Because of the use of modern super-alloys and better prosthesis designs, implant fracture is a problem that has largely been solved. Occasionally a fracture is seen in a prosthesis made with a suboptimal design or poor manufacturing methods. Cast implants with large grain size, and numerous impurities and asperities are predisposed to fracture. Although uncommon with modern implants, the majority of remaining fractures occur in the region of the femoral head and neck and are related to implant selection. The calcar femorale is normally composed of dense, strong bone that supports and protects the hip prosthesis. In the rare case when an arthroplasty is performed in a patient with a deficient calcar (due to tumor, unstable intertrochanteric or subtrochanteric fracture or during revision surgery) a calcar replacing prosthesis, designed to withstand the high stresses of this region should be selected to prevent fatigue fracture. There is also a higher incidence of prosthetic fracture with some modular implants but this has largely been resolved due to better engineering design. The remainder of prosthetic fractures usually occur in ceramic total hip components. This was an unacceptably common mode of failure with first generation ceramic implants. Up to 13.4% of ceramic femoral heads failed due to fracture, hence many companies stopped their sales. However, the newer (third generation) manufacturing techniques of hot isostatic pressing, laser marking and proof testing have lowered the fracture rate to approximately 0.004%. Isostatic pressing helps increase the density of the ceramic to the ideal © 2008, Woodhead Publishing Limited

276

Joint replacement technology

3.98 g/cm, and optimizes the ceramic microstructure to increase its overall strength. Laser marking serial numbers decreases the notch effect. Finally, the components are proof tested past the limits of physiologic load (at least eight times body weight). These manufacturing methods have increased the strength of ceramic implants and decreased their failure rate.32

12.4

Dislocation

Dislocation after THA is the second most common reason for revision hip surgery. Dislocation is associated with increased costs when revision arthroplasty is required.33 The incidence of dislocation after primary arthroplasty has been reported between 0.6% and 7%, but most published reports for primary arthroplasty report 2±4%.34 Revision arthroplasty (compared to primary) is associated with a much higher dislocation rate; 7.4% in a review of the Mayo Clinic experience by Alberton et al.35 Several factors have been associated with increased rates of dislocation, including female sex, patient age over 70 years, or surgery performed for avascular necrosis, fractures, non-union, inflammatory arthropathy, neurologic disorders and alcohol and drug dependence. The risk of dislocation extends over the lifetime of the prosthesis. Berry et al. found the risk of dislocation is 1% at 1 month, 1.9% at 1 year, and continues to increases at a rate of 1% every 5 years to a maximum of 7%.36 Three-quarters of dislocations occur within the first year after surgery, but this long-term follow-up study calls into question the low dislocation rates found in short-term follow-up studies. The most common direction of dislocation is posterior. This occurs when the lower extremity the hip joint in flexion, adduction, and internal rotation. Anterior dislocation occurs much less often, and is due to hip extension with external rotation and abduction. These extreme positions should be avoided post-operatively. Abduction wedges, braces, and knee immobilizers can be used to help reinforce patient compliance. The factors that affect hip stability include the surgical approach, component position, prosthesis head±neck ratios, and abductor muscle function. The compressive and stabilizing force for the hip joint is provided by the abductor muscles. Function is optimized by re-creating the mechanical hip center-of-rotation, with anatomic trochanteric offset and appropriate leg lengths. A greater trochanter that is positioned either too superiorly or medially renders the abductors weak and the hip prone to dislocation. Similarly, fractures of the greater trochanter, avulsion of the gluteus medius tendon, or dysfunction of the central nervous system (such as post-cerebrovascular accident, Parkinson's disease, etc.), or the peripheral nerves (superior gluteal nerve paralysis) can lead to hip instability. As most dislocations occur posteriorly, a posterior approach that compromises the posterior capsule is associated with a higher dislocation rate than an anterior or lateral approach. In a meta-analysis by Kwon et al.37 the shortterm dislocation rates for an anterolateral approach was 0.7%, for lateral 0.43%, © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

277

and 1.01% for the posterior approach. The risk of posterior dislocation can be minimized by performing a meticulous multilayered capsular repair (capsulorrhaphy). This fact was supported by Goldstein et al.,38 and Kwon et al.'s37 meta-analysis showed a relative risk reduction of 8.21 by performing a soft-tissue repair after a posterior approach to the hip. Surgical factors that lead to increased hip stability include increasing femoral head size, restoration of soft-tissue tension, and proper component positioning. Increasing the head±neck ratio is advantageous because it allows a greater degree of motion before primary impingement. In a cadaveric study by Bartz et al.,39 smaller head sizes were shown to decrease the effective hip range of motion prior to impingement and dislocation. Increasing the femoral head size from 22 to 28 mm increased the hip range of motion by 7.6ë prior to dislocation. Hedlundh et al.40 reported a 2.3 times greater recurrent dislocation rate for hips with a 22 mm vs. a 32 mm head. Increasing the head size also improves stability by increasing the distance the femoral component must travel to disengage the acetabular component ± the drop-distance. Unfortunately, increasing the femoral head size also increases the volumetric wear rate because for a given range of motion, increasing the femoral head diameter increases the contact surface. The elevated particulate load can lead to accelerated osteolysis. The ramifications of this, however can be offset to some degree by using ultra-highly crosslinked polyethylene, metal-on-metal or ceramic bearing surfaces. Impingement between the femur and pelvis (external impingement) can also lead to dislocation. Prevention is accomplished through proper placement of the acetabular shell, debriding major osteophytes, and restoring the natural anatomic offset (lateralization of the femur). Restoring offset also increases abductor tension, the crucial factor in creating a stable joint. Increasing abductor tension can be achieved by lateralizing the acetabular component, using a high-offset acetabular liner, increasing the femoral neck length, a high-offset femoral stem or by advancing the abductor mechanism via trochanteric osteotomy. Lateralizing the acetabular component can increase the joint reaction forces by increasing the body-weight moment-arm and hence the work generated by the abductor muscle complex. Lateralization of the hip center increases the torsional stresses on the femoral stem thereby potentially affecting stem stability and longevity. Using a high-offset acetabular liner has the same biomechanical effect, but it also increases the leg-length. Use of an increased neck-length also increases leglengths, offset and abductor tension. High offset femoral components have a lower neck-shaft angle, or a lower neck take-off point, and therefore generate offset without increasing leg-length. The soft-tissue sleeve can be tightened without lengthening the extremity by advancing the greater trochanter, either through an osteotomy, or by advancing the repair in a transtrochanteric approach. Non-union of the trochanter in this approach compromises the abductor integrity and has been shown to lead to a six-fold increased risk of dislocation.41 © 2008, Woodhead Publishing Limited

278

Joint replacement technology

Implant position affects stability. Optimal positioning has been proposed by Barrack et al.42 from computer modelling to be 45ë of cup abduction with 20ë of anteversion. The femoral stem should be placed in 15ë of anteversion.42 Similar studies by D'Lima et al.43 demonstrated that closure of the actabular component to 35ë resulted in markedly decreased motion before impingement or dislocation. Stability could be achieved only with increased acetabular and femoral component anteversion. By contrast, opening the cup to 45ë resulted in markedly increased stability at all positions of femoral and acetabular version. Excessively abducting the acetabular component results in edge loading of the polyethylene and leads to early wear and excessive osteolysis and failure as demonstrated by Schmalzried et al.44 When a hip arthroplasty is performed through a posterior approach, the anteriorly displaced femur places the cup into relative retroversion. The opposite is true for an anterior approach. Aberrant anatomy (i.e. developmental dysplasia of the hip, retroverted acetabulum) can also lead to component malposition. Placement of the components outside of these limits, which can be exacerbated by excessive anteversion of the femoral and acetabular components, leads to impingement and higher dislocation rates. Lipped or oblique liners re-orient the acetabular face to optimize stability. They do not change the offset or leg-length, but highly elevated lips (20ë) can influence the effective version and/or abduction angle, thereby augmenting stability. The trade-off to these components is they can decrease range of motion and increase the chance of neck-liner impingement. Dislocation is best treated by prevention. This includes meticulous patient and implant selection, re-creation of appropriate soft tissue tensions, intra-operative trial reduction with documentation of stability and range of motion, and meticulous surgical repair. Nevertheless, joint instability often may be persistent. The surgeon should determine the direction of dislocation, individual, and relative positioning of the components, and evaluate for neurovascular injury, fracture, or component failure. The surgeon should also evaluate for the presence of infection. The hip can usually be reduced in the emergency room with muscular relaxation and conscious sedation. Occasionally, fluoroscopic guidance can assist with reduction. Should an open reduction be required, the patient and surgeon should be prepared to perform a complete revision, as indicated. After successful closed reduction, an abduction brace should generally be applied for a minimum of six weeks to allow capsular healing. Non-compliant patients can be placed in a hip spica cast. Immobilization (or limitation of motion) with limited weight-bearing for six weeks and abductor strengthening effectively treats two-thirds of dislocations, but is more effective in patients who dislocate within a year of the initial procedure.41 Patients who dislocate more than one year from the initial procedure are more likely to develop chronic instability and require surgical correction. This may reflect the role of bearing surface wear in generating an unstable joint. © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

279

Surgical treatment for chronic instability is highly individualized and should be directed toward the cause of instability. Infections clearly should be treated with debridement, antibiotics and revision of the components (as discussed in Chapter 15, Section 15.2). Malpositioned components should be revised or occasionally treated with either lipped or oblique liners. Any loose or broken components should also be revised. Instability due to polyethylene wear is treated with femoral head and acetabular liner exchange (preferably to a more durable combination). Impingement should be treated by resection of osteophytes, increasing femoral offset and/or exchange of components as the operative findings dictate. Inadequate soft tissue tensions can be improved by increasing head/neck lengths and offset with stems or offset acetabular liners. If the modular components have been optimized and instability persists, a trochanteric advancement can be performed. Woo and Morrey (1982) demonstrated that revision surgery is effective approximately two-thirds of the time.41 In situations of persistent instability, when operative treatment fails or when patient health and physical demands limit surgical options, salvage procedures are indicated. These include placement of a constrained acetabular component, bipolar arthroplasty or girdlestone resection arthroplasty. Constrained devices capture the femoral head to improve stability to the detriment of range of motion and increased shear forces at the acetabular shell±bone interface (Fig. 12.3).

12.3 This patient had a primary THA with decreased femoral offset and increased acetabular anteversion (a and b). The patient sustained multiple hip dislocations despite brace treatment. Placement of a constrained acetabular liner (c) failed after four years due to accelerated wear with failure of the locking ring mechanism. Placement of a lateralizing polyethylene liner and large femoral head with increased neck length rendered the hip stable (d). © 2008, Woodhead Publishing Limited

280

12.5

Joint replacement technology

Stress shielding

Bone is in a constant state of flux. This remodeling process allows bone to react to its environment and stressors. According to Wolff's law, bone is formed and strengthens along lines of mechanical stress. The corollary is that bone devoid of stress, atrophies (like most tissues in the body). Wolff's law is clinically apparent by the formation of osteophytes around an arthritic joint (increased stress causes hypertrophy of the bone) and under rigid internal fixation plates (bone atrophy). It also manifests in osteoporosis of bedridden, non-functional or neurologically impaired patients with atrophic bone. In the setting of THA, stress in the proximal femur is shared by both the host bone and the metal implants. Cemented or rigidly fixed femoral components that have distal fixation and are constructed of stiff material (relative to cortical bone) can be expected to support the majority of the patient's weight, thereby relieving stress on the proximal femur. This causes resorption of proximal femoral bone. By contrast, a less-rigid femoral implant (such as titanium alloy versus cobalt chrome alloy) that gains proximal fixation (proximally porous coated stems) shares more stress with a greater length of the femur. Stress in the proximal femur helps prevent proximal resorption. This helps prevent insufficiency fractures, improves implant stability and facilitates revision surgery by improved bone stock. The process of stress shielding is seen around acetabular, femoral, and total knee implants. Each of these supports bone and shields the bone from stress. In the proximal tibia and distal femur, stress shielding is seen under the articular surfaces, especially in metal-backed components. In the proximal femur, it is seen in the greater and lesser trochanters and along cementless stems that are distally well fixed. It is seen variably around acetabular components that distribute weight-bearing stress. Stress shielding is present to a lesser degree around cemented components, because of the presence of a lower-modulus intermediate material, poly(methylmethacrylate). Prevention or stress shielding is accomplished by selecting implants that allow native bone to support as much of the patient's weight as possible (loadsharing implants). Distal fixation of long stem femoral components should be used only if necessary for implant stability. The use of a collar on the femoral component to distribute load to the proximal femur is theoretically beneficial in mitigating adverse bone remodeling and stress shielding. However, the surgical precision necessary for this to occur is often difficult to achieve. Medical treatment with bisphosphonate medications has been used to preserve bone mass in osteoporotic patients. Bisphosphonates have been shown to preserve bone mass around total joint implants, but they inhibit normal osteoclast function and bone remodelling. This may lead to detrimental effects on the stability of cementless stems and overall durability of the arthroplasty. Furthermore, these medications are not without potential adverse systemic © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

281

effects. Currently, the use of bisphosphonates to prevent stress-shielding after joint replacement is controversial.

12.6

Comment on surgical failure

Clearly, surgical technique is a crucial factor in successful arthroplasty surgery. As stated above, poor surgical technique can lead to fracture (either intraoperative or post-operative), accelerated implant wear, instability, impingement or stiffness, or limb-length discrepancy. Poor technique, in TKA, can result in soft tissue disruption (i.e., patellar tendon avulsion) or laceration of critical structures (medial collateral ligaments, posterior cruciate ligament) or neurovascular structures. It can result in persistent knee pain, patellar maltracking and early failure due to abnormal mechanics. Proper alignment and rotation of the TKA is critical to surgical success. Over 30 years ago, Lotke and Ecker45 recognized the importance of coronal alignment in implant survival. They noted five failures, all via fracture of the medial tibial plateau. Four of the five failures were in knees placed in excessive varus. Ritter et al.46 evaluated 421 knee arthroplasty patients with up to 13-year follow-up. They found eight failures, five in patients placed in varus (less than 4ë anatomic valgus), and three in patients placed in 5±8ë of anatomic valgus. No failures were reported in the patients placed in valgus. They concluded that malalignment is a significant contributor to mechanical failure, but varus malalignment is better tolerated than valgus.46 Similarly, Jeffery et al.47 reported an increased rate of failure in knees in which the axis of alignment (Maquet's line ± from the center of the femoral head to the center of the talus) failed to pass within the middle third of the knee. One-third of their knees were malaligned. The failure rate for the malaligned knees was 24% (compared with 3%) at 8 years.47 When the weight-bearing line of the lower extremity passes through the center of the knee, the prosthesis is evenly loaded. If, on the other hand, stresses are not shared evenly over the prosthesis, one side experiences lift-off, while the other experiences excessive compression. This leads to excessive wear, particulate debris, osteolysis, cement de-bonding and component failure. Component rotation is also a critical component to success. In a cadaveric study, femoral rotation was found to correlate best with the transepicondylar axis of the femur. Internal or external rotation of 5ë resulted in increased femorotibial joint forces as well as abnormal shear at the patellofemoral articulation. This leads to post-operative pain, patellofemoral subluxation, and early failure.48

12.7

Summary

Despite the fact that total joint replacement is an effective operation for relieving pain and improving function, there are still issues related to implant wear and the adverse effects of particulate debris, including periprosthetic osteolysis and © 2008, Woodhead Publishing Limited

282

Joint replacement technology

implant loosening. Better bearing surfaces will undoubtedly improve implant longevity; however, a bearing surface that allows a lifetime of normal activity, including impact-loading sports, has not yet been achieved. Recurrent dislocation is the second most common problem that has recently been addressed by careful pre-operative planning, meticulous surgical technique, and the use larger diameter alternative bearing surfaces. The cause of instability of a joint replacement should be investigated and corrected. Implant failure due to fracture is uncommon due to better designs and manufacturing techniques; however newer modular prostheses and those made with inferior designs or materials, and poor supporting bone stock are risk factors. Stress shielding is a manifestation of Wolff's law and although not a major clinical problem, can lead to compromised bone stock for revision surgery. Periprosthetic fractures are extremely challenging for the reconstructive surgeon. These should be classified as to the location, the type of prosthesis in situ, the functionality and stability of the implant, and the quality of the surrounding bone. It is preferable, if possible, to fix fractures associated with stable, well-functioning implants rather than to deal with the fracture and perform an implant revision simultaneously.

12.8

Future trends

Total joint replacement is changing. No longer is a `one size fits all' mentality pervasive. The options for joint replacement are individualized for each patient. This has lead to concepts such as choosing the most functional, least expensive prosthesis for the level of activity of the patient (so-called `implant matching') to minimally invasive surgery (MIS) in which incisions and surgical dissections are kept to those that are necessary to accomplish the procedure only. Hopefully the concept of MIS will facilitate a quicker rehabilitation, early discharge, and return to normal activities, but this has not yet been conclusively shown. Modularity of implants will provide the surgeon with a host of options to reconstruct normal biomechanics `on the spot'. Newer bearing surfaces employing highly crosslinked polyethylene, metal-on-metal and ceramic-on-ceramic articulations may allow a wider range of activities. However, enthusiasm should be tempered by the unavailability of long-term follow-up. Computer-assisted surgery has the potential to improve implant alignment with `on-line' feedback in the operating room, but this concept is still in its infancy and multi-plane alignment cannot easily be assessed. One of the most exciting trends in modern arthroplasty surgery is multi-modal pain management, that is, the use of numerous adjunctive agents and modes of anesthesia and analgesia intra- and peri-operatively. Hopefully, this will allow the operative experience to be better tolerated by the patient and facilitate earlier more effective rehabilitation and return to function.

© 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

12.9

283

References

1. National Hospital Discharge Survey, 1991±2004; available from the US Department of Health and Human Services; Centers for Disease Control and Prevention; National Center for Health Statistics. Available from: http://www.aaos.org/ Research/stats/patientstats.asp 2. McKellop HA (1995) Wear modes, mechanisms, damage, and debris: separating cause from effect in the wear of total hip replacements, in Galante JO, Rosenberg AG, Callaghan JJ (eds): Total Hip Revision Surgery, New York, Raven Press, pp. 21±39. 3. McKellop HA, Campbell P, Park S-H, Schmalzried TP, Grigoris P, Amstutz HC, Sarmiento A (1995) The origin of submicron polyethylene wear debris in total hip arthroplasty. Clin Orthop, 311, pp. 3±20. 4. Wright T, Goodman SB, eds (2001) Implant Wear in Total Joint Replacement. Rosemont, IL: American Academy of Orthopaedic Surgeons. 5. Willert HG, Semlitsch M (1977) Reactions of the articular capsule to wear products of artificial joint prostheses. J Biomed Mater Res, 11, pp. 157±164. 6. Willert HG, Buchorn G, Buchorn U, and Semlisch M (1980) Implant Retrieval Conference, 1±2 May, Gaithersberg, MD. 7. Goodman SB, Huie P, Song Y, Schurman D, Maloney W, Woolson S, Sibley R (1998) Cellular profile and cytokine production at prosthetic interfaces. Study of tissues retrieved from revised hip and knee replacements. J Bone Joint Surg Br, 80(3), pp. 531±539. 8. Goodman SB, Chin RC, Chiou SS, et al. (1989) A clinical±pathological± biochemical study of the membrane surrounding loosened and nonloosened joint arthroplasty. Clin Orthop, 244, pp. 182±187. 9. Kadoya Y, Revell PA, Kobayashi A, al-Saffar N, Scott G, Freeman MA (1997) Wear particulate species and bone loss in failed total joint arthroplasties. Clin Orthop, 340, pp. 118±129. 10. Goodman SB (2005) Wear particulate and osteolysis. Orthop Clin North Am, 36(1), pp. 41±48. 11. Goodman SB, Trindade M, Ma T, Genovese M, Smith RL (2005) Pharmacologic modulation of periprosthetic osteolysis. Clin. Orthop, 430, pp. 39±45. 12. Jacobs JJ, Roebuck KA, Archibeck M, Hallab NJ, Glant TT (2001) Osteolysis: basic science. Clin Orthop, 393, pp. 71±77. 13. Purdue PE, Koulouvaris P, Potter HG, Nestor BJ, Sculco TP (2007) The cellular and molecular biology of periprosthetic osteolysis. Clin Orthop, 454, pp. 251±261. 14. Saleh KJ, Thongtrangan I, Schwarz EM (2004) Osteolysis: medical and surgical approaches. Clin Orthop, 427, pp. 138±147. 15. Talmo CT, Shanbhag AS, Rubash HE (2006) Nonsurgical management of osteolysis: challenges and opportunities. Clin Orthop, 453, pp. 254±264. 16. Ries M (2003) Complications in primary total hip arthroplasty: avoidance and management: wear, in Donald C, Ferlic MD (eds) Instructional Course Lectures, Volume 52. Rosemont, IL, American Academy of Orthopaedic Surgeons, pp. 257±265. 17. Kim SR, Callaghan JR, Ahn PB, Brown TD (1995) Fracture of the acetabulum during insertion of an oversized hemisherical component. J Bone Joint Surg [Am], 77-A, pp. 111±376. 18. Sharkey PF, Hozack WJ, Callaghan JJ, Kim YS, Berry DJ, Hanssen AD, LeWallen DG (1999) Acetabular fractures associated with cementless acetabular component insertion. J Arthroplasty, 14, pp. 426±431. © 2008, Woodhead Publishing Limited

284

Joint replacement technology

19. Haidukewych GJ, Jacofsky DJ, Hanssen AD, Lewallen DG (2006) Intraoperative fractures of the acetabulum during primary total hip arthroplasty. J Bone Joint Surg [Am], 88-A, pp. 1952±1956. 20. Lesh ML, Schneider DJ, Gurvinder D, Barclay D, Jacobs CR, Pellegrini VD (2000) The consequences of anterior femoral notching in total knee arthroplasty. J Bone Joint Surg [Am], 82-A, pp. 1096±1101. 21. Shawen PJ, Klemme WR, Topoleski LT, Xenos JS, Orchowski JR (2003) Osteoporosis and anterior femoral notching in periprosthetic supercondylar femoral fractures. J Bone Joint Surg, 85-A, pp.115±121. 22. Kraay MJ, Goldberg VM, Figgie MP, Figgie HE 3rd (1997) Distal femoral replacement with allograft/prosthetic reconstruction for treatment of supercondylar fractures in patient with total knee arthroplasty. J Arthroplasty, 7(1), pp. 7±16, 23. Figgie MP, Goldberg VM, Figgie HE 3rd, Sobel M (1990) The results of supracondylar fracture aboce total knee arthroplasty. J Arthroplasty, 5(3), pp. 267± 276. 24. Merkel KD, Johnson EW (1986) Supracondylar fracture of the femur after total knee arthroplasty. J Bone Joint Surg [Am], 68-A, pp. 29±43. 25. Culp RW, Schmidt RG, Hanks G, Mak A, Esterhai JL, Heppenstall RB (1987) Supracondylar fracture of the femur following prosthetic knee arthroplasty. Clin Orthop, 222, pp. 212±222. 26. Healy WL, Siliski JM, Incavo S (1993) Operative treatment of distal femoral fractures proximal to total knee replacements. J Bone Joint Surg [Am], 75-A, pp. 27±34. 27. Ritter MA, Thong AE, Keating EM, Faris PM, Meding JB, Berend ME, Pierson JL, Davis KE (2005) The effect of femoral notching during total knee arthroplasty on the prevalence of postoperative femoral fractures and on clinical outcome. J Bone Joint Surg [Am], 87-A(11), pp. 2411±2414. 28. Kelly MA (2004) Extensor mechanism complications in total knee arthroplasty, in Helfer D and Greene W (eds) Instructional Course Lectures, Volume 53. Rosemont, IL, American Academy of Orthopaedic Surgeons, pp. 193±199. 29. Peterson CA, Lewallen DG (1996) Periprosthetic fracture of the acetabulum after total hip arthroplasty. J Bone Joint Surg [Am], 78-A(8), pp. 1206±1213. 30. Lewis PL, Rorabeck CH (1997) Periprosthetic fractures, in Engh FA, Rorabeck CH, (eds) Revision Total Knee Arthroplasty. Baltimore, MD, Williams & Wilkins, pp. 275±295. 31. Hozack WJ, Goll SR, Lotke PA, Rothman RH, Booth RE Jr (1998) The treatment of patellar fractures after total knee arthroplasty. Clin Orthop, 236, pp. 123±127. 32. Willmann G (2000) Ceramic femoral head retrieval data. Clin Orthop, 379, pp. 22± 28. 33. Sanchez-Sotelo J, Haidukewych GJ, Boberg CJ (2006) Hospital cost of dislocation after primary total hip arthroplasty. J Bone Joint Surg [Am], 88-A(2), pp. 290±294. 34. Barrack RL (2003) Dislocation after total hip arthroplasty: implant design and orientation. JAAOS, 11(2), pp.89-99. 35. Alberton GM, High WA, Morrey BF (2002) Dislocation after revision hip arthroplasty. J Bone Joint Surg [Am], 84-A(10), pp. 1788±1792. 36. Berry DJ, von Knoch M, Schleck CD, Harmsen WS (2004) The cumulative longterm risk of dislocation after primary Charnley total hip arthroplasty. J Bone Joint Surg [Am], 86-A(1), pp. 9±14. 37. Kwon MS, Kukowski M, Mulhall KJ, Macaulay W, Brown TE, Saleh KJ (2006) Does surgical approach affect total hip arthroplasty dislocation rates? Clin Orthop, 447, pp. 34±38. © 2008, Woodhead Publishing Limited

Failure mechanisms in joint replacement

285

38. Goldstein WM, Gleason TF, Kopplin M, Branson JJ (2001) Prevalence of dislocation after total hip arthroplasty through a posterolateral approach with partial capsulotomy and capsulorrhaphy. J Bone Joint Surg [Am], 83-A (Suppl 2-1), pp. 2±7. 39. Bartz RL, Noble PC, Kadakia NR, Tullos HS (2000) The effect of femoral component head size on posterior dislocation of the artificial hip joint. J Bone Joint Surg [Am], 82-A(9), pp. 1300±1307. 40. Hedlundh U, Ahnfelt L, Hybbinette CH, Wallinder L, Weckstrom J, Fredin H (1996) Dislocations and the femoral head size in primary total hip arthroplasty. Clin Orthop, 333, pp. 226±233. 41. Woo RG, Morrey BF (1982) Dislocations after total hip arthroplasty. J Bone Joint Surg [Am], 64-A(9), pp. 1295±1306. 42. Barrack RL, LAvernia C, Ries M, Thornberry R, Tozakoglou E (2001) Virtual reality computer animation of the effect of component position and design on stability after total hip arthroplasty. Orthop Clin North Am, 32(4), pp. 569±577. 43. D'Lima DD, Urquhart AG, Buehler KO, Waler RH, Colwell CW Jr (2000) The effect of the orientation of the acetabular and femoral components on the range of motion of the hip at different head±neck ratios. J Bone Joint Surg [Am], 82-A(3), pp. 315±321. 44. Schmalzried TP, Guttman D, Grecula M, Amstutz HC (1994) The relationship between the design, position, and articular wear of acetabula components inserted without cement and the development of pelvic osteolysis. J Bone Joint Surg [Am], 76-A(5), pp. 677±688. 45. Lotke PA, Ecker ML (1977) Influence of positioning of prosthesis in total knee replacement. J Bone Joint Surg [Am], 59-A(1), pp. 77±79. 46. Ritter ML, Faris PM, Keating EM, Meding JB (1994) Postoperative alignment of total knee replacement: its effect on survival. Clin Orthop, 299, pp. 153±156. 47. Jeffery RS, Morris RW, Denham RA (1991) Coronal alignment after total knee replacement. J Bone Joint Surg [Br], 73-B(5), pp. 709±714. 48. Miller MC, Berger RA, Petrella AJ, Karmas A, Rubash HE (2001) Optimizing femoral component rotation in total knee arthroplasty. Clin Orthop, 392, pp. 38±45.

© 2008, Woodhead Publishing Limited

13

Predicting the lifetime of joints: clinical results L R Y D , Karolinska University Hospital/Huddinge, Sweden

13.1

Introduction

The most natural and direct way of assessing the results after a joint replacement procedure is in terms of the well-being of the patient. How does it feel? Do you have any pain? How far can you walk? How much can you flex your knee? These are common questions that patients are asked. Such assessment is, of course, an obvious part of any scientific report on joint replacements and a variety of scores that include such direct results have been devised. The Hospital for Special Surgery score (HSS score; Ranawat and Insall, 1976) and the Knee Society Score (KSS score; Ewald, 1989) are well-known examples. Another way of assessing the results after these procedures is by focusing on the longevity of the improvements. Questions such as `For how long does a joint prosthesis last?' `Can I count on being able to play golf in 5 years' time?' and `Are walks in the mountains OK?' are quite relevant to the `results' after joint replacements procedures. The introduction of artificial joints represented a significant step forward in the annals of surgery regarding the question of longevity; unlike any other kind of volume surgery in those days, joint replacement surgery depended on biomaterials. Large pieces of metal and polymers were introduced into the body and unlike living material, there is no biological turn-over. Instead biomaterials wear, and eventually they wear out. This chapter will deal with this second mode of assessment of results of joint replacement surgery. How long do joint replacements last in the clinical setting and how has this been addressed? Two novel lines of research will be discussed: national joint replacement registries and radiostereometric analysis (RSA). Apart from overall and basic results from the national registries and RSA, attempts will be made at addressing some of today's most burning issues with regards to hip and knee replacement surgery. Examples of such issues are: · for the hip ± age, cement vs. non-cement (hydroxyapatite, HA), screw fixation of the cup and various operative techniques; · for the knee ± age, patella vs. no patella replacement and mobile vs. allpolyethylene tibial component. © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

13.2

287

National joint replacement registries

With the introduction of artificial joints in the late 1960s and early 1970s, the orthopaedic community was faced by a number of new issues. First, the immediate clinical results in terms of well-being were stunningly good (Charnley, 1964). Patients reported no pain and virtually normal function, which by far exceeded the results after arthrodeses and osteotomies. Hence, the operations quickly became very popular. Secondly, again, biomaterials were used and there was little knowledge about how such materials would fare in the long run. Thirdly, artificial joints need to be manufactured and a whole new industry evolved. The orthopaedic community was faced with the necessity of collaborating with commercial parties that produced a plethora of new devices of different designs and unknown qualities. In order to address these issues, new methods were needed. In the early 1970s, discussions were started in the Swedish Orthopaedic Society to launch a national registry where all joint replacements performed in the country were to be registered. In 1975 the first registry, the National Knee Registry, was launched and situated in Lund. Some years later it was followed by the National Hip Registry, which was housed in Gothenburg. There was no involvement from the government or health authorities, but instead reports to the registries were done on a collegial and voluntary basis. The history and development of these registries are beyond the scope of this presentation. Suffice it to say that they have been extremely successful and have had an enormous impact. The homogeneity and the non-competitive setting of the Swedish orthopaedic community in those days provided the necessary catalyst and today these registries have found a multitude of followers, both joint replacement registries in other countries as well as national registries in other medical disciplines. A similar joint registry was started in Finland in 1980 (Puolakka et al., 2001), in Norway in 1987 (Havelin, 1999), in Australia in 1992 (Graves et al., 2004), in Denmark 1995 (Lucht, 2000), and in 1997 in New Zealand (Rothwell, 1999). In Sweden this has resulted in the National Competence Centre, Orthopaedics (NKO, www.nko.se), which, among other things, works toward the implementation of a common webbased platform for all registers. A national registry is different from other types of scientific reports in a number of ways. First, numbers are huge and, therefore, the amount of information on each patient has to be kept to a minimum. Early attempts at including radiographic and clinical result variables were soon abandoned and apart from demographic data and operative details at the index operation, for many years the only result variable registered was: Revision YES or NO. In later years deployment of Internet-based platforms in combination with modern computing and statistical know-how is now tipping the scale towards more ambitious data collection such as quality-of-life results and radiography (Garellick et al., 1998; Soderman et al., 2000a). © 2008, Woodhead Publishing Limited

288

Joint replacement technology

Further, results from national registries do not mirror the situation at specially interested departments or clinics, as is the case for most scientific reports, but rather reflect the situation in an entire professional community. Results depend not on the skills of individual surgeons, but reflect the organisation, level of training and education of a whole professional body. In fact, results may represent factors in a society not easily connected with orthopaedic outcome, such as the organisation of public and private transportation, community-aid for the elderly and overall management of osteoporosis, to mention but a few aspects. The power of national registries lies in the large numbers that provide for a statistical strength not found anywhere else. The impact of demographic data can be analysed, development over time and of different operative techniques can be assessed and the track record of all implants on the market can be monitored. The last, in fact, has been, and still is, a major output variable of the registries in an attempt to fend off commercial pressures by hard and non-biased data. Following a publication by Tew and Waugh (1982), who suggested joint implants should be followed in ways similar to malignant diseases, i.e. as survival analyses, the results from the Swedish registries were presented as 5- or 10-year survivals or, the reverse, as cumulative revision rates. This way of presenting the results has since become the standard in the field.

13.2.1 Validity of data A most important issue in the area of national registers is the degree of compliance of the participating hospitals, and efforts have been made to assess this factor. The registries are based on different conditions of participation and most are voluntary, as in Sweden, Norway and Denmark, Australia and New Zealand. In Finland and Scotland (Meek et al., 2006) the registries are compulsory and government run while a recently launched UK register (excluding Scotland) was awarded to a private party (Philipson et al., 2005). The comprehensiveness of data seems to be good. In Norway, registration seems to be better from small hospitals than from university departments, but overall, well over 95% of surgeries were reported (Arthursson et al., 2005). Similar numbers have been reported from the Swedish Hip Registry (Soderman et al., 2000b). In the 2005 report from the Swedish Hip Registry (www.jru.orthp.gu.se/documents), participation was given as 100%.

13.2.2 What have the Swedish hip and knee registries shown? The incidence of the procedures has increased over the years. This increase has been close to linear for hip arthroplasty with a small tendency to level off in recent years. In contrast, the increase in knee arthroplasty is a bit exponential © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

289

13.1 The yearly incidence of knee replacements in Sweden subdivided into three different diagnoses (OA = osteoarthritis, RA = rheumatoid arthritis) with permission from the Swedish Knee Arthroplasty Register; ß 2007 SKAR).

(Fig. 13.1). Likewise, there has been a change in the mean age of patients at operation. For hips there has been a slow decrease in age over the years from not quite 70 years in 1992 to about 68.5 in 2004. For the knee, there was an increase in mean age from 65 years in 1975 to about 71 years in 1995 after which time there has been a slow decrease again. Age at operation has a decisive impact on survivorship for both hips and knees. For hips, divided into four age groups, <50, 50±59, 60±75 and >75 years, survivorship after 13 years was 71%, 84%, 92% and 96% respectively for women and 74%, 79%, 88% and 94% respectively for men. For total knee arthroplasty (TKA) in osteoarthrotic knees, the cumulative revision rate at 10 years for the age group <65 years was 8% while for the age group >75 years it was 3% (Fig. 13.2). The overall prosthetic survival (all revisions, all diagnoses) has improved over the years. Analysed by 5-year cohorts, total hip arthroplasty (THA) has improved from a 10-year survival percentage of 85% for the 1979 cohort to 93% for the 1995 cohort (Fig. 13.3). The improvement for TKA has been identical © 2008, Woodhead Publishing Limited

290

Joint replacement technology

13.2 Ten-year survival of tri-compartmental replacement of the knee for osteoarthrosis in three different age groups (with permission from the Swedish Knee Arthroplasty Register; ß 2006 SKAR).

while the results of unicompartmental knee arthroplasty (UKA) has not improved in the same way over the years (Fig. 13.4). For the hip, mechanical loosening is the overwhelmingly most common cause for revision; about 75%. Other common causes are deep infection, 7.2%, dislocation, 7% and fractures, 5.5%. For the knee, the picture is somewhat more varied. Loosening is still the most common cause of revision and infection, and instability has been about as common as in the hip. Patellar problems, wear and progress of the disease (UKA) have, however, been reported as the cause of revision in rather high numbers. It is interesting that wear is not reported separately in the hip registry while it is a separate entity in the knee. This would suggest that the wear problem, which is generally perceived as a larger problem in the hip, is reflected as loosening (osteolysis) in the hip registry, while it means mechanical breakdown of the prosthesis proper in the knee registry. `Dislocation' is reported as the cause of revision in 7.1% of hips in Sweden. The registry stopped reporting closed reduction after hip dislocation in the year © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

291

13.3 The overall improvement of ten-year survival after any kind of total hip replacement (THR) in Sweden (with permission from the National Swedish Hip Registry).

2000 so one can only hypothesise that early revision takes place after a (unknown) number of closed reductions. Unlike all other results, dislocation shows a clear increase in successive 5-year cohorts and this is presently of particular concern in the country (Fig. 13.5). Factors such as greater patient mobility, larger volume of surgery ± meaning more but less experienced surgeons ± have been the suggested cause. The registry shows that an anterior© 2008, Woodhead Publishing Limited

292

Joint replacement technology

13.4 The improvement in prosthetic survival, expressed as Cumulative Revision Rate (CRR), from six different time periods for three different combinations of prosthetic concept and diagnosis. Note how results improve with time except for unicompartmental knee arthroplasty (UKA) (with permission from the Swedish Knee Arthroplasty Register; ß 2006 SKAR).

lateral approach to the hip reduces the risk of dislocation by about 50% and there are combinations of cup and stem designs with significant impact, both positive and negative, on this particular complication. Cement or no cement remains a controversial issue. In Sweden, both hip- and knee prostheses are put in with cement in the vast majority of cases. Since the mid-1990s, there has been a slow increase up to about 10% for non-cemented, hybrid and inverted hybrids together, with about equal shares of each. For knees, there was a period from 1982 to 1995 with non-cemented implants used in up to 25% of cases. Since 2000, only 1±2% of knees have been inserted without cement. The use of cement is so dominant in the knee registry that a comparison between the two is unwarranted. In the hip registry, crude results for noncemented implants are inferior. For the older cohort, 1979±1991, the 20-year survival of non-cemented implants for osteoarthritis (OA), revised because of loosening, was 50%, and the corresponding 25-year survival for all cemented implants was 78%. However, in the more recent patient group, 1992±2004, there is a decisive hump on the curve, indicating that modern non-cemented designs perform better, possibly on a par with cemented ones (Fig. 13.6). The impact of © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

13.4 continued © 2008, Woodhead Publishing Limited

293

294

Joint replacement technology

13.5 Cumulative revision rates of THA for dislocation for six different time periods. Note progressively more frequent revisions in later series (with permission from the National Swedish Hip Registry; ß 2006 Swedish Hip Arthroplasty Register).

cement vs. non-cement can be expressed as `revision burden' (RB), i.e. the proportion of revisions among all operations during one year. For cemented hips, in the years 1979±2005, the RB is 7.9% (Fig. 13.7), while for noncemented it is 19.1%. The more recent reversed hybrids, are performing extremely well with an RB at 6.2%, again possibly reflecting the superior performance of uncemented HA cups. Using cement, operative technique has been shown to be of importance. Consecutive analyses from the hip registry in the years 1997±1999 have shown that high-pressure lavage, distal femoral plug and retrograde cement injection © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

295

13.6 Survival curves for uncemented implants from two time periods. Note the hump on the curve representing the later series indicating excellent results for the later implants (with permission from the National Swedish Hip Registry; ß 2006 Swedish Hip Arthroplasty Register).

have significant positive effects on revision. Recently, the use of a proximal seal during cementation, resulting in higher pressure during cement injection, yielded a significantly better 13-year survival. These modern cementation procedures are now almost totally implemented throughout the country and this points to an important effect and benefit from the registry. Since both the hip and the knee registries are seen with considerable pride, as collegial contributions to society, the results are quickly disseminated throughout the orthopaedic community. Hence, the registries serve as nationwide bench-marking factors. The use of a patellar button in the knee is another hotly contended issue. © 2008, Woodhead Publishing Limited

296

Joint replacement technology

13.7 Illustration of the development of the number of cemented primary hip replacements (234 584) and their revisions (20 244) expressed as revision burden (RB) in Sweden (with permission from the National Swedish Hip Registry; Swedish Hip Arthroplasty Register).

Analysis of clinical outcome, i.e. `patient satisfaction' after the operation with or without patellar button provides, in a number of well-conducted random controlled trials (RCTs), gave no decisive verdict (Abraham et al., 1988; Bourne and Burnett, 2004). In Sweden in 2004, 11% of the TKAs were provided with a patellar button. This is down from a peak of 80% in 1985 and over the years, not quite 30% of all TKAs had a patellar button. For many years the knee registry showed no difference between the two groups in terms of survivorship. In the last couple of years, however, the registry reports a higher risk for revision if the © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

297

13.8 Diagram showing the cumulative revision rate for TKA for osteoarthosis when depending on the use of a patellar replacement at the primary procedure (with permission from the Swedish Knee Arthroplasty Register; 2006 SKAR).

patella is not resurfaced, with a risk ratio of 1.4 for OA knees (Fig. 13.8). For rheumatoid knees this higher risk could not be found. Since all operating units in Sweden report to the registries, the registries can be used for bench-marking purposes. Looking at 5-year survival in the cohort operated between 1992 and 1997, 11% of the units performed significantly below standard, while 32% performed above standard, with a span between 90 and 99.5% survival. The same numbers for the cohort operated between 1998 and 2004, were 9% below and 31% above standard with a span from 94 to 99.6%. Hence, fewer units were substandard and the span had shrunk, indicating better consistency and an improvement of the overall performance of the entire professional body (Fig. 13.9). Are the registries representative? That depends on the position of the beholder. Garellick et al. (2000) published an interesting report. At the joint replacement unit in Gothenburg, Sweden, a randomised trial had been conducted © 2008, Woodhead Publishing Limited

13.9 Ten-year survival of primary THR for each participating operating unit in Sweden. Each bar represents one hospital. Note rather conspicuous overall differences between units. (with permission from the National Swedish Hip Registry). © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

299

with a comparison between the Charnley stem/Ogee cup and the Spectron/metalbacked cup, examining altogether 410 procedures. The 11-year survivorship, with revision as the end point, was 93.2% for the Charnley hips and 95.9% for the Spectrons. For all 410 hips it was 94.5%. When patient dissatisfaction was included as an end point the results decreased to 86.3%. The corresponding numbers for 14 053 Charnleys from the hip registry during the same time period was 92.1% and for the 726 Spectrons 88.6%. Hence the specially dedicated department performed `better' and the authors speculate on performance bias.

13.3

Radiostereometric analysis

While registries can handle huge amounts of rather simple data, radiostereometric analysis (RSA) is an exceedingly precise instrument to measure motion between so-called rigid bodies. As it turned out (see below) the higher resolution of RSA as compared with conventional radiography proved enough to open the field of joint implant loosening to objective measurements. Hence, RSA, in some respects, represents the opposite of national registries. While the registries handle large numbers of patients, the strength of RSA lies in the need for only few patients to conduct a valid study. Further, while registry survival data take a long time to acquire, i.e. until revision surgery is performed, RSA results, as we shall see, are relevant already after a couple of years. RSA represents a combination of a number of previously known principles into one integrated `set'. The bones of the skeleton are exceedingly irregular in shape and one precise point is difficult to identify from different directions. Hence, there is a need to mark skeletal parts by, for example, metal objects, and to make precise measurements, as has been known since the 1960s (Lysell, 1969). Stereometry, as a mode to obtain three-dimensional information, had been used in geodesics for decades and had found its use also in radiography in the middle of the last century (Hallert, 1970). Finally, the mathematics involved, rigid body kinematics, had been known for centuries (Euler, 1776). First, and perhaps foremost, however, the evolution of modern information technology (IT) helped to promote the development of RSA. Since RSA calculates motion between hard and rigid bodies, application to the skeleton was a natural way to implement the new tool. The first RSA system was presented by Selvik in 1974 (Selvik, 1974). Selvik used tantalum balls to mark skeletal parts and clusters of tantalum balls (minimum of three balls) represent the bones, implants, fracture fragments or others parts of interest. Radiographs were obtained with the use of calibration objects employing either a biplanar or a convergent-ray configuration. Radiographs were digitised, initially by using a Wild A8 instrument for geodetic precision measurements. Using in-house software, 2D coordinates were transformed into 3D ones followed by the calculation of motion between a reference object and a study object (implant, fracture fragment, etc.) (Selvik, 1974). The © 2008, Woodhead Publishing Limited

300

Joint replacement technology

original RSA was cumbersome and time consuming. With the increasing interest in RSA, software packages have been developed and today two comprehensive commercial packages exist, the RSA-CMS out of Leiden, the Netherlands (Vrooman et al., 1998), and UmRSA from UmeaÊ, Sweden (KaÈrrholm et al., 1997). Other units have developed extensive and successful in-house systems (Gill et al., 2006). These modern and comprehensive packages make RSA much easier and one package has ventured into the definition of parts of interest not by markers but by the configuration of, say, an implant on radiographs, so-called model-based RSA (Valstar et al., 1997). RSA is, however, not a `black box' where radiographs are fed in and results come out `automatically'; the method still demands considerable skill from personnel. Methodologically, the resolution of RSA is unsurpassed, but the cost and complicity of the method has made researchers go to look for simpler techniques (Bragdon et al., 2006; Devane et al., 1997; Ilchmann, 1997). Over the years RSA has been applied to a multitude of skeletal disorders, such as fractures (Ahl et al., 1988; Mattsson and Larsson, 2004; Ragnarsson et al., 1992), growth disturbances (KaÈrrholm et al., 1983) and spinal surgery (Axelsson et al., 1992; Zoega et al., 1998). Arguably the most important application of RSA has been found in the study of joint implants. It is a known fact in the field, not the least from the joint registries, that mechanical, noninfectious loosening is the most common mode of failure of these devices. Loosening implies motion and hence it was reasonable to apply this new tool to joint replacements. The application of RSA to joint implants can be subdivided into three phases. The first phase was purely exploratory: mechanical loosening had long ago been identified as a major reason for revision and, prior to RSA, such loosening was conceived of as a sudden, often traumatic event (Willert et al., 1974). Early RSA reports decisively showed that measurable motion over time, i.e. migration, was present for many, if not all, implants (MjoÈberg et al., 1984; Nilsson et al., 1989; Ryd et al., 1983, 1986; Wykman et al., 1988). Further, by the application of loads to the replaced joint, implants could be made to displace relative to the bone they were inserted into, so-called inducible displacement (Ryd, 1986; Ryd et al., 1987, 1990). Both migration and inducible displacement were new concepts at a time when implants were looked upon as securely fixed to bone until they suddenly loosened and became symptomatic, necessitating revision surgery. The second phase involved correlations to clinical factors, notably whether implants that `moved' on RSA also became clinically loose. Further research was, indeed, able to link loosening to certain migratory patterns. After having followed a large number of patients for long times and identifying a number of cases of loosening, it was eventually established that a pattern of large and continuous migration pointed to future loosening. Hence, for the femoral component in the hip, KaÈrrholm and coworkers reported in 84 cases of primary and © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

301

13.10 RSA migration of 143 non-revised (open boxes) and 15 revised (filled boxes) tibial components of different designs. The arrows represent the point in time when the implant became symptomatic (with permission from Journal of Bone and Joint Surgery (Br)).

revision cemented stems that those that eventually loosened had already migrated more and continuously in the initial period (KaÈrrholm et al., 1994a). Ryd and coworkers reported similar findings regarding the tibial component in the knee; all 15 out of 158 primary cases that were revised for loosening migrated continuously. Further, these cases had also already migrated significantly more after six months when compared with the cases that migrated only initially and that never developed loosening (Fig. 13.10). For the 15 revised cases, abnormal migration was found very early despite the fact that clinical symptoms occurred as late as after 10 years in some cases (Ryd et al., 1995). For both hips and knees, decisive findings were present already after two years. Also for the acetabular cup, a predictive power of RSA has been shown (Stocks et al., 1995). Once the relation between RSA findings and clinical outcome was established, RSA could be used instead of revision as an end point as a surrogate variable. The advantage is obvious; instead of studies involving hundreds of © 2008, Woodhead Publishing Limited

302

Joint replacement technology

patients over 5±10 years, small cohorts could be studied for two years and burning issues of the time could be addressed and answered in short periods of time, representing the third phase of RSA in implant research.

13.3.1 Fixation of implants Contemporary joint implants are bonded to bone either with or without cement. Fixation without cement can be augmented by the application of coatings to the implant surface, notably HA. RSA has proved ideal in studies regarding these different fixation concepts. In the knee, cemented implants show a small amount of initial migration (0.2±0.5 mm) and usually stabilise after 6±12 months (Ryd et al., 1986, 1987). This is somewhat in contrast to uncemented components, which tend to migrate more initially in many reports (Carlsson et al., 2005; Nilsson et al., 1991; Ryd et al., 1990). In the hip, similar findings have been reported with rather more migration without cement regarding stem fixation (KaÈrrholm et al., 1994b; Wykman, 1989). In the acetabulum, cemented cups fare well (Flivik et È nsten and Carlsson, 1994). Uncemented cups behave differently al., 2005; O depending on design. The PCA cup performed poorly (Thanner et al., 1999) while the Harris-Galante cup, involving a titanium mesh, performed similarly to È nsten and Carlsson, 1994). cemented cups (O It is important to emphasise that cemented femoral stem fixation is based on one of two mutually exclusive principles: either bonding of cement to gridblasted prosthetic surfaces or non-bonding to polished prosthetic surfaces (the Exeter principle). The predictive power of subsidence is only valid for the former while subsidence of the polished stem inside the cement mantle is part of the rationale of this concept (Alfaro-Adrian et al., 2001). Non-cemented fixation of knee prostheses was popular in the early 1980s and was promptly tested using RSA. The PCA prosthesis without cement displayed larger migration and larger inducible displacement than the same prosthesis inserted with cement. The conclusion was that consistent bony ingrowth did not take place (Ryd et al., 1990). A non-cemented design using screws for additional fixation showed less micromotion and it was concluded that sufficient bony ingrowth had occurred, at least in some cases (Ryd et al., 1993). A non-cemented all-polyethylene design, not aiming at bone ingrowth, showed large micromotion, with inducible displacement of as much as 5 mm (Ryd et al., 1988). This noncemented, all-polyethylene implant was subsequently provided with a metalbacking including a rather generous stem. This new device showed significantly less migration and inducible displacement in exactly the way a stem would be expected to work, i.e. less varus±valgus and anterior±posterior tilt but the same degree of rotation (Albrektsson et al., 1990). So from an RSA point of view, the addition of stems to tibial component designs was beneficial. HA has shown beneficial effects for both hips and knees. In the knee, tibial components of the Osteonics design did not show ominous continuous migration © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

303

in one single case with HA coating (Toksvig-Larsen et al., 2000). Similar findings were reported by Regner et al. (2000) when comparing the HA-coated Freeman-Samuelson design with the non-coated Miller-Galante II prosthesis. Similarly, HA-coated Tricon tibial components showed better RSA results than their cemented counterparts (Nilsson et al., 1999). In the hip, HA coating in combination with screws in acetabular components proved successful both in primary and revision cases (Nivbrant and KaÈrrholm, 1997). Similar finding were reported by Thanner (1999). For the stem, finally, the clinical success reported by Geesink in the late 1980s (Geesink, 1990) has been confirmed by a number of RSA studies (Grant et al., 2005; SoÈballe et al., 1993). The behaviour of the cement itself has been studied. The early low-viscosity cements were not particularly successful clinically and their RSA results did not represent an improvement over ordinary cement (MjoÈberg et al., 1987). The infamous Boneloc, which behaved catastrophically poorly in Denmark, showed large migration on RSA both in the hip and the knee (Nilsson and Dalen, 1998; Thanner et al., 1995). Finally, newer compositions of cement have been tested and found safe; an example of good market introduction (Kienapfel et al., 2004). In a series of publications, Uvehammar and coworkers have explored the influence of articulating surface design in the knee. Concerning fixation, they found small differences between flat, curved and posterior stabilised designs but they found larger differences regarding the kinematic behaviour (Saari et al., 2005; Uvehammer et al., 2000, 2001). Using mobile tibial polyethylene inserts in the knee is an ingenious way of reconciling two mutually exclusive design concepts; a lack of constraint to protect the interface from excessive shear forces and congruent articulating surfaces to protect the polyethylene from wear (Goodfellow and O'Connor, 1978). RSA studies on such prostheses have shown that the bearings do move as expected but the migratory patterns of the entire implant are the same as controls without a mobile bearing, hence suggesting that the interface remains uninfluenced both in uncemented and cemented knees (Hansson et al., 2005; Henricson et al., 2006).

13.3.2 Wear Wear of the polyethylene inserts is one of the main problems of joint replacement technology. Excessive wear, in the end, gives rise to the obvious calamity of the joint's `wearing out', i.e. the polyethylene breaks and metal articulates against metal. Measurements of wear have been performed rather extensively by Bragdon and coworkers. In a series of reports they found RSA to be applicable and more precise than other more simple methods (Bragdon et al., 2002, 2004, 2006; Gill et al., 2006). It has also been shown that examination in the supine or in the weightbearing position is of no importance (von Schewelov et al., 2004). Wear has also © 2008, Woodhead Publishing Limited

304

Joint replacement technology

been studied in the knee and Gill and coworkers report that wear of 100 mm3/year from the AGC design should be clinically insignificant (Gill et al., 2006). Wear products appear to give rise to osteolysis, where bone is undermined and loosening follows. This problem seems to be more pronounced in the hip. There is some evidence that cement acts not only to bond implants to bone but also to seal the interface. Without cement, wear particles gain access to the interface and this appears to be especially detrimental in the hip where `silent osteolysis' is a problem (Schmalzried et al., 1992). So the problems of wear and osteolysis and fixation blend together into an issue that can be studied by RSA. Some uncemented cups have had problems related to the liner and wear because of poor fixation of the polyethylene within the metal shell (Rohrl et al., 2006). The Omnitfit cup was studied and showed both `catastrophic' wear and its biological consequence, osteolysis (von Schewelov et al., 2004). In a recent study the usual bone preparation of the acetabulum, with retention of the subchondral bone plate, has been challenged precisely for the sealing effects of cement. A significantly `better interface' (Fig. 13.11) was found on radiography without the loss of acetabular stability as measured by RSA (Fig. 13.12) when the subchondral bone plate was removed (Flivik et al., 2006). While knee implants are being inserted with a precision of 2±3ë, hip implants, both stems and cups, can be positioned with a rather large variability.

13.11 RSA migration of the acetabular cup when the subchondral bone plate was retained or removed (with permission from Clinical Orthopaedics and Related Research). © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

305

13.12 Radiograph showing radiolucency at the interface of an acetabular cup two years after implantation into a bone bed with the subchondral bone retained (a). The interface at two years when the subchondral bone was removed shows no radiolucency (b) (with permission from Clinical Orthopaedics and Related Research).

Data are now accumulating that the rotational position of the stem is precarious. Hence, the more stems are inserted in relative retroversion, with reference to the previous neck anteversion, the more they migrate in posterior rotation postoperatively and this excessive rotation leads to loosening (Hermann et al., 1998) Data suggest that this mode of loosening may represent an important mode of failure for hip stems (Flivik et al., 2007; Gill et al., 2002; Hauptfleisch et al., 2006). With the realisation that loosening is not a mechanical but a biological process (Fig. 13.10), the concept of loosening needs to be reworked. Loosening may be addressed by interventions after the operation, physiotherapy and possibly medication may have altered the development of the interface. Hence, Hilding and coworkers were able to show, in a few publications, that gait had an effect, notably not, as expected, in the varus±valgus dimension, but rather while looking at the knee sideways: larger flexion moments were recorded in knees that showed ominous migration curves. Interestingly, these differences were already present before the operations, suggesting that training may be a possibility (Hilding et al., 1995, 1996, 1999). The ultimate proof of the biology of the interface would be by medication. Subsidence, and the postoperative migration of virtually all tibial implants, must be a reflection of bone resorption. If such resorption could be influenced, this could be beneficial in the long run. In a randomised series of 50 TKAs, Hilding and coworkers were indeed able to show that six months of medication with Clodronate, an osteoclast inhibitor, starting three weeks before the operation, significantly diminished postoperative migration (Hilding et al., 2000). Hilding has recently published a four-year report and differences between the groups were still to be found, although to a smaller degree than after two years (Hilding and Aspenberg, 2006). © 2008, Woodhead Publishing Limited

306

13.4

Joint replacement technology

Future trends

Both national registries and RSA represent `new' ways of accessing data after surgical interventions in destroyed joints. Both systems provide data of a novel type which in some ways is unique, yet does not replace more conventional modes of research such as the RCT. Both systems are now well established in the field; more and more national registries are started and RSA recently had a dedicated issue of Clinical Research and Related Orthopaedics (Volume 448, 2006). The commercial activities inherently connected with joint replacement surgery necessitate objective data. Competition in the market, both the manufacturing of implants as well the provision of the operative procedures at hospitals, creates pressures that need to be balanced. The public increasingly demands open and transparent assessment of overall results. The three modes of assessment , RSA, RCT and Registry (the three `R's'), are likely to become more established into one clinical package in the future, representing an introductory pathway that new implants will have to walk. Following preclinical investigations, new devices should be subjected to small RSA trials, followed by regular RCTs, possibly of multicentre design. Finally, the market track record is monitored in national registers after fully fledged clinical introduction. Only then will the safety of new implants be guaranteed while, at the same time, the improvement represented by implant evolution into new devices will be identified.

13.5

References

Abraham, W., Buchanan, J. R., Daubert, H., Greer, R. B. & Keefer, J. (1988) Should the patella be resurfaced in total knee arthroplasty? Efficacy of patellar resurfacing. Clin Orthop, 236, 128±34. Ahl, T., DaleÂn, N. & Selvik, G. (1988) Mobilisation after operation of ankle fractures. Good results of early motion and weight bearing. Acta Orthop Scand, 59, 302±306. Albrektsson, B. E., Ryd, L., Carlsson, L. V., Freeman, M. A., Herberts, P., Regner, L. & Selvik, G. (1990) The effect of a stem on the tibial component of knee arthroplasty. A roentgen stereophotogrammetric study of uncemented tibial components in the Freeman±Samuelson knee arthroplasty. J Bone Joint Surg Br, 72, 252±8. Alfaro-Adrian, J., Gill, H. S. & Murray, D. W. (2001) Should total hip arthroplasty femoral components be designed to subside? A radiostereometric analysis study of the Charnley Elite and Exeter stems. J Arthroplasty, 16, 598±606. Arthursson, A. J., Furnes, O., Espehaug, B., Havelin, L. I. & SoÈreide, J. A. (2005) Validation of data in the Norwegian Arthroplasty Register and the Norwegian Patient Register: 5,134 primary total hip arthroplasties and revisions operated at a single hospital between 1987 and 2003. Acta Orthop, 76, 823±8. Axelsson, P., Johnsson, R. & StroÈmqvist, B. (1992) Effect of lumbar orthosis on intervertebral mobility. A roentgen stereophotogrammetric analysis. Spine, 17, 678±81. Bourne, R. B. & Burnett, R. S. (2004) The consequences of not resurfacing the patella. Clin Orthop Relat Res, 166±9. Bragdon, C. R., Malchau, H., Yuan, X., Perinchief, R., KaÈrrholm, J., BoÈrlin, N., Estok, D. © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

307

M. & Harris, W. H. (2002) Experimental assessment of precision and accuracy of radiostereometric analysis for the determination of polyethylene wear in a total hip replacement model. J Orthop Res, 20, 688±95. Bragdon, C. R., Estok, D. M., Malchau, H., KaÈrrholm, J., Yuan, X., Bourne, R., Veldhoven, J. & Harris, W. H. (2004) Comparison of two digital radiostereometric analysis methods in the determination of femoral head penetration in a total hip replacement phantom. J Orthop Res, 22, 659±64. Bragdon, C. R., Martell, J. M., Greene, M. E., Estok, D. M., 2nd, Thanner, J., KaÈrrholm, J., Harris, W. H. & Malchau, H. (2006) Comparison of femoral head penetration using RSA and the Martell method. Clin Orthop Relat Res, 448, 52±7. È nsten, I. (2005) Cemented tibial component Carlsson, A., BjoÈrkman, A., Besjakov, J. & O fixation performs better than cementless fixation: a randomized radiostereometric study comparing porous-coated, hydroxyapatite-coated and cemented tibial components over 5 years. Acta Orthop, 76, 362±9. Charnley, J. (1964) The bonding of prostheses to bone by cement. J Bone Joint Surg, 46B, 518±529. Devane, P. A., Horne, J. G., Martin, K., Coldham, G. & Krause, B. (1997) Threedimensional polyethylene wear of a press-fit titanium prosthesis. Factors influencing generation of polyethylene debris. J Arthroplasty, 12, 256±66. Euler, L. (1776) Formulae generales pro translatione quaqunque corporum rigidorum. Novi commentarii academie scientiarum Petropolitanae. Orell FuÈssli Turici. Basel, 1968; 9: 84±98. ed. Ewald, F. C. (1989) The Knee Society total knee arthroplasty roentgenographic evaluation and scoring system. Clin Orthop, 248, 9±12. È nnerfalt, R., Kesteris, U. & Ryd, L. (2005) Migration of the Flivik, G., Sanfridsson, J., O acetabular component: effect of cement pressurization and significance of early radiolucency: a randomized 5-year study using radiostereometry. Acta Orthop, 76, 159±68. Flivik, G., Kristiansson, I., Kesteris, U. & Ryd, L. (2006) Is removal of subchondral bone plate advantageous in cemented cup fixation? A randomized RSA study. Clin Orthop Relat Res, 448, 164±72. Flivik, G., Hermann, K. & Ryd, L. (2007) The importance of adequate stem anteversion for rotational stability in THA. An RSA-study with 5-year followup Proceedings of the 8th EFORT Congress, 8, F253. Garellick, G., Malchau, H. & Herberts, P. (1998) Specific or general health outcome measures in the evaluation of total hip replacement. A comparison between the Harris hip score and the Nottingham Health Profile. J Bone Joint Surg Br, 80, 600±6. Garellick, G., Malchau, H. & Herberts, P. (2000) Survival of hip replacements. A comparison of a randomized trial and a registry. Clin Orthop Relat Res, 357, 157±67. Geesink, R. G. T. (1990) Hydroxyapatite-coated total hip prosthesis. Clini Orthop, 261, 39±58. Gill, H. S., Alfaro-Adrian, J., Alfaro-Adrian, C., McLardy-Smith, P. & Murray, D. W. (2002) The effect of anteversion on femoral component stability assessed by radiostereometric analysis. J Arthroplasty, 17, 997±1005. Gill, H. S., Waite, J. C., Short, A., Kellett, C. F., Price, A. J. & Murray, D. W. (2006) In vivo measurement of volumetric wear of a total knee replacement. Knee, 13, 312± 17. Goodfellow, J. & O'Connor, J. (1978) The mechanics of the knee and prosthetic design. J Bone Joint Surg, 60-B, 358±69. Grant, P., Aamodt, A., Falch, J. A. & Nordsletten, L. (2005) Differences in stability and © 2008, Woodhead Publishing Limited

308

Joint replacement technology

bone remodeling between a customized uncemented hydroxyapatite coated and a standard cemented femoral stem A randomized study with use of radiostereometry and bone densitometry. J Orthop Res, 23, 1280±5. Graves, S. E., Davidson, D., Ingerson, L., Ryan, P., Griffith, E. C., McDermott, B. F., McElroy, H. J. & Pratt, N. L. (2004) The Australian Orthopaedic Association National Joint Replacement Registry. Med J Aust, 180, S31±4. Hallert, B. (1970) X-ray Photogrammetry. Basic Geometry and Quality, Amsterdam, Elsevier. Hansson, U., Toksvig-Larsen, S., Jorn, L. P. & Ryd, L. (2005) Mobile vs. fixed meniscal bearing in total knee replacement: a randomised radiostereometric study. Knee, 12, 414±18. Hauptfleisch, J., Glyn-Jones, S., Beard, D. J., Gill, H. S. & Murray, D. W. (2006) The premature failure of the Charnley Elite-Plus stem: a confirmation of RSA predictions. J Bone Joint Surg Br, 88, 179±83. Havelin, L. I. (1999) The Norwegian Joint Registry. Bull Hosp Joint Dis, 58, 139±47. Henricson, A., Dalen, T. & Nilsson, K. G. (2006) Mobile bearings do not improve fixation in cemented total knee arthroplasty. Clin Orthop Relat Res, 448, 114±21. Hermann, K. L., Flivik, G., Egund, N., Ryd, L. & Jonsson, K. (1998) Correlation betwee the rotational position of the femoral stem in cemented total hip replacement and movement around its vertical axis at roentgenstereophotogrammetry. Acta Orthop, 69, 9±10. Hilding, M. & Aspenberg, P. (2006) Postoperative clodronate decreases prosthetic migration: 4-year follow-up of a randomized radiostereometric study of 50 total knee patients. Acta Orthop, 77, 912±16. Hilding, M. B., Lanshammar, H. & Ryd, L. (1995) A relationship between dynamic and static assessments of knee joint load. Gait analysis and radiography before and after knee replacement in 45 patients. Acta Orthop Scand, 66, 317±20. Hilding, M. B., Lanshammar, H. & Ryd, L. (1996) Knee joint loading and tibial component loosening. RSA and gait analysis in 45 osteoarthritic patients before and after TKA. J Bone Joint Surg Br, 78, 66±73. Hilding, M. B., Ryd, L., Toksvig-Larsen, S., Mann, A. & StenstroÈm, A. (1999) Gait affects tibial component fixation. J Arthroplasty, 14, 589±93. Hilding, M., Ryd, L., Toksvig-Larsen, S. & Aspenberg, P. (2000) Clodronate prevents prosthetic migration: a randomized radiostereometric study of 50 total knee patients. Acta Orthop Scand, 71, 553±7. Ilchmann, T. (1997) Radiographic assessment of cup migration and wear after hip replacement. Acta Orthop Scand Suppl, 276, 1±26. KaÈrrholm, J., Hansson, L. I., Svensson, K. (1983) Prediction of growth pattern after ankle fractures in children. Journal of Pediatric Orthopaedics, 3, 319±325. KaÈrrholm, J., Borssen, B., LoÈwenhielm, G. & Snorrasson, F. (1994a) Does early micromotion of femoral stem prostheses matter? 4±7-year stereoradiographic follow-up of 84 cemented prostheses. J Bone Joint Surg Br, 76, 912±17. KaÈrrholm, J., Malchau, H., Snorrasson, F. & Herbert, P. (1994b) Micromotion in femoral stems in total hip arthroplasty. A randomized study of cemented, hydroxyapatitecoated and porous coated stems with roentgen stereophotogrammetric analysis. J Bone Joint Surg, 76-A, 1692±1705. KaÈrrholm, J., Herberts, P., Hultmark, P., Malchau, H., Nivbrant, B. & Thanner, J. (1997) Radiostereometry of hip prostheses. Review of methodology and clinical results. Clin Orthop Relat Res, 94±110. Kienapfel, H., Hildebrand, R., Neumann, T., Specht, R., Koller, M., Celik, I., Mueller, © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

309

H.H., Griss, P., Klose, K. J. & Georg, C. (2004) The effect of Palamed G bone cement on early migration of tibial components in total knee arthroplasty. Inflamm Res, 53 Suppl 2, S159±63. Lucht, U. (2000) The Danish Hip Arthroplasty Register. Acta Orthop Scand, 71, 433±9. Lysell, E. (1969) Motion of the cervical spine. Acta Orthop Scand, 1±40. Mattson, P. & Larsson, S. (2004) Unstable trochanteric fractures augmented with calcium phosphate cement. A prospective randomized study using radiostereometry to measure fracture stability. Scand J Surg, 93, 223±8. Meek, R. M., Allan, D. B., McPhillips, G., Kerr, L. & Howie, C. R. (2006) Epidemiology of dislocation after total hip arthroplasty. Clin Orthop Relat Res, 447, 9±18. MjoÈberg, B., Hansson, L. I. & Selvik, G. (1984) Instability, migration and laxity of total hip prostheses. A roentgen sterophotogrammetric study. Acta Orthop Scand, 55, 141±5. È nnerfaÈlt, R. (1987) Low- versus high-viscosity MjoÈberg, B., Rydholm, A., Selvik, G. & O bone cement. Fixation of hip prostheses analysed by roentgen stereophotogrammetry. Acta Orthop Scand, 58, 106±8. Nilsson, K. G. & Dalen, T. (1998) Inferior performance of Boneloc bone cement in total knee arthroplasty: a prospective randomized study comparing Boneloc with Palacos using radiostereometry (RSA) in 19 patients. Acta Orthop Scand, 69, 479±83. Nilsson, K. G., BrobaÈck, L. G. & KaÈrrholm, J. (1989) Does a stem on the tibial component improve fixation of the uncemented Freeman±Samuelson prosthesis? Acta Orthopedica Scandinavica, 60, 34. Nilsson, K. G., KaÈrrholm, J., Ekelund, L. & Magnusson, P. (1991) Evaluation of micromotion in cemented vs uncemented knee arthroplasty in osteoarthrosis and rheumatoid arthritis. Randomized study using Roentgen stereophotogrammetric analysis. J Arthroplasty, 6, 265±78. Nilsson, K. G., KaÈrrholm, J., Carlsson, L. & Dalen, T. (1999) Hydroxyapatite coating versus cemented fixation of the tibial component in total knee arthroplasty: prospective randomized comparison of hydroxyapatite-coated and cemented tibial components with 5-year follow-up using radiostereometry. J Arthroplasty, 14, 9± 20. Nivbrant, B. & KaÈrrholm, J. (1997) Migration and wear of hydroxyapatite-coated pressfit cups in revision hip arthroplasty: a radiostereometric study. J Arthroplasty, 12, 904±12. È nsten, I. & Carlsson, A. S. (1994) Cemented versus uncemented socket in hip O arthroplasty. A radiostereometric study of 60 randomized hips followed for 2 years. Acta Orthop Scand, 65, 517±21. Philipson, M. R., Westwood, M. J., Geoghegan, J. M., Henry, A. P. & Jefferiss, C. D. (2005) Shortcomings of the National Joint Registry: a survey of consultants' views. Ann R Coll Surg Engl, 87, 109±12; discussion 112. Puolakka, T. J., Pajamaki, K. J., Halonen, P. J., Pulkkinen, P. O., Paavolainen, P. & Nevalainen, J. K. (2001) The Finnish Arthroplasty Register: report of the hip register. Acta Orthop Scand, 72, 433±41. Ragnarsson, J. I., Eliasson, P., KaÈrrholm, J. & LundstroÈm, B. (1992) The accuracy of measurements of femoral neck fractures. Conventional radiography versus roentgen stereophotogrammetric analysis. Acta Orthop Scand Suppl, 63, 152±6. Ranawat, C. S. & Insall, J. N. (1976) Duocondylar knee arthroplasty. Clin Orthop, 120, 76±92. Regner, L., Carlsson, L., KaÈrrholm, J. & Herberts, P. (2000) Tibial component fixation in porous- and hydroxyapatite-coated total knee arthroplasty: a radiostereometric © 2008, Woodhead Publishing Limited

310

Joint replacement technology

evaluation of migration and inducible displacement after 5 years. J Arthroplasty, 15, 681±9. Rohrl, S. M., Nivbrant, B., Snorrason, F., KaÈrrholm, J. & Nilsson, K. G. (2006) Porouscoated cups fixed with screws: a 12-year clinical and radiostereometric follow-up study of 50 hips. Acta Orthop, 77, 393±401. Rothwell, A. G. (1999) Development of the New Zealand Joint Register. Bull Hosp Joint Dis, 58, 148±60. Ryd, L. (1986) Micromotion in knee arthroplasty. A roentgen stereophotogrammetric analysis of tibial component fixation. Acta Orthop Scand Suppl, 220, 1±80. Ryd, L., BoegaÊrd, T., Egund, N., Lindstrand, A., Selvik, G. & Thorngren, K.-G. (1983) Migration of the tibial component in successful unicompartmental knee arthroplasty. A clinical, radiographig and roentgen stereophotogrammetric study. Acta Orthop Scand, 54, 408±16. Ryd, L., Lindstrand, A., Rosenquist, R. & Selvik, G. (1986) Tibial component fixation in knee arthroplasty. Clin Orthop Relat Res, 213, 141±9. Ryd, L., Lindstrand, A., Rosenquist, R. & Selvik, G. (1987) Micromotion of conventionally cemented all-polyethylene tibial components in total knee replacements. A roentgen stereophotogrammetric analysis of migration and inducible displacement. Arch Orthop Trauma Surg, 106, 82±8. Ryd, L., Albrektsson, B. E., Herberts, P., Lindstrand, A. & Selvik, G. (1988) Micromotion of noncemented Freeman±Samuelson knee prostheses in gonarthrosis. A roentgen-stereophotogrammetric analysis of eight successful cases. Clin Orthop Relat Res, 229, 205±12. Ryd, L., Lindstrand, A., StenstroÈm, A. & Selvik, G. (1990) Porous coated anatomic tricompartmental tibial components. The relationship between prosthetic position and micromotion. Clin Orthop Relat Res, 251, 189±97. Ryd, L., Carlsson, L. & Herberts, P. (1993) Micromotion of a screw-fixed, titanium meshed tibial component. An in vivo roentgen stereophotogrammetric study of the Miller-Galante prosthesis. Clini Orthop, 295, 218±25. Ryd, L., Albrektsson, B. E., Carlsson, L., Dansgard, F., Herberts, P., Lindstrand, A., Regner, L. & Toksvig-Larsen, S. (1995) Roentgen stereophotogrammetric analysis as a predictor of mechanical loosening of knee prostheses. J Bone Joint Surg Br, 77, 377±83. Saari, T., Tranberg, R., Zugner, R., Uvehammer, J. & KaÈrrholm, J. (2005) Changed gait pattern in patients with total knee arthroplasty but minimal influence of tibial insert design: gait analysis during level walking in 39 TKR patients and 18 healthy controls. Acta Orthop, 76, 253±60. Schmalzried, T. P., Jasty, M. & Harris, W. H. (1992) Periprosthetic bone loss in total hip arthroplasty. Polyethylene wear debris andthe concept of the effective joint space. J Bone Joint Surg, 74-A, 849±63. Selvik, G. (1974) Roentgen stereophotogrammetry. A method for the study of the kinematics of the skeletal systems. University of Lund, Sweden, 1974. Reprint in: Acta Arthop Scand 1989; 60 (Suppl. 232). SoÈballe, K., Toksvig-Larsen, S., Gelineck, J., Fruensgaard, S., Hansen, E. S., Ryd, L., Lucht, U. & Bunger, C. (1993) Migration of hydroxyapatite coated femoral prostheses. A roentgen stereophotogrammetric study. J Bone Joint Surg Br, 75, 681±7. Soderman, P., Malchau, H. & Herberts, P. (2000a) Outcome after total hip arthroplasty: Part I. General health evaluation in relation to definition of failure in the Swedish National Total Hip Arthoplasty register. Acta Orthop Scand, 71, 354±9. © 2008, Woodhead Publishing Limited

Predicting the lifetime of joints: clinical results

311

Soderman, P., Malchau, H., Herberts, P. & Johnell, O. (2000b) Are the findings in the Swedish National Total Hip Arthroplasty Register valid? A comparison between the Swedish National Total Hip Arthroplasty Register, the National Discharge Register, and the National Death Register. J Arthroplasty, 15, 884±9. Stocks, G. W., Freeman, M. A. R. & Moilonen, T. (1995) Prediction of late acetabular loosening by early measurement of proximal migration. J Bone Joint Surg, 77-B, 853±61. Tew, M. & Waugh, W. (1982) Estimating the survival time of knee replacements. J Bone Joint Surg, 64-B, 579±82. Thanner, J. (1999) The acetabular component in total hip arthroplasty. Evaluation of different fixation principles. Acta Orthop Scand Suppl, 286, 1±41. Thanner, J., Freij-Larsson, C., KaÈrrholm, J., Malchau, H. & Wesslen, B. (1995) Evaluation of Boneloc. Chemical and mechanical properties, and a randomized clinical study of 30 total hip arthroplasties. Acta Orthop Scand, 66, 207±14. Thanner, J., KaÈrrholm, J., Malchau, H. & Herberts, P. (1999) Poor outcome of the PCA and Harris±Galante hip prostheses. Randomized study of 171 arthroplasties with 9year follow-up. Acta Orthop Scand, 70, 155±62. Toksvig-Larsen, S., Jorn, L. P., Ryd, L. & Lindstrand, A. (2000) Hydroxyapatiteenhanced tibial prosthetic fixation. Clin Orthop Relat Res, 192±200. Uvehammer, J., KaÈrrholm, J. & Brandsson, S. (2000) In vivo kinematics of total knee arthroplasty. Concave versus posterior-stabilised tibial joint surface. J Bone Joint Surg Br, 82, 499±505. Uvehammer, J., Regner, L. & KaÈrrholm, J. (2001) Flat vs. concave tibial joint surface in total knee arthroplasty: randomized evaluation of 39 cases using radiostereometry. Acta Orthop Scand, 72, 257±65. Valstar, E. R., Spoor, C. W., Nelissen, R. G. H. H. & Rozing, P. M. (1997) Roentgen stereophotogrammetric analysis of metal-backed hemispherical cups without attached markers. J Orthop Res, 15, 869±73. È nsten, I. & Carlsson, A. (2004) Catastrophic failure of Von Schewelov, T., Sanzen, L., O an uncemented acetabular component due to high wear and osteolysis: an analysis of 154 omnifit prostheses with mean 6-year follow-up. Acta Orthop Scand, 75, 283± 94. Vrooman, H. A., Valstar, E. R., Brand, G. J., Admiraal, D. R., Rozing, P. M. & Reiber, J. H. (1998) Fast and accurate automated measurements in digitized stereophotogrammetric radiographs. J Biomech, 31, 491±8. Willert, H. G., Ludwig, J. & Semtlisch, M. (1974) Reaction of bone to methacrylate after hip arthroplasty. J Bone Joint Surg, 56-A, 1368±82. Wykman, A. (1989) Cemented and Non-cemented Total Hip Arthroplasty, Stockholm, Sweden. Wykman, A., Selvik, G. & Goldie, I. (1988) Subsidence of the femoral component in the noncemented total hip: a roentgen stereophotogrammetric analysis. Acta Orthopedica Scandinavica, 59, 635±7. Zoega, B., KaÈrrholm, J. & Lind, B. (1998) Plate fixation adds stability to two-level anterior fusion in the cervical spine: a randomized study using radiostereometry. Eur Spine J, 7, 302±7.

© 2008, Woodhead Publishing Limited

Part III

The device biological environment

© 2008, Woodhead Publishing Limited

14

The healing response to implants used in joint replacement P A R E V E L L , University College London, UK

14.1

Introduction

The replacement of joints with synthetic materials has been carried out with varying success for over 70 years, early attempts focusing on the hip (and shoulder, surprisingly; see Chapter 24). At first, at the hip, for example, one side only of the joint was replaced, by a metal cup (Smith-Peterson) on the acetabular side or by a metal ball on a stem inserted into the upper femur after resection of the femoral head (Thomson, Moore).1 The Judet brothers introduced acrylic as a component material but without great success. This material did, however, subsequently find use in a different form by Charnley who performed total joint replacements at the hip using a metal-stemmed small-headed femoral component articulating with a small plastic acetabular cup with an acrylic, poly(methylmethacrylate) (PMMA), as `bone cement' to fix these implants to the bone.1 Subsequently, prostheses have been developed to replace most of the other joints with varying degrees of success. Methods of fixation of the implants to the skeleton have progressed through coatings which allow bone ingrowth to bioactive coatings for so-called osseo-integration, both of which methods require no use of cement. There have been some developments in `cement' technology and various cementless methods are in current use. Basically, however, the materials implanted remain as various metal alloys, polymers providing bearing surfaces (polyethylene) or cements (PMMA), and ceramics, again as bearing surfaces. Bioactive coatings, most notably hydroxyapatite (HA), constitute the last broad category of materials to which the body reacts in joint replacement surgery. Any insult to the body, be it chemical, physical or biological, is met by an inflammatory response. This may be transient, where the injurious agent is trivial. However, if the insult is sustained, the inflammation will pass into a prolonged phase and persist for a long period of time. It is self-evident that the implantation of a large foreign body in the form of a prosthetic joint component into an equally large wound made in bone by such traumatic processes as sawing, drilling and reaming will give rise to a sustained tissue response, which hopefully will result in healing with stability, but may not do so in some individuals. © 2008, Woodhead Publishing Limited

316

Joint replacement technology

Inflammation is said to occur in acute and chronic forms, though this division is artificial in that there are processes going on as a continuum once initiated by the insult. Thus, in the early phase of the response, there is an increase in the permeability of blood vessels, leakage of fluid into the tissues from the blood vessels and recruitment of polymorphonuclear leucocytes (granulocytes) to the affected area, with an incidental (passive) extravasation of red blood cells. This acute inflammation does not persist for more than a few days, passing on into one of two main pathways. In both there is recruitment of monocytes from the circulation and these cells become macrophages, cells which engulf (phagocytose) any dead tissue, foreign material or microorganisms, when they get into the tissues. At this stage, in the case of a tissue such as bone, fibroblasts are also recruited and new blood vessels are formed, the resultant combination of cells and their products being known as granulation tissue. It has been suggested that this should be called `vascularised undifferentiated mesenchymal tissue' in the context of bone healing, a term that acknowledges this tissue has the propensity to undergo osteogenesis but may, depending upon local conditions, form bone, cartilage and fibrous tissue. The fibroblasts form collagenous connective tissue initially. The revascularised area undergoes healing with new bone formation, the collagen being type I, the kind that is present in bone, and variable amounts of residual fibrous tissue are formed, depending on local conditions. The alternative to this progression is the recruitment of lymphocytes to the damaged area which occurs when the insult continues. The inflammation then persists and passes into the chronic phase, best defined as a form of continuing inflammatory response in the presence of an attempt at healing. This chronic inflammatory response in relation to an implanted prosthesis will be discussed in detail in Chapter 15. The present account will describe the different features that occur in relation to prosthetic components which are `healed' in place and are functioning normally. This chapter will consider the immediate reaction occurring on implantation of foreign prosthetic material, usually in bulk, and then describe the known responses to different implants when healed in place in bone. Factors influencing the long-term result include the material properties of the implant, such as modulus, surface roughness and porosity, as well as its chemistry, including bioactivity, and the local mechanical environment in which it is placed. Factors thus influencing the appearances of the tissue related to a stable implant include the different materials used, whether `inert' or bioactive, porous or smooth, as well as the site of implantation which influences the biomechanics and loading considerations.

14.2

Immediate response to prosthesis placement

There are few descriptions of the early response to implanted joint prostheses, for the obvious reason that little material exists. Revision of an arthroplasty in © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

317

the first week is extremely unlikely as death within the perioperative period is exceptional. Most animal studies will also not deal with the acute phase after implantation since they should be preceded by adequate in vitro cell culture experiments to exclude toxicity. Devices should not be placed experimentally in animals if there is any evidence of incompatibility from these cell culture studies, particularly if this toxic effect were to be manifest within a short time of implantation. Also, in terms of the legal and ethical requirement to keep numbers of animals being tested to a minimum at all times, there is usually little or no justification for studying the transient acute inflammatory stage of the reaction to an implanted material. An exception is in biocompatibility and toxicity studies of novel materials which have been screened successfully in vitro but need to be examined for the purposes of satisfying the safety criteria of regulatory affairs bodies. The basis of the description in this chapter is a longstanding personal experience observing the response to apparently normally functioning artificial joints retrieved at autopsy, as well as experimental animal implantation studies and the information available in the literature. These all show that an acute inflammatory phase occurs which is entirely in keeping with the general principles stated briefly above. Three examples of this early phase after implantation seen personally were from individuals dying at 3, 7 and 10 days postoperatively from causes unrelated to the surgical procedure and due to the onset of completely unpredictable and previously undiagnosed fatal other disease (myocardial infarction (two cases) and perforated diverticulitis (one case)). The samples of bone clearly showed the presence of fragmented bone trabeculae with a polymorphonuclear leucocyte infiltrate and extravasation of red cells (local haemorrhage) in the bone at its interface with the implant in samples available at 3 and 7 days. New bone formation was already visible at 7 days (Fig. 14.1). There

14.1 Bone immediately adjacent to a tibial prosthetic component 7 days after implantation. There is residual haemorrhage (red blood cells) (bottom right and bottom left, top left). A large piece of dead bone (centre left) shows the presence of appositional new bone being formed by osteoblasts. © 2008, Woodhead Publishing Limited

318

Joint replacement technology

14.2 Macrophages on the surface of Bioglass 6 days after experimental intraosseous implantation in the rabbit.

was an absence of red cells and granulocytes but new bone formation was present in samples obtained at ten days after implantation. Similar changes are seen in animals and have been noted personally in relation to titanium alloy (TiAlV), stainless steel, ultra-high molecular weight polyethylene (UHMWPE), PMMA, poly(ethylmethacrylate/butylmethacrylate), HA (used as a coating or a porous bone substitute material), bioactive glass, glass ionomer cement, several epoxy resins, polyurethane and a polyurethane±nanohydroxyapatite composite. A macrophage infiltrate occurs as part of the ongoing process after the acute phase, as described in Section 14.1. The presence of macrophages on the surface of biomaterials is considered by some as a beneficial if not essential part of the process of implant incorporation. It is certainly seen not only with inert but also with bioactive materials which subsequently become incorporated into bone, as illustrated for bioactive glass in Fig. 14.2. It is the subsequent reaction to different materials which may vary, according to the influences mentioned above. These aspects will be considered separately in subsequent sections which will deal with the materials finding most use in joint replacement.

14.3

Remodelling of bone around implants

After the initial inflammatory response in relation to an implanted prosthetic component, there follows a period of healing which commences with the formation of granulation tissue, that is, newly formed vascular fibrous tissue. Healing in any site continues with the formation of variable amounts of fibrous © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

319

14.3 Photograph of the surface of bone adjacent to a polyethylene peg showing a ridged appearance of smooth bone forming an inner cortex-like structure. The surrounding trabecular bone can be clearly distinguished. Macerated bone sample (from Revell, Pathology of Bone, 1986, Springer, Berlin, pp. 203±234, with permission).

tissue, with or without scar formation.2,3 The process is not basically different in bone, except that the connective tissue cells in this context are capable of differentiating in several different ways depending on local conditions, including oxygen tension and mechanical loading.3 The changes in relation to an implant are in many ways similar to those seen in fracture healing. Thus, where the fracture is well fixed, there is healing with intra-membranous osteogenesis to form a hard callous, this initial newly formed bone being altered later by remodelling of the area so that it comes to resemble the original bone in structure.3 In the case of a prosthetic device, the bone remodels to the shape of the device, as is shown in Fig. 14.3, where can be seen the assumption of the shape of a fixation peg used in a tibial component at the knee. The remodelling of bone around a flanged peg is clearly shown in the radiograph and histology of a human implant retrieval (Fig. 14.4). A cross-section through a different design of peg is shown in Fig. 14.5 in which it can be seen that there is fibrous tissue present between implant and remodelled bone. Where there is a failure of union of a bone fracture, the gap between the broken ends of bone is filled with fibrous tissue and if there is long-term mobility between the bone ends, cartilaginous change occurs. The bones thus may be capped by both fibrous tissue and cartilage with the formation of a pseudo-arthrosis.3 The presence of cartilage within the fibrous or osseous tissue of fracture callus is a sure sign of ineffective fixation and movement. The same is almost certainly true of the tissue around an implanted prosthesis. Next to a prosthetic joint component there is formation of fibrous tissue and new bone, the latter occuring both as appositional new bone on existing living © 2008, Woodhead Publishing Limited

320

Joint replacement technology

14.4 (a) Radiograph of a slice through the upper part of a tibia containing a flanged polyethylene peg on the underside of a tibial prosthetic knee component. Note the presence of fine bone trabeculae between some of the flanges of the peg. (b) Bone growth around a flanged polyethylene peg inserted into the tibia, showing the presence of fine bone trabeculae and bone marrow between the flanges of the peg. This is a similar to but not the same case as that shown in (a).

bone trabeculae and as woven bone within the vascular fibrous tissue between existing osseous structures. Any dead bone fragments may be progressively removed by osteoclasts but equally well may provide a scaffold on which new bone is deposited, acting therefore like bone graft fragments. There are some differences among species, but personal experience shows that new bone formation is present as soon as one week after implantation of biocompatible

14.5 Cross-section of bone which contained a metal peg (removed before sectioning). There is a thick fibrous tissue layer between implant and bone. The bone around the peg has become remodelled to form an inner `cortex' in continuity with the surrounding trabecular bone. © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

321

14.6 Fibrous tissue layer between a tibial plateau and bone. Note there are occasional macrophages (arrow) present on the surface of the tissue adjacent to the implant.

metal or polymer in people and animals. With time, the bone around the implant is remodelled so that the stable device is surrounded by bone from which it may be separated by fibrous tissue or cartilage (Figs 14.5, 14.6 and 14.7). Bone growth right up to the implant surface occurs with certain materials, for example, TiAlV alloy and commercially pure titanium (cpTi) (Fig. 14.8). While there may be the perception that a stable interface will show only bone with some intervening fibrous tissue in places, there is also evidence that cartilage formation occurs in the absence of frank loosening and movement of the implant within the bone. This is particularly true of those forms of cementless fixation where there is no

14.7 Chondroid change adjacent to an uncemented polyethylene implant, which was removed before sectioning (top). Note the underlying bone shows abundant new bone formation and has remodelled to form a thin subchondral bony end plate beneath the cartilage (from Blaha et al., J Bone Joint Surg 1982, 64B. 326±335, with permission). © 2008, Woodhead Publishing Limited

322

Joint replacement technology

14.8 (a) Bone growth in intimate contact with a TiAlV pin inserted into the intercondylar region of the lower femur of a rabbit. There is little fibrous tissue between this bone and the metal. Note the continuity of the surrounding circle of bone with the trabecular bone of the lower femur on the right. (b) Higherpower view of the interface between the same implant and bone, showing intimate contact with absence of fibrous tissue layer.

bone ingrowth (see Section 14.5). Much has been made in the past of a radiolucent line around an implant as seen on X-ray examination. This space between implant and bone is due to the presence of fibrous tissue, but radiolucent lines are seen over parts of the interface of implants that are not loose.2±4 The significance of radiolucent lines will be dealt with further in Chapter 15. © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

323

14.9 Loose fibrous tissue between bone and implant, showing the presence of small numbers of macrophages and multinucleate giant cells at the surface immediately related to the site of the implant (impl, space at top).

That there may be macrophages present on the implant side of any fibrous tissue has been reported even when the implant is well fixed5 (Figs 14.6 and 14.9). These may be present as isolated cells, or form a layer and have associated fibroblastic cells present to form a synovium-like membrane. The presence of this layer of cells is illustrated in Fig. 14.10. Both type A (macrophages) and type B cells (fibroblasts) are present. The relationship between these cells has been well described for the true synovium6±9 (Figs 14.10 and 14.11) and with respect to the interface between fibrous tissue and implant.10 This is a topic of considerable importance when mechanisms of loosening in relation to wear debris are considered in Chapter 15.

14.4

The cemented joint prosthesis

Since the early total joint replacements, fixation to the skeleton has been by the incorporation of a layer of so-called bone cement between implant and bone. This material, PMMA, remains in use to the present day. PMMA cures in situ, being adminstered as a dough made up of polymer and monomer, with dimethylparatoluidine and benzoyl peroxide in small amounts present to aid the polymerisation process and radiographic contrast materials (barium sulphate or zirconium dioxide) as well as antibiotics (gentamycin) also being incorporated. Details of acrylic cements are provided in Chapter 10. The polymerisation of the dough is an exothermic reaction and high temperatures are produced locally in the tissue. Evidence of bone death is seen in autopsy-retrieved tissues adjacent to cemented prosthetic joint implants in humans.2,3,11 Experimental implantation in the dog has also shown the presence of bone death for a short distance away from PMMA (Fig. 14.12a). This bone death is considered to be due to the heat of © 2008, Woodhead Publishing Limited

324

Joint replacement technology

14.10 (a) Formation of a multicellular layer adjacent to an implant, showing the resemblance of this layer of cells to the surface synovial cell layer of a joint, even though this tissue was situated deep within the bone. The black material is not metal debris but Indian ink applied to the surface in the laboratory to aid orientation on microscopy and indicate which side of the tissue was next to the implant.

© 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

325

14.10 (continued) (b) Staining for proline-4-hydroxylase, a fibroblast marker, by immunohistochemistry of the synovium-like layer next to an implant. The positive labelling appears black in this monochromatic picture. Note: a number of cells, located deep to the surface, contain this protein (arrows) which is a precursor of collagen.

14.11 Immunohistochemical staining of true synovial membrane for type IV collagen, showing the presence of this protein around the deep cells of the surface lining cell layer. Note there is no basement membrane present, but the basement membranes of a blood vessel is clearly labelled. Fibronectin, laminin, type IV collagen and various other proteins show this distribution in true synovium and in the synovium-like layer next to an implant. © 2008, Woodhead Publishing Limited

14.12 (a) Tissue adjacent to poly(methylmethacrylate) inserted into dog bone so as to cure in situ. Note dead bone (shown by the presence of empty osteocyte lacunae) with some new bone formation, 14 days after implantation. The trabecula is more or less surrounded by bone cement and may also have had a jeopardised blood supply. Not all the bone death is related to the exotherm because appositional new bone has also subsequently died. (b) Living bone and bone marrow adjacent to poly(ethylmethacrylate)-n-butyl methacrylate, cured in situ 7 days prior to recovery from dog bone. A fine line of newly formed bone is seen just below the surface (arrows). (c) Living original bone (with cells in osteocyte lacunae) and viable cellular bone marrow immediately adjacent to poly(ethylmethacrylate)-n-butyl methacrylate 14 days after implantation. This bone is too well developed to have formed in 14 days. The difference may be due to the lower exotherm of this material. © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

327

polymerisation of the PMMA. Very little or no necrosis of bone was found adjacent to poly(ethylmethacrylate)/n-butylmethacrylate (PEM/BMA) cement used experimentally in the same dog study (Fig. 14.12b,c) and this material has a much lower exotherm on polymerisation.12±14 There is some leaching of unpolymerised monomer from bone cement at the time of curing and this may have a local effect as well as the cardio-respiratory depressive effects reported in the clinical and experimental literature. The BMA monomer is less toxic than methyl methacrylate monomer as well as having a lower exotherm.12,13 More macrophages were present on the surface of PMMA than PEM/BMA implanted in rat muscle, a difference that may be related to differences in the surfaces of the two materials, since more macrophages are recruited to roughened or highly contoured surfaces (PMMA) than to smooth ones (PEM/BMA).13±15 While the term bone cement is used for PMMA in joint replacement, it should be emphasised that this material in no way acts as an adhesive between implant and bone. Naked-eye examination and histological studies of stable normally functioning prostheses retrieved at autopsy show that the polymer is extruded into the spaces between bone trabeculae as rounded or finger-like processes (Fig. 14.13a). These serve effectively to increase the surface area of the implant in contact with bone tissue, so providing more interlock in the short term after implantation, and a morphology suited to bone growth and keying in at the interface when the implant is healed into place and mechanically stable (Fig. 14.13b). That the total engulfment of bone trabeculae by PMMA may jeopardise their blood supply and give rise to bone necrosis is a possibility, though the author has observed viable trabeculae completely embedded in bone cement. That loss of blood supply rather than the heat of polymerisation is responsible for the bone death shown in Fig. 14.12a cannot be excluded, especially since there is evidence of death of appositional new bone, which cannot have been an event occurring at implantation. As in the case of pegs and cementless implant surfaces (see Section 14.3), new bone growth occurs around the implant and the bone is remodelled in response to the loads applied locally. Autopsy retrieval studies have illustrated that remodelling of the bone occurs around the cemented implant and that new bone forms around or adjacent to the cement.2,3,16±18 Viewed in cross-section, the bone forms a structure like a rim of thin bone around the cemented stem in the upper femur at the hip. It is considered that bone and implant are in places in direct contact though over most of the interface there is a fibrous tissue layer of variable thickness.17 Descriptions of the interface between bone and cemented joint prostheses can be found in the literature.17±19 On both sides of the hip, intimate contact of bone with cement without any interposed soft tissue has been observed even after 17.5 years of implantation.19 These studies all document the long-term compatibility with bone of cement in bulk form. Apart from issues around polymerisation and dispersion of monomer leached from the dough at implantation, the problems resulting from bone cement usage © 2008, Woodhead Publishing Limited

328

Joint replacement technology

14.13 (a) Magnified naked-eye appearance of the interface between bone cement and bone in a retrieval specimen from man. The bosselated-appearing bone cement (top) has been teased out of the surrounding fibrous tissue covering the trabecular bone (bottom) to illustrate graphically the relationship between bone and cement. (b) Low-power microscopical photograph to show the presence of finger-like processes extending from the bone cement (top) into the bone (bottom). There is extensive fibrous tissue in relation to the cement. Note cement is soluble in solvents used in tissue processing and its former presence is denoted by the large spaces. © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

329

relate to its fragmentation and the generation of wear debris, either in the implant bed, or through three body wear, when acrylic particles become interposed between the bearing surfaces of the prosthetic joint. These aspects are described in Chapter 15.

14.5

The uncemented prosthetic joint component

Endoprostheses may be inserted without the use of bone cement to fix them in bone. There are two main strategies to achieving this fixation, namely, the development of porous surface into which bone will grow and the use of coatings of bioactive material on which bone forms. More recently HA bioactive coatings have been applied to porous surfaces, combining both approaches. The following sections will deal with these different solutions.

14.5.1 Porous metal surfaces Alternatives to cemented fixation have been sought and used particularly for the younger and more physically active patient requiring joint replacement. One cementless method was the application of a so-called porous coating to the surface of the implant. Such a coating comprised sintered beads or a small intertwined wire-like structure (fibre mesh) attached at the surface of the implant and these coatings were made in cobalt±chrome or titanium alloy respectively. Clinical results have been good, though it is not completely clear the extent to which bone growth occurs between the beads or fibre mesh.20±22 The size of pores did not seem to make any difference to bone ingrowth in one study where this was evaluated.21 The author has personally had the opportunity to examine both types of porous coating and has observed bone growth between the beads and the meshlike structure (Fig. 14.14). While there may be ingrowth of bone, equally well in other areas on the same porous implants, there may be no bone ingrowth with only fibrous tissue present instead. This is the experience of some authors in the literature,23,24 while others report ingrowth of bone in nearly all the explanted prostheses of the various designs and types examined.25,26 The mesh design was superior to beads in one large series in which these were compared.27 The exact reason for this inconsistency in findings among different workers may be related to individual factors in individual cases, such as the precise placement of the prosthesis in relation to the load transfer in bone. There are also differences at different points on the surface of any given prosthesis, and these are presumed to relate to local factors, such as micromotion and stress shielding. It is likely that this is the explanation for bone ingrowth along part of a surface and its absence from other areas on the same implant, as has also been seen personally. It seems likely that a large mismatch between the material properties of a metal, for example cobalt±chrome and bone, may give rise to bone loss due to stress © 2008, Woodhead Publishing Limited

330

Joint replacement technology

14.14 (a) Compensated polarisation microscopy of an undecalcified plastic embedded section of bone adjacent to a porous coated implant. There is bone growth between the beads of CoCr alloy in this case. (b) Compensated polarisation microscopy of an undecalcified plastic embedded section of bone adjacent to a porous coated implant. In this case, the coating is of the fibre mesh type and made of TiAlV alloy. Bone ingrowth is clearly seen between the fibres of the coating.

shielding. By contrast, bone formation has been noted by the author in relation to the polished distal part of a cobalt±chrome femoral stem (Fig. 14.15a,b). Bone was not only remodelled around this stem but was in intimate contact with the metal at many sites around the circumference, as demonstrated by undecalcified plastic embedded sectioning and microscopy (Fig. 14.15c). This appearance may be compared with that seen next to titanium-based materials which are considered to be bioactive (as shown in Fig. 14.8). Clearly the response to an inert material with a high modulus such as CoCr (see Chapter 4) may be different © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

331

14.15 (a) Bone in direct contact with the polished distal part of a CoCr femoral stem component of a hip. This is a macerated sample of a cross section of the stem cut with an Exakt system saw. Separation of the stem from the bone was not possible by firm pressure, though a formal biomechanical `push-out' test of interface shear strength was not performed (from Freeman et al., J Arthroplasty 2003, 18, 224±226, with permission). (b) Higher-power view of the interface between metal and bone shown in (a). (c) Histological section of the bone growth onto the surface of the CoCr stem shown in (a) and (b). This is a plastic embedded undecalcified section with the implant in situ, prepared with an Exakt system. © 2008, Woodhead Publishing Limited

332

Joint replacement technology

depending on loading, surface finish and other factors. Ideas over inertness and bioactivity may not be as clear-cut as is sometimes suggested. The early versions of porous coated implants in which beads were used suffered from loss of the beads from the surface.28 This is not now seen with current porous coated devices. The findings when tissue adjacent to failed uncemented porous coated implants is examined will be discussed in Chapter 15, but the most frequent observation is the presence of wear debris which may be metal, polyethylene or both. This appearance is not different from that seen with cemented prostheses. Enhancement of bone ingrowth into porous implants and in experimental models has been achieved with the use of transforming growth factors (TGF), bone morphogenetic proteins (BMP) and insulin-like growth factors (IGF) released locally.29±32 The addition of an HA coating to the porous implant surface also increases bone ingrowth.21 Both these aspects are described in a subsequent section of this chapter.

14.5.2 Porous tantalum Recently, a porous tantalum material has been developed to enable bone ingrowth and osseo-integration. This material has found use on the acetabular side of the hip joint, as well as in spinal surgery. Bone ingrowth has been demonstrated histologically in humans and other animals.33±36 There were similar results found for percentage bone ingrowth with two different porosities of the tantalum device in dogs, and 80% ingrowth was seen after a year. When trabecular (porous) tantalum was used to augment severe acetabular defects in revision surgery, only one out of a total of 28 hips required further revision for recurrent instability at an average follow-up of 3.1 years.37 There was only one revision for aseptic loosening out of 60 consecutive patients (mean follow-up, 42 months) undergoing revision total hip replacement in which a porous tantalum uncemented acetabular cup was used.38 In another radiological study, gaps up to 5mm were filled with bone and no acetabular component had migrated after 24 weeks.39 All 86 components studied showed no radiolucent lines or osteolysis, and there were no dislocations or other complications.

14.5.3 Polymer pegs and screws There are interesting insights into the healing reaction around implants when tissue is examined adjacent to polymers inserted without a grouting of bone cement or a bioactive layer. The changes are also found in relation to large surfaces of polymers and metals in direct contact with bone tissue. Thus, bone remodels around a peg or screw, as described above (Section 14.3), but the tissue at the interface between material and bone is either bone, cartilage or fibrous tissue. These have a distinctive distribution, the cartilage being located on the uppermost aspect of the tissue between the threads of a large screw or the © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

333

14.16 High-power view of the tissue between the threads of a screw-shaped fixing peg showing the outgrowth of bone which is covered by fibrous tissue and cartilage. Note that the cartilage is situated on the upper part of the tissue, that is on the loaded surface in relation to the underside of the thread or flange of the screw.

horizontal short plates of a flanged peg, and therefore in relation to the underside of a loaded component.3 This may best be illustrated for a large flanged peg used to prevent rotation and lateral movement of a polyethylene tibial component at the knee, as seen in Fig. 14.4.40 This chondroid change in the fibrous tissue between implant and bone may also be seen at other sites, for example, on the surface of the tissue related to the underside of an uncemented screw-shaped fixing pin/peg (Fig. 14.16) or in areas under an uncemented tibial component at the knee, be it metal or polyethylene (the latter occurring, historically, when such plastic components were not metal backed with a tibial tray). Cartilaginous differentiation occurs wherever there is micromotion or loading between the hard implant material (metal or plastic) and hard tissue (bone) (see Fig. 14.7). It has also been observed in relation to the surface of plastic patellar components that are loaded at the knee.41

14.6

Bioactive surfaces on prostheses

The surface of the prosthetic component may be coated with a layer of hydroxyapatite, Ca10(PO4)6(OH)2, which is similar in chemical composition to the mineral component of bone. Bone grows directly into contact with this material so that there is chemical bonding and the coating is then said to be bioactive. The © 2008, Woodhead Publishing Limited

334

Joint replacement technology

coating is usually applied by a plasma spraying technique at high temperature. In the pioneering work of Furlong and Osborn,42 the coating was thick (200 m), though it subsequently has varied between 50 and 200 m depending on the implant design and manufacturer.43±45 Coatings over 80 m are prone to fragment, giving rise to local peri-implant reactions, as described in Chapter 15, so that thinner layers of 50±75 m are usually preferred. Some modern HA coatings are deliberately manufactured to include tricalcium phosphate as a component. Details of the optimum characterisitics of HA coatings such as percentage crystallinity and porosity as well as thickness have been provided elsewhere and need not be described further here.45 HA coating is also added to porous (fibre mesh) implant surfaces of the type described in Section 14.5.1, in which application it enhances bone ingrowth in the short term, though long term there were not clear differences between HA coated and uncoated porous implants when mechanical tests of bonding strength were used.46 The following description deals with the appearances of the interface between bone and the HA coating without discussing details of the early changes that are thought to occur, except to say that calcium and phosphate ions are considered to be released from the surface, providing locally high concentrations of these elements. Proteins (e.g., fibronectin, vitronectin) and integrins (e.g., 5 1, v 3) bind to the HA surface and promote osteoblast attachment, so enabling appositional bone formation.47 The formation of appositional bone in relation to HA appears to be no different from the process seen in normal bone formation and remodelling. Histological studies of well-fixed HA coated implants have been performed on various types of hip and knee joint replacements at autopsy, and details are provided in the review by Dumbleton and Manley.45 The duration of implantation of retrieved samples has ranged from 5 months to over 6 years and more than 50 examples have been reported. A number of experimental implantation studies in animals have also been carried out. The experience of the author is in line with the reported findings both at the hip and knee in retrieved samples in humans as well as in various animal implantation studies performed personally. The naked-eye appearances of bone growth onto an HA-coated titanium alloy (TiAlV) tibial tray are illustrated in Fig. 14.17. Undecalcified methyl methacrylate embedded sections prepared using an Exakt system allow visualisation of the relationship between implant, HA coating and attached bone. The typical appearance is shown in Fig. 14.18. Mature bone is in intimate contact with the HA coating and often forms long, foot-like areas of attachment as it spreads along the surface. There is usually evidence of continuing bone remodelling seen as active osteoblasts laying down osteoid and osteoclastic resorption lacunae on the surfaces of this bone. Osteoblastic activity on the HA surface itself and the presence of osteoclasts on this surface may be seen occasionally in retrieved implants from humans48 (Revell, unpublished personal observations) but are more likely to be seen in experimental implantation © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

335

14.17 (a) Naked-eye appearance of a retrieved tibial plateau which had an hydroxyapatite (HA) coating on its underside. Note the presence of a large amount of bone firmly fixed to the implant. (b) Cross-section of part of an HAcoated TiAlV alloy tibial tray from a knee replacement showing the firm attachment of bone over the whole surface. The section has been prepared as a macerated specimen after cutting with an Exakt saw. This is from the same case as shown in (a).

14.18 High-power view of bone in intimate contact with HA coating on a TiAlV prosthetic component in man. The HA is seen at the bottom of the picture. Note there are numerous osteoblasts forming appositional bone in the upper part of the picture, while more mature bone with osteocytes in lacunae is seen adjacent to the HA. Plastic embedded undecalcified section, toluidine blue stain. © 2008, Woodhead Publishing Limited

336

Joint replacement technology

14.19 Bone ongrowth with complete incorporation into the lower femur of a rabbit of an HA-coated TiAlV pin, 3 months after implantation. Plastic embedded undecalcified section stained with toluidine blue and viewed by compensated polarisation microscopy.

studies, where earlier stages in the process of incorporation are captured. Bone grows around the device and remodels to incorporate it in just the same way as with cemented and uncemented prosthetic components, as described above. This is illustrated in Fig. 14.19 which shows the formation of an inner cortex-like structure in intimate contact with the HA coating of a titanium alloy (TiAlV) peg in the lower femur of a rabbit. The amount of bone ongrowth, or apposition, varies in the different series in the literature between 30 and 80%. An early stage in the bone ongrowth is described in a single case dying three weeks after implantation by Bloebaum and colleagues, when 10% of the surface of the HA coated femoral component and 20% of the similarly coated acetabular component showed new bone formation.49 Rabbit implantation studies in our own laboratories have shown a gradual increase in bone contact over the first weeks after implantation and equilibrium being reached at 60% at 10 days to six weeks, with a consistent, and as yet unexplained further increase to 80% at three and six months.50 That there was no evidence of loss of HA coating in humans over time has been reported,51 though the general consensus seems to be that some thinning of the HA layer is seen when this is measured on retrieved prostheses and that there is loss of the coating in places so that bone, fibrous tissue or bone marrow is seen in direct contact with the underlying metal of the implant (Fig. 14.20).45,48 Comparisons have been made between HA-coated and other implants with respect to attachment. Increased attachment to the HA coating was found in one © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

337

14.20 Human implant retrieval specimen of well-fixed HA-coated TiAlV femoral stem, showing thinning and loss of HA coating in areas so that intertrabecular bone marrow and fatty fibrous tissue is in contact with the metal in one localised place (shown by double headed arrow). Note there is an osteoclast present just below this arrow. Plastic embedded undecalcified section, toluidine blue stained.

study when a comparison was made with porous coated implants.51 The implants were of the same design but had different surfaces. No significant difference was noted among the bone to implant contact ratios in the three zones of DeLee and Charnley on the acetabulum in one study, and the predominant areas of bone contact were near the rim of the acetabular cup and around the fixing spikes.52 That differences in loading may affect the amount of bone ongrowth is apparent when comparisons are made between implants or different areas on the same implant are compared. HA is able to promote bone formation across gaps of up to 1 mm when stable, and even of 0.5 mm in the presence of micromotion.53,54 The former reference shows how the addition of micromotion to a model of the bone response to an HA coated implant in the dog will give rise to fibrous tissue formation, while bone ongrowth occurs effectively when the device is stable. The effects of changes to the HA coating by the introduction of Mg ions by an ion beam embedding method have been reported by Howlett, Revell and their colleagues.55,56 This method physically implants magnesium ions into the most superficial 100 nm of the HA coating without changing the chemistry; that is to say, the implanted ions are not chemically bonded into the HA molecules but have an interstitial form of incorporation. Using otherwise identical implants of HA-coated metal cylinders in the rabbit femur, increased bone formation was found with Mg-implanted HA coatings compared with ordinary HA coatings. Moreover, the interfacial shear strength measured by the push-out test was greater for the magnesium implanted HA (Mg±HA)-coated cylinders.56 Bone growth into a slot in the side of cylinders was greater when the floor of the slot had an Mg±HA coating compared with an HA coating (Fig. 14.21).55 © 2008, Woodhead Publishing Limited

338

Joint replacement technology

14.21 Bone growth into a slot in the side of a cylindrical metal implant placed in the lower rabbit femur for six weeks. The floor of the slot is HA-coated and has additionally been ion beam implanted with magnesium ions. Plastic embedded undecalcified section, toluidine blue stained.

Other aspects of HA coating will not be considered further here. The mechanisms of failure and the effects of HA particles and larger fragments which have broken off the bulk material, as well as the possible role of delamination will be discussed in Chapter 15.

14.7

Adjunctive methods or treatments and their effects

While the above sections have dealt with the main bone responses to the implanted prosthesis, for completeness it is felt necessary to add a section on the role of other adjunctive measures that may be employed to ensure stable bone healing and defect repair. Thus, there may be large defects in the bone due to the underlying disease process which mitigate against joint replacement in the absence of some other strategy to provide sufficient bone for fixation of the device. A typical example is the occurrence of large pseudocysts, which form just deep to the articular surface in osteoarthritis and are part of the classical pathological appearances of this disease process. This may be a particular problem on the acetabular side of the hip joint at primary replacement surgery. The loss of bone may also be considerable at revision surgery where there is © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

339

aseptic loosening and osteolysis or where there is loosening due to infection which may also be very locally destructive. The enhancement of bone formation around the implant in these situations may be achieved by bone grafting or the use of bone substitute materials. Sometimes large cortical grafts are needed in the reconstruction of a joint at revision surgery. The tissue response to bone grafts will be outlined in the following section. A subsequent section will briefly describe synthetic bone substitute materials, mainly porous HA and HA used as large granules or particles.

14.7.1 Bone grafts in joint replacement The use of autograft bone to fill large defects in relation to joint replacement surgery remains the gold standard, as in other aspects of orthopaedics. Cancellous bone provides not only a scaffold, by way of the fragments of bone, but also bone marrow elements (depending on source, but certainly present when iliac crest is used) and osteogenic factors such as BMPs and transforming growth factor beta (TGF ). The repertoire of the tissue in response to bone graft is effectively the same as that seen in fracture healing and the response at the traumatised (drilled, sawn and reamed) implant bed described above (Sections 14.2 and 14.3). Thus there is likely to be some haemorrhage locally (assuming the bone at the graft site is itself viable), followed by inflammation and the formation of granulation tissue with the vascularisation which occurs in this process. Appositional bone formation occurs in relation to the fragments of dead bone which constitute the graft itself, while de novo bone formation within the collagenous fibrous tissue also occurs, as shown in Fig. 14.22. Resorption of the bone graft fragments, to a greater or lesser extent, and remodelling with new bone formation, gives rise to incorporation of the graft with an eventual bone morphology and orientation in line with the local mechanical requirements. This process may take up to a year to complete in humans, and even after this time it may still be possible to detect the original bone graft fragments by examination by polarised light microscopy, when collagen fibre lamellar patterns and reversal lines together with the presence of empty osteocyte lacunae are the tell-tale signs of the graft. Cortical bone when used in large blocks becomes incorporated only at the edges where there is an interface with the original bone. There is slow resorption of the lamellar bone followed by new bone formation where there is revascularisation. However, unlike cancellous bone, a cortical bone graft is a combination of necrotic bone from the donor site and living remodelled bone encroaching into this bone from the periphery. A further description is in the chapter by Nather in the book edited by him, which describes non-vascularised cortical grafts and also provides comparison with the work of others using vascularised grafts.57 One major limitation to the use of autologous bone is that there may not be sufficient bone available from the donor site(s) to fill large defects. Other © 2008, Woodhead Publishing Limited

340

Joint replacement technology

14.22 Appositional new bone formation in relation to a fragment of bone graft which has empty osteocyte lacunae. Note there is also fine lacy new woven bone formation in the lower part of the picture.

significant problems include chronic pain at the donor site, infection, nerve or blood vessel damage and local fracture. The use of allografts provides a means around these difficulties, since the bone may be either cadaveric or obtained at the time of orthopaedic surgery, most notably joint replacement, when the femoral head, for example, is retained to provide bone graft material. Details of bone allograft preparation, setting up a bone bank, and more clinical aspects are also provided in the book mentioned above.58 Safety issues include the need for screening of donors to exclude HIV, Hepatitis B, Hepatitis C and appropriate investigation of the material itself to exclude microbial contamination, followed by an appropriate quarantine period after screening. The histological appearances of the healing bone around an allograft are not different from those with an autograft. In a study performed personally by the author and colleagues, there were no differences in the bone ingrowth into identical defects in a peg inserted into the iliac crest between autograft, allograft and a 50/50 mixture of these two materials.59 New bone extended 2:04  0:89 mm in relation to allograft, 2:29  0:52 mm for allograft/autograft composite and 2:21  0:61 mm in relation to autograft bone. Furthermore the mean appositional rate (MAR) estimated with the aid of double tetracycline labelling was not significantly different in relation to any of the three different bone grafts, which also did not have an MAR different from the healing bone surrounding the peg. Large cortical allografts show a similar pattern of changes to cortical autografts, though hyalinisation of blood vessels occurs, resorption and living bone formation occur at a slower rate if at all60 and the graft remains a largely dead piece of bone. Immunological considerations occur when vascularised allografts are used since these will contain antigenic components, namely bone © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

341

marrow cells. Such a vascularised graft transplanted where there is a large histocompatibility difference will show rapid rejection, which is a vascular phenomenon involving immune-mediated damage to the endothelium followed by thrombosis of the vessel.61

14.7.2 Bone substitute materials HA can be used as a porous bone substitute material or in granular (particulate) form to provide a scaffold for new bone formation. HA has the disadvantage of being brittle. Bioactive glasses and glass ceramics are also useful in promoting bone formation. Bone is formed in apposition to these materials and de novo in the interstices between HA or bioactive glass particles.62±64 There is evidence from experiments in rabbit that the amount of bone growth into porous HA is affected by macroporosity, microporosity and chemical composition.65±71 Thus there are differences between 60%, 70% and 80% macroporous implants, the presence of microporosity enhances bone ingrowth, and the incorporation of magnesium and silicon independently increases bone formation. While the earlier synthetic materials were derived from animal bone which was deproteinised to leave the mineralised scaffold, the HA materials developed more recently are synthesised de novo mostly by precipitation and sintering, with control of the various aspects in terms of material properties (porosity, chemical composition) to provide a more reproducible product, which may be in the form of granules or blocks that are either dense or macroporous. Bone forms in apposition to the HA material in just the same way as it does to bone graft, growing into the porosities of a bulk material (Fig. 14.23) and onto the surfaces of granules (Fig. 14.24). New woven bone is seen between the granules of HA and bioactive glass and in contact with the material surface (Fig. 14.24). As with the HA coating on an implant, there is evidence of osteoclastic resoption of the HA when used either as a porous bulk material (Fig. 14.25) or as granules. Osteoblastic activity also continues so that the area in which the bone substitute is placed undergoes remodelling according to the local loading requirements.

14.7.3 Enhancement of bone formation Further enhancement of bone formation and implant integration can be brought about by the use of various growth factors. The histological appearances are not different morphologically from those reported above, but histomorphometry studies show increased rates of bone formation and amounts of bone formed. Damien showed the beneficial effect of using IGFs (rhIGFI, rhIGFII) locally in combination with porous HA in the rabbit.72,73 BMPs, osteogenic protein and TGF have all been examined and shown to promote bone formation. A source of references to this body of work can be found in the review by Damien and Revell.74 © 2008, Woodhead Publishing Limited

342

Joint replacement technology

14.23 Plastic embedded undecalcified section of porous hydroxyapatite implant placed experimentally in the lower femur of rabbit, showing extensive bone formation to fill the porosities (arrows).

14.24 Plastic embedded undecalcified section of Bioglass particles (bp) implanted experimentally into rabbit femur, showing new bone formation on and between the particles after 7 days. The mineralised bone shows as lacy black structures with uncalcified newly formed osteoid of woven bone appearing grey in colour in this monochromatic picture of a von Kossa stained section. © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

343

14.25 Bone covering a strut of a porous HA implant has been partly resorbed exposing the underlying biomaterials. An osteoclast is seen on the surface of the hydroxyapatite (small arrow) and there is evidence of previous osteoclastic resorption of the bone giving a scalloped appearance (large arrow).

14.8

Summary

This chapter has considered the responses of bone to the implantation of foreign materials used in joint replacement. These are present as large devices in equally large defects made in bone. The response in terms of initial inflammation followed by healing and the development of a stable relationship between the implant and bone has many aspects in common whatever the material, and there are likenesses to the changes in bone healing in other circumstances, for example, fracture healing. There may be direct contact between implant and bone, which, while this occurs mostly with bioactive and porous surfaces, is also sometimes seen with bioinert materials, namely metal alloys. Bone remodels to fit the shape of the implant in general and the response is influenced by local conditions, such as load bearing. Thus, the tissue next to the biomaterial of the prosthesis, or the fixing cement, may be bone, cartilage or fibrous tissue, and the latter may also show a layer of synovial cells closely similar to that lining the true synovial cavity. The bone has a large capacity to repair damage including fracturing of trabeculae and bone necrosis resulting at the time of implantation. Where there is insufficient bone to permit simple implantation, owing to the primary disease process or because of bone loss following previous replacement surgery, bone grafts and bone substitute materials may be used.

© 2008, Woodhead Publishing Limited

344

14.9

Joint replacement technology

References

1. www.utahhipandknee.com/history.htm 2. Revell PA: Tissue reactions to joint prostheses and the products of wear and corrosion, in Current Topics in Pathology, vol 71. Berlin, Springer-Verlag, 1982; pp. 73±102. 3. Revell PA: Necrosis and healing in bone, Chap 9 in Pathology of Bone, Berlin, Springer-Verlag, 1986; pp. 203±234. 4. Freeman MAR, Bradley GW, Revell PA: Observations upon the interface between bone and polymethylmethacrylate cement. J Bone Jt Surg 1982; 64B: 489±493. 5. Levack B, Freeman MAR, Revell PA: The presence of macrophages at the bonePMMA interface of well-fixed prosthetic components. Acta Orthop Scand 1987; 58: 384±387. 6. Revell PA: The synovial lining cells. Rheumatol Int 1989; 9: 49±51. 7. Pollock LE, Lalor P, Revell PA: Type IV collagen and laminin in the synovial intimal layer: an immunohistochemical study. Rheumatol Int 1990; 9: 277±280. 8. Revell PA, Al-Saffar N, Fish S, Osei D: Extracellular matrix of the synovial intimal cell layer. Ann Rheum Dis 1995; 54: 404±407. 9. Stevens CR, Mapp PI, Revell PA: A monoclonal antibody (Mab67) marks type B synoviocytes. Rheumatol Int 1990; 10: 103±106. 10. Lalor PA and Revell PA: The presence of a synovial layer at the bone-implant interface. An immunhistological study demonstrating the close similarity to the true syovium. Clinical Materials 1993; 14: 91±100. 11. Willert H-G, Semlitsch M: Problems associated with the anchorage of artificial joints. In: Chaldach M, Hofmann D, eds. Advances in Artificial Hip and Knee Joint Technology. Berlin, Springer-Verlag, 1976, pp. 325±346. 12. Revell P, Braden M, Weightman B, Freeman M: Experimental studies of the biological response to a new bone cement: II Soft tissue reactions in the rat. Clinical Materials 1992; 10: 233±238. 13. Revell P, George M, Braden M, Weightman B, Freeman M: Experimental studies of the biological response to a new bone cement: I Toxicity of n-butylmethacrylate monomer compared with methylmethacrylate monomer. J Mater Sci: Mater Med 1992; 3: 84±87. 14. Revell PA, Braden M, Freeman MAR: Review of the biological response to a novel bone cement containing poly(ethylmethacrylate) and n-butyl methacrylate. Biomaterials 1998; 19: 1579±1586. 15. Taylor SR, Gibbons DF: Effect of surface texture on the soft tissue response to polymer implants. J Biomed Mater Res 1982; 61: 997±1001. 16. Jasty M, Maloney WJ, Bragdon CR, Haire T, Harris WH: Histomorphological studies of the long-term skeletal responses to well fixed cemented femoral components. J Bone Joint Surg 1990; 71A: 1220±1229. 17. Schmalzried TP, Kwong LM, Jasty M, Sedlacek RC, Haire TC, O'Connor DO, Bragdon CR, Kabo JM, Malcolm AJ, Harris WH: The mechanism of loosening of cemented acetabular components in total hip arthroplasty. Analysis of specimens retrieved at autopsy. Clin Orthop Relat Res 1992; 274: 60±78. 18. Willert HG, Ludwig J, Semlitsch M: Reaction of bone to methacrylate after hip arthroplasty: a long-term gross, light microscopic, and scanning electron microscopic study. J Bone Joint Surg 1974; 56A: 1368±1382. 19. Schmalzried TP, Maloney WJ, Jasty M, Kwong LM, Harris WH: Autopsy studies of the bone-cement interface in well-fixed cemented total hip arthroplasties. J Arthroplasty 1993; 8: 179±188. © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

345

20. Bauer TW, Stulberg BN, Ming J, Geesink RGT: Uncemented acetabular components: histologic analysis of retrieved hydroxyapatite-coated and porous implants. J Arthroplasty 1993; 8: 167±177. 21. Bloebaum RD, Bachus KN, Rubman MH, Dorr LD: Postmortem comparative analysis of titanium and hydroxyapatite porous-coated femoral implants retrieved from the same patient. A case study. J Arthroplasty 1993; 8: 203±211. 22. Collier JP, Mayor MB, Chae JC, Surprenant VA, Surprenant HP, Dauphinais LA: Macroscopic and microscopic evidence of prosthetic fixation with porous-coated materials. Clin Orthop Relat Res 1988; 235: 173±180. 23. Cook SD, Scheller AD, Anderson RC, Haddad RJ Jr.: Histologic and microradiographic analysis of a revised porous-coated anatomic (PCA) patellar component. A case report. Clin Orthop Relat Res 1986; 202: 147±51. 24. Cook SD, Barrack RL, Thomas KA, Haddad RJ Jr.: Quantitative analysis of tissue growth into human porous total hip components. J Arthroplasty 1988; 3: 249±262. 25. Engh CA, Zettl-Schaffer KF, Kukita Y, Sweet D, Jasty M, Bragdon C: Histological and radiographic assessment of well functioning porous-coated acetabular components. A human postmortem retrieval study. J Bone Joint Surg 1993; 75A: 814±824. 26. Pidhorz LE, Urban RM, Jacobs JJ, Sumner DR, Galante JO: A quantitative study of bone and soft tissues in cementless porous-coated acetabular components retrieved at autopsy. J Arthroplasty 1993; 8: 213±225. 27. Jasty M, Bragdon CR, Haire T, Mulroy RD Jr, Harris WH: Comparison of bone ingrowth into cobalt chrome sphere and titanium fiber mesh porous coated cementless canine acetabular components. J Biomed Mater Res 1993; 27: 639± 644. 28. Ranawat CS, Johanson NA, Rimnac CM, Wright TM, Schwartz RE: Retrieval analysis of porous-coated components for total knee arthroplasty. A report of two cases. Clin Orthop Relat Res 1986; 209: 244±248. 29. Goodman SB, Song Y, Chun L, Regula D, Aspenberg P: Effects of TGF on bone ingrowth in the presence of polyethylene particles. J Bone Joint Surg 1999; 81B: 1069±1075. 30. Sumner DR, Turner TM, Purchio AF, Gombotz WR, Urban RM, Galante JO: Enhancement of bone ingrowth by transforming growth factor-beta. J Bone Joint Surg 1995; 77A: 1135±1147. 31. Bostrom MP, Aspenberg P, Jeppsson C, Salvati E: The enhancement of bone ingrowth using bone morphogenetic protein 2. Clin Orthop Rel Res 1998; 350: 221± 228. 32. Damien E, Revell PA: Enhancement of the bioactivity of orthopaedic biomaterials: role of growth factor, ion substitution and implant architecture, Chap 29 in Bone Grafts and Bone Substitutes ed A Nather, World Scientific, New Jersey, 2005; pp. 459±488. 33. Adams JE, Zobitz ME, Reach JS Jr, An KN, Lewallen DG, Steinmann SP: Canine carpal joint fusion: a model for four-corner arthrodesis using a porous tantalum implant. J Hand Surg [Am] 2005; 30: 1128±1135 34. Barrere F, van der Valk C, M Meijer G, Dalmeijer RA, de Groot K, Layrolle P: Osteointegration of biomimetic apatite coating applied onto dense and porous metal implants in femurs of goats. J Biomed Mater Res B 2003; 67: 655±665. 35. Bobyn JD, Stackpool GJ, Hacking SA, Tanzer M, Krygier JJ Characteristics of bone ingrowth and interface mechanics of a new porous tantalum biomaterial. J Bone Joint Surg 1999; 81B: 907±914. © 2008, Woodhead Publishing Limited

346

Joint replacement technology

36. Bobyn JD, Poggie RA, Krygier JJ, Lewallen DG, Hanssen AD, Lewis RJ, Unger AS, O'Keefe TJ, Christie MJ, Nasser S, Wood JE, Stulberg SD, Tanzer M: Clinical validation of a structural porous tantalum biomaterial for adult reconstruction. J Bone Joint Surg 2004; 86-A (Suppl 2): 123±129. 37. Sporer SM, Paprosky WG: The use of a trabecular metal acetabular component and trabecular metal augment for severe acetabular defects. J Arthroplasty 2006; 21: 83± 86. 38. Unger AS, Lewis RJ, Gruen T: Evaluation of a porous tantalum uncemented acetabular cup in revision total hip arthroplasty: clinical and radiological results of 60 hips. J Arthroplasty 2005; 20: 1002±1009. 39. Macheras GA, Papagelopoulos PJ, Kateros K, Kostakos AT, Baltas D, Karachalios TS: Radiological evaluation of the metal±bone interface of a porous tantalum monoblock acetabular component. J Bone Joint Surg 2006; 88B: 304±309. 40. Blaha JD, Insler HP, Freeman MAR, Revell PA, Todd RC: The fixation of proximal tibial polyethylene prosthesis without cement. J Bone Joint Surg 1982; 64B: 326± 335. 41. Freeman MAR, MacInnes T, Revell PA: The histology of `reactive lines' in wellfixed components. J Arthroplasty 2003; 18: 224±226. 42. Furlong RJ, Osborn JF: Fixation of hip prostheses by hydroxyapatite ceramic coatings. J Bone Joint Surg 1991; 73B: 741±745. 43. Konttinen YT, Zhao D, Beklen A , Ma G, Takagi M, KivelaÈ-rajamaÈki M, Ashammakhi N, Santavirta S: The microenvironment around total hip replacement prostheses. Clin Orthop Rel Res 2005; 430: 28±38. 44. Geesink RG: Osteoconductive coatings for total joint arthroplasty. Clin Orthop Rel Res 2002; 395: 53±65. 45. Dumbleton J, Manley MT: Hydroxyapatite-coated prostheses in total hip and knee arthroplasty. J Bone Jt Surg 2004; 86A: 252±254. 46. Dean JC, Tisdel CL, Goldberg VM, Parr J, Davy D, Stevenson S: Effects of hydroxyapatite tricalcium phosphate coating and intracancellous placement on bone ingrowth in titanium fiber metal implants. J Arthroplasty 1995; 10: 830±838. 47. Kilpadi KL, Chang PL, Bellis SL: Hydroxylapatite binds more serum proteins, purified integrins, and osteoblast precursor cells than titanium or steel. J Biomed Mater Res 2001; 57: 258±267. 48. Bauer TW, Geesink RC, Zimmerman R, McMahon JT: Hydroxyapatite-coated femoral stems. Histological analysis of components retrieved at autopsy. J Bone Joint Surg 1991; 73A: 1439±1452. 49. Bloebaum RD, Merrell M, Gustke K, Simmons M: Retrieval analysis of a hydroxyapatite-coated hip prosthesis. Clin Orthop Rel Res 1991; 267: 97±102. 50. Zhang XS, Revell PA, Evans SL, Tuke MA, Gregson PJ: In vivo biocompatibility and mechanical study of novel bone-bioactive materials for prosthetic implantation. J Biomed Mater Sci 1999; 46: 279±286. 51. Coathup MJ, Blunn GW, Flynn N, Williams C, Thomas NP: A comparison of bone remodelling around hydroxyapatite-coated, porous-coated and grit-blasted hip replacements retrieved at post-mortem. J Bone Joint Surg 2001; 83B: 118±232. 52. Tonino A, Oosterbos C, Rahmy A, TheÁrin M, Doyle C: Hydroxyapatite-coated acetabular components histological and histomorphometric analysis of six cups retrieved at autopsy between three and seven years after successful implantation. J Bone Joint Surg 2001; 83A: 817±825. 53. Soballe K: Hydroxyapatite ceramic coating for bone implant fixation. Mechanical and histological studies in dogs. Acta Orthop Scand Suppl 1993; 255: 1±58. © 2008, Woodhead Publishing Limited

The healing response to implants used in joint replacement

347

54. Soballe K, Overgaard S, Hansen ES, Brokstedt-Rasmussen H, Lind M, Bunger C: A review of ceramic coatings for implant fixation. J Long Term Eff Med Implants 1999; 9: 131±151. 55. Zhang XS, Revell PA, Evans P, Tanner KE, Howlett CR: Magnesium-ion implantation of HA-coated implants enhances bone ingrowth in rabbits. 24th Annual Meeting Society for Biomaterials, San Diego, USA. 22±26 April 1998. 56. Revell PA, Damien E, Zhang XS, Evans P, Howlett CR: The effect of magnesium ions on bone bonding to hydroxyapatite coating on titanium alloy implants. Key Eng Mater 2004; 254±256: 447±450. 57. Nather A: Healing of large, non-vascularised, cortical autologous bone transplants: an experimental study in adult cats, Chap 7, in Bone Grafts and Bone Substitutes; Basic Science and Clinical Applications, Nather A (ed.), World Scientific; New Jersey, pp. 119±136. 58. Nather A (ed.) Bone Grafts and Bone Substitutes; Basic Science and Clinical Applications. World Scientific; New Jersey. 59. Samuelson KM, Freeman MAR, Levack B, Rassmussen GL, Revell PA: Homograft bone in revision acetabular arthroplasty. J Bone Joint Surg 1988; 70B: 367±372. 60. Enneking WF, Mindel ER: Observation on massive retrieved bone allografts. J Bone Jt Surg 1991; 73A: 1123±1142. 61. Kumta SM, Leung PC, Fu LK: Bone allotransplantation: future directions, Chap. 14, in Bone Grafts and Bone Substitutes; Basic Science and Clinical Applications, Nather A (ed.), World Scientific; New Jersey, pp. 243±254. 62. Mushipe MT, Revell PA, Shelton JC: Cancellous bone repair using bovine trabecular bone matrix particulates. Biomaterials 2002; 23: 365±370. 63. Mushipe MT, Revell PA, Shelton JC: The effects of bovine trabecular bone matrix particulates on cortical bone repair. J Mater Sci Mater Med 2002; 13: 99±105. 64. Revell PA, Damien E: The need for new materials for use in bone in man, Chap 27 in Bone Grafts and Bone Substitutes; Basic Science and Clinical Applications, Nather A (ed.), World Scientific; New Jersey, pp. 431±443. 65. Hing KA, Best SM, Tanner KE, Bonfield W, Revell PA: Quantification of bone ingrowth within bone-derived porous hydroxyapatite implants of varying density. J Mat Sci Mater Med 1999; 10: 663±670. 66. Hing KA, Revell PA, Smith N, Buckland T: Effect of silicon level on rate, quality and progression of bone healing within silicate-substituted porous hydroxyapatite scaffolds. Biomaterials 2006; 27: 5014±5026. 67. Hing KA, Best SM, Tanner KE, Bonfield W, Revell PA: Mediation of bone ingrowth in porous hydroxyapatite bone graft substitutes. J Biomed Mater Res A 2004; 68: 187±200. 68. Hing K, Annaz B, Saeed S, Revell P, Buckland T: Microporosity enhances bioactivity of synthetic bone graft substitutes. J Mater Sci Mater Med 2005; 16: 467±475. 69. Revell PA, Damien E, Zhang XS, Evans P, Howlett CR: The effect of magnesium ions on bone bonding to the hydroxyapatite coating on titanium alloy implants. Bioceramics 2003; 16: 447±450. 70. Damien E, Hing K, Saeed S, Revell PA: A preliminary study on the enhancement of the osteointegration of a novel synthetic hydroxyapatite scaffold in vivo. J Biomed Mater Res A 2003; 66: 241±246. 71. Patel N, Best SM, Bonfield W, Gibson IR, Hing KA, Damien E, Revell PA: A comparative study on the in vivo behaviour of hydroxyapatite and silicon substituted hydroxyapatite granules. J Mater Sci Mater Med 2002; 13: 1199±1206. © 2008, Woodhead Publishing Limited

348

Joint replacement technology

72. Damien E, MacInnes T, Revell PA: In vivo effects of insulin like growth factor-II on de novo bone formation in the presence of hydroxyapatite in rabbit femur. Bone 2001; 28: S140. 73. Damien E, Hing K, MacInnes T, Revell PA: Insulin like growth factor-I (IGF-1) increases the bioactivity of porous hydroxyapatite (PHA) in vivo in rabbits. J Pathol 2001; 193: 6A.

74. Damien E, Revell PA: Enhancement of the bioactivity of orthopaedic biomaterials: role of growth factors, ion substitution and implant architecture, Chap 29 in Bone Grafts and Bone Substitutes; Basic Science and Clinical Applications, Nather A (ed.), World Scientific; New Jersey, pp. 459±488.

© 2008, Woodhead Publishing Limited

15

Biological causes of prosthetic joint failure

P A R E V E L L , University College London, UK

15.1

Introduction

Joint replacement using indwelling prosthetic components has been one of the major advances in medicine in terms of success and promoting well-being in those with severely debilitating disease. Failure of a replacement joint with loosening can occur because of infection or for a variety of reasons in the absence of infection. It is virtually impossible to separate mechanical and altered load-bearing effects from those related to the presence of large amounts of wear debris once the process of aseptic loosening is well advanced. It is not only impossible to put a figure on the proportion of total joint replacements that are successful over 10 or more years, but also difficult to give a generalisation for the incidence of aseptic loosening. Clearly the failure of any prosthetic joint will depend on the particular joint being replaced (anatomical site), as well as the design and the individual surgeon. This said, it seems likely from a broad view of the literature that 90±95% of hip replacements are successful for 10±15 years, while knees are a little less successful than this in terms of 10-year results. Inevitably, a proportion of synthetic joints fail and this is for a variety of reasons. There may be fracture of the implant or of the bone into which it is implanted, or dislocation of the joint. Stress shielding is the process in which load, and therefore stress, is redistributed from the remaining upper part of the femur when a femoral component of a total hip replacement is present. Thus the upper part of the cortex is bypassed and load is carried through the metal stem of the implant. That all the load on a joint and related bone may be carried through the implant may also give rise to local osteoporosis, a form of disuse atrophy, in the lateral residual trochanteric part of the femur. Further mechanical effects include micromotion, a process in which movement at a micrometre scale gives rise to a local tissue response. The main biological causes of joint implant loosening are infection and loosening in the absence of infection. Such aseptic loosening is the largest single reason for joint revision surgery. Another biological question to be considered is whether the presence of foreign material in bulk or particulate form may give rise to the development of tumours or © 2008, Woodhead Publishing Limited

350

Joint replacement technology

developmental abnormalities (in younger individuals having a joint replacement while still of child-bearing age). Such matters are of concern to the regulatory affairs agencies in different parts of the world, and are on regular review. The topic will not be discussed in this chapter as the possibility of developing cancer or causing a genetic abnormality are not per se failures of the implant. The following description will be confined to the biological causes of prosthetic joint failure. It should be noted also that the various mechanisms are not necessarily mutually exclusive. Thus, stress shielding, micromotion and the generation of an excessive amount of wear debris with its accompanying cellular reaction may all be present and contribute to the overall appearance in any individual. Mullhall and colleagues recently analysed the reasons for the failure of 318 knee replacements and stated that 64.4% had more than one cause identifiable.1 Mechanical factors and details of the tribology of joints will not be described in this chapter as they are well described by other contributors (in particular, Chapters 1, 2, 12 and 18). In terms of the number of individuals coming to revision surgery, aseptic loosening is by far the largest group. A major part of this chapter will be devoted therefore to this topic. Infection will also be covered.

15.2

Infection

The percentage of cases developing infection, after joint replacement, has decreased considerably since the early days of the procedure, when it was 10%, down to values of around 1% in several series.2,3 This improvement is mostly related to the use of prophylactic antibiotics, laminar flow facilities and other precautions taken at the time of implantation. After aseptic loosening, which is considered in detail elsewhere in this chapter, infection is the single most difficult and clinically challenging complication of joint replacement surgery. Infections may be acute and fulminating, usually occurring within a month of primary arthroplasty, or they may be indolent, becoming obvious only after many months, and manifesting clinically as progressive pain. Those occurring early are assumed to arise as a result of direct infection, at the time of surgery, or shortly afterwards from the wound or any drain that may have been used. A cutoff time of three months has been suggested to differentiate between acute and chronic infections,4 though this may be a somewhat artificial and arbitrary separation. Some chronic infections according to this definition may also arise at the time of surgery but take longer to develop. Haematogenous infection may result several years after surgery, and is due to spread of organisms from a source remote from the replaced joint, for example, dental, dermal, urinary or respiratory tract sites.3±5 There needs only to be a bacteraemia, not a septicaemia, for haematogenous infection to occur. Contributing factors to this form of deep infection are the coexistence of disease such as rheumatoid arthritis, or © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

351

old age, or corticosteroid treatment, in all of which there is some depression of immune function.6 Where the classical signs of infection are present, including severe joint pain, fever, chills or a draining sinus onto the skin, there should be no problem in recognition of periprosthetic infection. But the condition is more often gradual in onset and the symptoms non-specific, making differentiation from aseptic loosening, a haematoma in the joint or a mechanical problem such as instability extremely difficult.3 Under these circumstances, a number of investigations is required, namely X-ray examination and blood tests, to include full white cell count (WCC), erythrocyte sedimentation rate (ESR) and C reactive protein (CRP) level.3,7,8

15.2.1 Blood tests in the diagnosis of periprosthetic infection In fulminant septic arthritis the WCC will be elevated, but it may be normal or only slightly raised where the infection is lower grade and indolent.7,8 There is a much stronger chance of diagnosing infection successfully if ESR and CRP are both elevated (ESR > 30 mm/hr; CRP > 10 mg/litre) even though each of these is a non-specific test. ESR may take up to a year to return to normal after major surgery such as joint replacement.9 The values of both can be raised in various other inflammatory conditions. If only one of these test results is raised then procedure to joint aspiration is suggested according to an algorithm provided by Urban and Garvin7 citing work by Spangehl and colleagues.10 Both ESR and CRP may be elevated in the immediate post-operative period for reasons other than the surgery itself, such as a postoperative chest infection like bronchopneumonia. Up-to-date discussions of these aspects and the so-called sensitivity and specificity of these test methods are provided in the contribution of Bauer and colleagues3 and that of Revell et al.8 The ability to detect a case that really is infected is referred to as the sensitivity, while the ability to exclude cases that are not infected is the specificity. Both are expressed as an index number less than one or as a percentage. Serum levels of the cytokine interleukin 6 (IL6) have been measured and found to be elevated by Di Cesare and colleagues11 in individuals with deep periprosthetic infection. A high level of discriminatory function was claimed for this test, but IL6 is produced in various conditions other than infection and serum levels may also therefore be raised in these. Serum IL6 levels are elevated for up to 3 days after surgery.12

15.2.2 Imaging methods Radiological study is useful to exclude other pathological processes such as marked osteolysis or a fracture. There are no specific features relating to infection in and around prosthetic joints. Imaging with radio-isotopes is similarly © 2008, Woodhead Publishing Limited

352

Joint replacement technology

non-specific. A discussion of this topic is provided in the review by Bauer and his colleagues.3 Briefly, technetium-99m, indium-111 and gallium-67 may all be used, and since they subserve different functions, for example, indium-111 labels white cells while gallium-67 is bound to serum transferrin, better results are obtained when these methods are combined and when the findings are considered alongside the ordinary radiological appearances. Positron emission tomography using fluorodeoxyglucose (FDG-PET scan) is under evaluation and may provide a reliable means of imaging an infective focus.3

15.2.3 Joint fluid examination Joint aspiration may aid diagnosis, but has been called into question as a reliable procedure. This aspect is discussed clearly by Bauer and colleagues,3 who refer to two previous studies by Barrack and coworkers13,14 which gave conflicting results. The studies were separated by four years and the first was on the hip with the second on the knee. The predictive value for hip aspirate examination was 15% while that for the knee was 85%. Differences may be a result of the differences in obtaining aspirate from these two joints, false positives being far commoner in hip aspirates. When Bayesian statistical methods are used and prevalence is considered as a part of the equation, a different result is obtained, since the predictive value for knees changes to 75%, but that for the hip remains at 15%.3 Gram stains for microorganisms on synovial fluid have been used but are considered to have poor sensitivity.3,6,7,15,16 Coming to the examination of the cells present in synovial fluid on aspiration or at revision surgery, there is some confusion as to the values above which infection can be diagnosed. In the diseased joint not having an implant in place, a raised WCC of up to 50 000 cells/mm3 is present in non-infectious inflammatory disorders such as crystal arthropathy (gout, calcium pyrophosphate dehydrate (CPPD) deposition) and acute flares of rheumatoid arthritis while a total count of 200 000±500 000 cells/mm3 may be seen in septic arthritis. Between 65% and 85% polymorphonuclear leucocytes are found in the differential count in rheumatoid arthritis, gout and pseudogout. It is important not to rely on the differential count alone but to bear in mind the total cell count as well. Thus, a differential count for neutrophil polymorphonuclears of 65% or greater set against a 200 000 (or greater) total count is highly likely to be due to infection.16 Work by Mason et al.17 and Trampuz et al.18 suggests that high values for synovial fluid total white cells with over 60% neutrophils are suitable criteria for the diagnosis of infection in relation to an implant. Further references to this subject are available in the article by Urban and Garvin7 and that of Bauer et al.3 From a practical point of view, it is important to ensure adequate mixing of synovial fluid samples with diluent, particularly where the fluid is viscous. Secondly, the acetic acid containing diluent used in haematological cell counting © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

353

precipitates the proteins present in synovial fluid and gives rise to falsely depressed cell counts, so that saline should be used for the dilution of synovial fluid.16

15.2.4 Laboratory examination of tissue samples: histopathology The histopathologist has a role to play in assisting the surgeon in the recognition of infection at the time of revision surgery. This is by evaluating the extent to which there is a neutrophil infiltrate in the interface tissues using frozen sections and conventional light microscopy. However, the criteria for the identification of infection have not been clearly defined by this method. The presence of polymorphonuclear leucocytes is always an indicator of acute or active inflammation, but determining how many cells need to be seen for their recognition to be of any predictive use is more difficult. Values of five polymorphonuclear cells per high-power field (HPF) in each of five such fields are described19,20 while, for others, one cell present in each of ten HPF was acceptable for the diagnosis of infection.21 Confusion results when the same researchers provide different results and altered criteria in subsequent papers, as has occurred with two groups of workers, details of which are provided elsewhere.3 Clear criteria are required where cell counting is being performed as part of a routine screening procedure. They are less important when a frozen section diagnosis is called for only when there is a suspicion of infection, perhaps after the surgeon is well into the operative procedure. It has been this author's experience to work in this way, and although cell counts have been performed, the report to the surgeon has given a confirmation, or otherwise, of an intra-operative impression. All too often, the role of the histopathologist is seen as that of performing a test, when more correctly his/her function is to provide an opinion on the basis of the evidence available from inspecting the changes in tissue down the microscope. Cell counting, in this type of response, is an adjunct to the reaching of a diagnostic opinion, not the answer that it would be considered to be when examining peripheral blood (cf. the peripheral blood white cell count). In all histopathological diagnosis, the frozen section appearances need afterwards to be compared with paraffin wax-embedded sections both from the frozen tissue block and from other material. The paraffin wax sections provide a clearer picture and give the opportunity to examine a larger sample of tissue. Fehring and McAlister22 compared the intra-operative frozen section result with that available after examination of all the formalin-fixed routinely paraffin-wax embedded material. The specificity changed little (89.5% to 86%) but the sensitivity was radically improved when fixed material was studied rather than frozen sections (18.2% to 82%), the disadvantages of rapid diagnosis and sampling error thus being manifest. While this study is over a decade old, it does point out the difference between frozen section and routine histological © 2008, Woodhead Publishing Limited

354

Joint replacement technology

diagnosis in terms of specificity. Another aspect of this and some other studies in this area is that the findings of microbiological culture of joint fluid and/or tissue obtained at the time of surgery are taken as the standard for determining whether any case was an example of infection. Nearly all work in this area has been performed as part of a retrospective review of clinical cases. However, a prospective study of the various diagnostic tests and their specificities and sensitivities, as well as the positive and negative predictive values, has been performed by Spangehl and his colleagues on 202 cases.10 These authors conclude that the combination of normal ESR and CRP values is reliable for predicting the absence of infection, that where either of these blood tests shows an elevated level, aspiration should be performed, and that intra-operative frozen section is useful where it remains equivocal as to whether infection is present on the basis of blood tests.

15.2.5 Laboratory examination of tissue samples: microbiology At the time of revision surgery, tissue should be sent not only for histological study, but also for microbiological examination. The literature is confusing as to the meaning and interpretation of series assessing culture for microorganisms of tissue obtained intra-operatively. The results from different studies are summarised in the review article by Bauer and colleagues.3 Between 1.9% and 5.8% of cases were considered to show negative culture results yet be infected in four of these reviewed studies, while there was an extremely wide variation (5.3± 90%) among six other contributions in respect of cases that were culture positive but where this was considered to be due to contamination. Thirty per cent of 142 hips treated with revision arthroplasty had at least one positive intra-operative culture, but only one of these developed a clinically important infection in the series reported by Padgett et al.,23 indicating a high percentage of false positives. The prospective analysis by Spangehl et al.10 showed that microbiological culture of tissue obtained at operation on 180 hips had a mean sensitivity value of 0.94 and a specificity of 0.97 with positive and negative predictive values of 0.77 and 0.99 respectively. These authors also looked at the results of swabbing the prosthesis in 168 of the cases, which resulted in lower mean sensitivity and higher positive predictive values, though the 95% confidence limits for these results changed very little. An investigation by Tunney and colleagues24 compared the results of culturing operative tissue samples with those from material obtained from the implant surface by sonication in a sterile environment. Organisms were cultured from tissue in 5 out of 120 cases, while sonication yielded 26 positives, with the same bacteria grown where both samples were positive. A similar smaller study by Neut et al.25 compared the results of culturing tissue with those using scrapings from the implant surface in 26 cases. Bacteria were grown from 41% of the tissue samples and this increased to 64% when prolonged culture was used, but extensive culture of surface scrapings © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

355

yielded an even higher value of 86% positive. The presence of a biofilm was detected on the implant surface and the potential role of this in providing anchorage and protection for the microorganisms is described. The aggregation of microorganisms together with the matrix they have secreted is referred to as a biofilm. While lipopolysaccharide (endotoxin) is produced by Gram-negative organisms, Gram-positive bacteria also make a glycocalyx or slime layer which is protective.8,26 It is these bacteria, particularly Staphylococcus aureus, Staphylococcus epidermidis and Pseudomonas species, which are prevalent in periprosthetic infections. Differences have been suggested in the susceptibility of different metal implant materials to biofilm formation with titanium alloy considered to have lower infection rates than stainless steel.27,28 While these observations are interesting, they have been performed in animals with fracture fixation devices and no details of the surface roughness is given. Until such time as different metals with the same surface roughness (ra value) are compared with respect to susceptibility to infection, a clear picture of the effect of one metal compared with another cannot be given. While a considerable period has passed since it was written, the reader would do well to see the paper by Gristina, which deals with many of the different aspects of the interaction between different materials and bacteria, including biofilms, and includes sections on different metals.29 Costerton describes how eight out of ten cases of supposed aseptic loosening showed the presence of bacterial biofilms on the prosthetic components even though they had never shown positive bacterial cultures from either aspirates or from removed devices. He goes so far as to suggest that large numbers of lowgrade biofilm infections may be misdiagnosed as aseptic loosening and are revised without the antibacterial precautions that would be used if they were correctly attributed.30 The question also arises as to whether endotoxin associated with wear particles may play a role in aseptic loosening. In this respect, the presence of significant levels of adherent endotoxin has been demonstrated on commonly used preparations of titanium particles as well as on titanium and titanium±alloy implant surfaces.30 Molecular methods are increasingly finding a use in microbiology and are of real relevance in relation to infections around implanted joint prostheses. The polymerase chain reaction (PCR) employs appropriate primers to amplify bacterial DNA, and the most common target for bacterial identification is the 16S rRNA gene, present in nearly all species of bacteria.31±33 Examining preoperative synovial fluid aspirates using PCR yielded 32 out of 50 specimens positive for bacterial infection in the study by Mariani and colleagues.32 Standard microbiological culture performed on the same samples gave only six positive results for the presence of bacteria, and culture of intra-operative specimens identified nine additional infections. The work of Tunney et al.24 using sonication has already been mentioned. In a subsequent paper, this group examined the same 120 sonicates from hip implants at revision arthroplasty by © 2008, Woodhead Publishing Limited

356

Joint replacement technology

PCR and 72% of the samples were considered to be positive. This may be a reflection of a recurring problem with PCR used in this way for microbiological diagnosis, namely the apparently high prevalence of false-positive results. In relation to orthopaedic devices and infection, positive results have been obtained in cases otherwise considered to be aseptic loosening by Clarke and colleagues.33 These authors compared tissue obtained at primary surgery (21.4% PCR positive) with that from near to aseptically loosened implants (46% PCR positive), and concluded, because of the high levels in the primary surgeryderived material, that the PCR has poor specificity for diagnosing infection in revision total hip arthroplasty. It should be borne in mind, however, that there are technical reasons why false positive results occur with PCR for microorganisms, namely, the bacterial DNA may be derived from both viable and necrotic organisms, Taq polymerase (a reagent used in PCR) is derived by a recombinant method involving use of Escherichia coli so that trace levels of E. coli DNA may be present as a contaminant and, lastly, trace contamination by clinically irrelevant organisms may occur because of the broad sensitivity of PCR using 16S rRNA.

15.3

Aseptic loosening

Aseptic loosening of a prosthetic joint occurs where there is no clinical and laboratory evidence of infection. It is not a single entity and various factors may play a role in this form of loosening, as mentioned in the overview (Section 15.1). Reported failure rates due to aseptic loosening may be as high as 20%, though this depends on various things such as the duration of implantation and follow-up time at which any series is being reported. Only 5% of Charnley hips inserted with a revised cementing technique had aseptic loosening in a minimum follow-up time of 15 years.34 The presence of a radiolucent line or focal areas of bone loss (osteolysis) are sometimes described, being known as `radiological loosening' by some authors. However, what matters to the patient is whether the implant is functioning normally or has become loose and whether the joint is painful. There is a medical aphorism which says that it is best always to treat the patient and not their X-ray (radiograph). Throughout this chapter aseptic loosening is taken to mean a clinically significant problem with pain and loosening in the absence of demonstrable infection requiring revision surgery. All the work from our laboratories on interface membranes involves samples that have no infection as proven microbiologically. The loosening of a prosthetic joint component in the absence of demonstrable infection is the commonest cause of joint replacement failure. As mentioned in the overview of this chapter (Section 15.1), it is open to debate whether the biological cellular processes to be described here are solely responsible for this form of joint failure. It is noteworthy, for example, that another author in this book (Chapter 13) states clearly that at the hip, mechanical loosening is © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

357

overwhelmingly the most common cause for revision, and this is not necessarily disputed, though the question may additionally arise as to whether abnormal wear at the bearing surface might give rise to excessive particle generation and the bone loss in relation to these particles might in turn result in loosening of the device. Particulate debris is now considered to be a major contributing factor in the development and perpetuation of aseptic loosening. It is proposed to describe the cellular and biological processes giving rise to bone loss rather than provide a detailed description of wear particle generation. It is necessary, however, to mention the methods developed for the isolation of particles from tissues at the time of revision surgery and the ways in which these may be characterised as this gives important background information for the proper study and understanding of the cellular processes involved in bone loss in relation to particles. While there is overlap, as will be seen, the specific mechanisms relating to the failure of different materials and the generation of particles are described elsewhere in this book.

15.4

The isolation and characterisation of wear particles

It is well recognised that wear occurs at the bearing surfaces of every artificial joint, in just the same way as it does between moving parts of machinery. The tribology of joint replacement is dealt with competently elsewhere in this book as are the mechanical aspects of joint performance. Suffice it here to state that in just the same way that a car engine generates wear particles in the first period of driving and that this is followed by less wear, there is a wearing in period for prosthetic joints. The terms `running-in wear' and `steady-state wear' are used in tribology as is clearly described in Chapter 18. Wear rates decrease due to increased congruency and surface smoothness after this initial period when the acetabular cup has undergone creep and machined lines have been smoothed according to Yamac.35 The size and morphology of polyethylene particles in a series of hips were different in the first months after implantation compared with the longer term, and this was considered to relate to the wearing-in process. Studies using joint simulators and wear testing machines provide important information about the inital wear of different bearing surfaces and have seen prominence recently, resulting in the development of metal against metal and ceramic against ceramic bearings. A good source of reference for this aspect is available in Chapters 2 and 18.

15.4.1 Conventional light microscopy There is an extensive literature on wear particles, which was usefully reviewed by Savio and colleagues.36 The appearances of the different types of material are characteristic and have been described elsewhere by the author.37 Using © 2008, Woodhead Publishing Limited

358

Joint replacement technology

15.1 Photomicrograph of macrophages and multinucleate giant cells with phagocytosed intracellular polyethylene debris which appear as empty spaces, some of which are large flakes and others smaller particles.

conventional light microscopy, small polyethylene (PE) particles and larger shards of PE are seen as unstained (transparent) objects (Fig. 15.1) which are birefringent when viewed between crossed polars (polarisation microscopy). Small particles are seen intracellularly in macrophages on microscopy of tissue sections, while larger particles and flakes are engulfed by foreign body multinucleate giant cells (MNGC). Put in the simplest way, these giant cells are formed as a result of the fusion of macrophages when the object to be phagocytosed is large and/or indigestible. Other polymers have from time to time found use in arthroplasty surgery, including polyacetal of which the author has personal experience. Particles of this material are birefringent by polarisation microscopy and so resemble polyethylene. They show the same pattern of behaviour when viewed by compensated polarisation microscopy using a quarter wavelength (/4) plate within the optical system, and cannot be distinguished by this means. But it is possible to distinguish these two polymeric materials by determining their refractive indices, then mounting samples for investigation in a medium of intermediate refractive index and identifying the polymers using Becke's line as described by O'Shea and her colleagues.38 Polyethylene (PE) and polyacetal (Pac) were readily distinguished in the same samples from five knees with a Pac femoral and PE tibial component.38 This simple method could be developed and readily used to differentiate other polymers. More sophisticated analytical methods such as Fourier transformed infrared (FTIR) spectroscopy could also be used (see Section 15.4.2) though rarely is differentiation of two polymers in the same sample required. Metal particles appear as brown or black granules or as short needles by transmitted light microscopy, the appearances depending somewhat on the particular metal (Fig. 15.2). Particles of metal show a weakly birefringent halo on polarisation microscopy, a feature attributed to the formation of metal proteinates due to the surface of the particle interacting chemically with the tissue proteins. This phenomenon is described as form birefringence37,39 and is a © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

359

15.2 Photomicrograph of macrophages and multinucleate giant cells (MNGC arrowed) with phagocytosed intracellular metal debris seen as small black particles within the cells.

useful feature when trying to decide whether a speck of intracellular material might be metal debris. While large shards of metal are not seen, much smaller metal particles are also found in MNGCs as well as macrophages, and as will be seen later, giant cell formation takes place in cell culture when macrophages phagocytose metal particles of the size present in tissues. Bone cement (poly(methylmethacrylate), PMMA) is dissolved out of tissue by the solvents such as xylene used in processing, so that the sites where PMMA was present are seen as empty spaces, often containing foci of very fine granular material which is the radiographic contrast material incorporated in the cement. These granules are grey in the case of barium sulphate and black in that of zirconia, these latter particles also having a mulberry-like appearance.38,40 That this is the constitution of granules with these appearances has been shown by elemental analysis. The empty spaces may be seen in macrophages and MNGC as well as in tissue. PMMA may be seen as large fragments which have detached from the main bulk of materials, or as circular spaces representing beads or spheres, which may be agglomerated. These beads range from 45 to 150 m in diameter, and represent PMMA powder that has not been incorporated by polymerisation at the time of insertion. Bone cement is retained in the tissue when frozen sections are used and is readily visualised by the use of Sudan red or Oil red O stains (Fig. 15.3).37,39,41 Ceramic debris is small, usually less than 5 m in diameter, presenting as fine greyish-brown particles in the case of zirconia, and brown-green, black or brown granules in the case of alumina.37,39,42 The paper by Mochida and colleagues gives a source of references to ceramic particles, their isolation and characterization.42 © 2008, Woodhead Publishing Limited

360

Joint replacement technology

15.3 Photomicrograph of a bead of bone cement (C) within a macrophage, stained with Sudan red.

Hydroxyapatite (HA) is the other material that may be seen in tissues adjacent to implants, and although there is usually bone ongrowth (see Chapter 14), when separated from the surface coating layer, HA is pale grey and seen either as large pieces broken from the bulk material or as finer powdery granules within macrophages. Such material has been examined by Bauer and colleagues43 using transmission electron microscopy (TEM) and shown to contain calcium and phosphate with EDAX elemental analysis. Other materials that have been used include poly(etheretherketone) (PEEK) polymer and carbon fibre-reinforced (CFR) plastics, most notably CFR-PE and CFR-PEEK, which are seen as particles in tissues usually as their separate component materials, namely black carbon fibre fragments and birefringent flakes/particles of the polymer, rather than as intact composite.44

15.4.2 Ultrastructural studies: submicroscopic and nanometresized particles A little over a decade ago, it became apparent that the particles present in the tissues adjacent to loosened replacement joints were almost all less than 1 m in size, and therefore too small to be visible by light microscopy. Submicroscopic particles were identified in tissues using electron microscopy by Shanbhag and colleagues.45 Wear debris was then isolated from tissue samples in several © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

361

different laboratories including our own in London.35,46±48 The methods varied a little from centre to centre, but basically involved tissue digestion and particle separation using ultracentrifugation, with sucrose density and/or isopropanol gradients, to produce a sample for transmission electron microscopy (TEM) or scanning electron microsopy (SEM) examination and thereby visualise the particles. While the initial thrust of this work was to isolate PE wear debris, over 95% of which was shown to be less than 1 m in size, subsequent methods have been developed for separating out metal, bone cement and ceramic particles. The thesis of Yamac35 showed the systematic evolution from the PE isolation method of a procedure for separating metal debris, which similarly proved to be submicrometre-sized (range, 64±97%; mean, 84%; 10 cases studied) (Fig. 15.4). Metal particle isolation was also described by Campbell and her group46,49 among others. Of some importance is the recognition that metal particles may be as small as 10±70 nm,50 6 nm49 and therefore defined as nanoparticles, being less than 100 nm in maximum dimension.51 Revisiting the thesis of Yamac,35 it is clear that there were also nanoparticles present in her samples, the range of particle size being quoted by her down to 0.02 m in length (i.e., maximum dimension). Combining the PE and metal protocols provides a means of isolating and characterising particles in tissue or, in fact from a wear testing machine such as a joint wear simulator. An ISO standard has been produced with the author as lead scientist on the basis of the work cited above with all the necessary and appropriate discussions among various international colleagues.52 With the exception of the results reported by Doorn et al.49 and Case et al.,50 the metal particles found by isolation methods have been in the range of

15.4 Scanning electron micrograph of retrieved polyethylene particles from the interface tissue of a total knee replacement at revision surgery (from Kobayashi et al., Proc Instn Mech Engrs 1997, 211H, 11±15, with permission). © 2008, Woodhead Publishing Limited

362

Joint replacement technology

0.1±200 m.35,48 Yamac commented that there was not a noticeable difference in the morphology between small (0.1±3 m) particles of CoCr, TiAlV, TiAlNb and stainless steel. A corrugated surface was present in the case of titanium alloy and stainless steel particles larger than 3 m, while surface morphology of large CoCr particles was almost always smooth. This difference in appearance may be attributed to differences in their material properties, with CoCr being stiffer than either the titanium alloys or stainless steel. The appearance of grooves and cracks on the surfaces of larger metal particles has also been reported by Maloney et al.48 The morphology of particles by SEM gives important clues as to the wear mechanisms involved in particle generation, but it is not appropriate to explore this further here. A method for isolating and characterising PMMA wear debris in tissues at revision surgery was reported by Iwaki and other colleagues in our group.53 FTIR and energy dispersive X-ray analysis (EDAX) were used to prove that the particles were PMMA, by the demonstration of the specific IR spectrum and the presence of radio-opaque contrast material respectively. The particles, viewed by SEM, were 0:96  0:11 m in size, expressed as the equivalent circle diameter (ECD) and overall 67% of them were submicrometre in size. Shape factors (roundness and aspect ratio) were also calculated. The particles were considered to be smaller than previously thought and within the phagocytosable range. Iwaki and colleagues also have described the simultaneous successful isolation and characterisation of three different types of wear particle from cemented metal±polyethylene joints.54 It is clear that TEM or SEM should be used to characterise the particles and that these should be isolated from tissues next to failed replacement joints when studies of the pathological effects of wear debris are carried out. In practice, even those who have developed these methods, including our own group, do not then use them in each case that becomes available for study. Rather, the assumption is made that there are numerous particles present within the macrophages and MNGC present between bone and implant. The proportions of metal, PE, PMMA and other material debris are simply not characterised apart from using inaccurate and misleading light microscopy. On this basis, it is simply not known what particles are really present in an infiltrate of cells near to a loosened joint on which sophisticated immunohistochemical, molecular biological and proteomic methods have been applied. What has been studied, however, is the effects that various implant materials in nano- and microparticle form have on macrophage function in cell culture. It is these functional studies that lend credibility to the histological and tissue extraction methods used on interface tissues. These aspects will be considered in the next sections, in which cellular activity in relation to wear debris will be described. The results of in situ studies will be reported rather than the large literature on cell culture.

© 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

15.5

363

The cellular reaction to particulate wear debris

Wear debris is generated mainly from the bearing surfaces of the replacement joint and the particles are shed into the synovial fluid from where they find their way directly into the synovial lining cell layer. Other parts of the prosthesis may give rise to particles from abrasion of the stem of a femoral component, for example, or wear of a trunion between the femoral head and the upper part of a femoral stem. In addition, there may be wear and/or corrosion around screws. The main thrust of the following review will be the effects of those particles present initially in the joint fluid and subsequently deep within the peri-implant bone as well as at distant sites in the body. It is necessary first to consider the normal morphology of the joint lining, that is the synovial membrane.

15.5.1 The synovial lining cells of joints and at the implant± bone interface The normal synovial joint has a fibrous capsule inside which is a layer of specialised cells which produce the synovial fluid, responsible for cartilage nutrition and joint lubrication, as well as removing foreign material from the joint. The lining layer of cells, often called synoviocytes, increases in thickness through an increase in numbers of cells, a process called hyperplasia (the increase in numbers being as a response to increased functional demand). This is a non-specific response which occurs in various joint conditions, including inflammatory arthritis, marked degenerative joint disease, crystal-induced arthropathy (e.g. gout) and even in the presence of mechanical joint derangement. Strictly, it is not hyperplasia, because this process occurs as a result of local proliferation of cells, which does not occur in the syovial lining cells with the exception of occasional basally situated cells, as shown by Lalor and colleagues.55 On the basis of ultrastructural studies, the cells lining the joint were classified into type A and type B synoviocytes56,57 and Graabeck58 noted that the type B cells were deep to the type A synoviocytes. Others, including ourselves had shown that the type A cells were labelled with macrophage monoclonal antibody (MAB) markers.59±61 The type A cells were characterised with macrophage markers at the ultrastructural level by immunocytochemistry62 while fibronectin was localised to the synovial lining cell layer and shown to be produced by type B synoviocytes in LM and TEM immunocytochemical studies.63,64 Other extracellular matrix proteins have also been demonstrated in association with the deepest layer of cells, the type B synoviocytes, and these include type IV collagen, laminin, chondroitin sulphate, heparan sulphate, type V collagen and entactin,65,66 all of which are components of, or closely associated with, the basement membrane of epithelia at other sites. Ultrastructural immunolocalisation showed types IV and V collagen, laminin and heparan © 2008, Woodhead Publishing Limited

364

Joint replacement technology

sulphate to be present in the endoplasmic reticulum of type B synoviocytes as well as coating the outside of these cells, indicating local production.66 It seems likely that one function of the type B synoviocytes might be providing a basement membrane-like anchorage for the synovial lining. (There is no true basement membrane in the synovial lining, that is to say, it is not an epithelium.) The synovial lining cell layer also shows the presence of adhesion molecules, intracellular adhesion molecule (ICAM-1) and vascular cell adhesion molecule-I (VCAM-1), as well as CD44.67±71 A monoclonal antibody (MAB 67) was also found to be a marker of type B synoviocytes by Stevens et al.72 Using routine haematoxylin-eosin stains, Goldring and colleagues showed that a synovium-like structure may develop on the surface of the fibrous tissue between implant and bone (Fig. 15.5).73 Proof that this cellular layer is closely similar to, if not identical with, the true synovium comes from detailed characterisation studies. Thus, the cells in this layer are macrophages and fibroblasts arranged in a manner similar to the type A and type B cells of the true synovium.74 Moreover, the distribution of fibronectin, type IV collagen and laminin around the deeper fibroblastic cells is also like true synovium (Fig. 15.6) and these cells are also marked with MAB 67.75 Other basement membrane components (type V collagen and heparan sulphate) are present (Revell, unpublished findings). Both ICAM-1 and VCAM-1 are localised to the synovium-like layer of cells at the implant interface,76 as is CD44.77 Finally, prolyl-4-hydroxylase is expressed by the type B synoviocytes in true synovium78

15.5 Low-power photomicrograph of the bone±cement interface in aseptic loosening showing the synovium-like layer next to the (site of the) implant (top) and an underlying cellular infiltrate comprising macrophages and multinucleate giant cells. Bone is seen at the bottom of the picture and the large spaces (C) are the the sites of bone cement, which has been dissolved out by solvents in histological processing. © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

365

15.6 Photomicrograph of interface tissue showing cells in the synovium-like layer adjacent to an implant, labelled by immunohistochemistry with a monoclonal antibody against laminin. The basement membrane of blood vessels (BV) is marked, as is the surface of the more basally situated fibroblastic (F) cells. Macrophages and MNGC (M) are not labelled but contain abundant black metal particles.

and by a proportion of deeply situated cells in the synovium-like layer of the implant interface tissue.79 It has been noted in Chapter 14 that macrophages are found in small numbers on the surface of implanted biomaterials, whether in clinical use in humans80,81 or after experimental implantation in animals (Section 14.3).82,83 Differences in the surface of PMMA of the same chemical composition gave rise to differences in macrophage recruitment experimentally, materials for implantation having been prepared by prior separation of polymer beads into different sizes before polymerisation.82 More macrophages and giant cells were found on the surface of PMMA cured in situ in muscle of rats than on poly(ethylmethacrylate)/ butylmethacrylate (PEM/BMA), a difference which might be due to different chemistry, but both materials had few cells present on the surface when inserted as a pre-cured pellet. The surface of the in situ cured PMMA was bosselated or nobbly compared with the smooth surfaces of the other three implants83 and these results were in keeping with the previous study in this respect.82 It seems likely that wherever a foreign material is placed in the body, there will be macrophage recruitment and MNGC cell formation. This is further borne out by the rat `air pouch' model in which air is injected repeatedly into the subcutaneous tissue, giving rise to a space containing predominantly nitrogen, as the oxygen is absorbed, which remains inflated and becomes lined by macrophages. This model, first described over 25 years ago,84 has found use in inflammation research ever since. There is therefore strong evidence for the © 2008, Woodhead Publishing Limited

366

Joint replacement technology

acquisition of a lining layer of phagocytic cells adjacent to orthopaedic implants, often with the appearance and function of a synovial lining where an element of loosening is present.

15.5.2 The dissemination of particles to peri-implant tissues While macrophages and MNGC may be found in relation to implanted bulk material, such as a replacement joint, as has been seen above, it is the recruitment of these cells to a site in which there are numerous wear debris particles which is the most significant in terms of the effects on the bone adjacent to the implant and whether loosening of the device occurs. Particulate debris from the bearing surfaces is shed first and foremost into the synovial fluid and reaches the lining cells of the synovial membrane. It has long been considered that there is a steady state or equilibrium reached in which the phagocytosis of particles by these synovial cells is matched by their clearance from the joint which occurs through lymphatic vessels to the local lymph nodes.37,85,86 Lymphatic vessels have, however, only just recently been identified as present in syovium and the implant±bone interface using MAB staining and immunohistology.87 That the products of wear may be found in local lymph nodes and the spleen has been demonstrated by Case and his colleagues among others88±90 and the experimental dissemination of CoCr particles from bone to the spleen has been shown in the guinea pig by our own group.91 The relevance of the presence of biomaterials in lymphoid organs will be apparent in a later section. Here it is important to consider the process by which particles reach the interface between implant and bone and how the cellular response to their presence affects local bone and the fixation of the prosthesis. The presence of a synovium-like layer covering the fibrous tissue next to the implant implies that there is a fluid layer between tissue and biomaterial. It seems possible that there is continuity between the synovial fluid and this tissue fluid deep within the bone. Indeed such continuity has been demonstrated for the prosthetic hip in human bone specimens with a cement to bone interface.92 The effects of fluid pressure on the migration of particles is discussed by Aspenberg and Van der Vis.93 A further method of particle dissemination was suggested by Anthony and colleagues41 in the case of cemented implants. They believed that particles could track between a metal implant and the surrounding PMMA cement, then exit this space through cracks in the cement mantle to cause focal osteolysis distant from the joint space. Radiological evidence of loosening and bone loss has been classified by Gruen and colleagues94 for the hip. They divided the area around the prosthetic joint in the antero-posterior (AP) radiograph into seven zones and evaluated the sites at which changes were present in a large number of radiographs of replaced hip joints. Radiolucency between cement and stem was most often present in zone 1 followed by zone 7 (the two most proximal zones) and occurred in no © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

367

other zones, while radiolucency between bone and cement was seen most in zone 1, then zone 4 (near the end of the stem), followed by more or less equal numbers in the other zones, with a slight predominance for the medial rather than the lateral ones. The collection of excessive amounts of debris in the joint could be seen as giving rise to the first and most frequent evidence of bone loss occurring at the upper end of the implant adjacent to the synovium, as the accumulation of macrophages gives rise to local bone loss. That bone loss occurs deeper in the bone suggests that debris is conducted there in the fluid bathing the implant, as described above. The lack of visible radiological defects at the implant±cement interface in these deeper parts of the stem rather mitigates against the theory of particle dissemination between cement and implant. Cement particles may be released into the bone±cement interface locally where a macrophage and giant cell response causes local bone loss. It should also be noted, however, that macrophages are recruited to bone cement and have been observed in relation to bone-cement radiolucencies in the absence of particulate cement debris.80,81 The presence of focal osteolysis in stable hip replacements has been noted in relation to PMMA, and in the absence of PE debris at light microscopical level.95 There are multiple descriptions of the presence of polyethylene particles in the tissues adjacent to prosthetic joints deep within the bone. The mechanism of distribution to these sites is likely to be as described above. They clearly cannot have been generated locally in the case of PE debris derived from an acetabular component yet found in relation to the distal part of the femoral stem. Schmalzreid and colleagues described the encroachment of bone resorption in relation to PE debris from the periphery of the acetabular components and regarded this as a biological process related to macrophage activity, while they thought that femoral radiolucencies were due to stress-related remodelling.96 It is not proposed to give a detailed bibliography of the various publications in which the effects of PE wear debris are detailed. The excellent account of the tribology of joints in Chapter 2 includes details about the problem of excessive PE wear and the measures, such as crosslinking, taken to reduce this. Metal particles were considered problematic when metal against metal articulations were used, but then passed out of notice during the period when metal against PE articulations predominated. They have again come into prominence since metal±metal joints, of the so-called second generation, such as the resurfacing hip arthroplasty, have been developed. Metal against metal and ceramic against ceramic bearings have been used because of problems occurring with polyethylene. These aspects are not relevant to the biological aspects of prosthetic joint failure which is the subject of this chapter. That metal particles were present in considerable amounts even where metal articulated against PE is apparent in the literature and the black discoloration of tissue, often referred to as `metallosis', is testament to this. While the articulation may be metal±PE, many of the prosthetic joints had metal used in other sites, such as metal backs to the © 2008, Woodhead Publishing Limited

368

Joint replacement technology

acetabular component at the hip, or metal tibial trays in which a PE tibial plateau was located. Fretting of TiAlV or stainless steel screws, metal backs, and CoCr or TiAlV femoral components with bone were all identified even though the joints studied were ostensibly metal±PE joints in the work of Yamac.35 More details of first and second generation metal±metal joints will be considered later in respect of the development of hypersensitivity reactions (see Sections 15.8 and 15.9).

15.6

The role of macrophages and multinucleate giant cells

The macrophage is the cell that deals with foreign particulate material, together with multinucleate giant cells which are formed by the fusion of macrophages. Wherever there are particles, macrophages will be present, so that the histological appearance of the synovial membrane or the bone surrounding the prosthetic component (bone±implant interface) will be one of numerous macrophages with intermingled MNGCs (Figs 15.6 and 15.7). While the debris may be apparent within these cells, this is not necessarily the case, since the vast amounts of particles are not visible by light microscopy (see above). The macrophages and MNGCs may be organised into granulomata or occur as more uniform sheets of cells. The term granuloma is used by anatomical pathologists

15.7 Photomicrograph of the interface tissue of an aseptically loosened joint showing the presence of numerous macrophages (M) and MNGC (GC) marked with the monoclonal antibody against CD68. No large particles of debris are visible, those particles present being submicroscopic in size (from (modified after) Kadoya et al., Bone and Mineral 1994, 27, 85±96, with permission). © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

369

to describe a specific appearance in which nodule-like circumscribed collections of macrophages and MNGCs are present, with an associated collection or cuff of lymphocytes. This appearance is characteristic of certain diseases in which cellmediated immunity is involved, such as tuberculosis and sarcoidosis. While this appearance is seen in relation to orthopaedic implants, the term granuloma is often used more loosely in the relevant literature to mean a macrophage and MNGC infiltrate. For the most part, therefore, the term granuloma will not be used in this description of the pathophysiology of bone loss in relation to wear debris. This is not to deny that true granulomas are seen in some cases. Macrophages and MNGCs may be characterised in tissue sections using MAB and immunocytochemistry. The use of such a method is important because the morphology of a cell in routinely stained sections can be misleading; thus, spindle-shaped cells may be interpreted as fibroblasts, but macrophages can assume this shape as they migrate through the tissues. The best known macrophage marker is MAB against CD68, a 110 kDa transmembrane glycoprotein present on circulating monocytes and tissue macrophages (Fig. 15.7). This antibody also labels MNGC (Fig. 15.7).97,98 Other markers less often used are CD13, CD35, CD36.97,99 An MAB marker has been used to characterise cells in each study in our laboratories where cell function in relation to biomaterial particles or the interface has been studied. Some of the antibodies are listed further elsewhere.97±99 Using MAB markers it is possible to show that the macrophages and MNGCs in the implant interface are activated. Thus, these cells show the presence of surface human leucocyte antigen (HLA) class II molecules (HLA-DR)100,101 and express integrins, particularly CD11b (also known as M/ 2 integrin),102,103 both of which have been shown in cell culture studies to be expressed by macrophages on phagocytosis of wear particles in the theses of Altaf104 and Curtis.105 This work and that of Clarke106 demonstrates the expression of various cytokines by macrophages in functional studies of particle engulfment. Similar observations have been made by others, two of which are cited here.107,108 Suffice it to say that there is plentiful evidence from the literature, both orthopaedic and more generally immunological, that stimulated activated macrophages produce a large number of different cytokines and other inflammatory mediators. An excellent source of references to various cell culture studies in which cytokine release is induced with particles is the article by Archibeck and colleagues.109 The cytokines which have been demonstrated in macrophages related to particles in periprosthetis tissue samples, together with the relevant references, are as follows: · · · ·

Interleukin Interleukin Interleukin Interleukin

1 114 1 101,110,114±116,121,123,124 6115,116,123,124 10104,117,118,120,125,126

© 2008, Woodhead Publishing Limited

370 · · · · · · · ·

Joint replacement technology

Interleukin 11129 Interleukin 15100,119 Tumour necrosis factor 115,120,123,124 Transforming growth factor 112,131 Granulocyte-macrophage colony stimulating factor111 Macrophage colony stimulating factor112,130 Platelet-derived growth factor110 Epidermal growth factor127,131

An example of IL1 expression by macrophages and MNGCs in relation to particulate debris is shown in Fig. 15.8. Other mediators of inflammation have been shown in situ, namely, prostaglandin E2,73,121,122 various metalloproteinases/collagenases73,112,113,116,121,122 and inducible nitric oxide synthase (iNOS),128 the last of these being the first demonstration of iNOS in human macrophages in any context. The effects of these mediators and the ways that they interact are extremely complicated. Most of them are pro-inflammatory (for example, IL1, TNF, IL6) though anti-inflammatory activity is also present (IL10). The detailed analysis of all these mediators in the context of implant loosening is beyond the scope of this chapter, and may well be so complex as to be virtually impossible to unravel in the current state of knowledge and information. Clearly there is a need for in vitro studies of the effects of different cytokine combinations in the context of macrophage activation by particles before a clear understanding can be obtained. One aspect that has been profitably explored is the relationship of MNGC to osteoclasts and the role that various mediators may play in osteoclastogenesis.

15.8 Photomicrograph of the interface tissue in aseptic loosening showing labelling of macrophages and MNGC with monoclonal antibody against interleukin 1 (IL1 ). Occasional black metal particles are visible (from AlSaffar et al., J Mater Sci: Mater Med 1997, 8, 64±68, with permission). © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

15.7

371

Bone resorption and wear debris: osteoclasts, macrophages and multinucleate giant cells

Some of the cytokines known to be present in the interface membrane have direct effects on osteoclast and MNGC formation, namely TGF, M-CSF and GM-CSF. While M-CSF and TGF are considered to influence osteoclastogenesis and GM-CSF to promote MNGC formation, these differences may not be so clear-cut as at first thought. Additionally, IL1, TNF and IL6 have indirect effects through their action on other cells such as osteoblasts, which then in turn influence osteoclasts and bone resorption, as well as further promoting the macrophage-driven cellular reaction, being pro-inflammatory cytokines. An important question arises as to the cells which are responsible for the removal of bone which occurs where there is abundant wear debris present. The osteoclast is the cell specialised to fulfil this role and is seen on the bone side of the inflammatory infiltrate. It is likely that a significant amount of the bone loss is mediated by the classical bone resorptive activity of osteoclasts. However, nearly all the studies of the cellular infiltrates in tissue next to implants in people do not actually include the bone in the samples. Kadoya did obtain small amounts of bone from the areas of resorption at revision surgery and his work in our group shows interesting findings. Thus, while osteoclasts occupied 7:67  1:82% of the bone surface, macrophages covered 19:33  5:16% of the bone, and tissues retrieved from implantation sites with radiographic evidence of osteolysis had significantly higher macrophage surface than those without osteolysis (33:37  8:59% vs 5:29  1:34%).132,133 Osteolytic zones also showed significantly higher osteoblastic surface. It has long been known that monocytes/macrophages, MNGC and osteoclasts have the same lineage in terms of precursor cell in bone marrow, but the question arises how far back in this lineage the separation of cell types occurs in terms of their differentiation. In the light of the findings of Kadoya, the relationship between osteoclasts and MNGCs may not be quite so clear as previously understood. Kadoya showed that some of the markers for osteoclasts were also shared by MNGC in the infiltrate towards the bone side of the interface, while MNGC on the implant side of the interface tissue and in relation to the synoviumlike layer did not express these markers.134 That MNGCs were present on the bone surface and that these cells had a folded or ruffled border in relation to the bone, like that seen in osteoclasts, was demonstrated by TEM. These same cells, however, could also be demonstrated to contain submicrometre-sized metal particles in cytoplasmic membrane bound bodies. Athanasou has shown that debris-related macrophages are capable of resorbing bone in vitro, albeit at a low rate,135,136 and that, apart from the ultrastructural evidence of Kadoya, osteoclasts may themselves be capable of biomaterial particle phagocytosis.137 The presence of lymphocytes in the interface has not so far been mentioned and this will be given prominence in a later section. However, a further aspect of © 2008, Woodhead Publishing Limited

372

Joint replacement technology

bone resorption in relation to particles becomes apparent when the possible role of lymphocytes is considered. The receptor activator of nuclear factor-B ligand (RANKL)±RANK signalling system between osteoblasts and osteoclasts plays a central role in osteoclastogenesis and osteoclast activation and there are also links to the immune system. This complex area is reviewed by Boyce and Xing.138 RANKL has been demonstrated as present in the interface tissues.139,140 Moreover, interleukin 17 (IL17) has been shown to be produced by T helper lymphocytes in the implant interface where there is aseptic loosening117,118 and this cytokine in turn has a stimulatory effect on RANKL expression and hence osteoclastic activity. The possible role of T lymphocytes in the bone±implant interface will be considered in the next section.

15.8

Lymphocytes, sensitisation and aseptic loosening

The presence of lymphocytes in the infiltrate related to wear debris was noted over 25 years ago by Vernon-Roberts and Freeman.85 Some authors describe the presence of these cells but consider them to be rare95 or of little consequence, being less than 10%110 of the cells present, and regarded as passive bystanders recruited by the other cells, the macrophages, making a foreign body response. That an interaction between macrophages and lymphocytes may be taking place is a possibility, but not in the sense originally envisaged by early workers. This will be expounded below. The studies by Lalor and colleagues in London141±143 were among the first to recognise the likely significance of a lymphocytic infiltrate at the implant interface and to attempt an association of this with an identifiable immunological process. Thus, the lymphocytes in the infiltrate seen in relation to aseptically loosened hip joints from five cases were identified as being T cells using immmunohistochemistry and there were no B cells or plasma cells present. The authors concluded that such a response `implies type IV sensitivity (cell mediated immunity, contact sensitisation)'. Two of the individuals reported in this study were shown to be sensitised to titanium on skin testing, and the others considered also to be sensitised even in the absence of such a positive clinical dermatological test. Titanium sensitisation is a rare occurrence which has nevertheless been reported on occasions since by others in the context of joint replacement.50,144 The presence of T lymphocytes and the lack of B cells has been a recurring theme of our work ever since the early 1990s.97,99 Lymphocytes are the cells of the immune system which are the effectors of the host reactions and they make immunological responses specific. The B lymphocytes are responsible for the production of antibodies while T lymphocytes respond by releasing soluble mediators called lymphokines. When an immune reaction occurs in the body, it may have local tissue damaging or systemic effects known as hypersensitivity © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

373

reactions. Those reactions mediated by B cells have been called types I, II and III hypersensitivity, while type IV hypersensitivity is mediated by T lymphocytes. It was the finding of only T cells that led us to propose and investigate for titanium sensitisation in the cases already described.141±143 It has been well recognised in the past that sensitisation to metals occurs in some patients with a joint replacement. The cases described in the 1970s were mostly first-generation metal against metal hip replacements.145±148 Elves and colleagues did skin sensitisation tests on 50 patients with metal against metal hip joints145 and found that 38% of them were sensitive to one or more of the metals tested (Cr, Co, Ni, Mo, V, Ti). Fifteen out of 23 with joint loosening were sensitive, leaving 4 sensitive cases out of the remaining 27 who did not show loosening. They concluded that metal-on-metal implants may sensitise the patient to metals contained in the prosthesis, but also felt unable to state whether the loosening causes the sensitisation or vice versa. In the same issue of the journal, Benson et al.146 reported a high incidence (28%) of unexpected metal sensitivity on skin testing of patients with metal-to-metal (McKee) hip arthroplasties, but patients with metal-to-plastic (Charnley) prostheses (2.6%) had no greater incidence of metal sensitivity than a control group awaiting operation. Evans and colleagues had studied a small number of individuals in some detail a year earlier showing elevated levels of Co and Cr in tissues adjacent to CoCr±CoCr articulations.147 They also demonstrated that 9 out 14 individuals with loosened implants were sensitised (with 11 loose implants) whereas no sensitivity was present in 24 individuals with non-loose (well-fixed) prosthetic joints. This is one of the few papers in which histological appearances of periprosthetic tissues from sensitised individuals are described, and the authors noted tissue and bone necrosis in association with endarterial obliterative changes, as well as a macrophage and multinucleate giant cell reaction. Specifically, no mention is made of a lymphocytic infiltrate of the kind later described by our own group and others; occasional lymphocytes are noted in association with macrophages and giant cells and some blood vessels. Sensitisation to metal where there is a metal against polyethylene (M±PE) articulation has been noted. Nater and colleagues148 skin tested 66 individuals with metal±plastic hip joints and found sensitisation in four of these, one of whom was known to have been negative before surgery (along with 61 others who remained negative throughout). None of the individuals with contact sensitisation had problems with their implants. Other papers where metal sensitisation is recorded in relation to M±PE joints are those by Case and Lalor with their respective colleagues.50,143 Additionally, in a thorough investigation of a small number of M±PE joints, Pazzaglia et al.149 show the presence of metal particles, and this has been our experience over many years, namely that metal debris is present even where PE is the predominant wear material. Sensitisation to acrylic materials is well known in those handling these materials, such as orthopaedic surgeons and dental technicians, and occasional cases in patients © 2008, Woodhead Publishing Limited

374

Joint replacement technology

with cemented joint replacements have been described.150±152 The author's own experience has been to see tissue samples from the interface of loosened implants with metal or, on two occasions, acrylic sensitisation, one of these being sensitivity to para-toluidine, rather than PMMA itself. In all these cases, the predominant feature was a large increase in the numbers of lymphocytes present. In two metal sensitivity cases, tissue obtained after a period without an implant but with a `spacer' present showed that the T lymphocytes were no longer present. The next section of this chapter will describe the evidence for Tcell activity and an immunological process in implant loosening accumulated over many years from the study of tissue samples.

15.9

Evidence for immunological processes in loosening

The presence of lymphocytes in the bone±implant interface in aseptic loosening has long been held as significant by the author and his group, and has been already mentioned above.97,99,141,142 The lymphocytes occur as a diffuse infiltrate intermingled with the macrophages and MNGCs (Fig. 15.9) as well having a perivascular distribution (Fig. 15.10). These observations have been confirmed recently by Willert and co-authors.153,154 Immunohistochemistry shows the lymphocytes to be T cells, labelled with anti-CD2 and/or anti-CD3 MABs, and practically all authors agree that there are no B lymphocytes present98,99,124 though Willert et al.154 show B cells present in so-called periprosthetic tissues, which it should be noted are not interface but synovial membrane. Plasma cells feature in the descriptions by these authors153,154 and

15.9 Photomicrograph of the interface tissue in aseptic loosening showing the presence of a diffuse infiltrate of T lymphocytes, labelled with a monoclonal antibody against CD3. The synovium-like layer of cells is seen to the top and right (from Revell, J Histotechn 2006, 29, 287±295, where previously published in colour, with permission). © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

375

15.10 Photomicrograph of T lymphocytes in a perivascular distribution in the interface tissue in aseptic loosening, labelled with anti-CD3 monoclonal antibody; BV = blood vessel.

the other recent paper by Milosev and colleagues.155 These cells have not been seen by the present writer who notes, however, that mast cells have been reported and characterised by immunohistochemistry at the bone±joint interface in aseptic loosening.156 Mast cells and plasma cells may sometimes be difficult to tell apart in the absence of special stains. These studies relied at least partly on routine haematoxylin±eosin staining.153,155

15.9.1 T-cell subtypes In cases of aseptic loosening of M±PE joints, the numbers of T cells have been measured and calculated as 6±16%118 and 4±23%.104 By contrast, Hercus found 30 and 31% lymphocytes to be present in two cases of known nickel sensitivity (Fig. 15.11).118 T lymphocytes may be subdivided into different types according to their functional activity, then subtyped according to further current opinion within immunology. Thus there are T helper (TH) and T cytotoxic/suppressor (TC/S) cells and the former have been divided into TH1 and TH2 cells. Using MAB labelling of interface tissue, a predominance of TH cells over TC/S cells has been demonstrated (TH : TC/S ˆ 7.2 : 1).118 This predominance of helper over cytotoxic/suppressor cells has been confirmed by other studies within the London-based group.157 Identification of the subtype of T helper cell present might aid in deciding what sort of immunological process is taking place in the interface, since the TH1 cell is critical in the activation of macrophages and TC/S cells, being involved in cell-mediated immune reaction, while the TH2 response is engaged in B-cell activation and humoral immunity. Arora et al. could find no evidence © 2008, Woodhead Publishing Limited

376

Joint replacement technology

15.11 Photomicrograph showing heavy T-lymphocyte infiltration of the interface tissue in a case known to be sensitised to nickel on skin testing, retrieved at revision surgery. The T cells made up 31% of the cells in the interface in this case. Anti-CD3 monoclonal antibody labelling of T cells.

of a particular predominance as between these two subtypes of helper T cell, using immunohistochemistry to look at tissue from areas of osteolysis and those without this change.158 Ten per cent of the cells overall were T lymphocytes, which is a little lower than our findings in London104,118 and more in keeping with Jiranek et al.'s finding.110 A different result has been obtained using the PCR on interface tissue to characterise the cytokines present and provide a profile of the subtypes. These studies clearly show a predominance of TH1 over TH2 cells,157,159 a result which is in line with the findings of Weyand et al.160 Evidence from T-lymphocyte typing, therefore, points to a cell mediated or contact sensitisation process taking place in those cases of aseptic loosening examined. Protein extraction and western blotting of samples showed the presence of IL17, fractalkine and CD40 molecules which are also associated with TH1 activity.157

15.9.2 T-cell proliferation and maintenance If there is an active immune process taking place in relation to wear debris then further evidence of lymphocyte activation should be apparent in tissue sections, even though these are like a freeze frame picture (still) from a film rather than all of the movie. That this is the case is shown by the fact that the lymphocytes express HLA-DR and are proliferating, in a proportion of cells, as demonstrated by the presence of the nuclear protein marked with MAB Ki67 (Fig. 15.12).161 By contrast, the group working in Finland found T-cell numbers low and no evidence of T-cell activation as judged by the absence of interleukin-2 receptor (IL-2R) as well as the lack of IL2, IFN and TNF production by cells in their cases.162 T-cell activity in terms of proliferation and cell maintenance is dependent on IL2 both in cell culture and in vivo, and failure to show IL2 could well be taken to mean that any T cells present are not involved in an immunological process. © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

377

15.12 High-power view of macrophages containing metal debris (black granules) with closely associated lymphocytes the nuclei of which are all labelled (arrows) with a monoclonal antibody against a nuclear protein present in proliferating cells (Ki67) (from Revell & Jellie, J Mater Sci: Mater Med 1998, 9, 727±730, with permission).

However, there is an alternative molecule to IL2, namely IL15, which acts as a surrogate for this cytokine in vivo and is also able to sustain T cells in culture. Interleukin 15 has been shown to be abundant in macrophages and MNGC in the implant interface tissues in aseptic loosening by immunohistochemistry (Fig. 15.13) and the mRNA is expressed in the same cells.161,163 Interestingly, while IL2 could not be shown by immunohistochemistry, the IL2 receptor was demonstrable on some lymphocytes161 and, in a separate study, the mRNA for

15.13 Photomicrograph of the synovium-like layer and adjacent tissue showing labelling with monoclonal antibody against IL15. Note that some MNGC and macrophages containing metal debris are producing IL15 while many others are not. © 2008, Woodhead Publishing Limited

378

Joint replacement technology

IL2 was demonstrated.159 Using MAB immunohistochemistry staining together with sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE), western blotting and reverse transcriptase polymerase chain reaction (RT-PCR) methods, it has been shown that macrophages, MNGCs and endothelial cells in the interface express IL15 and the IL15 receptor (IL15R ) forming an autocrine feedback, while T lymphocytes interact with IL15 through the IL2 receptor (IL2R ),163 which, from work in other fields within immunology, it is known IL15 can `borrow' and use. Evidence is presented in the same paper for the concept that cell to cell contact occurs between macrophages and lymphocytes for this IL15-IL2R reaction, since IL15 seems to be present as a trimeric form bound at the macrophage cell membrane rather as an exportable smaller soluble product.163 The question as to whether there is IL2 present in the interface membrane remains unanswered, though also in a sense becomes irrelevant because of the demonstrated abundance of IL15.

15.9.3 T-cell memory and activation: antigen presentation If there is an active immune process taking place, then the lymphocytes should be shown to be memory cells rather than naive lymphocytes which have not been primed. This is the case since the lymphocytes are shown to be CD45RO positive (memory or primed cells) and rather than CD45RA (naive cells) (Fig. 15.14).79 The process by which these cells become primed involves the presentation of antigen to them by specialised phagocytic cells, the antigenpresenting cells (APCs). There are APCs in lymph nodes and the spleen known as dendritic cells, as well as in the skin, the Langerhans cells, but some macrophages without the dendritic morphology of these specialist cells also show evidence of antigen presentation. Particular receptors have to be engaged on the surface membranes of the cells involved in the antigen presentation

15.14 High-power view of T-lymphocytes in the interface tissues in aseptic loosening showing positive labelling with a monoclonal antibody CD45RO which marks primed memory cells. © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

379

process and these are known as the co-stimulatory molecules. When an APC ingests foreign material, such as a wear particle, it not only produces numerous cytokines and integrins (as described above) but also expresses large amounts of major histocompatibility complex (MHC)±peptide complexes at its surface. These HLA class II molecules are one of the features of macrophage activation. The T cell recognises these antigens, engaging the T-cell receptor (TCR) on its surface with the MHC±peptide in what is known as the first signal. No further progress is made in the absence of a secondary signal which is provided by costimulatory molecules, pairs of surface molecules expressed on the surfaces of the antigen-presenting cell (APC) and T cell. The costimulatory molecules are represented diagrammatically in Fig. 15.15. Of particular note are the CD80/ CD86 molecules on APCs with their CD28 counterligands on T cells. Successful antigen presentation and T-cell activation occurs when both primary and

15.15 Schematic to represent the co-stimulatory molecules present on the surface of the antigen-presenting macrophage and the T lymphocyte. The primary signal, the interaction between MHC class II-peptide complex and the T-cell receptor (TCR), does not lead to active presentation unless a secondary signal, between the other molecules also takes place. The CD80/86:CD28 pairing is the most investigated and shown to be present in the interface tissues of aseptic loosening (from Revell, J Histotechn 2006, 29, 287±295, with permission). © 2008, Woodhead Publishing Limited

380

Joint replacement technology

secondary signals are present. The presence of co-stimulatory molecules is taken as evidence of antigen presentation and T-cell activation in tissue sections and can be demonstrated in functional co-culture cell studies.104 Macrophage subpopulations in the implant interface tissues have been recognised for some time, both by the differences in cytokine and integrin expression at different levels of the interface,6,97±99,106,111,112,134,164 and by the use of particular markers, namely, the RFD series of MABs. Thus RFD1 recognises APC and RFD7 marks mature phagocytic macrophages.165 The findings by AlSaffar et al. were important in that they showed evidence of antigen-presentation in relation to wear debris at the implant interface for the first time, in that a proportion of macrophages and MNGCs were marked with RFD1 (Fig. 15.16).166 Furthermore, those samples having the highest percentages of RFD1 positive cells (70±90%) were those in which metal debris was a feature. Subsequently, antibodies against particular co-stimulatory molecules have become available and used to study interface tissues. Both CD80 and CD86 are present on macrophages and MNGCs as well as CD28 being found on the related T lymphocytes.167±169 There is a large predominance of CD86 over CD80 expression. The presence of CD40 on APCs and its counterligand, CD40L, on lymphocytes has also been shown at the bone±implant interface.126,170 Moreover, ICAM-1 and LFA-1 have been shown on interface macrophages and T cells.171 The expression of these costimulatory molecules in the context of particle phagocytosis has been demonstrated in various functional studies in cell culture.104,126,167,168 Most recently, the intracellular signalling molecules known as transcription factors in macrophages have been studied by Altaf, both in cell culture using a

15.16 Photomicrograph of the synovium-like layer and adjacent tissue at the implant interface in aseptic loosening, labelled with the monoclonal antibody RFD1 which marks antigen-presenting cells. The black material on the surface of the tissue is not wear debris but Indian ink painted on the tissue after retrieval but before processing to enable accurate orientation. © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

381

15.17 Low-power photomicrograph of the interface tissue in aseptic loosening showing the distribution of T-lymphocytes, labelled with monoclonal antibody against CD3. Note these cells are situated as a band towards the bottom of the synovium-like layer. Compare this distribution with that of antigen-presenting cells in Fig. 15.16.

cell line (U937) and in normal peripheral blood monocyte/macrophages as well as the expression of these molecules in situ by interface inflammatory macrophages.104 She systematically evaluated all the members of the NFB family of molecules (RelA, RelB, c-rel, p50, p52) showing these to be expressed by phagocytic cells in both the cell culture and tissue sample contexts using a variety of different techniques including immunohistochemistry, RT-PCR and fluorescent activated cell sorting (FACS) analysis. The molecule of particular relevance to the present discussion is RelB which is expressed by APCs, and not other macrophages, during their activation in the process of phagocytosis and antigen presentation. Expression of RelB has been demonstrated by Altaf both in interface inflammatory tissue and by cells phagocytosing wear debris in vitro.104 It is also interesting to note the similarity in the distribution of T lymphocytes (Fig. 15.17) and APC (RFD1 positive or CD80/86 expressing cells) (Fig. 15.16) which is towards the lower part of the synovium-like layer at the interface and in the perivascular region of the deeper tissue suggesting close interrelationship between these cells.

15.9.4 Lymphocyte migration into the interface tissue The lymphocytes and macrophages/MNGCs in the interface tissues are for the most part present as a result of migration from the blood vessels, though it is recognised that some lymphocyte proliferation occurs locally. Analysis of the molecules expressed by endothelial cells gives important additional information about the pathophysiology involved, since the adhesion molecules involved in inflammation are well known and now easily detected using MABs. The various stages in the process leading eventually to cell migration between the endothelial cells are illustrated in Fig. 15.18. Initial margination of the cells as they fall out of the axial stream in the blood vessel and their attachment to the endothelial cells is under the influence of selectins, with adhesion and eventual transmigration into the tissues involving integrins and cellular adhesion molecules (CAMs). The © 2008, Woodhead Publishing Limited

382

Joint replacement technology

15.18 Diagram to show the molecules involved in the interaction between cells in the circulating blood and endothelial cells as the former become attached to the latter then migrate through the vessel wall into the tissues. Cells fall out of the axial stream and come into contact with the endothelial cells, then are activated and attached before becoming adherent and migrating between endothelial cells.

endothelial cells of the vessels in the interface tissues in aseptic loosening have been shown to express P-selectin, E-selectin (ECAM-1), ICAM-1, VCAM-1 and CD44.77,118,164,171,172 A discussion as to which adhesion molecules may be involved with which integrins in the context of the interface membrane in aseptic loosening is available elsewhere.106,164 The presence of E-selectin (Fig. 15.19) is the finding of most interest and significance as it is this molecule that is known to be expressed on endothelial cells and that mediates the migration of T lymphocytes from blood vessels at sites of contact sensitisation in the skin.173 This finding of E-selectin expression therefore gives important information with respect to the pathophysiology of implant loosening.

15.10 Wear particles and corrosion products in distant organs: systemic effects The discussion so far has been confined to the local changes around the implanted joint, with a brief mention of lymphatic drainage of joints in Section 15.5. Thus, Case and his colleagues and others88±90 have described the presence of material derived from hip joints in lymph nodes and the spleen. There is considerable further evidence of dissemination of wear material from joints in the case of silicone derived from silastic finger and toe joints to local lymph nodes, from hip joints to pelvic lymph nodes and from the shoulder. A total of 15 separate references is available in the review by Al-Saffar and Revell.99 Dissemination of biomaterials from joints to other organs has also been reported, namely the spleen, liver, kidney and lung.88,90,174,175 © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

383

15.19 Low-power photomicrograph of the interface tissue in aseptic loosening to show the expression of E-selectin by endothelial cells (arrows) in the two blood vessels present. E-selectin is labelled with a specific monoclonal antibody. Note the presence of abundant macrophages and MNGCs containing metal debris (from Revell et al., Proc Instn Mech Engrs 1997, 211H, 187±197, with permission).

Elemental analysis shows the presence of metal in ionic form in blood, synovial fluid and urine of individuals having replacement joints and that the levels are higher for those having metal against metal articulations.176±180 It is assumed that these high metal concentrations are ionic in form, most likely bound to proteins, and this has been a cause for concern recently in the literature. They may be derived directly from the implant itself where corrosion is present, or they may presumably originate from the numerous particles present, also most likely by a corrosive process. It is known that the particles generated from metal±metal hip articulations are nanoparticles, both from isolation studies49 and joint simulators.181 The smaller a particle the greater is its relative surface area and thus the greater its chemical reactivity. Contact sensitisation to metals in proven cases where skin testing is positive has always been considered to be haptenic, which is to say that the metal ions are too small to illicit an immune response but that combination of the ions with larger molecules, namely proteins, enables immune processing and sensitisation. Contact sensitisation can be initiated in experimental animals by simple skin painting with soluble metal salts. It is in the light of these aspects that concern has been expressed over the possible adverse effects which may result in the long term from metal±metal articulations. That there are changes in lymphocyte populations in peripheral blood of those with metal implants in place, in the absence of frank sensitisation judged by skin testing, has been shown by Granchi et al.182 and Hart et al.180 In both cases, circulating T-lymphocyte levels were reduced, though no adverse effects of this were noted clinically. It is currently unknown from clinical studies whether a sensitisation process plays a part in aseptic loosening and/or osteolysis. The concern expressed over the issue of metal sensitisation in patients receiving second generation metal against metal (M±M) joint replacements has already been mentioned. Nine out of 165 patients having primary cementless M± © 2008, Woodhead Publishing Limited

384

Joint replacement technology

M total hip replacements had an osteolytic lesion and these individuals had a higher incidence of cobalt sensitivity on patch testing than controls. Perivascular lymphocytes and macrophages were a feature of the histological picture in retrieved samples of periprosthetic tissue collected during revision arthroplasty from two of the cases with early osteolysis.183 Similar observations were made by Willert and others.153,154 The reader is here asked to recall that previous sections of this chapter have described these features, along with considerable other molecular and cellular pathological evidence for a T-cell mediated immunological process of contact sensitisation type in aseptic loosening (Sections 15.8 and 15.9). While metal particles were almost certainly involved in most of the cases studied, the work of the author's own group has been solely concerned with cases in which the articulating surfaces were metal and polyethylene, at the hip or knee. Under these circumstances, polyethylene wear debris is predominant and indeed the presence of osteolysis has been shown to be directly correlated with the number of polyethylene particles isolated from the tissue.184 However, these polyethylene particles need not necessarily be immunologically inert as it has been shown that various proteins including type I collagen, aggrecan proteoglycans and immunoglobulins are bound to polyethylene wear particles in aseptic loosening.185 The other question that arises is whether nanoparticles of a material are more toxic than microparticles of the same material. The toxicity of nanoparticles has been reviewed recently by the author, and although various of these are known to be toxic, causing lung disease and malignancy, there are others which appear relatively harmless.186 Carbon nanoparticles, as carbon black or diesel fume, have marked deleterious effects, while the nanoparticulate diamond form of carbon is less toxic than microparticles of diamond.104,125,186 No direct comparisons of carbon black and nanodiamond using the same cell culture system have been performed but it would seem likely that a clear difference between these two forms of nanoparticulate carbon would be shown. There is a need for careful comparative studies of the effects of different metals in nanoparticulate and microparticulate forms on macrophages in cell culture, but preliminary work suggests that nanoparticles are not necessarily more toxic.

15.11 Summary and conclusions This chapter has attempted to review the pathophysiology of loosening of replacement joints where these are considered to be related to biological cellular mechanisms. Infection in a replaced joint is a major cause for concern. The diagnosis of infection may be straightforward in many cases but the criteria for reaching such a diagnosis are not yet clearly agreed. The part played by radiology, blood tests and microbiology has been described. The usefulness, or otherwise, of histopathological examination of tissue has been discussed. Aseptic loosening is by far the most common reason for revision of a total © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

385

joint replacement and has been considered in most detail. There are various nonbiological factors that may contribute to, or even be the main culprits in, loosening of any particular case, but there is undoubtedly a significant contribution made by wear debris and the cellular reaction to this in the implant bed. The involvement of phagocytic cells in dealing with foreign material has been described and the likely interactions with lymphocytes, which may or may not give rise to an immunologically mediated process, has been discussed. Sensitisation to metals as demonstrated by positive skin or blood lymphocyte tests is known to be accompanied by replacement joint failure in some individuals. The question remains as to whether this process is also present in a proportion of individuals with aseptic loosening in the absence of other clearly defined and clinically recognised signs of sensitisation. Evidence from the various studies performed in the author's own laboratories and from other centres suggests that the cellular reactions detected in the tissues in aseptic loosening are indeed those of contact sensitisation, showing as they do all the signs of a type IV cell mediated immune reaction with TH1 cell involvement and active antigen presentation. If there remain problems in deciding exactly what the main pathogenetic process is in aseptic loosening, this could be compounded further by considering the possibility that not all cases considered to be aseptic are in fact occurring in the absence of infection. In this respect, it is worth being aware of the work on endotoxins in relation to wear particles, which holds that adherent endotoxin may be responsible for some of the phenomena observed in studies of the effects of particles.187±189 If such lipopolysaccharides are present on the surface of particles in vivo then the question of whether these are derived from a low-grade infective process or the presence of a biofilm arises and the boundary between aseptic loosening and septic loosening begins to become blurred. The conclusion at present has to be that there are recognisable processes in prosthetic joint failure with pain and loosening, some of which are related to the presence of microorganisms, some to mechanical effects and others to the presence of wear particles. That sensitisation occurs in some individuals is undeniable, but whether it is significant in many others remains a subject for further study requiring the sophisticated molecular techniques of modern biology coupled to the use of appropriate in vitro and in vivo models. Available evidence suggests that immunological processes may be more important than previously thought.

15.12 References 1. Mullhall K, Ghomrawi H, Scully S, Callaghan JJ, Saleh KJ. Current etiologies and modes of failure in total knee arthroplasty revision. Clin Orthop Rel Res. 2006; 446: 45±50. 2. Charnley J, Eftekhar N. Postoperative infection in total prosthetic replacement arthroplasty of the hip-joint. With special reference to the bacterial content of the air of the operating room. Br J Surg 1969; 56: 641±649. © 2008, Woodhead Publishing Limited

386

Joint replacement technology

3. Bauer TW, Parvizi J, Kobayashi N, Krebs V. Diagnosis of periprosthetic infection. J Bone Joint Surg 2006; 88A: 869±882. 4. Coventry MB. Treatment of infection occurring in total hip surgery. Orthop Clin N Am 1975; 6: 991±1003. 5. Garvin KL, Hanssen AD. Infection after total hip arthroplasty± Past, present and future. J Bone Joint Surg 1995; 70A: 1576±1588. 6. Al-Saffar N, Revell PA. Pathology of the bone±implant interfaces. J Long-term Effects Med Impl 1999; 9: 319±347. 7. Urban J, Garvin K. Infection after total hip arthroplasty. Curr Opin Orthop 2001; 12: 64±70. 8. Revell M, Stockley I, Norman P. Surgical management of the infected hip prosthesis. Chapter 19, in Limb D, Hay S, eds, Evidence for Orthopaedic Surgery, 1st edn. Shrewsbury: TFM Publishing Ltd, 2006: 233±243. 9. Forster IW, Crawford R. Sedimentation rate in infected and uninfected total hip arthroplasty. Clin Orthop Rel Res 1982; 168: 48±52. 10. Spangehl MJ, Masri BA, O'Connell JX, Duncan CP. Prospective analysis of preoperative and intraoperative investigations for the diagnosis of infection at the sites of two hundred and two revision total hip arthropalsties. J Bone Joint Surg 1999; 81A: 672±683. 11. Di Cesare PE, Chang E, Preston CF, Liu CJ. Serum interleukin-6 as a marker of periprosthetic infection following total hip and knee arthroplasty. J Bone Joint Surg 2005; 87A: 1921±1927. 12. Sakamoto K, Arakawa H, Mita S, Ishiko T, Ikei S, Egami H, Hisano S, Ogawa M. Elevation of circulating interleukin 6 after surgery: factors influencing the serum level. Cytokine 1994; 6: 181±186. 13. Barrack RL, Harris WH. The value of aspiration of the hip joint before revision total hip arthroplasty. J Bone Joint Surg 1993; 75A: 66±76. 14. Barrack RL, Jennings RW, Wolfe MW, Bertot AJ. The value of preoperative aspiration before total knee revision. Clin Orthop Relat Res 1997; 345: 8±16. 15. Della Valle CJ, Scher DM, Kim YH, Oxley CM, Desai P, Zuckerman JD, Di Cesare PE. The role of intraoperative Gram stain in revision total joint arthroplasty. J Arthroplasty 1999; 14: 500±504. 16. Revell PA. Examination of synovial fluid. Current Topics Pathol 1982; 71: 1±24. 17. Mason JB, Fehring TK, Odum SM, Griffin WL, Nussman DS. The value of white blood cell counts before revision total knee arthroplasty. J Arthroplasty 2003; 18: 1038±1043. 18. Trampuz A, Hanssen AD, Osmon DR, Mandrekar J, Steckelberg JM, Patel R. Synovial fluid leukocyte count and differential for the diagnosis of prosthetic knee infection. Am J Med 2004; 117: 556±562. 19. Lonner JH, Desai P, Di Cesare PE, Steiner G, Zuckerman JD. The reliability of analysis of intraoperative frozen sections for identifying active infection during revision hip or knee arthroplasty. J Bone Joint Surg 1996; 78A: 1553±1558. 20. Mirra JM, Amstutz HC, Matos M, Gold R. The pathology of the joint tissues and its clinical relevance in prosthesis failure. Clin Orthop Relat Res. 1976; 117: 221±240. 21. Pandey R, Berendt AR, Athanasou NA. Histological and microbiological findings in non-infected and infected revision arthroplasty tissues. The OSIRIS Collaborative Study Group. Oxford Skeletal Infection Research and Intervention Service. Arch Orthop Trauma Surg 2000; 120: 570±574. 22. Fehring TK, McAlister JA Jr. Frozen histologic section as a guide to sepsis in revision joint arthroplasty. Clin Orthop Relat Res 1994; 304: 229±237. © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

387

23. Padgett DE, Silverman A, Sachjowicz F, Simpson RB, Rosenberg AG, Galante JO. Efficacy of intraoperative cultures obtained during revision total hip arthroplasty. J Arthroplasty 1995; 10: 420±426. 24. Tunney MM, Patrick S, Gorman SP, Nixon JR, Anderson N, Davis RI, Hanna D, Ramage G. Improved detection of infection in hip replacements. A currently underestimated problem. J Bone Joint Surg Br 1998; 80B: 568±572. 25. Neut D, van Horn JR, van Kooten TG, van der Mei HC, Busscher HJ. Detection of biomaterial-associated infections in orthopaedic joint implants. Clin Orthop Relat Res.2003; 413: 261±268. 26. Wang ZM, Liu C, Dziarski R. Chemokines are the main proinflammatory mediators in human monocytes activated by Staphylococcus aureus, peptidoglycan, and endotoxin. J Biol Chem 2000; 275: 20260±20267. 27. Cordero J, Munuera L, Folgueira MD. Influence of metal implants on infection: an experimental study in rabbits. J Bone Joint Surg Br 1994; 76B: 717±720. 28. Arens S, Schlegel U, Printzen G, Ziegler WJ, Perren SM, Hansis M. Influence of materials for fixation implants on local infection; an experimental study of steel versus titanium dcp in rabbits. J Bone Joint Surg Br 1996; 78B: 647±651. 29. Gristina AG. Biomaterial-centered infection: microbial adhesion versus tissue integration. Science 1987; 237: 1588±1595. 30. Costerton JW. Biofilm theory can guide the treatment of device-related orthopaedic infections. Clin Orthop Relat Res 2005; 437: 7±11. 31. Tunney MM, Patrick S, Curran MD, Ramage G, Hanna D, Nixon JR, Gorman SP, Davis RI, Anderson N. Detection of prosthetic hip infection at revision arthroplasty by immunofluorescence microscopy and PCR amplification of the bacterial 16SrRNA gene. J Clin Microbiol 1999; 37: 3281±3290. 32. Mariani BD, Martin DS, Levine MJ, Booth RE Jr, Tuan RS. Polymerase chain reaction detection of bacterial infection in total knee arthroplasty. Clin Orthop Relat Res1996; 331: 11±22. 33. Clarke MT, Roberts CP, Lee PT, Gray J, Keene GS, Rushton N. Polymerase chain reaction can detect bacterial DNA in aseptically loosened total hip arathroplasties. Clin Orthop Relat Res 2004; 427: 132±137. 34. Madey SM, Callaghan JJ, OlejniczakJP, Goetz DD, Johnston RC. Charnley total hip arthroplasty with use of improved techniques of cementing. The results after a minimum of fifteen years of follow-up. J Bone Joint Surg 1997; 79A: 53±64. 35. Yamac T. The extraction and characterisation of wear particles from tissues around failed orthopaedic implants of different designs. PhD Thesis, University of London; 1999. 36. Savio JA, Overcamp LM, Black J. Size and shape of biomaterial wear debris. Clin Mater 1994; 15: 101±147. 37. Revell PA. Tissue reactions to joint prostheses and the products of wear and corrosion. Curr Top Pathol 1982; 71: 73±101. 38. O'Shea S, Swettenham KJ, Revell PA. A simple optical method for differentiation of two types of polymeric wear debris in tissue samples. J Mater Sci: Mater Med 1992; 3: 391±396. 39. Revell PA, Al-Saffar N, Kobayashi A. Biological reaction to debris in relation to joint prostheses. Proc Inst Mech Eng 1997; 211H: 187±197. 40. Willert HG, Ludwig J, Semlitsch M. Reactions of bone to methacrylate after hip arthroplasty: a long term gross, light microscopic and scanning electron microscopic study. J Bone Joint Surg Am 1974; 56A: 1368±1382. 41. Anthony PP, Gie GA, Howie CR, Ling RS. Localised endosteal bone lysis in © 2008, Woodhead Publishing Limited

388

42. 43. 44. 45. 46. 47. 48. 49.

50. 51. 52. 53.

54.

55. 56. 57. 58.

Joint replacement technology relation to the femoral components of cemented total hip arthroplasties. J Bone Joint Surg 1990; 72B: 971±979. Mochida Y, Boehler M, Salzer M, Bauer TW. Debris from failed ceramic-onceramic and ceramic-on-polyethylene hip prostheses. Clin Orthop Rel Res 2001; 389: 113±125. Bauer TW, Geesink RCT, Zimmerman R, McMahon JT. Hydroxyapatite-coated femoral stems: histological analysis of components retrieved at autopsy. J Bone Joint Surg 1991, 73A: 1439±1452. Revell PA. The biological effects of carbon fibre reinforced polyetheretherketone implants. 119th Meeting of American Orthopaedic Association, Pittsburgh, April 2006. Shanbhag AS, Jacobs JJ, Glant TG, Gilbert JL, Black J, Galante JO. Composition and morphology of wear debris in failed uncemented total hip replacement (THR). J Bone Joint Surg 1994; 76-B: 60±67. Campbell P, Ma S, Yeom B, McKellop HA, Schmalzried TP, Amstutz HC. Isolation of predominantly submicron-sized UHMWPE wear particles from periprosthetic tissues. J Biomed Mater Res 1995; 29: 127±131. Margevicius KJ, Bauer TW, McMahon JT, Brown SA, Merritt K. Isolation and characterization of debris in membranes around total joint prostheses (TJR). J Bone Joint Surg 1994; 76-A: 1664±1675. Maloney WJ, Smith RL, Schmalzried TP, Chiba J, Huene D, Rubash H. Isolation and characterization of wear particles generated in patients who have had failure of a hip arthroplasty without cement. J Bone Joint Surg 1995; 77-A: 1301±1310. Doorn PF, Campbell PA, Worrall J, Benya PD, McKellop HA, Amstutz HC. 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 1998; 42: 103±111. Case CP, Langkamer VG, James C, Palmer MR, Kemp AJ, Heap PF, Solomon L. Widespread dissemination of metal debris from implants. J Bone Joint Surg 1994; 76-B: 701±712. Revell PA. The biological effects of nanoparticles. Nanotechnology Perceptions 2006; 2: 283±298. ISO 17853. Method of extraction, separation and quantification of polymer and metal wear debris. Iwaki H, Kobayashi A, Kadoya Y, Revell PA, Al-Saffar N, Yamac T, Scott G, Freeman MAR, Rehman I. The size, shape and number of PMMA bone cement particles in failed total joint replacement. J Bone Joint Surg 1999; 81B Suppl I : 84± 85. Iwaki H, Miyaguchi M, Kobayashi A, Kadoya Y, Yamac T, Revell PA, Freeman MAR, Yamano Y. The size, shape and number if three kinds of wear particles in cemented hip arthroplasty. Transactions of the 6th World Congress on Biomaterials, 1434 April 2000; Hawaii. Lalor PA, Mapp PI, Hall PA, Revell PA. Proliferative activity of cells in the synovium as demonstrated by a monoclonal antibody, Ki67. Rheumatol Int 1987; 7: 183±186. Barland P, Novikoff AB, Hamerman D. Electron microscopy of the human synovial membrane. J Cell Biol 1962; 14: 207±220. Ghadially FN. The Fine Structure of Synovial Joints. 1983. Butterworths, London. Graabeck PM. Ultrastructural characteristics and endocytis functions of synoviocytes. 1988 PhD thesis, Aarhus University.

© 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

389

59. Hogg N, Palmer DG, Revell PA. Mononuclear phagocytes of normal and rheumatoid synovium identified with monoclonal antibodies. Immunology 1985; 56: 675±681. 60. Palmer DG, Selvendran Y, Allen C, Revell PA, Hogg N. Features of synovial membrane identified with monoclonal antibodies. Clin Exp Immunol 1985; 59: 529±583. 61. Forre O, Jhoen T, Lea T, Dubloug JH, Mellbye OT, Natvig JB, Pahle J, Solheim BG. In situ characterisation of mononuclear cells in rheumatoid tissues, using monoclonal antibodies. Scand J Immunol 1982; 16: 315±319. 62. Mapp PI, Revell PA. Ultrastructural characterisation of macrophages (type A cells) in the synovial lining. Rheumatol Int 1988; 8: 315±319. 63. Mayston V, Mapp PI, Davies PG, Revell PA. Fibronectin in the synovium of chronic inflammatory joint diease. Rheumatol Int 1984; 3: 129±133. 64. Mapp PI, Revell PA. The production of fibronectin by synovial intimal cells. Rheumatol Int 1985; 5: 229±237. 65. Pollock LE, Lalor P, Revell PA. Type IV collagen and laminin in the synovial intimal layer: an immunohistochemical study. Rheumatol Int 1990; 9: 277±280. 66. Revell PA, Al-Saffar N, Fish S, Osei D. Extracellular matrix of the synovial intimal cell layer. Ann Rheum Dis 1995; 54: 404±407. 67. Hale LP, Martin ME, McCollum DE, Nunley JA, Springer TA, Singer KH, Haynes BF. Immunohistologic analysis of the distribution of cell adhesion molecules within the inflammatory synovial microenvironment. Arthritis Rheum 1989; 32: 22±30. 68. Koch AE, Burrows JC, Haines GK, Carlos TM, Harlan JM, Leibovich SJ. Immunolocalisation of endothelial and leucocyte adhesion molecules in human rheumatoid and osteoarthritic tissues. Lab Invest 1991; 64: 313±20. 69. Szekanecz Z, Haines GK, Lin TR, Harlow LA, Goerdt S, Rayan G, Koch AE. Differential distribution of intercellular adhesion molecules (ICAM- 1, ICAM-2 and ICAM-3) and the MS-1 antigen in normal and diseased human synovial. The possible pathogenetic and clinical significance in rheumatoid arthritis. Arthritis Rheum 1994; 37: 221±31. 70. Wilkinson LS, Edwards JCW, Poston R, Haskard DO. Cell populations expressing VCAM-1 in normal and diseased synovium. Lab Invest 1993; 68: 82±88. 71. Henderson KJ, Edwards JC, Worrall JG. Expression of CD44 in normal and rheumatoid synovium and cultured synovial fibroblasts. Ann Rheum Dis 1994; 53: 729±734. 72. Stevens CR, Mapp PI, Revell PA. A monoclonal antibody (MAB 67) marks type B synoviocytes. Rheumatol Int 1990; 10: 103±106. 73. Goldring SR, Jasty M, Roelke MS, Rourke CM, Bringhurst FR, Harris WH. Formation of a synovial-like membrane at the bone±cement interface. Its role in bone resorption and implant loosening after total hip replacement. Arthritis Rheum 1986; 29: 836±842. 74. Revell PA, Lalor PA. Evidence for the development of a true synovial structure adjacent to orthopaedic implants. Pathol Res Pract 1991; 187: 753±754. 75. Lalor PA, Revell PA. The presence of a synovial layer at the bone±implant interface: an immunohistological study demonstrating the close similarity to true synovium. Clin Mater 1993; 14: 91±100. 76. N Al-Saffar N, Mah JTL, Kadoya Y, Revell PA. Neovascularisation and the induction of cell adhesion molecules in response to degradation products from orthopaedic implants. Ann Rheum Dis 1995; 54: 201±208. 77. McFarlane T, Revell PA. The expression of CD44 in archival paraffin embedded © 2008, Woodhead Publishing Limited

390

78.

79. 80. 81. 82. 83. 84. 85.

86. 87. 88. 89. 90. 91.

92. 93. 94.

Joint replacement technology interface tissues of failed orthopaedic implants. J Mater Sci: Mater Med 2004; 15: 315±319. Smith SC, Folefac VA, Osei DK, Revell PA. An immunocytochemical study of the distribution of proline-4-hydroxylase in normal, osteoarthritic and rheumatoid arthritic synovium at both the light and electron microscopy level. Br J Rheumatol 1998; 37: 287±291. Revell PA, Al-Saffar N. Inflammatory mediators in aseptic loosening of prostheses, in Downes S, Dabestani N, eds, Failure of Joint Replacement. A Biological, Mechanical or Surgical Problem. Institute of Orthopaedics, London, 1994; pp. 89±96. Freeman MAR, Bradley GW, Revell PA. Observations upon the interface between bone and polymethylmethacrylate cement. J Bone Joint Surg (Brit) 1982; 64: 489± 493. Levack B, Freeman MAR, Revell PA. The presence of macrophages at the bonePMMA interface of well-fixed prosthetic components. Acta Orthop Scand 1987; 58: 384±387. Bliss JP, Revell PA. Macrophage recruitment to experimentally implanted bone cement of differing surface contour. J Pathol 1988; 155: 342A. Revell P, Braden M, Weightman B, Freeman M. Experimental studies of the biological response to a new bone cement. II. Soft tissue reactions in the rat. Clin Mater 1992; 10: 233±238. Edwards JC, Sedgwick AD, Willoughby DA. The formation of a structure with the features of synovial lining by subcutaneous injection of air: an in vivo tissue culture system. J Pathol 1981; 134: 147±156. Vernon-Roberts B, Freeman MAR. Morphological and analytical studies of the tissues adjacent to joint prostheses: investigations into the causes of loosening of prostheses, in Shaldach M and Hohmann M, eds, Advances in Artificial Hip and Knee Joint Technology, Springer, Berlin, 1976; pp. 148±185. Revell PA. Pathology of Bone. Berlin, Springer, 1986; pp. 217±223. Jell G, Kerjaschki D, Revell P, Al-Saffar N. Lymphangiogenesis in the boneimplant interface of orthopedic implants: importance and consequence. J Biomed Mater Res 2006, 77A: 119±127. Case CP, Langkamer VG, James C, Palmer MR, Kemp AJ, Heap PF, Solomon L. Widespread dissemination of metal debris from implants. J Bone Joint Surg 1994; 76B: 701±712. Bae SC, Park CK, Jun JB, Kim SY, Bae DK. Multiple lymphadenopathy induced by wear debris after total knee replacement. Scand J Rheumatol 1996; 25: 388±390. Langkamer VG, Case CP, Heap P, Taylor A, Collins C, Solomon L. Systemic distribution of wear debris after hip replacement ± a cause for concern. J Bone Joint Surg 1992; 74B: 831±839. Revell PA, Gatti AM, Gambarelli A, Monari E, Hercus S, Saeed S, MacInnes T. Detection of CoCr particles in the spleen of guinea pigs six weeks after their intraosseous implantation. Proceedings of the 7th World Biomaterials Congress, Sydney, Australia, 2004; 753. Liebs T, Noble P, Alexander J, Monroe W. Interface pressures during cyclic loading of cemented femoral stems. Trans Orthop Res Soc 1997; 22: 208. Aspenberg P, Van der Vis H. Migration, particles, and fluid pressure. Clin Orthop Rel Res 1998; 352: 75±80. Gruen TA, McNeice GM, Amstutz HC. `Modes of failure' of cemented stem-type femoral components: a radiographic analysis of loosening. Clin Orthop Rel Res 1979; 141: 17±27.

© 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

391

95. Jasty MJ, Floyd WE, Schiller AL, Goldring SR, Harris WH. Localized osteolysis in stable, non-septic total hip replacement. J Bone Joint Surg 1986; 68A: 912±919. 96. Schmalzried TP, Maloney WJ, Jasty M, Kwong, Harris WH. Autopsy studies of the bone±cement interface in well-fixed cemented total hip arthroplasties. J Arthroplasty 1993; 8: 179±88. 97. Revell PA, Al-Saffar N, Kobayashi A. Biological reaction to debris in relation to joint prostheses. Proc Instn Mech Engrs 1997; 211H: 187±197. 98. Revell PA. Characterization of the cells and immunological reactions adjacent to aseptically loosened orthopaedic implants. J Histotechnol 2006; 29: 287±295. 99. Al-Saffar N, Revell PA. Pathology of the bone±implant interfaces. J Long Term Effects Med Impl 1999; 9: 319±347. 100. Revell PA, Jellie SE. Interleukin 15 production by macrophages in the implant interface membrane of aseptically loosened joint replacements. J Mat Sci: Mater Med 1998; 9: 727±730. 101. Al-Saffar N, Revell PA. Interleukin-1 production by activated macrophages surrounding loosened orthopaedic implants: a potential role in osteolysis. Br J Rheum 1994; 33: 309±316. 102. Kadoya Y, Revell PA, Kobayashi A, Al-Saffar N, Scott G, Freeman MAR. Wear particulate species and bone loss in failed total joint arthroplasties. Clin Orthop Rel Res 1997; 340: 118±129. 103. Clarke SA, Revell PA. Integrin expression at the bone/biomaterial interface. J Biomed Mater Res 2001; 57: 84±91. 104. Altaf H. The inflammatory response to particulate wear debris in the context of total hip replacement. PhD Thesis, University of London; 2007. 105. Curtis P. Signalling in macrophages following exposure to retrieved wear particles. PhD Thesis, University of London; 2002. 106. Clarke SA. Integrin expression at the bone biomaterial interface. PhD Thesis, University of London; 1999. 107. Howie DW, Haynes DR, Hay S, Rogers SD, Pearcy MJ. The effect of titanium alloy and cobalt chrome alloy wear particles on production of inflammatory mediators IL-1, TNF, IL-6, and prostaglandin E2 by rodent macrophages in vitro. Trans Orthop Res Soc 1992; 17: 344. 108. Lee SH, Brennan FR, Jacobs JJ, Urban RM, Ragasa DR, Glant TT. Human monocyte/macrophage response to cobalt±chromium corrosion products and titanium particles in patients with total joint replacements. J Orthop Res 1997; 15: 40±49. 109. Archibeck MJ, Jacobs JJ, Roebuck KA, Glant TT. The basic science of periprosthetic osteolysis. J Bone Joint Surg 2000; 82A: 1478±1489. 110. Jiranek WA, Machado M, Jasty M, Jevsevar D, Wolfe HJ, Goldring SR, Goldberg MJ, Harris WH. Production of cytokines around loosened cemented acetabular components. Analysis with immunohistochemical techniques and in situ hybridization. J Bone Joint Surg 1993; 75A: 863±879. 111. Al-Saffar N, Harris KA, Kadoya Y, Revell PA. Assessment of the role of GM-CSF in the cellular transformation and the devleopment of erosive lesions around orthopaedic implants. Am J Clin Path 1996; 105: 628±639. 112. Al-Saffar N, Revell PA. Differential expression of TGF alpha and MCSF/CSF-R (cfms) by multinucleated giant cells involved in pathological bone resorption at the site of orthopaedic implants. J Orthop Res 2000; 18: 800±807. 113. VidovszkyTJ, Cabanela ME, Rock MG, Berry DJ, Morrey BF, Bolander ME. Histologic and biochemical differences between osteolytic and nonosteolytic © 2008, Woodhead Publishing Limited

392

114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124.

125. 126. 127. 128. 129.

Joint replacement technology membranes around femoral components of an uncemented total hip arthroplasty. J Arthroplasty 1998; 13: 320±330. Westacott CI, Taylor G, Atkins R, Elson C. Interleukin-1 and production by cells isolated from membranes around aseptically loose total joint replacements. Ann Rheum Dis 1992; 51: 638±642. Al-Saffar, Revell PA, Khwaja HA, Bonfield W. Assessment of the role of cytokines in bone resorption in patients with total joint replacements. J Mater Sci: Mater in Med 1995; 6: 762±767. Takagi M, Konttinen YT, Lindy O, Sorsa T, Kurvinen H, Suda A, Santavirta S. Gelatinase/type IV collagenases in the loosening of total hip replacement endoprostheses. Clin Orthop Rel Res 1994; 306: 136±144. Hercus B, Saeed S, Revell PA. Expression profile of T cell associated molecules in the interfacial tissue of aseptically loosened prosthetic joints. J Mater Sci: Mater Med 2002; 13: 1153±1156. Hercus B. Modelling T lymphocyte reactions to biomedical materials. PhD thesis. University of London, 2005. Saeed S, Revell PA. Production and distribution of interleukin 15 and its receptors (IL-15R and IL-R2 ) in the implant interface tissues obtained during revision of failed total joint replacement. Int J Exp Path 2001; 82: 201±209. Merkel KD, Erdmann JM, McHugh KP, Abu-Amer Y, Ross FP, Teitelbaum SL. Tumor necrosis factor-alpha mediates orthopedic implant osteolysis. Am J Pathol 1999; 154: 203±210. Dorr LD, Bloebaum R, Emmanual J, Meldrum R. Histologic, biochemical, and ion analysis of tissue and fluids retrieved during total hip arthroplasty. Clin Orthop Rel Res 1990; 261: 82±95. Goodman SB, Chin RC, Chiou SS, Schurman DJ, Woolson ST, Masada MP. A clinical-pathologic-biochemical study of the membrane surrounding loosened and non-loosened total hip arthroplasties. Clin Orthop Rel Res 1989; 244: 182±187. Chiba J, Rubash HE, Kim KJ, Iwaki, Y. The characterization of cytokines in the interface tissue obtained from failed cementless total hip arthroplasty with and without femoral osteolysis. Clin Orthop Rel Res 1994; 300: 304±312. Goodman SB, Huie P, Song Y, Schurman D, Maloney W, Woolson S, Sibley R. Cellular profile and cytokine production at prosthetic interfaces. Study of tissues retrieved from revised hip and knee replacements. J Bone Joint Surg 1998; 80B: 531±539. Altaf H, MacFarlane T, Revell PA. The inflammatory potential of microparticles vs.nanoparticles in vitro. Proceedings of the 20th European Congress on Biomaterials, 371, Nantes, 27 September±1 October 2006. Altaf H, Revell PA. The characterisation of antigen presenting cells in the boneimplant interface and in response to biomaterial. Proceedings of the 7th World Biomaterials Congress, 370, Sydney, 17±21 May 2004. Bainbridge J, Al-Saffar N. Persistent expression of mitogenic/transforming factors at the site of failed orthopaedic implants: the impact on immune reactivity. J Mater Sci: Mater Med 1998; 9: 695±700. Moilanen E, Moilanen T, Knowles R, Charles I, Kadoya Y, Al-Saffar N, Revell PA, Moncada S. Nitric oxide synthase is expressed in human macrophages during foreign body inflammation. Am J Pathol 1997; 150: 881±887. Xu JW, Li TF, Partsch G, Ceponis A, Santavirta S, Konttinen YT. Interleukin-11 (IL-11) in aseptic loosening of total hip replacement (THR). Scand J Rheumatol 1998; 27: 363±367.

© 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

393

130. Xu JW, Konttinen YT, Waris V, Patiala H, Sorsa T, Santavirta S. Macrophagecolony stimulating factor (M-CSF) is increased in the synovial-like membrane of the periprosthetic tissues in the aseptic loosening of total hip replacement (THR). Clin Rheumatol 1997; 16: 243±248. 131. Xu JW, Ma J, Li TF, Waris E, Alberty A, Santavirta S, Konttinen YT. Expression of epidermal growth factor and transforming growth factor alpha in interfacial membranes retrieved at revision total hip arthroplasty. Ann Rheum Dis 2000; 59: 822±827. 132. Kadoya Y, Revell PA, Al-Saffar N, Kobayashi A, Scott G, Freeman MAR. The bone formation and bone resorption in failed total joint arthroplasties. Histomorphometric analysis with histochemical and immunohistochemical technique. J Orthop Res 1996; 14: 473±482. 133. Kadoya Y, Revell PA, Kobayashi A, Al-Saffar N, Scott G, Freeman MAR. Wear particulate species and bone loss in failed total joint arthroplasties. Clin Orthop Rel Res 1997; 340: 118±129. 134. Kadoya Y, Al-Saffar N, Kobayashi A, Revell PA. The expression of osteoclast markers on foreign body giant cells. Bone Mineral 1994; 27: 85±96. 135. Athanasou NA, Quinn J, Bulstrode CJK. Resorption of bone by inflammatory cells derived from the joint capsule of hip arthroplasties. J Bone Joint Surg 1992; 74B: 57±62. 136. Quinn J, Joyner C, Triffit JT, Athanasou NA. Polymethylmethacrylate-induced inflammatory macrophages resorb bone. J Bone Joint Surg 1992; 74B: 652±658. 137. Wang W, Ferguson DJ, Quinn JM, Simpson AH, Athanasou NA. Biomaterial particle phagocytosis by bone-resorbing osteoclasts. J Bone Joint Surg 1997; 79B: 849±856. 138. Boyce BF, Xing L. Biology of RANK, RANKL, and osteoprotegerin. Arthritis Res Ther 2007; 9(Suppl 1): S1. doi:10.1186/ar2165. 139. Mandelin J, Li TF, Liljestrom M, Kroon ME, Hanemaaijer R, Santavirta S, Konttinen Y. T imbalance of RANKL/RANK/OPG system in interface tissue in loosening of total hip replacement. J Bone Joint Surg 2003; 85B: 1196±1201. 140. Haynes DR, Crotti TN, Potter AE, Loric M, Atkins GJ, Howie DW, Findlay DM. The osteoclastogenic molecules RANKL and RANK are associated with periprosthetic osteolysis. J Bone Joint Surg 2001; 83: 902±911. 141. Lalor P, Revell PA. T-lymphocytes and titanium±aluminium±vanadium (TiAlV) alloy. Evidence for immunological events associated with debris deposition. Clin Mater 1993; 12: 57±62. 142. Lalor PA, Revell PA, Gray AB, Wright SG, Railton GT, Freeman MAR. Sensitivity to titanium. A cause of implant failure? J Bone Joint Surg 1991; 73B: 25±28. 143. Lalor PA, Gray AB, Wright S, Railton B, Freeman MAR, Revell P. Contact hypersensitivity to titanium hip prosthesis? A preliminary report. Contact Dermatitis 1990; 23: 193±194. 144. Parker AW, Drez D, Jacobs JJ. Titanium dermatitis after failure of metal-backed patellas. Am J Knee Surg 1993; 6: 129±131. 145. Elves MW, Wilson N, Scales T, Kemp HBS. Incidence of metal sensitivity in patients with total joint replacements. Brit Med J 1975; 4: 376±378. 146. Benson MKD, Goodwin PG, Brostoff J. Metal sensitivity in patients with joint replacement arthroplasties. Brit Med J 1975; 4: 374±375. 147. Evans EM, Freeman MAR, Miller AJ, Vernon-Roberts B. Metal sensitivity as a cause of bone necrosis and loosening of the prosthesis in total joint replacement. J Bone Joint Surg 1974; 56B: 626±642. © 2008, Woodhead Publishing Limited

394

Joint replacement technology

148. Nater JP, Brian RG, Deutman R, Mulder ThJ. The development of metal hypersensitivity in patients with metal-to-plastic hip arthroplasties. Contact Dermatitis 1976; 2: 259±261. 149. Pazzaglia UE, L Ceciliani L, Wilkinson MJ, Dell'Orbo C. Involvement of metal particles in loosening of metal-plastic total hip prostheses. Arch Orthop Trauma Surg 1985, 104: 164±174. 150. Waterman AH, Schrik J. Allergy in hip arthroplasty. Contact Dermatitis 1985; 13: 294±301. 151. Gil-Albarova J, Lacleriga A, Barrios C, Canadell J. Lymphocyte response to polymethyl methacrylate in loose total hip prostheses. J Bone Joint Surg 1992; 74B: 825±830. 152. Romaguera C, Grimalt F, Vilaplana J. Short communications: methyl methacrylate prosthesis dermatitis. Contact Dermatitis 1985; 12: 172±183. 153. Davies AP, Willert HG, Campbell PA, Learmonth ID, Case CP. An unusual lymphocytic perivascular infiltration in tissues around contemporary metal-onmetal joint replacements. J Bone Joint Surg 2005; 87A: 18±27. 154. Willert H-G, Buchhorn GH, Fayyazi A, Flury R, Windler M, Koster G, Lohmann CH. Metal-on-metal beaings and hypersensitivity in patients with artificial hip joints: a clinical and histomorphological study. J Bone Joint Surg 2005; 87A: 28±36. 155. Milosev I, Trebse R, Kovac S, Cor A, Pisot V. Survivorship and retrieval analysis of sikomet metal-on-metal total hip replacements at a mean of seven years. J Bone Joint Surg 2006; 88A: 1173±1182. 156. Al-Saffar N, Iwaki H, Revell PA. Direct activation of mast cells by prosthetic biomaterial particles. J Mat Sci: Mater Med 1998; 9: 849±853. 157. Hercus B, Saeed S, Revell PA. Expression profile of T cell associated molecules in the interfacial tissue of aseptically loosened prosthetic joints. J Mater Sci: Mater Med 2002; 13: 1153±1156. 158. Arora A, Song Y, Chun L, Huie P, Trindade, M, Lane Smith R, Goodman S. The role of the TH1 and TH2 immune responses in loosening and osteolysis of cemented total hip replacements. J Biomed Mater Res 2003; 64A: 693±697. 159. Hercus B, Revell PA. Phenotypic characteristics of T lymphocytes in the interfacial tissue of aseptically loosened prosthetic joints. J Mater Sci: Mater Med 2001; 12: 1063±1067. 160. Weyand CM, Geisler A, Brack ME, Bolander ME, Goronzy JJ. Oligoclonal T-cell proliferation and interferon-gamma production in periprosthetic inflammation. Lab Invest 1998; 78: 677±685. 161. Revell PA, Jellie SE. Interleukin 15 production by macrophages in the implant interface membrane of aseptically loosened joint replacements. J Mater Sci: Mater Med 1998; 9: 727±730. 162. Li TF, Santavirta S, Waris V, Lassus J, Lindroos L, Xu JW, Virtanen I, Konttinen Y. No lymphokines in T-cells around loosened hip prostheses. Acta Orthop Scand 2001; 72: 241±247. 163. Saeed S, Revell PA. Production and distribution of interleukin 15 and its receptors (IL-15R and IL-2R ) in the implant interface tissues obtained during revision of failed total joint replacement. Int J Exp Path 2001; 82: 201±209. 164. Clarke SA, Revell PA. Integrin expression at the bone/biomaterial interface. J Biomed Mater Res 2001; 57: 84±91. 165. Poulter LW, Campbell DA, Munro C, Janossy G. Discrimination of human macrophages and dendritic cells by means of monoclonal antibodies. Scand J Immunol 1986; 24: 351±357. © 2008, Woodhead Publishing Limited

Biological causes of prosthetic joint failure

395

166. Al-Saffar N, Revell PA, Kobayashi A. Modulation of the phenotypic and functional properties of phagocytic macrophages by wear particles from orthopaedic. J Mater Sci: Mater Med 1997; 9: 727±730. 167. Bainbridge JA, Revell PA, Al-Saffar N. Costimulatory molecule expression following exposure to orthopaedic implants wear debris. J Biomed Mater Res 2001; 54: 328±334. 168. Altaf H, Saeed S, Bhatt R, Revell PA. The assessment of antigen presenting cells in the bone±implant interface. Biomaterialen 2003; 4: 86. 169. Farber A, Chin R, Song Y, Huie P, Goodman S. Chronic antigen-specific immunesystem activation may potentially be involved in the loosening of cemented acetabular components. J Biomed Mater Res 2001; 55: 433±441. 170. Bhatt R, Saeed S, Altaf H, Revell PA. In vitro assessment of interactions between T-cells and antigen-presenting cells (Apcs) when challenged with biomaterials: the CD40±CD40L interaction. Proceedings of the 7th World Biomaterials Congress, Sydney, Australia, 2004, 488. 171. Al-Saffar N, Kadoya Y, Revell PA. The role of newly formed vessels and cell adhesion molecules in the tissue response to wear products from orthopaedic implants. J Mater Sci: Mater Med 169; 5: 813±818. 172. Altaf H, Revell PA. The characterisation of antigen presenting cells in the bone implant interface & in response to biomaterials. Proceedings of the 7th World Biomaterials Congress, Sydney, Australia, 2004, 686. 173. Norris P, Poston RN, Thomas DS, Thornhill M, Hawk J, Haskard DO. The expression of endothelial leucocyte adhesion molecule-1(ELAM-1), and intracellular adhesion molecule-1 (ICAM-1) in experimental cutaneous inflammation: a comparison of ultra-violet B erythema and delayed hypersensitivity. J Invest Dermat 1992; 96: 763±770. 174. Bauer TW, Saltarelli M, McMahon JT, Wilde AH. Regional dissemination of wear debris from total knee prostheses. J Bone Joint Surg 1993; 75A: 106±111. 175. Urban RM, Jacobs JJ, Tomlinson MJ, Gavrilovic J, Black J, Peoc'h M. Dissemination of wear particles to the liver, spleen and abdominal lymph nodes of patients with hip and knee replacement. J Bone Joint Surg 2000; 82A: 457±476. 176. Cracchiolo A III, Revell P. Metal concentration in synovial fluids of patients with prosthetic knee arthroplasty. Clin Orthop Rel Res 1982; 170: 169±174. 177. Sargeant A, Goswami T, Swank M. Ion concentrations from hip joints. J Surg Orthop Adv 2006; 15: 113±114. 178. Dorr L, Bloedbaum R, Emmanuel J, Meldrum RM. Histologic, biochemical, and ion analysis of tissue and fluids retrieved during total hip arthroplasty. Clin Orthop Rel Res 1990; 261: 82±95. 179. Daniel J, Ziaee H, Pradhan C, Pynsent PB, McMinn DJW. Blood and urine metal ion levels in young and active patients after Birmingham hip resurfacing arthroplasty. J Bone Joint Surg 2007; 89B: 169±173. 180. Hart AJ, Hester T, Sinclair K, Powell JJ, Goodship AE, Pele L, Fersht NL, Skinner J. The association between metal ions from hip resurfacing and reduced T-cell counts. J Bone Joint Surg 2006; 88B: 449±453. 181. Brown C, Williams S, Tipper JL, Fisher J, Ingham E. Characterisation of wear particles produced by metal on metal and ceramic on metal hip prostheses under standard and microseparation simulation. J Mater Sci: Mater Med 2007; 18: 819±827. 182. Granchi D, Ciapetti G, Stea S, Cavedagna D, Bettini N, Bianco T, Fontanesi, G, Pizzoferrato A. Evaluation of several immunological parameters in patients with aseptic loosening of hip arthroplasty. Chir Organi Mov 1995; 80: 399±408. © 2008, Woodhead Publishing Limited

396

Joint replacement technology

183. Park Y-S, Moon Y-W, Lim S-J, Yang J-M, Ahn G, Choi Y-L. Early osteolysis following second-generation metal-on-metal hip replacement. J Bone Joint Surg 2005; 87A: 1515±1521. 184. Kobayashi A, Freeman MAR, Bonfield W, Kadoya Y, Yamac T, Al-Saffar N, Scott G, Revell PA. Number of polyethylene particles and osteolysis in total joint replacements. J Bone Joint Surg 1997; 79B: 844±848. 185. Wooley PH, Fitzgerald RH Jr, Song Z, Davis P, Whalen JD, Trumble S, Nasser S. Proteins bound to polyethylene components in patients who have aseptic loosening after total joint arthroplasty. A preliminary report. J Bone Joint Surg 1999; 81A: 616±623. 186. Revell PA. The biological effect of nanoparticles. Nanotechnology Perceptions 2006; 2: 283±298. 187. Bi Y, Seabold JM, Kaar SG, Ragab AA, Goldberg VM, Anderson JM, Greenfield EM. Adherent endotoxin on orthopedic wear particles stimulates cytokine production and osteoclast differentiation. J Bone Miner Res 2001; 16: 2082±2091. 188. Greenfield EM, Bi Y, Ragab AA, Goldberg VM, Nalepka JL, Seabold JM. Does endotoxin contribute to aseptic loosening of orthopedic implants? J Biomed Mater Res B Appl Biomater 2005; 72: 179±185. 189. Ragab AA, Van De Motter R, Lavish SA, Goldberg VM, Ninomiya JT, Carlin CR, Greenfield EM. Measurement and removal of adherent endotoxin from titanium particles and implant surfaces. J Orthop Res 1999; 17: 803±809.

© 2008, Woodhead Publishing Limited

16

Using drug delivery systems to enhance joint replacement D P P I O L E T T I , Ecole Polytechnique FeÂdeÂrale de Lausanne, Switzerland

16.1

Why do we need to improve the outcome of orthopaedic implants?

Aseptic loosening accounts for more than two-thirds of hip revisions in Sweden.1 After total joint arthroplasty, a radiolucent zone is frequently observed at the interface of bone and implant.2 This radiolucent zone is associated with a progressive peri-implant bone resorption. The implant fixation is affected inducing therefore a risk of aseptic loosening. Two hypotheses are generally used to explain peri-implant bone resorption. The first hypothesis focuses on a biological reaction to wear particles, e.g. Horikoshi et al.,3 while the second hypothesis is based on biomechanical considerations, e.g. Huiskes and Nunamaker.4 Indeed, studies over the past two decades have strongly implicated osteoclasts as the major cause of the bone lysis leading to implant failure.5,6 Correspondingly, orthopaedic practice recently entered a new area by considering use of drugs to enhance the fixation of implants. Decreasing the catabolic bone activity could be a good strategy to avoid peri-implant bone loss. The drug of choice so far is of bisphosphonate type and several clinical trials have been performed and effectively showed a reduction of peri-implant bone loss in treated groups.7±9 Systemic injections of a drug may then potentially be interesting to control the bone remodelling around orthopaedic implant. However, this mode of delivery may not be optimal. The failure of an orthopaedic implant is strongly correlated to its bone fixation.10 When an implant is used without cement, stability immediately after the surgery must be obtained, a process called primary fixation, followed by a long-term fixation, a process called secondary fixation. A race on fixation quality is then engaged with the concept that the faster the secondary implant fixation is obtained, the better will be the outcome for the implant. The primary fixation being ensured by the press-fit technique, the drug should then target the secondary fixation with the goal of reducing bone loss. Secondary fixation indeed already happens during the first months following the surgery. This has © 2008, Woodhead Publishing Limited

398

Joint replacement technology

been shown in a clinical study where up to 14% bone loss arose during the first three months after total hip arthroplasties.11 Recent clinical studies have shown that systemic bisphosphonate treatment following a prosthesis implantation reduced peri-implant bone loss only three months after the start of the treatment.7,8 When compared with the early bone loss arising during the initial period of three months, systemic delivery of bisphosphonate would be ineffective to induce a rapid secondary fixation. Moreover, the proximal femur, the region where most of the bone resorption occurs, is reached with difficulty by bisphosphonate systemically injected as shown by the smaller decrease in proximal femoral fracture prevention compared with vertebra fracture prevention in a Phase III clinical studies for osteoporosis treatment with systemic bisphosphonate treatment.12 Since the targeted skeleton site is limited to the peri-implant bone and the drug should be rapidly available, it makes sense to use the implant itself as a drug delivery system enabling it to overcome the limitations of systemic delivery. Moreover, using local delivery reduces the amount of drug needed, decreasing the potential side-effect of the drug. With this idea of associating a drug and an implant to enhance the bone fixation, the uncemented implants category remains the optimal. Traditionally, uncemented hip prostheses are selected for young patients with an acceptable bone stock. The results of the Australian and Swedish registers demonstrate that for young patients, problems result essentially because of the wear particles liberated in function of the tribological characteristics of the femoral head and the socket.13,14 For this category of patients, the initial fixation and furthermore the secondary fixation are not the main difficulties. Nevertheless we can imagine a reduction in thigh pain due to the relative initial instability of the femoral component. We can conclude that the adjunction of a drug for this class of patients will be more potent if the effect continues over many years to prevent long-term bone resorption. For older patients with reduced bone quality, the commonly accepted solution for the femoral stem remains the cemented one. With the progression of the osteoporosis, we are confronted with very old patients suffering from osteoarthritis or femoral neck fracture. For this high morbid group of patients, the cemented phase of the surgical procedure remains dangerous.15 In this situation, the benefit of an uncemented femoral stem with fixation performance enhanced by the apposition of a specific drug could be easily demonstrated by a short-term follow-up study.16

16.2

What is the clinical situation for orthopaedic implants used as drug delivery systems?

Orthopaedic implants used as drug delivery systems have mainly targeted infection.17 A combination of bone cement and antibiotics can then be

© 2008, Woodhead Publishing Limited

Using drug delivery systems to enhance joint replacement

399

considered as a precursor approach in drug delivery systems for orthopaedic applications. Therapy of bone infections (osteomyelitis) was justified because of the poor accessibility of the infection site by common systemically administered antibiotics. Therefore, to improve therapy, resorbable calcium phosphate (CaP) ceramics materials,18 polymers such as methylmethacrylate either as beads19 or as cement20 have been used as carriers for antibiotics. They release effective drug amounts at the site of infection for several months and the systemic drug concentration remains low. Among the various antibiotics, vancomycin21,22 and gentamycin have been extensively investigated and proved efficacious in human osteomyelitis.23

16.3

Is the research for orthopaedic drug delivery systems advanced enough to translate it to clinical applications?

Corresponding to the clinical applications, most of the studies on drug delivery systems for orthopaedic applications were done for cements associated with antibiotics.24±28 This approach has been extended to improve the properties of bone substitutes by associating an osteogenic factor with a synthetic material. In this attempt, growth factors, such as transforming growth factor,29 platelet-derived growth factor,30 bone morphogenetic proteins (BMP),31 growth hormone32 and insulin-like growth factor-133 have been investigated successfully. Osteoarticular disorders associated with increased osteoclastic bone resorption (as observed in osteoporosis, Paget's disease, bone lytic tumours, periodontal disease, etc.) often lead to pathological fractures. They are widely treated by systemic administration of bisphosphonates, which are potent inhibitors of osteoclast activity. Association of CaP materials with bisphosphonates would increase the efficiency of bisphosphonate by being locally released and decreasing significant secondary effects (nephrotoxicity) observed after systemic treatments. In this objective, ceramic hydroxyapatite implants have been used in dental surgery. Denissen et al.34 reported that bisphosphonates could be beneficial in preventing the alveolar bone destruction associated with periodontal disease. He demonstrated the potential of bisphosphonatecomplexed hydroxyapatite implants on the repair of alveolar bone. The next step in the use of an orthopaedic implant as a drug delivery system is to combine the implant, the CaP and the bisphosphonate. Coating of orthopaedic implants with CaP is routinely performed, and the combination of CaP and bisphosphonate has a good potential. In particular, combined with CaP, bisphosphonate molecules can be released at very low concentrations,35 enabling evaluation of the biological activity of bisphosphonate-loaded materials using in vitro bone resorption assays. Bisphosphonate-loaded CaP materials were found to decrease the number and activity of osteoclastic cells.36 Indeed, in

© 2008, Woodhead Publishing Limited

400

Joint replacement technology

a pit resorption assay, osteoclastic resorption activity was markedly reduced. In addition, bisphosphonate-loaded CaP exhibited a dose-dependent inhibitory effect on osteoclastic activity similar to that observed with bisphosphonate solutions. These results clearly indicate that CaP matrices are suitable carriers for bisphosphonate, providing a bioactive drug delivery system whose release kinetics is compatible with the inhibition of bone resorption. The effect of implant coated with CaP and Zoledronate ± the latest of the bisphosphonate generation of drugs ± has been evaluated in in vivo studies. Both on osteoporotic and normal rats, implants used as a bisphosphonate delivery system induced a denser peri-implant trabecular architecture compared with normal implants as well as having a higher mechanical stability.37,38 Similar results have been obtained with Zoledronate39 or other bisphosphonates.40±42 A summary of bisphosphonates used in a drug delivery system for orthopaedic applications is given in Table 16.1. The combination of CaP and bisphosphonates does not change the surgical practice and needs only slight adjustments in the manufacturing process. Orthopaedic implants coated with CaP and bisphosphonate could then be easily translated to routine clinical practice. Bisphosphonate targets the catabolic aspect of bone remodelling. Beside bisphosphonates, different approaches have been proposed to control catabolic bone remodelling, such as local delivery of anti-tumour necrosis factor (TNF) therapy43 or calcitonin.44 An interesting review of the possible therapeutical approach for controlling the bone catabolic process has been published.45 In order to maintain bone quality around the implant, the anabolic aspect of bone remodelling should also be considered. Indeed, one of the first studies for an orthopaedic implant used as drug delivery system was performed by adding transforming growth factor beta-1 to the hydroxyapatite coating.46 A positive effect was observed on the amount of peri-implant bone ingrowth. The actual trend for anabolic process has focused on the use of BMP.47 The BMP, usually rhBMP-2, can be delivered by combining it with a CaP coating of the implant. Either hip implants48 or intramedullary nails49 were tested to evaluate the local delivery of BMP. Bone ingrowth and accelerated bone healing have been observed. Interestingly, it has been proposed to combine the use of BMP with bisphosphonate in order to act both on catabolic and on anabolic aspects of bone remodelling.50 There are two limitations for the use of rhBMP in clinical orthopaedic practice: the first one is the cost and the second is the supraphysiological dose needed to observe a therapeutical effect. A solution for these two problems may be to use not the BMP protein but its gene code. Indeed, this approach has been proposed by developing a DNA controlled-release coating for gene transfer.51 In vivo gene transfection of the peri-implant cells to upregulate the production of BMP could then be obtained. Beside CaP implant coating with BMP or bisphosphonate, different works have been performed combining metallic implant, polymer coating and drugs © 2008, Woodhead Publishing Limited

Table 16.1 List of works associating bisphosphonate and calcium±phosphate carrier used for orthopaedic applications Targeted application

Bisphosphonate

Carrier

Main performance

Reference

Alveolar bone destruction

(3-Dimethylamino-1hydroxypropylidene)1,1-P-C-P

Hydroxyapatite

In vitro: release affects osteoclasts but not osteoblasts

Denissen et al.34

Bone resorption around orthopaedic implant

Zoledronate

Calcium-deficient hydroxyapatite

In vitro: bisphosphonate release can be controlled

Roussie©re et al.35

Bone substitute in degenerative bone disease

Zoledronate

-tricalcium phosphate, calcium-deficient hydroxyapatite, hydroxyapatite

In vitro: inhibition of osteoclastic activity

Josse et al.36

Osteoporotic bone around orthopaedic implant

Zoledronate

Hydroxyapatite

In vivo: increase of implant mechanical stability

Peter et al.37

Bone resorption around orthopaedic implant

Zoledronate

Hydroxyapatite

In vivo: increase implant osseo-integration

Peter et al.38

Bone resorption around orthopaedic implant

Zoledronate

Hydroxyapatite

In vivo: bone augmentation around implant

Tanzer et al.39

Bone resorption around orthopaedic/dental implant

Pamidronate

Calcium-immobilised titanium implant

In vivo: new bone formation around implant

Kajiwara et al.40

Osteoporotic bone around orthopaedic implant

Ibandronate

Hydroxyapatite

In vivo: increase osseointegration surface implant

Kurth et al.41

© 2008, Woodhead Publishing Limited

WPTF3007

402

Joint replacement technology

such as TGB-beta1 or IGF-I.52 Bone mechanical properties were increased when local delivery of these growth factors was used. An original approach has been proposed by coating the metallic implant with collagen and different proteins.53 A controlled release of the proteins was obtained. Proteins or drugs can be effectively loaded on metallic implants by impregnating them with a polymer inside the pore of a titanium implant surface.54 This approach may allow better control of the release of the drug. From a general point of view, it can be anticipated that most of the developed biomaterials are or will be combined with either growth factors or drugs in order to functionalise them more and not just use them as filling materials for cavities.55±57

16.4

Will drug delivery systems be the future for orthopaedic implants?

While implants used as drug delivery systems are well developed for cardiovascular applications (drug-eluting stents) and are entering the market for diabetes management (insulin pump), this approach is still at its infancy for orthopaedic applications. Based on the growing evidence obtained in different in vivo studies, it seems quite clear that the orthopaedic implant used as a drug delivery system induces faster bone healing as well as a more mechanically stable situation for the implant. Orthopaedic implants are primarily designed to support mechanical load in the skeleton. As for the normal bone remodelling process, bone formation occurs where the skeleton is mechanically stimulated. It would then be beneficial to correlate the drug delivery with the mechanical situation surrounding the implant. This approach has been recently proposed by designing release of growth factors in response to a mechanical signal.58 The amount of drug concentration to be coated in order to obtain a controlled periimplant bone remodelling could also be obtained using computer simulation.59,60 As mentioned, orthopaedic implant industries are entering a new field by considering more biologically oriented products. This new field also has some implications from the regulatory process. Indeed, an orthopaedic implant used as a drug delivery system is considered as a combination product by the US Food and Drug Administration (FDA) (http://www.fda.gov/oc/combination/) and correspondingly the registration of these new implants followed a different process from traditional orthopaedic implants. The combination product will be evaluated on its primary mode of action and the regulation process will then depend on it. It would then be advantageous for an orthopaedic implant used as a drug delivery system to have its primary mode of action related to the implant part and not its drug action. Orthopaedic implants used as a drug delivery system represent the future in orthopaedic development. This is of importance not only for orthopaedic companies but also for pharmaceutical companies, as the combination of drugs and implants has been identified as the future of the pharmaceuticals.61 © 2008, Woodhead Publishing Limited

Using drug delivery systems to enhance joint replacement

16.5

403

References

1. Malchau H, Herberts P, Eisler T, Garellick G, Soderman P. The Swedish total hip replacement register. J Bone Joint Surg Am. 2002; 84-A Suppl 2: 2±20. 2. Reckling FW, Asher MA, Dillon WL. A longitudinal study of the radiolucent line at the bone±cement interface following total joint-replacement procedures. J Bone Joint Surg Am 1977; 59: 355±358. 3. Horikoshi M, Macaulay W, Booth RE, Crossett LS, Rubash HE. Comparison of interface membranes obtained from failed cemented and cementless hip and knee prostheses. Clin Orthop Relat Res 1994: 69±87. 4. Huiskes R, Nunamaker D. Local stresses and bone adaption around orthopedic implants. Calcif Tissue Int 1984; 36 Suppl 1: S110±117. 5. Haynes DR, Crotti TN, Zreiqat H. Regulation of osteoclast activity in peri-implant tissues. Biomaterials 2004; 25: 4877±4885. 6. Pioletti DP, Kottelat A. The influence of wear particles in the expression of osteoclastogenesis factors by osteoblasts. Biomaterials 2004; 25: 5803±5808. 7. Nehme A, Maalouf G, Tricoire JL, Giordano G, Chiron P, Puget J. Effect of alendronate on periprosthetic bone loss after cemented primary total hip arthroplasty: a prospective randomized study. Rev Chir Orthop Reparatrice Appar Mot 2003; 89: 593±598. 8. Venesmaa PK, Kroger HP, Miettinen HJ, Jurvelin JS, Suomalainen OT, Alhav EM. Alendronate reduces periprosthetic bone loss after uncemented primary total hip arthroplasty: a prospective randomized study. J Bone Miner Res 2001; 16: 2126± 2131. 9. Yamaguchi K, Masuhara K, Yamasaki S, Nakai T, Fuji T. Cyclic therapy with etidronate has a therapeutic effect against local osteoporosis after cementless total hip arthroplasty. Bone 2003; 33: 144±149. 10. Mjoberg B. Fixation and loosening of hip prostheses. A review. Acta Orthop Scand 1991; 62: 500±508. 11. Venesmaa PK, Kroger HP, Miettinen HJ, Jurvelin JS, Suomalainen OT, Alhava EM. Monitoring of periprosthetic bmd after uncemented total hip arthroplasty with dualenergy x-ray absorptiometry ± a 3-year follow-up study. J Bone Miner Res 2001; 16: 1056±1061. 12. Liberman UA, Weiss SR, Broll J, Minne HW, Quan H, Bell NH, Rodriguez-Portales J, Downs RW, Jr., Dequeker J, Favus M. Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. The alendronate phase III osteoporosis treatment study group. N Engl J Med 1995; 333: 1437±1443. 13. Graves SE, Davidson D, Ingerson L, Ryan P, Griffith EC, McDermott BF, McElroy HJ, Pratt NL. The Australian Orthopaedic Association National Joint Replacement Registry. Med J Aust 2004; 180: S31±34. 14. Soderman P, Malchau H, Herberts P, Zugner R, Regner H, Garellick G. Outcome after total hip arthroplasty: Part II. Disease-specific follow-up and the Swedish national total hip arthroplasty register. Acta Orthop Scand 2001; 72: 113±119. 15. Gierer P, Landes J, Grubwinkler M, Gradl G, Lob G, Andress HJ. The femoral neck fracture in the elderly patient ± cemented or cementless hip arthroplasty? Zentralbl Chir 2002; 127: 514±518. 16. Andress HJ, von Ruckmann B, Zwonitzer R, Kahl S, Ringling M, Lob G. Changes in bone density of the femur after cement-free implantation of a modular hip prosthesis with a long shaft. Unfallchirurg 2001; 104: 622±628. 17. Wu P, Grainger DW. Drug/device combinations for local drug therapies and © 2008, Woodhead Publishing Limited

404

Joint replacement technology

infection prophylaxis. Biomaterials 2006; 27: 2450±2467. 18. Radin S, Campbell JT, Ducheyne P, Cuckler JM. Calcium phosphate ceramic coatings as carriers of vancomycin. Biomaterials 1997; 18: 777±782. 19. Walenkamp GH, Kleijn LL, de Leeuw M. Osteomyelitis treated with gentamicinPMMA beads: 100 patients followed for 1±12 years. Acta Orthop Scand 1998; 69: 518±522. 20. Stabile DE, Jacobs AM. Local antibiotic treatment of soft tissue and bone infections of the foot. J Am Podiatr Med Assoc 1990; 80: 345±353. 21. Gautier H, Daculsi G, Merle C. Association of vancomycin and calcium phosphate by dynamic compaction: in vitro characterization and microbiological activity. Biomaterials 2001; 22: 2481±2487. 22. Obadia L, Amador G, Daculsi G, Bouler JM. Calcium-deficient apatite: influence of granule size and consolidation mode on release and in vitro activity of vancomycin. Biomaterials 2003; 24: 1265±1270. 23. Yamashita Y, Uchida A, Yamakawa T, Shinto Y, Araki N, Kato K. Treatment of chronic osteomyelitis using calcium hydroxyapatite ceramic implants impregnated with antibiotic. Int Orthop 1998; 22: 247±251. 24. El-Ghannam A, Ahmed K, Omran M. Nanoporous delivery system to treat osteomyelitis and regenerate bone: gentamicin release kinetics and bactericidal effect. J Biomed Mater Res B Appl Biomater 2005; 73: 277±284. 25. Penner MJ, Masri BA, Duncan CP. Elution characteristics of vancomycin and tobramycin combined in acrylic bone-cement. J Arthroplasty 1996; 11: 939±944. 26. von Frauhofer JA, Polk HC, Jr, Seligson D. Leaching of tobramycin from pmma bone cement beads. J Biomed Mater Res 1985; 19: 751±756. 27. Walenkamp GH, Vree TB, van Rens TJ. Gentamicin-PMMA beads. Pharmacokinetic and nephrotoxicological study. Clin Orthop Relat Res 1986: 171±183. 28. Yu D, Tsai CL, Wong J, Fox JL. Hydroxyapatite cement based drug delivery systems: drug release in vitro. J Formos Med Assoc 1991; 90: 953±957. 29. Blom EJ, Klein-Nulend J, Wolke JG, van Waas MA, Driessens FC, Burger EH. Transforming growth factor-beta1 incorporation in a calcium phosphate bone cement: material properties and release characteristics. J Biomed Mater Res 2002; 59: 265±272. 30. Lee YM, Park YJ, Lee SJ, Ku Y, Han SB, Klokkevold PR, Chung CP. The bone regenerative effect of platelet-derived growth factor-bb delivered with a chitosan/ tricalcium phosphate sponge carrier. J Periodontol 2000; 71: 418±424. 31. Suh DY, Boden SD, Louis-Ugbo J, Mayr M, Murakami H, Kim HS, Minamide A, Hutton WC. Delivery of recombinant human bone morphogenetic protein-2 using a compression-resistant matrix in posterolateral spine fusion in the rabbit and in the non-human primate. Spine 2002; 27: 353±360. 32. Guicheux J, Gauthier O, Aguado E, Pilet P, Couillaud S, Jegou D, Daculsi G, Heymann D. Human growth hormone locally released in bone sites by calciumphosphate biomaterial stimulates ceramic bone substitution without systemic effects: a rabbit study. J Bone Miner Res 1998; 13: 739±748. 33. Laffargue P, Fialdes P, Frayssinet P, Rtaimate M, Hildebrand HF, Marchandise X. Adsorption and release of insulin-like growth factor-i on porous tricalcium phosphate implant. J Biomed Mater Res 2000; 49: 415±421. 34. Denissen H, van Beek E, van den Bos T, de Blieck J, Klein C, van den Hooff A. Degradable bisphosphonate-alkaline phosphatase-complexed hydroxyapatite implants in vitro. J Bone Miner Res 1997; 12: 290±297. 35. RoussieÁre H, Montavon G, Samia LB, Janvier P, Alonso B, Fayon F, Petit M, © 2008, Woodhead Publishing Limited

Using drug delivery systems to enhance joint replacement

36. 37.

38. 39. 40. 41. 42. 43.

44. 45. 46. 47. 48. 49.

50.

405

Massiot D, Bouler JM, Bujoli B. Hybrid materials applied to biotechnologies: Coating of calcium phosphates for the design of implants active against bone resorption disorders. J Mater Chem 2005; 15: 3869±3875. Josse S, Faucheux C, Soueidan A, Grimandi G, Massiot D, Alonso B, Janvier P, Laib S, Pilet P, Gauthier O, Daculsi G, Guicheux JJ, Bujoli B, Bouler JM. Novel biomaterials for bisphosphonate delivery. Biomaterials 2005; 26: 2073±2080. Peter B, Gauthier O, Laib S, Bujoli B, Guicheux J, Janvier P, van Lenthe GH, Muller R, Zambelli PY, Bouler JM, Pioletti DP. Local delivery of bisphosphonate from coated orthopedic implants increases implants mechanical stability in osteoporotic rats. J Biomed Mater Res A 2006; 76: 133±143. Peter B, Pioletti DP, Laib S, Bujoli B, Pilet P, Janvier P, Guicheux J, Zambelli PY, Bouler JM, Gauthier O. Calcium phosphate drug delivery system: influence of local zoledronate release on bone implant osteointegration. Bone 2005; 36: 52±60. Tanzer M, Karabasz D, Krygier JJ, Cohen R, Bobyn JD. The Otto Aufranc award: bone augmentation around and within porous implants by local bisphosphonate elution. Clin Orthop Relat Res 2005; 441: 30±39. Kajiwara H, Yamaza T, Yoshinari M, Goto T, Iyama S, Atsuta I, Kido MA, Tanaka T. The bisphosphonate pamidronate on the surface of titanium stimulates bone formation around tibial implants in rats. Biomaterials 2005; 26: 581±587. Kurth AH, Eberhardt C, Muller S, Steinacker M, Schwarz M, Bauss F. The bisphosphonate ibandronate improves implant integration in osteopenic ovariectomized rats. Bone 2005; 37: 204±210. Tengvall P, Skoglund B, Askendal A, Aspenberg P. Surface immobilized bisphosphonate improves stainless-steel screw fixation in rats. Biomaterials 2004; 25: 2133± 2138. Schwarz EM, Campbell D, Totterman S, Boyd A, O'Keefe RJ, Looney RJ. Use of volumetric computerized tomography as a primary outcome measure to evaluate drug efficacy in the prevention of peri-prosthetic osteolysis: a 1-year clinical pilot of etanercept vs. placebo. J Orthop Res 2003; 21: 1049±1055. Aldini NN, Caliceti P, Lora S, Fini M, Giavaresi G, Rocca M, Torricelli P, Giardino R, Veronese FM. Calcitonin release system in the treatment of experimental osteoporosis. Histomorphometric evaluation. J Orthop Res 2001; 19: 955±961. Rodan GA, Martin TJ. Therapeutic approaches to bone diseases. Science 2000; 289: 1508±1514. Sumner DR, Turner TM, Purchio AF, Gombotz WR, Urban RM, Galante JO. Enhancement of bone ingrowth by transforming growth factor-beta. J Bone Joint Surg Am 1995; 77: 1135±1147. Seeherman H, Wozney JM. Delivery of bone morphogenetic proteins for orthopedic tissue regeneration. Cytokine Growth Factor Rev 2005; 16: 329±345. Bragdon CR, Doherty AM, Rubash HE, Jasty M, Li XJ, Seeherman H, Harris WH. The efficacy of bmp-2 to induce bone ingrowth in a total hip replacement model. Clin Orthop Relat Res 2003: 50±61. Schmidmaier G, Wildemann B, Cromme F, Kandziora F, Haas N, Raschke M. Bmp2 coating of implants increase biomechanical strength and accelerate bone remodeling in fracture treatment. 48th Orthopedic Research Society, Dallas, Feb, 2002: p. 0332. Seeherman H, Li J, Blake C, Gavin D, Wozney J, Bouxsein ML. Histology indicates bisphosphonate limit transient resorption without decreasing bone induction in nonhuman primate core defects treated with rhbmp-2/acs. 23rd Annual Meeting of ASBMR, Oct 12±16 2001, Phoenix.

© 2008, Woodhead Publishing Limited

406

Joint replacement technology

51. Labhasetwar V, Bonadio J, Goldstein S, Chen W, Levy RJ. A DNA controlledrelease coating for gene transfer: transfection in skeletal and cardiac muscle. J Pharm Sci 1998; 87: 1347±1350. 52. Wildemann B, Sander A, Schwabe P, Lucke M, Stockle U, Raschke M, Haas NP, Schmidmaier G. Short term in vivo biocompatibility testing of biodegradable poly(D,L-lactide)-growth factor coating for orthopaedic implants. Biomaterials 2005; 26: 4035±4040. 53. Puleo DA. Release and retention of biomolecules in collagen deposited on orthopedic biomaterials. Artif Cells Blood Substit Immobil Biotechnol 1999; 27: 65±75. 54. Agrawal CM, Pennick A, Wang X, Schenck RC. Porous-coated titanium implant impregnated with a biodegradable protein delivery system. J Biomed Mater Res 1997; 36: 516±521. 55. Heller J. Poly(ortho esters) ± a bioerodible polymer system specifically designed for drug delivery. Eur Pharmac Contractor 2003: 86±88. 56. Price JS, Tencer AF, Arm DM, Bohach GA. Controlled release of antibiotics from coated orthopedic implants. J Biomed Mater Res 1996; 30: 281±286. 57. Saito N, Murakami N, Takahashi J, Horiuchi H, Ota H, Kato H, Okada T, Nozaki K, Takaoka K. Synthetic biodegradable polymers as drug delivery systems for bone morphogenetic proteins. Adv Drug Deliv Rev 2005; 57: 1037±1048. 58. Lee KY, Peters MC, Anderson KW, Mooney DJ. Controlled growth factor release from synthetic extracellular matrices. Nature 2000; 408: 998±1000. 59. Peter B, Pioletti DP, Terrier A, Rakotomanana LR. Orthopedic implant as drug delivery system: a numerical approach. Comp Meth Biomech Biomed Eng 2001; 4: 505±513. 60. Zygourakis K, Markenscoff PA. Computer-aided design of bioerodible devices with optimal release characteristics: a cellular automata approach. Biomaterials 1996; 17: 125±135. 61. Dubin CH. A one-two punch: drug/medical device combination products are taking healthcare in a new direction. Is the pharmaceutical industry prepared? Drug Deliv Tech 2004; 4.

© 2008, Woodhead Publishing Limited

17

Sterilization of joint replacement materials

A I A N U Z Z I and S M K U R T Z , Exponent, Inc., USA

17.1

Introduction

17.1.1 Clinical importance of maintaining sterile joint replacement materials There is clinical importance in manufacturing sterile joint replacement materials and maintaining the sterility of those materials. One obvious purpose of choosing effective sterilization methods is the reduction of infection. In general, the risk of joint infection is relatively low for knee (1±2%), hip (0.3±1.3%), and shoulder (<1%) arthroplasty (Hanssen and Rand, 1999; Lidgren et al., 2003; Sia et al., 2005; Sperling et al., 2001), but the results of deep infection can be devastating. Complications related to deep infection of total joint replacement may include pain, reoperation, potential loss of the prosthesis, and, in some cases, loss of the limb or life (Sia et al., 2005). The manufacturer is responsible for producing sterile joint replacement prostheses and providing them in packaging that will maintain the sterility of the components over time. The method of sterilization and packaging should be chosen carefully such that the implant materials do not degrade. Care should also be taken to select sterilization and packaging methods that do not alter the chemical composition or mechanical properties of the prosthesis. The purpose of this chapter is to review sterilization methods that are relevant to medical devices and joint replacement prostheses in particular. Relevant established national and international standards and their terminology are reviewed. Lastly, the effects of sterilization methods and packaging systems are discussed as they pertain to the mechanical properties and biocompatibility of the sterilized implant.

17.1.2 Testing and regulation The ISO defines `sterile' as `free from viable microorganisms' (ISO 11135: 1994). The term `sterilization' is defined as `validated process used to render a product free of all forms of viable microorganisms' (ISO 11135:1994). The © 2008, Woodhead Publishing Limited

408

Joint replacement technology

purpose of sterilizing joint replacement materials is to inactivate any microbiological contaminants that may be present on the surface of the device (Dorman-Smith, 1997). The ISO standard defines inactivation as `loss of ability of microorganisms to grow and/or multiply' (ISO/TS 11139, `Sterilization of health care products ± Vocabulary,' definition 2.21). In addition, the reduction of abundant dead bacteria on the prosthesis can prevent pyrogenicity and other undesirable effects on a patient with a reduced immune system (Dorman-Smith, 1997). The Food and Drug Administration (FDA) provides guidance for good manufacturing practice (GMP) via national and international standards for the production and proper sterilization procedures that should be followed for the manufacture of joint replacement materials that will be marked as `sterile.' A `clean' surface differs from one that is `sterile' in that a clean surface is one in which the bulk material extends all the way to the surface without change in chemical composition and without any foreign species attached to the surface (Kasemo and Lausmaa, 1988). Perfectly clean surfaces usually have unsaturated chemical bonds on the surface, which attract surface contaminants that stick to the surface (i.e. adsorb) via strong bonds (Kasemo and Lausmaa, 1988). The density and nature of the new adsorbed layer of contaminants depends on the pretreatment history (including sterilization procedures) (Kasemo and Lausmaa, 1988), the mechanisms of which are described in more detail in the sections that follow. Clean surfaces with unsaturated bonds have a higher surface energy than surfaces that have contaminants (Kasemo and Lausmaa, 1988). After a period of time, multiple layers of contaminant molecules may bond to the surface (Kasemo and Lausmaa, 1988), potentially affecting the biocompatibility of the device. The FDA provides guidance for the evaluation of medical device sterility. These documents are publicly available and can be found through the FDA online search database. In 1997, the Center for Devices and Radiological Health (CDRH) decided that the safety and effectiveness of a device manufacturer's sterilization process would best be ensured through compliance with the Quality System (QS) regulation rather than through a 510(k) review. Changes in sterilization procedures would require a submission of a new 510(k) only if the changes in the sterilization procedure results in the alteration of properties/ specifications of a device or result in a lower sterility assurance level (SAL); otherwise, appropriate documentation of the change should be maintained in accordance with QS regulation requirements. On 30 August 2002, the FDA issued the guidance document, `Updated 510(k) Sterility Review Guidance K90-1; Guidance for Industry and FDA.' The purpose of this document was to provide additional guidance to help FDA review staff and 510(k) sponsors differentiate between various types of nontraditional methods of sterilization, how they are employed and how they should be handled. Traditional and non-traditional methods are described in more detail in the sections that follow. © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

409

For every method of sterilization, the FDA Office for Device Evaluation reviews the following information for devices that are labeled as sterile: · The sterilization method that will be used. · A description of the method that will be used to validate the sterilization cycle (not the validation data itself). · Description of the packaging that will be used to maintain the device sterility (not the package integrity testing data itself). · For ethylene oxide (EtO) sterilized devices, the maximum levels of EtO and ethylene chlorhydrin (ECH) residuals that remain on the device (the recommended standard is ANSI/AAMI/ISO 10993-7:1995 Biological Evaluation of Medical Devices ± Part 7: Ethylene oxide sterilization residuals'). · If the product is labeled `pyrogen free,' a description of the method used to make this determination should be included. · The sterility assurance level specification (SAL), which should be 10ÿ6 for all devices except those that only contact intact skin (10ÿ3). · Where radiation sterilization is used, the radiation dose should be reported.

17.1.3 Sterility assurance levels The ISO defines `sterile' as `free from viable microorganisms,' but recognizes that `[i]n practice no such absolute statement regarding the absence of microorganisms can be proven' (ISO 11135:1994). The presence of viable microorganisms can be expressed in terms of probability, as the nature of microbial death can be described as an exponential function (ISO 11135:1994). The probability of viable organisms being present on a product after unit sterilization, the sterility assurance level (SAL), can be estimated based on this function (ISO 11135:1994). SAL for medical devices is recommended by the FDA to be 10ÿ6, or a probability of one in a million that the item is non-sterile. Validation Validation is a systems approach to test a sterilizer and product to determine if the process is performing properly (as reviewed by Gillis, 1981). The process of validation includes: documentation of the manufacturing process; installation qualification on all equipment; metrology on all process sensors, indicators, recorders, and controllers; and performance qualification of the system. Performance qualification is often mistakenly portrayed as validation because it involves the physical performance of the sterilizer as well as the biological performance of the process on the product. Performance qualification should be retested at least annually, as well as after any major change to the sterilizer or any change to the product or packaging. It should also be retested after any failure in the testing of the product that may be explained by failure of the process equipment. © 2008, Woodhead Publishing Limited

410

Joint replacement technology

The validation process must be documented thoroughly, including the protocols, procedures, specifications, and reports; each of which must have appropriate review and approval signatures for quality assurance (Gillis, 1981). The protocol provides the technical detail of the scope of the study and points to the supportive documentation. The procedures describe the stepby-step processes for each task in the total validation process. Specifications describe the material or standard processes (e.g. ISO standards), and are tested using the protocol process. Lastly, reports are produced using the experimental data (raw or summary data) that were generated by performing the protocol, and they should reference the protocol and detailed procedures used. Any deviations from the procedures should be detailed in the report. The actual validation testing program is segmented into three phases: equipment installation qualification, calibration, and performance qualification. The first phase, equipment installation qualification, is where: each piece of equipment used in the sterilization process must be examined for accuracy, completeness and discrepancies in the engineering drawings; manufacturer's specifications are reviewed; the operation of the equipment is evaluated to ensure that it is functioning properly; spare part listings are identified; and those components that are critical to the control of the process are identified. The second step, calibration, is where all sensors, indicators, controllers, and recorders that are determined to be critical to the process are calibrated. This step should indicate the precision and accuracy of these devices. The frequency of calibration is determined based on the performance history of the device and how critical it is to the process; the frequency of validation may be before and after every use, or it could occur every 3±6 months. The last step, performance qualification, is where the product and process equipment are tested together. This step should be conducted only after the sterilization cycle has been developed for the product and the equipment has been evaluated and calibrated such that meaningful data are produced. The performance is qualified by placing parameter sensors throughout the sterilizing chamber and inside the product packaging in places where the sterilization of the chamber and/or product is most difficult to sterilize. A qualified system is one that demonstrates acceptable control over the desired parameters within the sterilizing chamber and product, as well as one that demonstrates acceptable microbial lethality. The completion of the three phases described above results in a validated sterilization process.

17.1.4 Packaging Packaging must be considered carefully with respect to the device that is being sterilized, as well as the type of sterilization method being used. The packaging acts as a biological barrier to maintain a sterile product, but an appropriate packaging must be selected such that it does not block the sterilant or inhibit the © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

411

sterilization process in some way (Gillis, 1981). The environment inside the packaging must also be considered, as it should not allow the passage of, or contain, contaminants or reactive substances that could change the devices being sterilized. The importance of proper selection of these processes can be demonstrated using the history of sterilization and packaging of ultra-high molecular weight polyethylene (UHMWPE) as an example, which is described in more detail in the sections that follow.

17.2

Sterilization techniques and their suitability

According to the FDA guidance document entitled `Updated 510(k) Sterility Review Guidance K90-1; Guidance for Industry and FDA,' the FDA considers there to be two categories of sterilization methods: traditional and non-traditional. Traditional methods include: · · · · ·

dry heat sterilization; moist heat sterilization; EtO with devices placed in a fixed chamber; radiation (gamma and electron beam); liquid chemical sterilants for sterilizing single-use devices incorporating materials of animal origin. Non-traditional methods of sterilization include:

· · · · · · · ·

EtO not using a fixed chamber; high-intensity light; chlorine dioxide; ultraviolet light; combined vapor and gas plasma; vapor systems (e.g. peroxide or peracetic acid); filtration methods; limited use of a liquid peracetic acid system in endoscopy and with metal instruments.

In addition, the Office of Device Evaluation may consider non-traditional methods that employ a unique or novel sterilant, where the device manufacturer should follow additional procedures in their 510(k) submission. Sterilization methods that are relevant to total joint replacement prostheses (moist heat or autoclave, irradiation, ethylene oxide, and a few non-traditional methods) are described in detail in the sections that follow.

17.2.1 Moist heat (autoclave) Autoclave sterilization (moist heat) is a commonly used method for sterilization of medical devices (Dorman-Smith, 1997). The process is carried out in an © 2008, Woodhead Publishing Limited

412

Joint replacement technology

autoclave, where the temperature is raised to exceed 121 ëC, typically ranging from 121 to 141 ëC with pressure in the range 206±368 kPa (An et al., 2005). The duration of the process depends on the temperature chosen, as well as the configuration and packaging of the device (Dorman-Smith, 1997); however, the minimum recommended standard for sterilization is exposure to steam at 1 bar (equivalent to 121 ëC for 15 min) (An et al., 2005). Biological indicators, which are defined as `test system[s] containing viable microorganisms providing a defined resistance to a specified sterilization process' per ISO 11138-1:2006 (`Sterilization of Health Care Products ± Biological Indicators ± Part 1: General Requirements'), are used. The organism chosen as the indicator must be as resistant as the most difficult to kill organism that is found on the device in the natural bioburden (Dorman-Smith, 1997). ISO 11138-1:2006 also recommends that the test organism be a strain that is suitable for handling without containment facilities, does not need specific containment procedures for handling, and does not have specific transport/mailing requirements. The test organism should also be one sufficiently stable to maintain its resistance characteristics for the duration of the stated shelf-life when transported and stored in accordance with label directions. Traditionally, test organisms have been bacterial spores, typically derived from the Bacillus or Geobacillus species. Viability of the indicators is tested, which involves incubation that can take up to seven days. The product cannot be released until it has been verified that the indicators are not viable (Dorman-Smith, 1997). The temperatures in the autoclave can be too severe for use on certain plastics (which could result in changes of material properties), but it is suitable for other materials utilized in total joint replacement prostheses, including devices made of metals or ceramics (Dorman-Smith, 1997). It is an inexpensive process that can quickly sterilize porous materials and accessible surfaces and spaces (An et al., 2005). Disadvantages of the technique include the requirement that the materials have pronounced stability in conditions of high heat and humidity, and in particular they must have resistance to hydrolytic decomposition (An et al., 2005). Steam itself can also carry contaminants to implant surfaces and lead to a lesser degree of biocompatibility (An et al., 2005), which is described in more detail for relevant materials in the sections that follow.

17.2.2 Irradiation Sterilization by irradiation destroys microorganisms by ionization, where the ionization penetrates the implants and contaminating organisms (An et al., 2005). The penetrating radiation releases energy, which forms ionized, excited atoms and molecules along the path of the radiation; this energy is several orders of magnitude greater than the energy contained in a covalent bond, which makes possible a variety of chemical reactions in the product as well as the microorganisms (An et al., 2005). Gamma irradiation involves exposure of the © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

413

materials to be sterilized to gamma rays from cobalt-60 or cesium-137 radionucleotides (Dorman-Smith, 1997). Alternatively, an electron beam (e-beam) generator may be used (Dorman-Smith, 1997). A typical minimum dosage for sterilization of medical devices is 25 kGy, with dosage ranging from 25 to 40 kGy (Kurtz, 2004). For bone allografts (which are typically used in revision surgeries to fill voids in bone), tissue banks apply gamma irradiation doses that may range from 15 to 35 kGy (Nguyen et al., 2006). It is not recommended practice or required to use biological indicators with irradiation, but indicators such as Bacillus pumilus may be used (Dorman-Smith, 1997). For parametric release of a load, dosimeters registering the amount of radiation falling on them can be utilized; they should be placed in a region that will be exposed to the minimum dose of the load or in a reference location that has a known relationship to the minimum dose (Dorman-Smith, 1997). Irradiation processes are advantageous because they do not cause surface contamination (An et al., 2005). Another advantage is the reproducibility, which allows for the use of physical indicators (dosimeters) that can be included in or on the packaging to demonstrate that the minimum dose has been achieved. The dosimeter can be read immediately and the product released quickly (DormanSmith, 1997). In general, biological indicators are not required (and hence the incubation period that is required for other sterilization processes does not apply), except for the occasional spot checks or for process validation (Dorman-Smith, 1997). Other advantages include high efficiency, minimal thermal effects, and that it can be conducted on packaged and/or sealed products (An et al., 2005). Disadvantages of irradiation include the fact that both gamma irradiation and e-beam sterilization can impose considerable restraints on the sizes of packaging that can be put through the system due to the effects that backscatter and energy absorption have on penetration distance (Dorman-Smith, 1997). Sterilization times can be excessive. There is also the problem of polymer degradation, which could result in poor performance of the implant (An et al., 2005).

17.2.3 Ethylene oxide EtO sterilization has been commercially available since the 1980s (Bruck and Mueller, 1988). EtO is a toxic gas that causes an internal chemical reaction in the microorganisms, which is generally ascribed to alkylating functional groups of enzymes, other proteins, and nucleic acids (Bruck and Mueller, 1988). The medical device is prepared and exposed to EtO in a sealed chamber, which may be up to 100% EtO at sub-atmospheric pressures (Dorman-Smith, 1997). Gas mixtures including chlorofluorocarbon, carbon dioxide, or nitrogen may also be used at pressures well above atmospheric pressure (Dorman-Smith, 1997). It is required that the packaging be porous to EtO, while at the same time the seals must maintain sterility and withstand the rate of pressure changes that can be imposed by the process (Dorman-Smith, 1997). © 2008, Woodhead Publishing Limited

414

Joint replacement technology

17.1 Gas-permeable packaging for ethylene sterilization of Durasul highly crosslinked UHMWPE components, which was used by Centerpulse, Inc. (Austin, TX). Reprinted with permission from The UHMWPE Handbook: UltraHigh Molecular Weight Polyethylene in Total Joint Replacement, S.M. Kurtz, `Packaging and sterilization of UHMWPE', pp. 37±51, Copyright Elsevier (2004).

The process is carried out in a sealed chamber for a sufficient time to allow the gas to penetrate all of the packaging material, as well as the enclosed devices, followed by a removal of residual gas (Dorman-Smith, 1997). An example of contemporary packaging that is used for EtO sterilization is shown in Fig. 17.1. An effective sterilization process depends on the temperature, gas concentration, exposure time, and relative humidity (An et al., 2005). Compared with other methods, EtO sterilization is carried out at relatively low temperatures (i.e., from 20 to 50 ëC), which makes the process ideal for materials with low melting temperature (plastics) (Dorman-Smith, 1997). Materials being sterilized with EtO need to be placed in a well-ventilated room or placed in an aerator after sterilization for least 48 h at 50 ëC under reduced pressure (Bruck and Mueller, 1988). Similar to irradiation sterilization, biological indicators must be included with the load during EtO sterilization (Dorman-Smith, 1997). The viability of the biological indicator for each load has to be tested before the load can be released, which usually takes up to seven days of incubation (Dorman-Smith, 1997). A concern with EtO sterilization is device toxicity, which may be caused by process residuals (i.e., EtO and ethylene chlorohydrin (ECH)). EtO is a strong alkylating agent (Bruck and Mueller, 1988), and it may cause concern for hemolytic reactions (An et al., 2005). Some materials resist absorption of EtO, while others can be desorbed fairly rapidly by heating and air washing the material (Dorman-Smith, 1997). According to ISO10993-7:1995, `Biological Evaluation of Medical Devices ± Part 7: Ethylene oxide sterilization residuals,' the amount of product residuals depends on the sterilization process parameters, with factors such as: material composition; packaging, EtO sterilization cycle; © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

415

Table 17.1 Average and maximum daily dose for ethylene oxide and ethylene chlorohydrin residuals for permanent contact devices (per ISO 10993-7) Residual

Average daily dose

Ethylene oxide (EtO) Ethylene chlorohydrin (ECH)

0.1 mg/day 2 mg/day

First 24 hours First 30 days 20 mg 12 mg

60 mg 60 mg

Lifetime 2.5 g 50 g

aeration parameters such as temperature, load density, and configuration, air flow, loading pattern, surface area of the product and aeration time; and sample retrieval. Maximum allowable residues for devices sterilized with EtO are designed in ISO 10993-7; specifically, the allowable limit for devices with permanent contact (i.e. `whose single, multiple or long-term use or contact exceeds 30 days') have the dose requirements outlined in Table 17.1. As outlined in ISO 10993-7, the amount of residuals in EtO sterilized devices can be determined by two extraction methods: simulated use extraction or exhaustive extraction. Simulated use extraction methods produce results that are expressed in terms of delivered dose, while exhaustive extraction is intended to cover the entire residual content of a device. The latter produces results that would tend to represent a dose greater than or equal to one that the patient might receive, and is the recommended method for evaluating permanent contact devices. For measurement of EtO, extraction procedures include thermal extraction followed by headspace gas analysis. Solvent extraction with either headspace gas analysis of the solvent extract, chromatography of the solvent extract, or preparation of the bromohydrin derivative of EtO is determined using a more sensitive gas chromotography detector. Water is typically used for ECH extraction. Small devices may be placed in a vial and subjected to the extraction procedures in their entirety, while larger devices require taking representative samples of material components. In the latter case, it may be necessary to take representative portions from several locations. The advantages of EtO make the process suitable for sterilization of most plastics, due to the low melting temperatures (Dorman-Smith, 1997). A large quantity of product can be treated at the same time (Dorman-Smith, 1997), and EtO sterilization offers excellent penetration characteristics with minimal polymer damage (An et al., 2005; Costa et al., 1998). Disadvantages of the technique include long processing and aeration times (owing to the amount of time required for penetration into complex or tortuous device pathways), which can substantially increase costs for devices that require a quick turnaround. This process is also confounded by the requirement for biological indicators with each load and the subsequent time for incubation/testing them (Dorman-Smith, 1997). A large number of variables also need to be controlled (Dorman-Smith, 1997). The lethality of EtO gas and toxicity of residuals, as well as environ© 2008, Woodhead Publishing Limited

416

Joint replacement technology

mental considerations, are of concern (Dorman-Smith, 1997). Residuals of EtO sterilization have been implicated in EtO burns (e.g. Cardenas-Camarena, 1998; Karacalar and Karacalar, 2000), hemolysis (Anand et al., 2003) or potentially in allergic reactions (Bellucci, 2004).

17.2.4 Non-traditional methods As described above, the FDA lists non-traditional methods of sterilization: EtO not using a fixed chamber, high-intensity light, chlorine dioxide, ultraviolet light, combined vapor and gas plasma, vapor systems (e.g., peroxide or peracetic acid), filtration methods, and limited use of a liquid peracetic acid system in endoscopy and with metal instruments. The methods relevant to total joint replacement prostheses are described below. Plasma or radio-frequency glow discharge (RFGD) is a method that utilizes gas plasma consisting of electrons, ions or neutral particles, where the ionized gas deactivates the organisms on the surface of the implant (Bruck and Mueller, 1988). The low-temperature gas plasmas are generated with gases (e.g., argon, nitrogen, or helium) that are stimulated with radio-frequency waves under a deep vacuum, where the plasma can be used for surface cleaning to ensure biocompatibility, change surface characteristics (e.g., water affinity), and for chemical barrier enhancement (An et al., 2005; Bruck and Mueller, 1988). Two methods that have been used in the orthopaedic industry are Sterrad (Johnson & Johnson Medical, Inc., Irvine, CA) and Plazlyte (Abtox Inc., Mundelein, IL) (Kurtz et al., 1999). Gas plasma (Plazlyte system) has been used by DePuy Orthopaedics, Inc. (Warsaw, IN), since the 1990s to routinely sterilize UHMWPE components (Kurtz, 2004; Kurtz et al., 1999); see Fig. 17.2. However, modification of polymeric surface chemistry and appearance remain a concern (Lerouge et al., 2002). Advantages of gas plasma methods include shorter processing time, smaller effect on material dimensions/properties, no toxic by-products, and easier disposal (An et al., 2005). Disadvantages include that the ability to penetrate into lumens and through materials is limited so that the inside surfaces of the device are not sterilized (Rutala et al., 1998). Cost efficiency is also an issue (An et al., 2005). Sterilization using ultraviolet (UV) light has also been utilized for medical devices, where the wavelength ranges from 210 to 328 nm and bactericidal effect peaks at 240±280 nm (An et al., 2005). The optimal wavelength range can be generated by most mercury-vapor lamps (An et al., 2005). The biological mechanism for this sterilization method is that the UV light targets microbial DNA, where thymine dimers are created that result in non-coding regions in the bacterial DNA and subsequent cell death (Delgado and Schaaf, 1990). An effective dose of 16-second dynamic UV exposure was effective in sterilizing various dental implants, stainless steel cortical bone screws, and polysulfone polymer healing caps with at least 90% success rate (Delgado and Schaaf, 1990). © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

417

17.2 Contemporary packaging for gamma sterilization of Enduron TM conventional UHMWPE components, used by DePuy Orthopaedics (Warsaw, IN). Reprinted with permission from The UHMWPE Handbook: Ultra-High Molecular Weight Polyethylene in Total Joint Replacement, S.M. Kurtz, `Packaging and sterilization of UHMWPE', pp. 37±51, Copyright Elsevier (2004).

Disadvantages of the technique include that it is suitable only for outer surfaces of the device and has a limited penetration effect, as well as a lack of surface cleaning effect (An et al., 2005). Investigational methods of sterilization include supercritical fluid carbon dioxide (SCF), which is a method of carbon dioxide sterilization. As demonstrated in a rat animal model, titanium implants sterilized with supercritical phase carbon dioxide had comparable biocompatibility to implants sterilized using other methods (Hill et al., 2006). Recent studies also indicate that inactivation of bacteria follows linear kinetics (allowing for estimation of SAL) and that the process appears to be `gentle' in that the morphology and protein profiles of the inactivated bacteria are largely unchanged (White et al., 2006). The technique is advantageous because it offers both sterilization and cleaning, where bacterial debris and other contaminants may be removed (An et al., 2005). The time to sterilization is comparable to steam autoclaving and significantly faster than EtO or irradiation sterilization (An et al., 2005). Although recent studies have demonstrated inactivation of bacterial loads, results have been inconsistent and the mechanism of inactivation is not known (White et al., 2006). It is thought that the method of inactivation could include acidification, lipid modification, inactivation of essential enzymes, or extraction of intracellular substances (White et al., 2006).

17.3

Issues with sterilization of joint replacement materials

As described briefly in the preceding sections, sterilization processes may substantially alter chemical and/or material properties of total joint replacement © 2008, Woodhead Publishing Limited

418

Joint replacement technology

constituents, thus affecting the performance and/or biocompatibility of the implant in vivo. In the sections that follow, materials that are commonly used in total joint replacement prostheses are described with respect to the material changes that have been reported with varying sterilization procedures.

17.3.1 Metals A perfectly clean metal (M) exposed to air will rapidly form an oxide layer (MxOy) on the surface by reaction with oxygen. This passivating surface oxide layer is usually desired, and in metals used for medical implants it will stop growing in the thickness range of 1±10 nm, or 5±50 atomic layers thick (Kasemo and Lausmaa, 1988). Metals with surface oxide layers may still be regarded as `clean.' However, the oxidized surface contains unsaturated chemical bonds (high surface energy) that attract air impurities such as hydrocarbons or other molecules (Kasemo and Lausmaa, 1988). After about 1 min most of the unsaturated chemical bonds will be saturated, with one molecular monolayer adsorbed on the surface (Kasemo and Lausmaa, 1988). It has been reported that steam autoclaving increases the thickness of the oxide layer on various surgical metal alloy surfaces. For example, after steam Ê thick have been observed, which is ten autoclaving oxide layers up to 650 A times thicker than non-autoclaved implants (Lausmaa et al., 1985). Ti surfaces have increased oxide layer thickness due to autoclaving and contain contaminants such as fluorine, alkali metals, and silicon (Lausmaa et al., 1985). In another study (Baier et al., 1982), cast surgical vitallium discs and subperiosteal implants with three different surfaces (which ranged from clean, high-energy metals to uniformly low-energy organic layers) were subjected to steam sterilization; hydrophobic organic and hydroscopic salt contaminants were observed over the implant surfaces that were exposed to steam sterilization. Surfaces of commercially pure titanium (cp-Ti) implants were altered following sterilization treatments, especially during steam sterilization where surface oxide layers were thicker and were contaminated with Fe and Cl (An et al., 2005; Keller et al., 1990). The changes in metal surface chemistry could have detrimental effects on tissue responses during the early stages of implantation (An et al., 2005), particularly with respect to initial cell±implant interaction at the tissue±implant interface. This is of particular importance with respect to `cementless' joint replacement components, where surface morphology (e.g. porous coatings) or calcium phosphate coatings (e.g. hydroxyapatite) are utilized to promote bony fixation to the implant surface. In these components, cell attachment and spreading during the initial phases are necessary for proper fixation of the implant. In a cell culture study, cp-Ti samples that were sterilized with different methods and then cultured with fibroblasts demonstrated the greatest amount of cell attachment when the surfaces were sterilized with UV light, followed by © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

419

exposure to EtO, RFGD, and lastly steam autoclaving (Keller et al., 1990). The differences in cell attachment were attributed to the amount of surface contamination, which was greatest for the surfaces that were steam sterilized (Keller et al., 1990). Murine fibroblast attachment and spreading was decreased on cp-Ti discs that were sterilized with UV light, EtO, or steam autoclaving, respectively; cell attachment and spreading was also poorer with increases in the number of cycles (one, five, or ten) for both EtO and steam autoclaving (Vezeau et al., 1996). Sterilization procedures that increase surface energy and improve wettability have been demonstrated to improve cell attachment. Ti surfaces subjected to RFGD cleaning and sterilization exhibited enhanced osteoblast attachment, with the highest level of cell attachment noted on surfaces that were treated for 1 min (Swart et al., 1992). Ti surfaces that were treated for longer periods, however, exhibited inorganic contaminants that may have interfered with cell attachment (Swart et al., 1992). Osteoneogenesis was observed in an animal model that had Ti implants sterilized with RFGD or UV light, while osteoneogenesis was not observed in implants sterilized using autoclaving (Swart et al., 1992). Sterilized surfaces that contained oxygen, carbon, and nitrogen contaminants had altered implant surface energies, which affected cell attachment and spreading in several different cell types (An et al., 2005; Campbell and von Recum, 1989; Keller et al., 1990; Matlaga et al., 1976; Ungersbock et al., 1994). These studies underscore the importance of obtaining sterile, clean implant surfaces. Although cell attachment may be affected by sterilization method, the effect on phenotypic expression of the cells is less clear. To investigate this, bone cells (osteoblasts) can be cultured on the surfaces and monitored for increases in osteocalcin production or alkaline phosphatase activity (both of which are associated with bone mineralization). Cell culture studies investigating rat calvarial osteoblast-like cells indicate that combinations of surface roughness and sterilization procedures may affect phenotypic response of osteoblast-like cells, where plasma cleaned surfaces showed the highest level of osteocalcin expression and alkaline phosphatase activity on the smoothest high-energy surfaces (Stanford et al., 1994). However, differences in phenotypic expression were not observed with different surface roughness conditions for surfaces sterilized with other methods (i.e., UV light, autoclave, and EtO) (Stanford et al., 1994). Human osteoblast precursor cells did not demonstrate differences in alkaline phosphatase activity or osteocalcin production during the first seven days that they were cultured on cp-Ti surfaces that were sterilized with plasma-glow discharge versus controls, but at day 10 alkaline phosphatase activity was higher for control surfaces versus the glow-discharged surfaces (Youngblood and Ong, 2003). In addition, utilization of multiple sterilization procedures for metals may be cause for concern with respect to the initial effects of cell attachment and cell spreading, again related to contaminants on the implant surfaces (An et al., © 2008, Woodhead Publishing Limited

420

Joint replacement technology

2005). Titanium surfaces that were contaminated by sterilization and were autoclaved repeatedly demonstrated levels of bacterial colonization that were three to four orders of magnitude greater than other modes of sterilization (Drake et al., 1999). Sterilization methods have also demonstrated differences in healing times in animal models (An et al., 2005). Ti surfaces that were treated with UV light and RFGD demonstrated faster healing in the initial five-week to three-month period compared to autoclaving in an in vivo pig model, although there was no notable difference after a six-month period (Budd et al., 1991). Supercritical carbon dioxide sterilization of titanium implants inserted in a rat animal model demonstrated a similar biological response to other forms of sterilization such as UV radiation, steam autoclave, EtO, and RFGD, as indicated by the similarity in the thickness and density of the foreign body capsule (Hill et al., 2006).

17.3.2 Polymers A common example of the importance of selecting proper sterilization procedures and packaging material is the history of UHMWPE and the surrounding controversy that occurred during the 1990s (as reviewed in Kurtz, 2004). Up to 1995, sterilization of UHMWPE was conducted in air at 25±40 kGy using gamma irradiation. The packaging consisted of a box (with literature and stickers containing lot and catalog information for attachment to patient medical records) that typically contained nested, polymeric packages and an inner foam insert (Fig. 17.3). Air-permeable packaging was replaced in the mid-1990s by barrier packaging with a low oxygen environment (Fig. 17.4). By 1998, gamma irradiation in reduced oxygen, or sterilization using non-ionizing methods (e.g., EtO or gas plasma), was being used by all major orthopaedic manufacturers. The impetus for the switch was accumulating evidence that gamma in air sterilization plus long-term storage in oxygen environments promoted oxidative chain scission of the UHMWPE and subsequent degradation of physical, chemical and mechanical properties of UHMWPE. As reviewed by Costa and Bracco (2004), UHMWPE oxidative degradation can be induced by high-energy radiation in the presence of oxygen. Free radicals are produced during irradiation, which subsequently react with oxygen at a rate dependent upon the concentration of radicals and the amount of available oxygen. E-beam irradiation has a higher dose rate compared with gamma irradiation (10 kGy/s versus 1±10 kGy/h, respectively), and the time for irradiation is in the order of seconds for e-beam irradiation and in the order of hours with gamma irradiation. The amount of oxidation of UHMWPE was lower for the e-beam method than with gamma irradiation for the same dosage because of the longer irradiation times and the ability of oxygen to diffuse into the material. During shelf storage of UHMWPE components that were gamma sterilized and stored in air, the oxidative degeneration that occurs results in increased density © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

421

17.3 Historical air-permeable packaging used with gamma sterilization (Wright Medical, Arlington, TN). Reprinted with permission from The UHMWPE Handbook: Ultra-High Molecular Weight Polyethylene in Total Joint Replacement, S.M. Kurtz, `Packaging and sterilization of UHMWPE', pp. 37± 51, Copyright Elsevier (2004).

and crystallinity, along with a loss of mechanical properties associated with progressive embrittlement (Kurtz, 2004). Even though gamma sterilization and storage in air is no longer standard practice, estimates indicate that at least 2 million US patients may have been implanted with UHMWPE components that were sterilized in air from 1980 to 1989, and an additional 2 million patients are estimated to have been implanted with air-sterilized UHMWPE components in the US from 1990 to 1995. Therefore, oxidative degradation of UHMWPE components will continue to be relevant well into the century.

17.4 Contemporary nitrogen-filled barrier packaging for gamma sterilization of UHMWPE components used by Zimmer, Inc. (Warsaw, IN). Reprinted with permission from The UHMWPE Handbook: Ultra-High Molecular Weight Polyethylene in Total Joint Replacement, S.M. Kurtz, `Packaging and sterilization of UHMWPE', pp. 37±51, Copyright Elsevier (2004). © 2008, Woodhead Publishing Limited

422

Joint replacement technology

Today, gamma sterilization of UHMWPE is used in conjunction with barrier packaging, which generally consists of evacuating the air from the packaging and backfilling the packaging with an inert gas (e.g., nitrogen or argon). The barrier packaging consists of polymer laminates or metallic foils that block gas diffusion, the goal of which is to reduce the oxidative degradation that can occur with long-term shelf storage. Exclusion of oxygen in the packaging prevents UHMWPE degradation, even if it was originally sterilized in air, until the package is opened and the device is implanted. Because it contains free radicals, the UHMWPE will be subject to in vivo oxidation as it is exposed to oxygen species in the body. Current alternatives to ionizing sterilization methods include EtO and gas plasma. Other options such as steam sterilization or heat and steam sterilization are not generally considered owing to changes that can make the polymer component mechanically weaker or can cause physical or dimensional changes in the components (An et al., 2005). On one hand, irradiation of polyethylene (PE) causes crosslinking of the polymer and results in improved tribological properties that are not observed with other sterilization methods. For example, over 3.5 to 5 million cycles (each million cycle refers to about a year of use in vivo for the average patient), the wear rate of UHMWPE sterilized with gamma irradiation in air was 18:5  0:9 mm3/million cycles compared with 40  0:6 mm3/million cycles for UHMWPE sterilized in EtO (McKellop et al., 1999). Similarly, a reduction in wear rate that exceeded 50% was observed in gamma-irradiated UHMWPE versus EtO sterilization or non-sterilized materials (Wang et al., 2001). Retrospective clinical studies therefore suggest that the tribological improvements observed in gamma-irradiated UHMWPE components are related to improved clinical performance. In a hip simulator study, it was concluded that crosslinking with gamma irradiation produced desirable wear characteristics but the ideal case is to avoid both immediate and long-term oxidation (McKellop et al., 2000). On the other hand, sterilization of UHMWPE components using alternative methods may have benefits as well because they avoid long-term oxidative effects. In a retrieval study, EtO-sterilized components showed significantly less surface damage and delamination than gamma-sterilized components, while gamma-sterilized components had reduced ductility, decreased toughness, and increased crystallinity (White et al., 1996). Retrieved PE components of 1635 knees indicated that PE knee components that were gamma irradiated in air have a high incidence of delamination and cracking, which at times led to complete wear through the bearing (Williams et al., 1998). On the other hand, EtO sterilization generated greater numbers of wear particles compared with gamma irradiation (particle sizes were not reported) (Affatato et al., 2002). In a Low Contact Stress Total Knee System (DePuy Orthopaedics, Inc.), gas plasma sterilization did not introduce free radicals that subsequently oxidize with extended shelf aging as in gamma-air irradiation (McNulty et al., 2002). © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

423

Low-temperature hydrogen peroxide gas plasma sterilization of UHMWPE led to improved wear performance of UHMWPE compared with the non-sterile control, while there was little oxidation observed and no change in density (while increases in density and oxidation were observed with gamma irradiation) (Goldman and Pruitt, 1998). In addition to the mechanical and tribological performance of polymer implants with respect to sterilization methods, the potential physiological effects should also be considered. Oxidized UHMWPE activated a significantly higher percentage of granulocytes (73:35  5:2%) in cell culture compared with UHMWPE that was not oxidized (21:5  3:8%) (Reno et al., 2003), which may be influenced by increased granulocyte activation and adverse tissue response in vivo. UHMWPE may also release chemical eluates, which have been demonstrated to contribute to periprosthetic osteolysis (An et al., 2005). Specifically, EtO sterilization of UHMWPE resulted in reduced cell viability (evidenced by a reduction in (3)H-thymidine uptake) at lower elute concentrations compared with gamma-irradiated UHMWPE (Wang et al., 2001). In a murine macrophage cell culture study, release of TNF- (the cytokine most implicated in the osteolytic process, Green et al., 2000) was stimulated at lower volumetric concentrations of wear particles derived from UHMWPE crosslinked with 5 Mrad (50 kGy) of gamma irradiation compared with UHMWPE crosslinked with 10 Mrad (100 kGy) of gamma irradiation (Ingram et al., 2004). The authors concluded that the lower concentration of wear particles that is needed to stimulate cells for crosslinked materials may negate the benefit of the lower wear volume.

17.3.3 Biological materials Biological materials can be used in conjunction with joint arthroplasty prostheses. For example, it is often necessary to utilize bone allografts during revision surgeries in order to fill large bone defects (see, for example, Lotke et al., 2006). Although gamma irradiation has been used to sterilize bone allografts for many years, gamma irradiation affects the mechanical and biological properties of bone allografts by splitting polypeptide chains and thus degrading collagen in the matrix (Nguyen et al., 2006). There is a dose-dependent decrease in mechanical properties for cortical bone (above 25 kGy) and cancellous bone (above 60 kGy) (Nguyen et al., 2006). In addition, bone graft remodeling may be affected by potential effects of gamma irradiation on osteoclast activity, apoptosis of osteoblasts, and bacterial product remnants that may induce inflammatory reactions (Nguyen et al., 2006). Tissue allografts present challenges for other methods of sterilization because of the potential adverse effects that heat or irradiation may have on the biomechanical properties of the tissue (An et al., 2005). EtO penetrates bone grafts, potentially causing intra-articular synovial and immune reactions (An et al., 2005) and therefore may not be suitable. © 2008, Woodhead Publishing Limited

424

Joint replacement technology

The clinical performance of bone allografts sterilized by different mechanisms indicates that the optimal method of bone allograft preparation is unclear. In a clinical study involving revision hip arthroplasties with impaction of bone allografts that were fresh-frozen, non-irradiated, and pulse-washed in normal saline, only one patient (0.7%) out of 138 (144 hips) developed a deep infection (mean follow-up time was four years) (Kwong et al., 2005). Comparatively, out of 123 acetabular components that were revised for aseptic loosening and treated with impaction of frozen, morselized, irradiated bone grafts, eight re-revisions (6.5%) were required due to deep sepsis (Buckley et al., 2005). In another study with a follow-up for a minimum of three years that investigated 127 patients (14 at revision total joint replacement) receiving `massive' bone allografts (size greater than 7 cm) that were sterilized by 25 kGy irradiation, none of the joint arthroplasty revision patients experienced bacterial infection; 9% of patients overall experienced infection, all of whom were being treated for malignant tumors with adjuvant therapy (Hernigou et al., 1993). The authors concluded that the rate of complications (which included non-union and fracture) was not significantly different from that reported for non-irradiated grafts (Hernigou et al., 1993). Selection of bone allograft sterilization methods should involve a careful consideration of the potential changes in biomechanical properties, risk of infection, and potential effects on bone graft incorporation/remodeling.

17.4

Conclusions

In conclusion, the relevant international standards and FDA guidance documents for sterilization of total joint replacement prostheses were reviewed. In vitro studies demonstrate that the proper selection of sterilization and packaging methods can have a profound effect on the mechanical properties and biocompatibility of the implant components. As clinical and retrieval data become more available, the in vivo performance of components sterilized using different methods will become more clear.

17.5

References

Affatato, S., Bordini, B., Fagnano, C., Taddei, P., Tinti, A., and Toni, A. (2002). Effects of the sterilisation method on the wear of UHMWPE acetabular cups tested in a hip joint simulator. Biomaterials 23, 1439±46. An, Y. H., Alvi, F. I., Kang, Q., Laberge, M., Drews, M. J., Zhang, J., Matthews, M. A., and Arciola, C. R. (2005). Effects of sterilization on implant mechanical property and biocompatibility. Int J Artif Organs 28, 1126±37. Anand, V. P., Cogdill, C. P., Klausner, K. A., Lister, L., Barbolt, T., Page, B. F., Urbanski, P., Woss, C. J., and Boyce, J. (2003). Reevaluation of ethylene oxide hemolysis and irritation potential. J Biomed Mater Res A 64, 648±54. Baier, R. E., Meyer, A. E., Akers, C. K., Natiella, J. R., Meenaghan, M., and Carter, J. M. (1982). Degradative effects of conventional steam sterilization on biomaterial surfaces. Biomaterials 3, 241±5. © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

425

Bellucci, A. (2004). Shortness of breath and abdominal pain within minutes of starting hemodialysis. Semin Dial 17, 417±21. Bruck, S. D., and Mueller, E. P. (1988). Radiation sterilization of polymeric implant materials. J Biomed Mater Res 22, 133±44. Buckley, S. C., Stockley, I., Hamer, A. J., and Kerry, R. M. (2005). Irradiated allograft bone for acetabular revision surgery. Results at a mean of five years. J Bone Joint Surg Br 87, 310±13. Budd, T. W., Bielat, K. L., Meenaghan, M. A., and Schaaf, N. G. (1991). Microscopic observations of the bone/implant interface of surface-treated titanium implants. Int J Oral Maxillofac Implants 6, 253±8. Campbell, C. E., and von Recum, A. F. (1989). Microtopography and soft tissue response. J Invest Surg 2, 51±74. Cardenas-Camarena, L. (1998). Ethylene oxide burns from improperly sterilized mammary implants. Ann Plast Surg 41, 361±6; discussion 366±9. Costa, L., and Bracco, P. (2004). Mechanisms of crosslinking and oxidative degeneration of UHMWPE. In The UHMWPE Handbook: Ultra-High Molecular Weight Polyethylene in Total Joint Replacement (S. M. Kurtz, ed.), pp. 245±261. Elsevier Academic Press, New York. Costa, L., Luda, M. P., Trossarelli, L., Brach del Prever, E. M., Crova, M., and Gallinaro, P. (1998). Oxidation in orthopaedic UHMWPE sterilized by gamma-radiation and ethylene oxide. Biomaterials 19, 659±68. Delgado, A. A., and Schaaf, N. G. (1990). Dynamic ultraviolet sterilization of different implant types. Int J Oral Maxillofac Implants 5, 117±25. Dorman-Smith, V. (1997). Sterilisation processes and residuals. In Biocompatibility Assessment of Medical Devices and Materials (J. H. Bradybrook, ed.), pp. 101±18. John Wiley & Sons, New York. Drake, D. R., Paul, J., and Keller, J. C. (1999). Primary bacterial colonization of implant surfaces. Int J Oral Maxillofac Implants 14, 226±32. Gillis, J. R. (1981). Sterilization validation. In Sterilization of Medical Products (E. R. L. Gaughran and R. F. Morrissey, eds), Vol. 2, pp. 3±10. Multiscience Publications Limited, Montreal. Goldman, M., and Pruitt, L. (1998). Comparison of the effects of gamma radiation and low temperature hydrogen peroxide gas plasma sterilization on the molecular structure, fatigue resistance, and wear behavior of UHMWPE. J Biomed Mater Res 40, 378±84. Green, T. R., Fisher, J., Matthews, J. B., Stone, M. H., and 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±7. Hanssen, A. D., and Rand, J. A. (1999). Evaluation and treatment of infection at the site of a total hip or knee arthroplasty. Instr Course Lect 48, 111±22. Hernigou, P., Delepine, G., Goutallier, D., and Julieron, A. (1993). Massive allografts sterilised by irradiation. Clinical results. J Bone Joint Surg Br 75, 904±13. Hill, C. M., Kang, Q. K., Wahl, C., Jimenez, A., Laberge, M., Drews, M., Matthews, M. A., and An, Y. H. (2006). Biocompatibility of supercritical CO2-treated titanium implants in a rat model. Int J Artif Organs 29, 430±3. Ingram, J. H., Stone, M., Fisher, J., and 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, 3511±22. © 2008, Woodhead Publishing Limited

426

Joint replacement technology

Karacalar, A., and Karacalar, S. A. (2000). Chemical burns due to blood pressure cuff sterilized with ethylene oxide. Burns 26, 760±3. Kasemo, B., and Lausmaa, J. (1988). Biomaterial and implant surfaces: on the role of cleanliness, contamination, and preparation procedures. J Biomed Mater Res 22, 145±58. Keller, J. C., Draughn, R. A., Wightman, J. P., Dougherty, W. J., and Meletiou, S. D. (1990). Characterization of sterilized CP titanium implant surfaces. Int J Oral Maxillofac Implants 5, 360±7. Kurtz, S. M. (2004). Packaging and serilization of UHMWPE. In The UHMWPE Handbook: Ultra-High Molecular Weight Polyethylene in Total Joint Replacement (S. M. Kurtz, ed.), pp. 37±51. Elsevier Academic Press, New York. Kurtz, S. M., Muratoglu, O. K., Evans, M., and Edidin, A. A. (1999). Advances in the processing, sterilization, and crosslinking of ultra-high molecular weight polyethylene for total joint arthroplasty. Biomaterials 20, 1659±88. Kwong, F. N., Ibrahim, T., and Power, R. A. (2005). Incidence of infection with the use of non-irradiated morcellised allograft bone washed at the time of revision arthroplasty of the hip. J Bone Joint Surg Br 87, 1524±6. Lausmaa, J., Kasemo, B., and Hansson, S. (1985). Accelerated oxide growth on titanium implants during autoclaving caused by fluorine contamination. Biomaterials 6, 23±7. Lerouge, S., Tabrizian, M., Wertheimer, M. R., Marchand, R., and Yahia, L. (2002). Safety of plasma-based sterilization: surface modifications of polymeric medical devices induced by Sterrad and Plazlyte processes. Biomed Mater Eng 12, 3±13. Lidgren, L., Knutson, K., and Stefansdottir, A. (2003). Infection and arthritis. Infection of prosthetic joints. Best Pract Res Clin Rheumatol 17, 209±18. Lotke, P. A., Carolan, G. F., and Puri, N. (2006). Technique for impaction bone grafting of large bone defects in revision total knee arthroplasty. J Arthroplasty 21, 57±60. Matlaga, B. F., Yasenchak, L. P., and Salthouse, T. N. (1976). Tissue response to implanted polymers: the significance of sample shape. J Biomed Mater Res 10, 391±7. McKellop, H., Shen, F. W., Lu, B., Campbell, P., and Salovey, R. (2000). Effect of sterilization method and other modifications on the wear resistance of acetabular cups made of ultra-high molecular weight polyethylene. A hip-simulator study. J Bone Joint Surg Am 82-A, 1708±25. McKellop, H. A., Shen, F. W., Campbell, P., and Ota, T. (1999). Effect of molecular weight, calcium stearate, and sterilization methods on the wear of ultra high molecular weight polyethylene acetabular cups in a hip joint simulator. J Orthop Res 17, 329±39. McNulty, D. E., Liao, Y. S., and Haas, B. D. (2002). The influence of sterilization method on wear performance of the low contact stress total knee system. Orthopedics 25, s243±6. Nguyen, H., Morgan, D. A., and Forwood, M. R. (2006). Sterilization of allograft bone: effects of gamma irradiation on allograft biology and biomechanics. Cell Tissue Bank 8, 93±105. Reno, F., Lombardi, F., and Cannas, M. (2003). UHMWPE oxidation increases granulocytes activation: a role in tissue response after prosthesis implant. Biomaterials 24, 2895±900. Rutala, W. A., Gergen, M. F., and Weber, D. J. (1998). Comparative evaluation of the sporicidal activity of new low-temperature sterilization technologies: ethylene oxide, 2 plasma sterilization systems, and liquid peracetic acid. Am J Infect Control 26, 393±8. © 2008, Woodhead Publishing Limited

Sterilization of joint replacement materials

427

Sia, I. G., Berbari, E. F., and Karchmer, A. W. (2005). Prosthetic joint infections. Infect Dis Clin North Am 19, 885±914. Sperling, J. W., Kozak, T. K., Hanssen, A. D., and Cofield, R. H. (2001). Infection after shoulder arthroplasty. Clin Orthop Relat Res, 206±16. Stanford, C. M., Keller, J. C., and Solursh, M. (1994). Bone cell expression on titanium surfaces is altered by sterilization treatments. J Dent Res 73, 1061±71. Swart, K. M., Keller, J. C., Wightman, J. P., Draughn, R. A., Stanford, C. M., and Michaels, C. M. (1992). Short-term plasma-cleaning treatments enhance in vitro osteoblast attachment to titanium. J Oral Implantol 18, 130±7. Ungersbock, A., Pohler, O., and Perren, S. M. (1994). Evaluation of the soft tissue interface at titanium implants with different surface treatments: experimental study on rabbits. Biomed Mater Eng 4, 317±25. Vezeau, P. J., Koorbusch, G. F., Draughn, R. A., and Keller, J. C. (1996). Effects of multiple sterilization on surface characteristics and in vitro biologic responses to titanium. J Oral Maxillofac Surg 54, 738±46. Wang, K. Y., Horne, J. G., Devane, P. A., Wilson, T., and Miller, J. H. (2001). Chemical eluates from ultra-high molecular weight polyethylene and fibroblast proliferation. J Orthop Surg (Hong Kong) 9, 25±33. White, A., Burns, D., and Christensen, T. W. (2006). Effective terminal sterilization using supercritical carbon dioxide. J Biotechnol 123, 504±15. White, S. E., Paxson, R. D., Tanner, M. G., and Whiteside, L. A. (1996). Effects of sterilization on wear in total knee arthroplasty. Clin Orthop Relat Res 164±71. Williams, I. R., Mayor, M. B., and Collier, J. P. (1998). The impact of sterilization method on wear in knee arthroplasty. Clin Orthop Relat Res 356, 170±80. Youngblood, T., and Ong, J. L. (2003). Effect of plasma-glow discharge as a sterilization of titanium surfaces. Implant Dent 12, 54±60.

© 2008, Woodhead Publishing Limited

Part IV

Specific joints

© 2008, Woodhead Publishing Limited

18

Hip replacement: tribological principles, materials and engineering D D O W S O N , University of Leeds, UK

18.1

Introduction

In earlier parts of this text fundamental aspects of biomechanics, tribology, materials science and the biological environment in which joint replacements function have been outlined. The development of specific joints including the hip, knee, ankle, shoulder, elbow, intervertebral disc and the temporomandibular joint are reviewed in subsequent chapters in Part IV. All the fundamental concepts mentioned have contributed to the successful and in some cases spectacular development of joint replacement, but in this chapter the significant engineering roles of tribology, materials, design and manufacture in hip replacement will be considered.

18.1.1 Historical background to tribological features of hip arthroplasty Total hip replacement, or hip arthroplasty, has developed over almost two centuries, but spectacular progress has been recorded during the last half century. In the early period surgeons attempted arthrodesis, or the interposition of various materials between the articulating bones, in their attempt to relieve pain and restore joint function. Fat and muscle, wood, ivory, gold foil, glass, celluloid, Bakelite, platinum, white wood, cobalt±chrome alloys, stainless steel, acrylic, polymers and ceramics had all been inserted between damaged or diseased bearing surfaces in the hip. These early but intriguing attempts to deal with joint disorders by inserting such an array of materials were largely unsuccessful. Scales (1967) has presented a useful history of these early years. Most of the developments in the first half of the 20th century were directed towards either femoral head or acetabular cup replacement; the simultaneous replacement of both in total hip arthroplasty being thought to be too difficult. The Smith-Peterson acetabular cup arthroplasty (Smith-Peterson, 1939) developed in Boston, USA, with cups or moulds made from a variety of materials ranging from glass to vitallium, is widely recognised as the outstanding cup replacement in the © 2008, Woodhead Publishing Limited

432

Joint replacement technology

pre-World War II era. The results were encouraging, with initial claims that 50% were satisfactory, followed in the mid-1950s by revised estimates that over 80% could be deemed to be satisfactory. A major move towards total hip replacement and the use of (metal-on-metal) arthroplasties was undertaken in 1938 by Wiles at the Middlesex Hospital, in London. Stainless steel was used for both components, although only six operations were performed and all the X-ray records were lost. Philip Wiles (1957) later developed an improved metal-on-metal hip replacement and his name is linked by many to the start of the impressive era of development of total hip arthroplasty in the second half of the 20th century. Hemi-arthroplasty, concentrated on the femoral head, continued to develop in the hands of the Judet brothers in the 1940s (acrylic) and Austin Moore in the 1950s (cobalt±chromium alloy). In the latter case the replacement head was integral with a long, curved, fenestrated stem. Thompson (1954) introduced a similar but solid stem and head made of vitallium which provided the basic form for many future total hip replacements.

18.1.2 The mid-20th century development of the McKee±Farrar (metal-on-metal) and Charnley (metal-on-polymer) total hip replacements George Kenneth McKee was a Senior Registrar with Philip Wiles in Norwich in the middle years of the 20th century. By the early 1950s he was actively developing metal-on-metal total hip replacements, initially using stainless steel, but later adopting CoCrMo alloy (McKee and Watson-Farrar, 1966). Screws were used to aid fixation of the early implants, but these were later replaced by methylmethacrylate cement. A modified Thompson femoral stem of 1.25 inches (31.75 mm) diameter was used initially, with 4 mm studs on the outer surface of the acetabular cup to aid fixation in acrylic cement. In later designs the head diameter was increased to 1.75 inches (44.45 mm). An illustration of the McKee±Farrar (metal-on-metal) total hip replacement is shown in Fig. 18.1. An observation of great significance from the tribological point of view (McKee, 1967) was that the cup was `hemispherical and lapped in to fit the sphere of the femoral portion, forming a pair, and they should only be used as such, and they are numbered to ensure correct pairing'. In recent times it has been widely acknowledged that a small, but finite clearance, or difference between head and cup diameters, is essential for the satisfactory functioning of metal-on-metal total hip replacements. This not only avoids the ill-effects of elastic distortions and equatorial contact between the components under load, but also minimises wear by promoting the formation of satisfactory lubricating films. Other forms of metal-on-metal total hip replacements were developed in the 1950s and 1960s, notably the Ring (1968, 1971) with its integral cup and long pelvic screw, the Sivash (1969) with its constrained head, the MuÈller, with three © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

433

18.1 McKee±Farrar metal-on-metal total hip replacement.

intriguing plastic pads within the bearing region of the acetabular cups to minimise friction (see Semlitsch and Willert, 1997) and the Stanmore (see DuffBarclay et al., 1966), with its horseshoe-shaped bearing area in the acetabular cup. The name George Kenneth McKee will nevertheless be recognised not only for his pioneering development of metal-on-metal bearing pairs for hip replacement, but also of total hip arthroplasty itself (see Reynolds and Tansey, 2007). His work stands out, alongside that of John Charnley, who developed the metal-on-polymer low-friction arthroplasty (LFA) which dominated the hip replacement field as the `Gold standard' for at least half a century. In April 1967 the Institution of Mechanical Engineers held a Symposium in London on `Lubrication and Wear in Living and Artificial Human Joints' (IME, 1967). Both John Charnley (1967) and Kenneth McKee (1967) presented papers and since the event was held during the first decade in which satisfactory total joint replacements emerged, their presentations make fascinating reading. Charnley (1967) estimated the number of total hip replacement operations in the United Kingdom during the previous five or six years to be about 4000, whereas today the corresponding figure would have been at least 200 000. Charnley focused his attention upon the reduction of friction in hip replacements. He was far from certain that chrome±cobalt±molybdenum alloy heads and cups sliding on each other could achieve low enough friction to avoid loosening with synovial fluid as the lubricant. He observed that such bearing © 2008, Woodhead Publishing Limited

434

Joint replacement technology

pairs and both stainless steel or poly(methylmethacrylate) (PMMA) heads on bone exhibited coefficients of friction of about 0.5, whereas stainless steel on normal articular cartilage yielded a much lower coefficient of about 0.05. He concluded that `the only chance of success in lubricating an animal joint would be by using surfaces which were intrinsically slippery on each other'. This focused attention upon the polymer poly(tetrafluorethylene) (PTFE), developed initially for high-temperature engineering bearing applications, but which was also found to be slippery with coefficients of friction of about 0.04. Charnley's (1961) announcement of his low-friction arthroplasty in the Lancet correctly identified the minimisation of frictional torque as an important feature for resisting acetabular cup loosening. He initially proposed the use of a stainless steel femoral head of diameter 7/8 inch (22.225 mm) integral with a stainless steel stem and a PTFE acetabular cup. The relatively small head minimised frictional torque while increasing the wall thickness of the cup available to accommodate wear. Charnley et al. (1969) later demonstrated theoretically that a head diameter of half the external cup diameter would maximise the wear life of the implant. There were thus two important aspects to the launch of the low-friction arthroplasty for the hip: the use of a small diameter metallic femoral head and the selection of a low-friction material for the acetabular cup. The former concept was based upon simple biomechanical principles while the latter introduced a biomaterial with one of the lowest known coefficients of friction for a synthetic bearing material. The poor wear characteristics of PTFE made it unsuitable for most engineering bearing applications in its bulk form, as they did for acetabular cups. In engineering applications fillers such as glass beads or fibres, carbon fibres or ceramics could be used to improve wear resistance, but Charnley (1967) failed to detect such improvements in the body. In the period 1959 to 1962 some 300 hip replacements using PTFE cups had been implanted, but in many cases excessive wear had allowed the metallic femoral heads to wear completely through the cup walls as shown in Fig. 18.2. It was at this stage that Charnley switched to ultra-high molecular weight polyethylene (UHMWPE) for the cup material as shown in Fig. 18.3. The UHMWPE presented higher coefficients of friction, but much lower wear than PTFE and it rapidly became the polymer of choice. Two forms of prostheses were thus dominating hip arthroplasty in the mid1960s; the Charnley (metal-on-polymer (UHMWPE)) and the McKee±Farrar (metal-on-metal). A number of the latter implants experienced premature loosening, while the former revealed better survivorship. The McKee (metal-onmetal) hip replacement was largely abandoned in favour of the Charnley (metalon-UHMWPE) low-friction arthroplasties by the mid-1970s. The Charnley LFA total hip replacement became regarded as the gold standard against which alternative designs were to be compared for the remainder of the 20th century. © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

435

18.2 Early Charnley low-friction arthroplasty showing penetration of the metallic femoral head into the PTFE acetabular cup.

18.3 Charnley total hip replacement with metallic stem and femoral head (7/8 in (22.225 mm)) and polyethylene (UHMWPE) acetabular cups.

18.1.3 Introduction by Boutin of ceramic-on-ceramic total hip replacements Ceramics entered the list of material pairs for total hip arthroplasty in France early in the 1970s. The use of bioceramics in the body had been developing throughout much of the 20th century, initially as bone substitute materials and then in the mid-1950s as dental implants. The introduction of ceramic-onceramic hip replacements is attributed to Pierre Boutin, working in Pau in © 2008, Woodhead Publishing Limited

436

Joint replacement technology

18.4 Pierre Boutin's ceramic-on-ceramic total hip replacement (illustration supplied by Dr Claude Rieker).

southern France. He took out a patent for dense alumina oxide (alumina) femoral heads and acetabular cups in 1970 (Boutin, 1972). The ceramic-on-ceramic material pair shown in Fig. 18.4 commended itself to surgeons primarily because of the lower friction and wear in the body than metal-on-polymer implants, but also because of concern for the toxicity of metallic and polymeric wear debris. Initially binary oxides such as alumina (Al2O3), zirconia (ZrO2) and titania (TiO2) were considered (Hulbert et al., 1970), but these were followed by more complex formulations. In Boutin's ceramic-on-ceramic prosthesis both head and cup were made from high density (99.5%) alumina. Experiments were carried out with both cemented (PMMA) 32 mm diameter and uncemented components. Characteristic features of the uncemented alumina acetabular cups were the 1 mm grooves on the outer diameter formed to promote fixation. In due course the heads were formed to fit on to small angled taper spigots machined on the metallic stems in Germany and in France. The design and manufacture of the taper spigots proved to be an important feature of stress reduction and the control of brittle fracture in many subsequent total hip replacements. Boutin and Blanquaert (1979) noted that only two preoperative head and four postoperative cup fractures occurred in 373 implants inserted between 1970 and 1973. The cup failures were attributed to © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

437

production and control faults, all with one batch of material, and no further fractures were witnessed during the subsequent seven years. Boutin and Blanquaert (1979) recorded a number of features of the ceramicon-ceramic hip replacement, which are of direct relevance to the role of tribology in ceramic joints. They found that if low wear was to be achieved, very low clearances between head and cup of about 10 m, sphericity less than 1±2 m and surface roughnesses of about 20 nm were required. They also drew attention to the importance of wetting angle between liquids and alumina `which positively influences the lubricating film'. For alumina the wetting angle was found to be 44ë, while for 316 L stainless steel, UHMWPE and Co±Cr±Mo alloy it was 72ë, 80ë and 87ë, respectively. The impressive introduction of bioceramics into hip joints by Boutin prompted a flurry of activity in Switzerland, Germany and later in Japan and the United Kingdom. These developments built upon the relatively low wear and friction of ceramic-on-ceramic material pairs compared with the McKee±Farrar (metal-on-metal) and Charnley's (metal-on-polyethylene) LFA. The inherent inertness of ceramics was attractive, but at the same time the problems of brittle fracture and fixation were addressed (Miller et al., 1996). These authors reported friction coefficients of 0.16, 0.08 (0.06±0.10) and 0.045 (0.03±0.06) for metalon-polymer (UHMWPE), alumina-on-polymer and alumina-on-alumina respectively. The friction of the ceramic pairing was only about 50% of that for metal-on-polymer, while significant improvements in the wear factors were also observed. A major factor determining the longevity of the Charnley LFA, which dominated the hip replacement field throughout the second half of the 20th century, proved to be osteolysis associated with polymeric wear debris as discussed elsewhere in this text. The modular ceramic femoral head on a metallic stem with a polyethylene cup thus emerged as a further material pair of interest in hip arthroplasty by the end of the 20th century.

18.2

Millennium prostheses

By the year 2000 four material combinations covered almost all forms of total hip replacement (THR) then available. Particular forms are sketched in Fig. 18.5. The growth in hip surgery for patients suffering from arthritis or from injuries and trauma in the second half of the 20th century was truly outstanding. Most of the THRs listed above offered good survivorship for at least a decade, but this still implied that revision surgery was required for an unsatisfactory number of patients. Furthermore, the very success of the Charnley gold standard prosthesis promoted interest in ceramic-on-polyethylene material pairs yielding smaller volumes of polyethylene wear debris. Younger, more active patients sought longer lasting, stable, total hip replacements. This re-awakened interest in hard-on-hard material pairs, such as metal© 2008, Woodhead Publishing Limited

438

Joint replacement technology

18.5 Principal material pairs in `Millennium' forms of total replacement hip joints.

on-metal, ceramic-on-ceramic and, more recently, ceramic-on-metal. Studies of the metal-on-metal combination were particularly active, partly because of renewed interest in the resurfacing concept (Amstutz and Le Duff, 2006; Grigoris et al., 2006; McMinn and Daniel, 2006), but also because of some © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

439

interesting revelations about the fundamental tribological behaviour of THRs. The basic tribological features of hard-on-hard THRs are outlined in the next section.

18.3

Introduction to the tribology of total hip replacements

The effectiveness of lubrication determines both the nature and magnitude of wear and friction in bearings. Three main approaches are available for the determination of the mode of lubrication in a bearing system. 1. The most direct method is to measure the film thickness by mechanical, optical, electrical or ultrasound techniques. This is very difficult, particularly in the case of hip replacements. It is, however, possible to apply and monitor variations in a small voltage difference applied to electrically insulated metallic heads and cups in a joint simulator. If the applied voltage is maintained throughout a full cycle, electrical insulation is indicated, whereas a fall to zero indicates that asperity contact has occurred. Electrical insulation of the metallic components can be attributed to separation of the surfaces by a fluid film, or the interposition of insulating surface layers formed by boundary lubricants or tribo-corrosion. 2. A second approach is to calculate the film thickness that could be generated by hydrodynamic action between smooth surfaces and to compare the result with some simple representation of the combined roughness of the two sliding surfaces. The ratio of theoretical minimum film thickness to a composite surface roughness, known as the lambda ratio, has been defined in equation (2.5) of Chapter 2. Lambda ratio …† ˆ

theoretical minimum film thickness composite surface roughness

ˆh

…hminimum † ÿ
2 i1=2 …Ra head †2 ‡ Ra cup

18:1

The definition of (Ra) is such that if lambda () is equal to or less than unity, boundary lubrication is indicated and studies of gears and rolling element bearings have indicated that values greater than about 3 roughly corresponded to fluid film lubrication. Intermediate values, between 1 and 3 are indicative of mixed lubrication. 3. The friction and wear characteristics of THRs are determined mainly by the mode of lubrication developed and the inherent material properties of the bearing materials. The concept of lubrication regimes was introduced in Section 2.1.6 of Chapter 2 and a Stribeck diagram is shown in Fig. 2.4. © 2008, Woodhead Publishing Limited

440

Joint replacement technology

18.6 Stribeck diagram for spherical bearings showing the influence of lubrication regimes (boundary, mixed and fluid-film) upon the coefficient of friction.

For hemispherical bearings the Stribeck diagram can be represented as shown in Fig. 18.6. The three dominant modes of lubrication outlined in Section 2.1.6 are determined for a spherical bearing by a dimensionless group of variables representing the geometry (diameter (d) and clearance, or difference between cup and head diameters (cd)) and the independent operating variables (viscosity (), angular velocity ( ) and load (w)), as indicated by the Sommerfeld number on the abscissa. For an implant with specified diameter (d) and diametral clearance (cd), the mode of lubrication is thus determined by the viscosity, angular velocity and load. If the product (viscosity  speed) is high and the load small, fluid-film lubrication is favoured and theoretically there would be no contact between the asperities on the opposing sliding surfaces. The frictional characteristics would then be determined solely by hydrodynamic principles. For more severe conditions of operation, where the load is relatively high and/or the viscosity and speed are low, boundary lubrication pertains and the applied load is transmitted entirely through asperity contacts between the contacting solids. The friction is then determined by contact mechanics and the inherent frictional properties of surface films on the solids generated either by physical adsorption or chemical reaction with constituents of the lubricant, or by tribo-corrosion. For intermediate values of the Sommerfeld number some of the applied load is supported by fluid-film action, but some is transmitted through asperity © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

441

contacts. This is known as mixed lubrication. Representative values of friction coefficients () for fluid film and boundary lubrication can differ by about two orders of magnitude at (10±3) and (10±1) respectively. Furthermore, the friction changes rapidly over this wide range with relatively small changes in the Sommerfeld number in the mixed lubrication regime. The Stribeck diagram is valuable in two ways in determining the mode of lubrication in either engineering or human bearings: 1. The magnitude of (). This gives an initial indication of the mode of lubrication since the differences between the coefficients of friction in the fluid film and boundary modes is so great. 2. The variation of friction coefficient with the Sommerfeld parameter. If the friction coefficient is high and it varies little as the Sommerfeld number is changed, or even shows a slight decline, boundary lubrication is indicated. If the friction falls rapidly as the Sommerfeld number is increased, mixed lubrication applies, while a gradual increase from a low level is indicative of fluid film lubrication.

18.3.1 Lubrication of metal-on-metal total hip replacements Voltage drop experiments on a joint simulator (Dowson et al., 2000) indicated that mixed lubrication was evident in some 36 mm diameter metal-on-metal total replacement hip joints. During part of the cycle full metallic contact was recorded, but periods of reduced voltage drop during much of the cycle were also evident, as shown in Fig. 18.7. Similar but less conclusive results were obtained for ceramic-on-ceramic THR by applying thin conducting films to the surfaces (Smith et al., 2001). These observations promoted several theoretical analyses of fluid-film lubrication in model hip joints (Jin et al., 1997, 2003; Chan et al., 1998, 1999; Jin and Dowson, 1999; Jalali-Vahid and Jin, 2002; Jalali-Vahid et al., 2003; Dowson, 2006). It soon became clear that if the diameter was large and the diametral clearance small and carefully selected, fluid-film lubrication could indeed contribute substantially to the low friction and low wear observed in some metal-on-metal prostheses. The availability of theoretical film thickness predictions, based initially upon studies of elastohydrodynamic lubrication in engineering situations, greatly assisted the interpretation of experimental simulator measurements of wear in prostheses of differing diameter and clearance (Chan et al., 1998; Dowson, 2003, 2006; Dowson et al., 2004a,b). Both running-in and steady-state wear rates were found to be dependent upon the theoretical elastohydrodynamic film thickness as shown in Figs 18.8 and 18.9. In these diagrams over one hundred values of the wear volumes and wear rates determined in simulator tests in several laboratories in the United Kingdom and North America were found to correlate quite well with the calculated elastohydrodynamic film thickness. © 2008, Woodhead Publishing Limited

442

Joint replacement technology

18.7 Voltage drop measurements for 36 mm diameter metal-on-metal total hip replacements in a hip joint simulator (simulated physiological walking cycle; diametral clearances 170, 157, 149, 130 m).

18.8 Empirical expression for volumetric running-in wear in metal-on-metal THR as a function of the theoretical elastohydrodynamic film thickness (data from North American and United Kingdom research laboratories). Head diameters, 22.225±60 mm; diametral clearances, 50±300 m. © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

443

18.9 Empirical expression for steady-state wear in metal-on-metal THR as a function of the theoretical elastohydrodynmic film thickness (data from North American and United Kingdom research laboratories). Head diameters, 22.225±60 mm; diametral clearances, 50±300 m.

Further confirmation of the influence of fluid film lubrication upon friction in metal-on-metal hip joints came from Stribeck curves measured over periods up to five million cycles by Unsworth (2006) and Vassiliou et al. (2006). These studies revealed a progressive development of the Stribeck curves over the first two to three million cycles, representing transition from boundary, through mixed and then into fluid film lubrication. The initial trace reflected boundary lubrication, followed at one million cycles by a mixed lubrication characteristic. The two, three and five million cycles results shown in Fig. 18.10 suggested that full fluid-film lubrication dominated frictional behaviour after a period comparable to that associated with the running-in process observed in several laboratories with metal-on-metal THR. The theoretical predictions of film thickness and the experimental measurements of both wear and friction all indicated that boundary lubrication governed the initial stages of articulation in metal-on-metal THR, but that for the larger diameter heads this gave way to a mixed form of lubrication in which the contribution of fluid-film lubrication increased as the head diameter increased and/or the clearance decreased. The powerful effect of increasing head diameter and hence improved effective film thickness is evident in Fig. 18.11. The prostheses with the two smallest diameter heads experienced boundary lubrication, with the increased wear in the 22.225 mm THR over that in the 16 mm prosthesis being close to the ratio of diameters and hence sliding distance. Thereafter the wear rate declined quite dramatically as the head diameter increased, owing to increasingly effective load support from hydrodynamic action in a mixed lubrication regime. If boundary © 2008, Woodhead Publishing Limited

444

Joint replacement technology

18.10 Development of friction factor with Sommerfeld Number over five million simulator cycles (data kindly supplied by Professor A. Unsworth, Centre for Bioengineering at Durham University; note definition of Sommerfeld number).

18.11 Influence of diameter upon volumetric running-in wear for metal-onmetal total hip replacements (hip joint simulator experiments). © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

445

Table 18.1 Mean theoretical film thicknesses and lambda () ratios for metal-onmetal heads of different diameters and clearances (mean load 2500 N; angular velocity (!) 1.5 rad/s; lubricant viscosity 0.0009 m Pa s) Diameter (mm) 16 22.225 28 36 54.5

Diametral clearance (m)

Composite roughness (nm)

Predicted film thickness (nm)

Lambda ratio ()

53±70 46±66 55±70 76±78 83±129

9.8 10.8 11.31 10.23 11.63

3.5 7.7 11.8 17.1 32.8

0.36 0.71 1.04 1.67 2.80

lubrication had persisted in all the prostheses the wear would have followed the dotted line, such that at a diameter of 54 mm the total wear would have been at least 40 times greater than the measured running-in wear. The mean theoretical film thicknesses and the lambda ratios for the five heads of different diameters are shown in Table 18.1. It is evident that a full understanding of the effect of head diameter and clearance upon the ability of fluid film lubrication to minimise wear in metal-onmetal implants requires the calculation of film thickness and this is considered in Section 18.3.2.

18.3.2 Predicting film thickness in metal-on-metal hip replacements The pressures generated in typical hip replacements readily cause local elastic deformations 10±100 times greater than the mean theoretical film thicknesses. Elastohydrodynamic action must therefore be considered. Furthermore the shear rates are typically very high at (106±107) 1/s and under these conditions the viscosity of the bovine serum adopted in the simulator tests which yielded the results displayed in Fig. 18.11, and even of synovial fluid in vivo, is little greater than that of water (0.001 Pa s). A value of 0.002 Pa s is frequently considered in calculations for in vivo conditions and will be adopted in the present analysis. These considerations indicate that iso-viscous-elastic action (see Hamrock and Dowson, 1981) governs the effective mode of elastohydrodynamic lubrication in these bearings. An approximate representation of the minimum film thickness (hmin) under these conditions has been presented by Hamrock and Dowson (1978): Hmin ˆ

hmin ˆ 7:43…1 ÿ 0:85eÿ0:3k †U 0:65 W ÿ0:21 Rx

© 2008, Woodhead Publishing Limited

18:2

446

Joint replacement technology

where



 d2  cd  Rx ˆ the effective radius in the entraining direction 1‡ 2cd d k ˆ ellipticity ratio (unity for spherical geometry) 2 3     u  cd  cd 6 7  U ˆ dimensionless speed parameter ˆ4 cd 5  2E0 d E0 Rx 2E0 d 1 ‡ d 2 3   2 2 w 4wcd 6 7 4w…cd † ˆ  W ˆ dimensionless speed parameter 4  cd 2 5 E0 d 4 E0 R2x E0 d 4 1 ‡ d With these simplifications the film thickness expression for a hard-on-hard hip joint becomes: ÿ0:21    d 2  cd 0:65 4wc2d hmin ˆ 1:40 18:3 cd 2E0 d E0 d 4 Or, in terms of the dimensional variables,   d 2:19 0:65 0:65 E0ÿ0:44 hmin ˆ 0:666 85 cd 0:77 w0:21

18:4

This expression shows clearly that the theoretical elastohydrodynamic film thickness is strongly influenced by the head diameter (d 2:19 ); is inversely dependent upon clearance (cd ÿ0:77 ) and only slightly affected by load (wÿ0:21 ). For simulators the viscosity () is generally fixed according to the ISO standard, (E0 ) is determined by bulk material properties and the angular velocity for a simulated walking cycle is most commonly set at 1 Hz. Since both load and entraining velocity vary, the film thickness determined by this `steady-state' entraining expression varies throughout the cycle of articulation, but squeeze film action preserves a much steadier film thickness which is close to the value determined from equation [18.4] for average values of load and speed (Chan et al., 1998; Jin and Dowson, 1999; Jalali-Vahid and Jin, 2002; Jin et al., 2003). Charts have been prepared for the estimation of theoretical film thickness for metal-on-metal hip replacements of different diameter and clearance by Dowson (2003, 2006) as illustrated for representative joint simulator and in vivo lubricant viscosities in Fig. 18.12. The estimates of mean film thickness for perfectly smooth bearing surfaces can now be compared to the measured surface roughnesses on heads and cups to provide an indication of the lubrication conditions in metal-on-metal joints. It is important to note that this procedure is applicable to both solid and surface replacement heads of spherical form, but care has to be taken if significant structural deformation occurs.

© 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

447

18.12 Film thickness predictions for metal-on-metal THR and two lubricant viscosities representative of diluted bovine serum (0.0009 Pa s) and synovial fluid (0.002 Pa s) at high shear rates (d ˆ 50 mm; ˆ 1:5 rad/s; w ˆ 1750 N).

18.3.3 Relationship between elastohydrodynamic film thickness (or lambda ratio) and the cumulative wear in metal-on-metal hip joints Both the running-in wear and the steady-state wear rates are related to the theoretical film thickness as shown in Figs 18.8 and 18.9. The wear should more correctly be related to the lambda ratio, as pointed out by Chan et al. (1999), but since most manufacturers of metal-on-metal THRs succeed in preparing components with impressively small and consistent initial surface roughnesses (heads  10 nm; cups  5 nm) the relationship with film thickness alone is quite strong. The empirical relationships for the running-in wear and steady-state wear rates shown in Figs 18.8 and 18.9 can be written as: Running-in volumetric wear V ˆ

94 1:49

‰h …nm†

Š

…mm3 †

Steady-state volumetric wear rate …h  7ÿ50 nm† 1:87 …mm3 †=106 cycles ˆ ‰h …nm†1:02 Š If running-in is completed in (r) million cycles the total wear volume (V) after (n) million cycles can be estimated from: © 2008, Woodhead Publishing Limited

448

Joint replacement technology

Table 18.2 Representative values of predicted `running-in' wear volume and `steady-state' wear rate for metal-on-metal total replacement hip joints h (nm) V (mm3) running-in V(mm3)/(106 cycles) steady-state

( Vˆ

5 8.54

10 3.04

15 1.66

20 1.08

25 0.78

30 0.59

40 0.39

0.36

0.18

0.12

0.09

0.07

0.06

0.04

94 ‰h …nm†1:49 Š

‡ …n ÿ r†

1:87

)

‰h …nm†1:02 Š

mm3

18:5

Representative values of both the running-in wear volumes and the steady-state wear rates based upon these relationships are shown in Table 18.2. As a rough guide running-in wear volumes range from 0.4 to 10 mm3 and steady-state wear rates are typically 0.04±0.36 mm3/106 cycles for the conditions considered. These ranges are very much smaller than the values associated with the volumetric wear of smaller diameter heads operating in the boundary lubrication regime. Representative traces of the cumulative wear for different predicted film thicknesses are shown in Fig. 18.13. These traces are illustrative and based upon

18.13 Predictions of cumulative wear in (metal-on-metal) total hip replacements (d ˆ 50 mm; ˆ 1:5 rad/s;  ˆ 0:002 Pa s; w ˆ 1750 N) (running-in completed after one million cycles). © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

449

a simple analysis of a wide range of simulator tests. They nevertheless provide a useful indication of the likely long-term wear characteristics of metal-on-metal THR based upon present-day laboratory data. It is quite likely that both the effective radius and the composite surface roughness will change and affect the wear rates after long periods of time. This would lead to changes and possibly convergence of wear and friction characteristics due to the normal process of `running-in' and the establishment of similar surface roughnesses in a universal adhesive wear process. The analysis can be refined as new results from longterm laboratory tests and in vivo observations become available.

18.4

Hard-on-hard total hip joint tribology

The analysis presented in Section 18.3.3 for metal-on-metal joints is illustrative of the general features of hard-on-hard THR tribology. Other material combinations in this category have attracted attention including ceramic-on-ceramic and, in recent years, ceramic-on-metal. Factors affecting the wear characteristics of this extended family of hard-on-hard THR include the intrinsic wear characteristics and the wear factors for the materials; the effective properties (elastic modulus, Poisson ratio and hardness) of the material combination; the surface topography and finish; wear particle geometry and size; corrosion and ion release. Indications of the effects of material properties of the major pairs of materials upon elastohydrodynamic film thickness and lambda ratio are shown in Table 18.3. It is interesting to note that when the harder ceramic materials are used the film thicknesses are smaller than those developed in metal-on-metal THR, but the lambda ratios are larger owing to the better surface roughness of the ceramic materials.

Table 18.3 Film thicknesses and lambda ratios for hard-on-hard total hip replacements (diametral clearances 150 m; viscosity 0.002 Pas; angular velocity 1.5 rad/s; average load 1750 N; effective modulus of elasticity 2:5  1011 (CoCRMo alloy); 4:0  1011 (alumina ceramic)) Material pair

Diameter (mm)

Metal-on-metal Metal-on-metal Ceramic-on-ceramic Ceramic-on-ceramic Ceramic-on-metal Ceramic-on-metal

36 50 36 50 36 50

© 2008, Woodhead Publishing Limited

Minimum film thickness (nm)

Ra (head) (nm)

19 38 15 31 17 35

10 10 2 2 2 2

Ra Ra Lambda (cup) (composite) ratio (nm) (nm) () 5 5 5 5 5 5

11 11 5 5 5 5

1.7 3.4 2.8 5.8 3.2 6.5

450

Joint replacement technology

18.4.1 Friction and wear of metal-on-metal total hip replacements Recent laboratory simulator tests and observations of clinical performance suggest that both large diameter monolithic and surface replacement forms of metal-on-metal prostheses can perform with much lower wear rates than the traditional smaller diameter hard-on-soft materials such as metal or ceramic heads-on-UHMWPE cups. The increased use of large diameter metal-on-metal heads has been promoted primarily by renewed interest in surface replacement femoral components, but also by a recognition of the significant potential of elastohydrodynamic lubrication for partial load support in a mild mixed lubrication regime. It is widely recognised that beneficial lubrication can be optimised by using the largest possible head diameters and the smallest practicable clearances (see Dowson et al., 2004a,b). This pair of geometrical features ensures that both the effective radius of the bearing in the loaded conjunction and the theoretical elastohydrodynamic film thicknesses are maximised (Chan et al., 1996, 1998, 1999; Dowson, 2003, 2006). While most of the quantitative data on wear of metal-on-metal hip joints has been derived from simulator tests, there is a growing need to relate the measurements to clinical performance. The problem is far from trivial, partly because of the great difficulty in measuring very small wear volumes on explanted joints, but also because of the limited number of metal-on-metal joints yet available. Morlock et al. (2006) have taken a valuable step in this direction by developing a measurement system for failed surface replacement implants. The mean measured wear rate of 0.012 mm3/day, or 4.4 mm3/year, was reasonably consistent with, but somewhat greater than the majority of simulator findings. Long-term studies will be necessary before the data on in vivo wear of monolithic and surface replacement metal-on-metal implants can be fully appraised. High-stress and low-speed pin-on-disc tests on high carbon CoCrMo alloys in the presence of serum by Streicher et al. (1996) revealed friction coefficients in the range 0.09±0.45. The same authors also used a pendulum machine to record a frictional torque of about 2.3 N m on the same materials in serum. The corresponding friction coefficient was 0.16. Tests were also carried out on implants of differing diameters on a Stanmore Mk III hip joint simulator with all three degrees of freedom simulated. The friction torques on 32 mm diameter (CoCrMo±UHMWPE) and (CoCrMo±CoCrMo) implants yielded values of 2.8 and 3.0 N m respectively. CoCrMo (metal-on-metal) pairs of diameters (28, 32 and 37 mm) showed torques of 3.5, 3.0 and 6.6 N m respectively. The friction between rubbing pairs of CoCrMo implant materials is greater than that of both ceramic-on-ceramic or ceramic-on-metal combinations and it is widely recognised that high friction in the equatorial region of some early forms of metal-on-metal joints contributed to their restricted survival rates. If good © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

451

18.14 Metal-on-metal THR (femoral stem forged Ti-6Al-4V, hydroxyapatite coated; femoral head and acetabular cup ± high carbon CoCrMo alloy-cast and heat treated; acetabular cup ± porous coating beads and hydroxyapatite coated) (illustration kindly supplied by Professor G. Isaac, De Puy International Ltd).

lubrication is promoted the frictional torques are modest and associated with shearing stresses in the lubricant rather than the much higher coefficients of friction associated with metal-to-metal contact. The essential features of the two current basic forms of monolithic and surface replacement metal-on-metal THRs are shown in Figs 18.14 and 18.15.

18.15 Metal-on-metal articular surface replacement hip joint (femoral head and acetabular cup ± high carbon CoCrMo alloy-cast and heat treated; acetabular cup ± porous coating beads and hydroxyapatite coated) (illustration kindly supplied by Professor G. Isaac, DePuy International Ltd). © 2008, Woodhead Publishing Limited

452

Joint replacement technology

18.4.2 Friction and wear characteristics of ceramic-on-ceramic and ceramic-on-metal total hip replacements It has long been known that bioceramics can form excellent counterfaces for UHMWPE acetabular cups. Willmann et al. (1996) found a 50% reduction in UHMWPE cup wear for alumina rather than CoCrMo alloy femoral heads in their ring-on-disc tests. Alumina-on-alumina components yielded less than onefortieth of the wear generated by metal-on-UHMWPE material pairs. Semlitsch et al. (1977) carried out tests on a tribometer consisting of a roller and a stationary counterface and found that in distilled water the combined wear rates of alumina ceramic-on-UHMWPE and alumina ceramic-on-ceramic were similar, while a CoCrMo alloy roller-on-UHMWPE produced about six times and metal-on-metal (CoCrMo) almost 200 times more wear. The corresponding friction coefficients were 0.05, 0.09, 0.21 and 0.35. It has also been reported that the friction of UHMWPE is about 40% greater when rubbed against zirconia rather than against alumina. The wear rate was found to exhibit similar behaviour with increases of about 50%. Zhou et al. (1997) compared the start-up and steady-state friction coefficients of four ceramic materials rubbing against themselves in a pin-on-disc machine. Carboxymethyl cellulose sodium salt solution was used as the lubricant and the direction of disc rotation could be reversed. The 5 mm diameter pins had spherical tips with radii of 1 m and the average roughnesses (Ra) of the alumina and zirconia specimens were 4 and 6 nm respectively. The discs of both materials had roughnesses of about 10 nm. The start-up friction coefficients were high and in the range 0.15±0.25 for alumina and 0.1±0.34 for zirconia. Much lower coefficients were recorded under steady-state conditions with the alumina values ranging from 0.002 to 0.003 and those for zirconia from 0.01 to 0.008. Both start-up and steady-state values were load dependent, but in the steady state it was clear that the friction of zirconia on itself was considerably higher than that of sliding pairs of alumina. The coefficient of friction of ceramic materials depends upon the physical properties of the material pair, load, lubricant, surface quality and test apparatus employed. It is particularly sensitive to the mode of lubrication established and unless this is known and carefully controlled, meaningful comparisons of the performance of different material pairs are difficult to achieve and the results can be misleading. This is well illustrated by Unsworth's (2006) and Vassiliou et al.'s (2006) findings that the friction factor for metal-on-metal THRs decreased by a factor of six (0.09±0.015) over the first three to five million cycles of simulator tests (Fig. 18.10). Saikko (1998) reported very similar coefficients of friction of 0.012 for 32 mm CoCrMo heads and UHMWPE cups. He also reported average values of the friction factor of 0.28 (CoCrMo-on-CoCrMo); 0.03 (CoCrMo-on-UHMWPE) and 0.002 (Al2O3-on-Al2O3). Kumar et al. (1991) found values of friction coefficients for UHMWPE against metal (CoCrMo), alumina and zirconia of 0.065, 0.054 and 0.04 respectively. © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

453

The range of friction coefficients recorded is large, being particularly dependent upon the test apparatus, testing conditions and lubricant. It is essential that all test conditions are recorded fully if the friction coefficients of different pairs of prosthetic materials are to be meaningful. Effective contributions to load support from elastohydrodynamic lubrication in a mixed lubrication regime established in many hard-on-hard implants can ensure much lower levels of friction than those associated with boundary lubrication action alone in metal or ceramic-on-UHMWPE hip replacements. Estimates of the lambda ratios in laboratory friction studies would greatly assist interpretation of the findings. While the excellent resistance to wear and scratching of alumina was established many decades ago, its low toughness has restricted its use in total joint replacements. When zirconia was explored as an alternative material, phase transformations and expansion during cooling proved to be troublesome. Toughening achieved through the addition of stabilising oxides resulted in the introduction of partially stabilised zirconia (PSZ). In the late 1970s the beneficial features of transformation toughening in which the use of yttria-stabilised tetragonal zirconia polycrystals (Y-TZP) resulted in bending strengths and toughnesses about 50% greater than alumina, but with lower hardness. Microseparation of the head from the cup during the swing phase can result in subsequent rim impact and the formation of wear stripes on the ceramic heads. This phenomenon has been studied in a modified hip joint simulator by Stewart et al. (2001, 2003) for both hot isostatic pressed (HIPed) alumina and ceramic matrix composites. The results exhibited good visual agreement with retrieved ceramic heads, but both simulator and in vivo operation resulted in some roughening and increased wear. The need to combine the good tribological properties of alumina with improved toughness thus resulted in the introduction of alumina±matrix composites. Zirconia toughened alumina (ZTA) was followed by zirconia-platelet toughened alumina (ZPTA) in the 1990s in which crack propagation was resisted impressively by the platelets. This material opened the way for much improved biomaterials for ceramic-on-ceramic THRs with impressive tribological properties and much improved toughness. Illustrations of recent forms of ceramic-on-polymer and ceramic-on-ceramic THRs are shown in Figs 18.16 and 18.17. Representative wear rates for 28 mm diameter THRs of various material combinations recorded in laboratory tests are shown in Table 18.4.

18.5

Wear particles and metal ions

The excellent tribological features of hard-on-hard total hip replacements have enabled them to be utilised for younger more active patients in recent years. Much of the development has been encouraged by the need to minimise wear volume since large volumes of polymeric debris have previously been linked to © 2008, Woodhead Publishing Limited

454

Joint replacement technology

18.16 Ceramic-on-polymer total hip replacement (femoral head ± zirconia toughened alumina; acetabular cup insert crosslinked polyethylene; metallic acetabular shell with porous coating-cementless fixation) (illustration kindly supplied by Mr Horst-O Esser, DePuy International Ltd).

18.17 Ceramic-on-ceramic total hip replacement (zirconia toughened alumina femoral heads and acetabular cup inserts; metallic acetabular shell with porous coating-cementless fixation) (Illustration kindly supplied by Mr Horst-O Esser, DePuy International Ltd). © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

455

Table 18.4 Representative wear rates for various 28 mm diameter material pairs Material pair

Volumetric wear rate (mm)3/year ( 106 cycles)

Metal (CoCrMo)-on-PTFE Ceramic (Al2O3)-on-PTFE Metal (CoCrMo)-on-UHMWPE Ceramic-on-UHMWPE Metal (CoCrMo)-on-metal (CoCrMo) Ceramic-on-ceramic

4000 (2334±4779) 4000 (3428±4469) 60 (20±500) 30 (alumina 30±150; zirconia 18) 4 (0.14±16)*, 0.9 (0.04±1.5)** 0.3 (0.03±0.6)

* Running-in, ** Steady-state.

osteolysis and component loosening. The hard-on-hard implants have yielded impressive reductions in wear volumes, typically to 1/40th±1/100th of that of UHMWPE. However, the metallic and ceramic wear particles from hard-onhard THR are much smaller than the UHMWPE debris, such that their numbers greatly exceed those of UHMWPE. The numbers are staggering. The metallic (CoCrMo) wear particles have diameters in the range 10±120 nm with a mean size of about 30 nm. The wear volume during running-in is in the range 1±4 mm3 and this is generated in about one million cycles. For a representative wear volume of 2.5 mm3/106 cycles the number of particles generated with each step is thus 177 million! In reality this simple calculation can considerably overestimate the number of particles since some of the larger particles contain substantial proportions of the wear volume. Even in the steady state with only about 0.1 mm3 of smaller wear debris of mean diameter 15 nm being produced in 106 cycles, the number of particles per step is of the order of 56 million. In 2001 a further form of hard-on-hard implant was proposed by Firkins et al. (2001) in which a ceramic femoral head was paired with a metallic acetabular cup. The overall wear rates recorded in simulator tests were much reduced, typically to about 0.01 mm3/106 cycles, with the mean metallic wear particle size being initially 17.57 nm and, after five million cycles, 6.11 nm. Ishida et al. (2007) found higher wear rates averaging 1.18 mm3/106 cycles during running-in and 0.20 mm3/106 cycles in the steady state. If wear rates of 0.5 mm3/106 cycles and particle diameters of 6 nm are considered, the number of wear particles generated at each step with ceramic-on-metal implants is 4400 million and a good deal higher than the level calculated for metal-on-metal THR. If the wear rate reported by Firkins et al. (2001) of 0.01 mm3/106 cycles is appropriate, the figure for wear particle generation falls back to 88 million per step. While the vastly reduced wear rates of hard-on-hard implants suggest that they may reduce the incidence of osteolysis, some anxiety persists about the long-term implications of metal ion release. Metal ion release is undoubtedly very high, particularly during the running-in period, but there appears to be little firm evidence to support these concerns over the time periods assessed. © 2008, Woodhead Publishing Limited

456

Joint replacement technology

Endo et al. (2002) introduced the concept of functional biological activity in which the volumetric wear rate, particle characteristics and macrophage response to the debris were combined to predict the overall response to wear of different material pairs in vivo. Comparisons of the tribological performance and functional biological activity of alternative material pairs in total replacement hip joints have been comprehensively reviewed by Fisher et al. (2006). The biological and clinical data have been assessed by Brown et al. (2006), Shetty and Villar (2006), Bhamra and Case (2006), Cobb and Schmalzreid (2006) and Visuri et al. (2006). The Scandinavian registers are particularly helpful for studies in this field.

18.6

Summary

Spectacular but essentially empirical advances in total hip replacement took place in the 20th century. Two main material pairs dominated the field, metalon-metal (McKee±Farrar) and metal-on-polymer (Charnley), but advantage was also taken of the inherent low wear features of alumina ceramics. The Charnley (metal-on-UHMWPE) low-friction arthroplasty was dominant throughout the last few decades of the century, such that it became the gold standard against which new designs were compared. At the time of the millennium the Charnley THR held a dominant position, and, while this situation persists, there is now a bewildering range of alternative joints available to the surgeon. Although many features contributed to the success of the Charnley prosthesis, tribology played a major role. A significant factor in implant longevity in the mid-20th century was loosening and this in turn was linked to the relatively high frictional torques associated with earlier designs of total hip replacements. Charnley addressed this problem spectacularly by twin biomechanical and biotribological means. The use of a small femoral head (7/8 inch or 22.225 mm) minimised the torque arm for frictional forces at the interface between the femoral head and acetabular cup. It also maximised the polyethylene cup thickness and hence extended the implant wear life. The second approach was to use the low-friction `dry' bearing material PTFE and later UHMWPE for the acetabular cup. It was recognised that the wear volume of the polymer would greatly exceed that of metal cups, but the frictional advantage was deemed to be paramount. Towards the end of the twentieth century progress with hard-on-hard, particularly metal-on-metal and ceramic-on-ceramic total replacement joints, was evident. It was, however, the move to larger diameter femoral heads that confirmed the potential of these material combinations. This development was initially promoted by renewed interest in surface replacement hip joints, but experiments revealed a remarkable potential enhancement in the tribological characteristics of such joints. The use of large diameters and opportunities to © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

457

manufacture both heads and cups with high precision and small clearance encouraged support of a proportion of the applied load by fluid film lubrication. The joints were found to be operating in the mixed lubrication regime in which both friction and wear could be reduced substantially through improved lubrication. Simple analysis revealed that the benefits could be optimised by making the femoral heads as big as possible and by keeping the clearance between head and acetabular cup as small as practicable. The latter constraint reflects the inevitable small deviations from perfect smoothness and sphericity of components at the time of manufacture and elastic distortions that occur when the cups are mounted in their bony foundations. The potential of large diameter hard-on-hard prostheses offers encouragement for their use in younger and more active patients, since both friction and wear are much reduced compared with earlier forms of implants. This in turn is expected to extend the life of total hip replacements since loosening of earlier forms of joints was mainly attributed to osteolysis promoted by relatively large volumes of polymeric wear debris. The penalties in using hard-on-hard implants are the substantial increase in the number of very small (nanometre) wear particles, probably exceeding one million with each step taken, and the high release of metal ions. Both are being actively investigated, but it will be many years before acceptable clinical confirmation of the results of current laboratory tests becomes available. Confirmation of the vast number of minute wear particles and very high metal ion release rates has been achieved, but assessment of the implications of these effects continues and the findings will need to be balanced against the apparent long-term tribological benefits of these new hip replacements. In the absence of guidelines on critical levels of wear debris and the influence of particle shape, it appears to be prudent to strive for minimum levels of both wear and friction. The recently promoted (ceramic-onmetal) form of hip replacement, with one or two orders of magnitude further reduction in wear volumes and negligible running-in, appear promising in this regard. An interesting feature of the trends in tribological aspects of hip joint design over the past half century is that they have moved further and further away from the natural configuration. Nature adopts a relatively soft bearing material, articular cartilage, while engineers have promoted the use of bearing materials such as polymers, metals and ceramics of ever-increasing hardness. There is therefore some continuing interest in developing softer, well-lubricated, lowfriction and wear materials with adequate longevity. Biomimetic concepts support this interest and further solutions might emerge from the use of highly irradiated polymers with very low wear rates, or the use of low elastic modulus bearing materials such as polyurethane in cushion form bearings (see Auger et al., 1993; Scholes et al., 2006). There are, of course, a number of significant problems to be overcome, including the longevity of the materials and the biological response to their wear debris. © 2008, Woodhead Publishing Limited

458

Joint replacement technology

Although the long-term resolution of problems leading to total joint replacement will probably be biological rather than mechanical, perhaps based upon stem cell research, impressive progress has been made towards optimising the tribological performance of engineering solutions to the problem.

18.7

References

Amstutz, H.C. and Le Duff, M.J., (2006), `Background of metal-on-metal resurfacing', Proc. Instn. Mech. Engrs., H(2), Eng. Med., 220, 85±94. Auger, D.D., Dowson, D., Fisher, J. and Jin, Z.M., (1993), `Friction and lubrication in cushion form bearings for artificial hip joints', Proc. Instn. Mech. Engrs., H(1), Eng. Med., 207, 25±33. Bhamra, M.S. and Case, C.P., (2006), `Biological effects of metal-on-metal hip replacements', Proc. Instn. Mech. Engrs., H(2), Eng. Med., 220, 379±384. Boutin, P., (1972), `Arthroplastie totale de la hanche par prostheses en alumine fritteÂ', Rev. Chir. Orthop., 58, 229±246. Boutin, P. and Blanquaert, D., (1979), `New materials used in total hip replacement', Cahiers d'Enseignement de la SOFCOT, No. 10, 27±44. Brown, C., Fisher, J. and Ingham, E., (2006), `Biological effects of clinically relevant wear particles from metal-on-metal hip prostheses', Proc. Instn. Mech. Engrs., H(2), Eng. Med., 220, 355±369. Chan, F.W., Bobyn, J.D., Medley, J.B., Krygier, J.J., Yue, S. and Tanzer, M., (1996), `Engineering issues and wear performance of metal hip implants', Clin. Orthop. Rel. Res., 333, 96±107. Chan, F.W., Medley, J.B., Bobyn, J.D. and Krygier, J.J., (1998), `Numerical analysis of time-varying fluid film lubrication of metal-metal hip implants in simulator tests', ASTM STP, 1346, 111±128 Chan, F.W., Bobyn, J.D., Medley, J.B., Krygier, J.J., and Tanzer, M., (1999), `Wear and lubrication of metal-on-metal hip implants', Clin. Orthop. Rel. Res., 369, 10±24. Charnley, J., (1961), `Arthroplasty of the hip ± a new operation', Lancet, 27 May, 1129± 1132. Charnley, J., (1967), `Factors in the design of an artificial hip joint', I. Mech. E. Proceedings, 3J, 181, 104±111. Charnley, J., Kamaangar, A. and Longfield, M.D., (1969), `The optimum size of prosthetic heads in relation to the wear of plastic sockets in total replacement of the hip', Med. Biol. Eng., 7, 31±39. Cobb, A.G. and Schmalzreid, T.P., (2006), `The clinical significance of metal ion release from cobalt±chromium metal-on-metal hip joint arthroplasty', Proc. Instn. Mech. Engrs. H, 220, 161±171. Dowson, D., (2003), `The relationship between steady-state wear rate and theoretical film thickness in metal-on-metal total replacement joints', Proceedings of the 29th Leeds-Lyon Symposium on Tribology, Elsevier, Tribology Series 41, 273±280. Dowson, D., (2006), `Tribological principles in metal-on-metal hip joint design', Proc. I. Mech. Eng, H, 220, 161±171. Dowson, D., McNie, C.M. and Goldsmith, A.A.J., (2000), `Direct experimental evidence of lubrication in a metal-on-metal total hip replacement', Proc. Instn. Mech. Engrs., C(1), 214, 75±86. Dowson, D., Hardaker, C., Flett, M. and Isaac, G. H., (2004a), `A hip joint simulator

© 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

459

study of the performance of metal-on-metal joints. Part I: The role of materials', J. Arthroplasty, 19(8), Suppl. 3, 118±123. Dowson, D., Hardaker, C., Flett, M. and Isaac, G. H., (2004b), `A hip joint simulator study of the performance of metal-on-metal joints. Part II: Design', J. Arthroplasty, 19(8), Suppl. 3, 124±130. Duff-Barclay, I., Scales, J.T. and Wilson, J.N., (1966), `The development of the Stanmore total hip replacement', Proc. Roy. Soc. Med., 59(10), 948±951. Endo, M., Tipper,J.L., Barton, D.C., Stone, M.H., Ingham, E., Farrar, R. and Fisher, J., (2002), `Comparison of wear, wear debris and functional biological activity of moderately crosslinked and non-crosslinked polyethylene in hip prostheses, Proc. Instn. Mech. Engrs., H, Eng. Med., 216(2), 111±122. Firkins, P.J., Tipper, J.L., Ingham, E., Stone, M.H., Farrar, R. and Fisher, J., (2001), `A novel low wearing differential hardness, ceramic-on-metal hip joint prosthesis', J. Biomechanics, 34, 1291±1298. Fisher, J., Jin, Z., Tipper, J., Stone, M. and Ingham, E., (2006), `Tribology of alternative bearings', Clin. Orthop. Rel. Res., 453, 25±34. Grigoris, P., Roberts, P., Panousis, K. and Jin, Z., (2006), `Hip resurfacing arthroplasty: the evolution of contemporary designs', Proc. Instn. Mech. Engrs., H(2), Eng. Med., 220, 95±105. Hamrock, B.J. and Dowson, D., (1978), `Elastohydrodynamic lubrication of elliptical contacts for materials of low elastic modulus-1-fully flooded conjunction', Trans. ASME, J. Lubrication Technol., 100(2), 236±245. Hamrock, B.J. and Dowson, D., (1981), Ball Bearing Lubrication, John Wiley & Sons, New York, 1±386. Hulbert, S.F., Young, F.A., Matthews, R.S., Klawitter, J.J., Talbert, C.D. and Sterling, F.H., (1970), `Potential of ceramic materials as permanent implantable skeletal prostheses', J. Biomed. Mater. Res., 4, 433±456. IME (1967), `Lubrication and wear in living and artificial human joints', I. Mech. E. Proc., 3J, 181, 1±172. Ishida, T., Clarke, I.C., Donaldson, T.D., Shirasu, H., Shishido, T and Yamamoto, K., (2007), `Ceramic-on-metal simulator wear and ion comparisons: 32 mm, 38 mm diameters', 53rd Annual Meeting of the Orthopaedic Research Society, San Diego, February 11±14, Poster 1671. Jalali-Vahid, D. and Jin, Z.M., (2002), `Transient elastohydrodynamic lubrication analysis of ultra-high molecular weight polyethylene hip joint replacements', Proc. Instn. Mech. Engrs., C, J. Mech. Eng. Sci., 216, 409±420. Jalali-Vahid, D., Jin, Z.M. and Dowson, D., (2003), `Isoviscous elastohydrodynamic lubrication of circular point contacts with particular reference to metal-on-metal hip implants', Proc. Instn. Mech. Engrs., J (5), J. Eng. Tribol., 217, 397±402. Jin, Z.M. and Dowson, D., (1999), `A full numerical analysis of hydrodynamic lubrication in artificial hip joint replacements constructed from hard materials', Proc. Instn. Mech. Engrs. C, 213, 355±370. Jin, Z.M., Dowson, D. and Fisher, J., (1997), `Analysis of fluid film lubrication in artificial hip joint replacements with surfaces of high elastic modulus', Proc. Instn. Mech. Engrs., H, Eng. Med., 211, 247±256. Jin, Z.M., Medley, J.B. and Dowson, D., (2003), `Fluid film lubrication in artificial hip joints', Proceedings of the 29th Leeds±Lyon Symposium on Tribology, Elsevier, Tribology Series 41, 237±256. Kumar, P., Oka, M., Ikeuchi, K., Shimizu, K., Yamamuro, T., Okumura, H. and Koyoura, Y., (1991), `Low wear rate of UHMWPE against zirconia (Y-PSZ) ceramic in © 2008, Woodhead Publishing Limited

460

Joint replacement technology

comparison to alumina ceramic and SUS 316L alloy', J. Biomed. Mater. Res., 25(7), 813±828. McKee, G.K., (1967), ` Developments in total hip joint replacement', I. Mech. E. Proc., 3J, 181, 85±89. McKee, G.K. and Watson-Farrar, J., (1966), `Replacement of arthritic hips by the McKee±Farrar prosthesis', J. Bone Joint Surg., 48B, 245±259. McMinn, D., and Daniel, J., (2006), `History and modern concepts in surface replacement', Proc. Instn. Mech. Engrs., H(2), Eng. Med., 220, 239±251. Miller, J.A., Talton, J.D. and Bhatia, S., (1996), `Alumina±alumina and alumina± polyethylene total hip prostheses', Chapter 4 in Clinical Performance of Skeletal Prostheses, Ed. by Hench, L.L. and Wilson, J., Chapman & Hall, 41±55. Morlock, M.M., Bishop, N., Ruther, W., Delling, G. and Hahn, M., (2006), Proc. Instn. Mech. Engrs., H(2), J. Eng. Med., 220, 333±343. Reynolds, L.A. and Tansey, E.M., Eds., (2007), Early Development of Total Hip Replacement, Wellcome Witness to Twentieth Century Medicine, Volume 29, 1± 167. Ring, P.A., (1968), `Complete replacement arthroplasty of the hip by the Ring prosthesis', J. Bone Joint Surg., 50B, 44±58. Ring, P.A., (1971), `Ring total hip replacement', Chapter 2 in Total Hip Replacement, Ed. by Jayson, M., Sector Publishing Limited, 26±46. Saikko, V.O, (1998), `A three-axis hip simulator for wear and friction studies', in Advances in Medical Tribology; Orthopaedic Implants and Implant Materials, Ed. Dowson, D., The Institution of Mechanical Engineers, 35±45. Scales, J.T., (1967), `Arthroplasty of the hip using foreign materials: a history', in Proc. I. Mech. E., 3J, 181, 63±84. Scholes, S.C., Burgess, I.C., Marsden, H.R., Unsworth, A., Jones, E. and Smith, N., (2006), `Compliant layer acetabular cups: friction testing of a range of materials and designs for a new generation of prosthesis that mimics a natural joint', Proc. Instn. Mech. Engrs., H(5), J. Eng. Med., 220, 583±596. Semlitsch, M. and Willert, H.G., (1997),`Clinical wear behaviour of ultra-high molecular weight polyethylene cups paired with metal and ceramic ball heads in comparison to metal-on-metal pairings of hip joint replacements', Proc. Instn. Mech. Engrs., H(1), J. Eng. Med., 211, 73±88. Semlitsch, M., Lehmann, M., Weber, H., Doerre, E. and Willert, H.G., (1977), `New prospects for a prolonged functional life-span of artificial hip joints by using the material combination polyethylene/aluminium oxide/metal', J. Biomed. Mater. Res., 11, 537±552. Shetty, V.D. and Villar, R.N., (2006), `Development and problems of metal-on-metal hip arthroplasty', Proc. Instn. Mech. Engrs., H(2), Eng. Med., 220, 371±377. Sivash, K.M., (1969), `The development of a total metal prosthesis for the hip joint from a partial joint replacement', Reconstr. Surg.Traumat., 11, 53±62. Smith, S.L., Dowson, D., Goldsmith, A.A.J., Valizadeh, R. and Colligan, J.S., (2001), `Direct evidence of lubrication in ceramic-on-ceramic total hip replacements', Proc. Instn. Mech. Engrs., C(3), 215, 265±268. Smith-Peterson, M.N., (1939), `Arthroplasty of the hip. A new method', J. Bone Joint Surg., 21(2), 269±288. Stewart, T.D., Tipper, J.L., Streicher, R.M., Ingham, E. and Fisher, J., (2001), `Long-term wear of HIPed alumina on alumina bearings for THR under microseparation conditions', J. Mater. Sci. Mater. Med., 12, 1053±1056. Stewart, T.D., Tipper, J.L., Insley, G., Streicher, R.M., Ingham, E. and Fisher, J., (2003), © 2008, Woodhead Publishing Limited

Hip replacement: tribological principles, materials and engineering

461

`Long-term wear in ceramic matrix composite materials for hip prostheses under severe swing phase microseparation', J. Biomed. Mater. Res., B, Appl. Biomater., 66 B, 567±573. Streicher, R.M., Semlitsch, M., SchoÈn, R., Weber, H. and Rieker, C., (1996), `Metal-onmetal articulation for artificial hip joints: laboratory study and clinical results', Proc. Instn. Mech. Engrs, H (3), J. Eng. Med., 210, 223±232. Thompson, F.R., (1954), `Two and a half years' experience with a vitallium intramedullary hip prosthesis', J. Bone Joint Surg., 36A, 489. Unsworth, A., (2006), `Tribology of artificial hip joints', Proc. Instn. Mech. Engrs, J(8), J. Eng. Tribol., 220, 711±718. Vassiliou, K., Elfick, A.P.D., Scholes, S.C. and Unsworth, A., (2006), `The effect of `running-in' on the tribology and surface morphology of metal-on-metal Birmingham hip resurfacing device in simulator studies', Proc. Instn. Mech. Engrs., H(2), J. Eng. Med., 220, 269±277. Visuri, T.I., Pukkala, E., Pulkkinen, P. and Paavolainen, P., (2006), `Cancer incidence and causes of death among total hip replacement patients: a review based on Nordic cohorts with a special emphasis on metal-on-metal bearings', Proc. Instn. Mech. Engrs., H(2), 220, 399±407. Wiles, P., (1957), `The surgery of the osteoarthritic hip', Br. J. Surg., 45, 488±497. Willmann, G., FruÈh, H.J. and Pfaff, H.G., (1996), `Wear characteristics of sliding pairs of zirconia (Y-TZP) for hip endoprostheses', Biomaterials, 17, 2157±2162. Zhou, Y.S., Ikeuchi, K. and Ohashi, M., (1997),`Comparison of the friction properties of four ceramic materials for joint replacements', WEAR, 210(1±2), 171±177.

© 2008, Woodhead Publishing Limited

19

Hip replacement: clinical perspectives M R E V E L L and E T D A V I S , Royal Orthopaedic Hospital, UK

19.1

Introduction

To achieve good results consistently is a challenge today just as it was for the pioneers of modern total hip replacement (THR) over four decades ago. At each stage along the patient's clinical journey, innovation and advances in technology promise incrementally to help improve performance. At the same time expectations from joint replacement are rising. Patients require more choices and more information. Widened indications for joint arthroplasty give rise to new challenges in selecting patients, assessing the best time for operating and preparing the patient for surgery. Digital radiographs are now commonplace and digital templating software allows detailed planning and sizing of components in a way not feasible a few years ago. Although less widely available, three-dimensional reconstructions from computerised tomograms bridge the interface between operative planning and the surgery itself; commercial platforms now provide navigated placement of implants from these images. Advances in modern anaesthesia have impacted directly on orthopaedic practice. Hypotensive techniques provide a bloodless field of operation and can reduce operating times. Selective nerve and lumbar plexus blocks or local infiltration techniques have been developed that allow mobilisation as early as a few hours after surgery. This leads to the possibility of a reduced thromboembolic risk and the expectation of early discharge. Such methods have usually required increased resources and personnel (such as more physical therapy input). While the transgluteal anterolateral and posterolateral surgical approaches remain most popular, recent initiatives have been towards reducing incision size and muscle dissection. Alterations and improvements in instrumentation have resulted. Mini-anterior, mini-posterior and two-incision techniques have been prominent in recent years. Computer-guided surgery is in its infancy. Surgeons can find it cumbersome and time-consuming to begin with. But to improve accuracy should mean to improve precision in the restoration of leg length, reduction in dislocation rates and benefits in wear profiles. © 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

463

Prosthetic engineering and design continue to evolve. Currently, modular designs offer interchangeability of components leading to advantages for storage, intra-operative decision making and revision. Modularity also produces new challenges for manufacturers; each component interface represents a potential for new implant failure (for example, galvanic corrosion, back side wear).1 In addition implants can be more expensive and inventories larger, requiring larger, more complex stock and storage areas. Bone conservation complements the minimally invasive philosophy, with the hope of preserving stock in younger patients so as to broaden revision options if they are ever required. The wide acceptance of metal-on-metal hip resurfacing has revolutionised bearings technology since 1995. The classical metal-on-polyethylene couple is now only one of a host of possibilities including metal-on-metal, oxinium or metal-on-highly-crosslinked polyethylene, ceramic-on-ceramic, ceramic-on-polyethylene or even metal-on-ceramic. With all combinations, the current trend is towards larger bearings as confidence in the materials and engineering tolerances grows (Figs 19.1 and 19.2). The tribology of the hip and the detailed arguments for this solution to the problem of wear at the bearing

19.1 Bilateral hip resurfacing. © 2008, Woodhead Publishing Limited

464

Joint replacement technology

19.2 A metal-on-metal total hip replacement.

surfaces are considered elsewhere in this book (see Chapters 2, 8 and 18). There are a large number of choices for implant fixation; an assortment of porous coatings and meshes for ingrowth of bone in cementless designs or a range of cements with varying viscosity, working characteristics and additives (such as antibiotics). Postoperatively, the return of the patient to normal activity has accelerated. Many implant manufacturers now boast that sporting activities are feasible with their implants. There does indeed appear to be an improvement in the function and activity scores for modern implants, particularly those with large bearings. Observers now seek more refined scoring systems, that allow differentiation between high-performance prostheses.2 The health economics of joint replacement vary according to local provision, but the market remains lucrative for manufacturers. The age threshold for © 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

465

intervention is steadily falling, so that more patients today are of working age. Return to work,3 including heavy manual work, has economic and societal benefits. Those at the older extreme of age are increasingly regarding the recovery periods and risks of surgery as acceptable and are not uncommonly wishing to undergo THR even in the tenth decade of life.4 It is clear that the clinician does not work in isolation. All these areas impact on surgical practice and on outcomes. But the journey begins and ends with the patient, and the question this year, as it would have been 40 years ago, is `Are we keeping our customers satisfied?'.

19.2

Problems with hip replacement at the beginning of the 21st century

19.2.1 Patient factors Hip replacement carries an increasing expectation of a safe and reliable result. Aspirations are driven by all involved in the activity: patients, clinicians, industry and, on occasion, politicians. It is not ethical or desirable to deny surgical intervention on the basis of age or particular co-morbidity. For all patients the decision to undergo surgery is a matter of how he or she balances the risks of intervention with the intended benefits. As confidence has grown in bearings and fixation methods, the perceived risks for younger patients appear to be lower. The risks in older patients and those with significant co-morbidities (including morbid obesity) remain high, but as numbers undergoing hip replacement continue to rise, demand for intervention remains high in all demographic and patient groups. Outcome in patients with high body mass index is thought to be associated with longer hospital stay, while opinion varies as to whether it is a risk factor for thrombo-embolic disease, dislocation, infection risk, bleeding or wound problems.5±12 Outcome in patients of young age13±15 is now regarded as acceptable in many cases so that the theshold for intervention in young patients is considerably lower than was usually the case even a decade ago. With the performance of more surgical procedures in young patients comes the expectation of more revision surgery in the future. This will possibly mean more complex revisions as those attending for repeated surgery do so at increasingly young age and with increasing expectation of safe surgery and a functional result. Outcome in the elderly4,16,17 is also felt to be acceptable to an increasing number of such patients. If multiple comorbidities have accumulated or there is poor physiological reserve to handle complications, risks may be relatively high, but, in the elderly, implant longevity is perceived as less of a problem. Consequently, there is a potential moral dilemma whether all patients should get the best hip replacement available or all patients should be entitled to a hip replacement that can reasonably be expected to meet their needs and is cheaper. © 2008, Woodhead Publishing Limited

466

Joint replacement technology

19.2.2 Problems associated with novel techniques In the middle of the 20th century advances in joint arthroplasty were led by clinicians in the main. As the industry developed and the commercial opportunities became clear, manufacturers moved to the fore with innovation, both in the use of new technologies and the provision of patient information. Patients also play a more prominent role in development, monitoring and evaluation of prostheses. New techniques and implants require responsible evaluation. Some innovations make the technical aspects of surgery more straightforward. Others can be technically more demanding or have a significant learning curve. For hip resurfacing, it is widely recognised that revision, particularly for fracture and malposition, are more common in early cases than later ones in large series.18±20 Many authors discuss the `learning curve' associated with minimally invasive surgery. These issues are important to acknowledge and it is vital that the reality of learning new techniques is recognised and accounted for in service planning. It is of interest that computer navigation, which is discussed in detail later in this chapter, appears to carry with it a learning period where operating times are initially increased, but may ultimately reduce the time taken to master other surgical techniques by providing immediate and detailed feedback for the surgeon.21 The emphasis of this chapter is that safe, reliable hip arthroplasty relies on core skills and insight that are by no means in the exclusive domain of the specialist arthroplasty surgeon, nor is there any evidence that the modern surgeon possesses these abilities in greater abundance than the pioneers of hip surgery. Rather, improvements in technology throw forward new challenges and changes, while vigilance remains vital. There are discussions in the literature about the relative merits of specialist centres as opposed to generalists, and also the reasons behind the continued contrast in results between centres that develop new techniques or prostheses that are better than the results obtained by those who follow operative procedures developed elsewhere. Similarly there are reported differences between the results of high-volume surgeons (that is those performing large numbers of operations) and low-volume surgeons. Familiarity may be the common thread that links these patterns.

19.3

Specific complications

19.3.1 Dislocation The current trend for implant design is for larger femoral heads to be used, this being made possible by improvements in the wear characteristics of newer bearings. Larger heads improve jump distance (the distance a femoral head must travel to dislocate) and range of movement to impingement according to Burroughs et al.22 and Fricka et al.,23 governing functional range of movement © 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

467

19.3 Some of the implant-related factors affecting hip prosthetic range of movement (after Yoshimine and Ginbayashi (2002),24).

and also relating to dislocation risk. On the acetabular side, the range of movement may be improved by cutting away sections of the rim or by reducing the section of a hemisphere desribed by the component (acetabular sector angle). This would correspond to increasing arc A in Fig. 19.3 (adapted from the paper by Yoshimine and Ginbayashi24). Most manufacturers offer an angled lip (10±20ë) in one sector of polyethylene implants. This can be rotated on an individual basis to the region of maximum perceived risk of dislocation while leaving other regions free of impingement. Hip resurfacing carries a relatively low head neck ratio as a result of conserving anatomical proportions. Yet the dislocation rate is very low in reported series. Jump distance may offer the explanation. In addition, it is possible, though to our knowledge it has never been investigated, that there is improved proprioception at the replaced hip from the retained native femoral neck when compared with a metal prosthetic stem. Computer navigation methods under evaluation include those guided by preoperative imaging of some kind and imageless techniques. A dedicated section on computer navigation is provided later in this chapter (Section 19.5). The early results suggest that dislocation rates might eventually be reduced using such methods.25

19.3.2 Leg length discrepancy Navigation also appears to be capable of reducing error in leg length restoration in hip replacement.25 Although widely used, preoperative templating has rarely been compared with non-templated hip replacement or emphasised specific© 2008, Woodhead Publishing Limited

468

Joint replacement technology

ally.26 Present-day computer templating platforms offer precise estimation of implant size.

19.3.3 Nerve palsy The possible causes of nerve palsy after total hip replacement are many and include traction, thermal damage and compression.27±33 The incidence varies between 0.6 and 3.7% of primary hip replacements according to these authors. Recovery is often incomplete and may take up to 2 years. Those with milder lesions and early signs of recovery tend to have a better outcome than those with complete lesions that have not shown signs of recovery before leaving hospital. An increased rate of nerve palsy has been reported in some series of minimally invasive hip surgery.34 These would appear often to be the result of traction injuries that to some degree may represent a manifestation of the surgeon's learning process. This has been discussed earlier, but the issue has also driven adaptations in instrumentation which is specially designed so that the procedure has to be carried out in a certain way.35

19.3.4 Infection It is likely that over time there will be an increase in the antibiotic resistance of bacteria in periprosthetic infections. The commonest organisms remain those in the genus Staphylococcus; with Staph. epidermidis and Staph. aureus36 highest among the pathogens. It is possible that the apparent rise in numbers of Staph. epidermidis infections represents improved detection.37 Such advances in sensitivity, however, can prove problematic. Polymerase chain reaction (PCR) methods are extremely sensitive and will on occasion give a positive indication for more than one pathogen. It may be that as in the case of Staph. epidermidis, PCR is simply unmasking multiple pathogenic organisms, hitherto undetected because of the phenomenon of one organism dominating in the culture enrichment broths of the conventional microbiological methods. That false positive results can occur with PCR methods is recognised and the likely explanations for this are described briefly in Chapter 15. Prevention of infection for the majority of hip replacements carried out in the developed world includes use of a laminar flow theatre and the use of antibiotics. Bone cement usually includes an antibiotic, often with gentamicin. Despite the perceived gradual shift in bacterial and resistance patterns, this form of microprophylaxis has remained largely unchanged for many years.38 For uncemented prostheses, attention is moving to the consideration of whether antibiotic may be added to the implant to good effect. Debridement is feasible in the earliest stages after the index procedure. Antibiotic suppression of infection without re-operation can have success, but is generally considered only for the very elderly or infirm. Revision of the infected © 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

469

total hip replacement can be carried out in one or more stages but these issues are beyond the remit of this chapter.

19.3.5 Thrombo-embolic disease A variety of alternative anticoagulant therapies are regarded as acceptable by the majority of surgeons for the prevention of thrombo-embolic disease; low molecular weight heparins, heparin, warfarin and aspirin among the most popular. It seems likely that a clear and definitive answer to any question of the benefit from their use would be possible only from a large randomised controlled study. Sufficient power would be feasible only if many centres were involved and the economics of such a project are prohibitive. Most dose regimes are in the prophylactic range where the risk of bleeding complications, including fatal bleeds, is finely balanced against the risk of death from a thrombo-embolic event. There is also no consensus as to the optimum length of prophylactic therapy.39,40 Early mobilisation as the simplest from of mechano-prophylaxis is often underestimated in its importance.

19.3.6 Loosening For cemented prostheses, the dominant philosophy is that the use of a polished stem with a double or triple taper and a spacer at the tip allows subsidence without cement fracture. Theoretically this type of device converts longitudinal forces into centripetal hoop forces that reinforce contact at the bone±cement interface. Avoidance of crack propagation in the cement mantle is important in lessening the possibility of cement particle generation in the implant bed, as is considered elsewhere in this book. For uncemented implants, a variety of materials have been used to encourage bone ingrowth. Overall, materials can be classified in terms of their biocompatibility and the nature of any reactive bone formation. For porous coated implants, pore size and shape influence host response. The optimum pore size is probably in the range 1±400 m. Rigid fixation provides the necessary mechanical environment, which may be achieved by press fit or supplementary fixation with screws or pegs. Implant integration can be further supplemented by osteoconductive coatings such as plasma-sprayed hydroxyapatite or tricalcium phosphate. It is not agreed whether such porous and osteo-integrative surfaces should cover the whole prosthesis or the proximal part. Proximal stress shielding does not appear to be a major issue with fully coated designs in which long-term follow-up is available.41

19.3.7 Wear The traditional low-friction metal-on-polyethylene bearing was adopted widely following the success of the Charnley hip replacement. Metal-on-metal bearings © 2008, Woodhead Publishing Limited

470

Joint replacement technology

were often associated with early failure and were largely abandoned for two decades. In the 1990s hip resurfacing came to new prominence. The Birmingham Hip Replacement (BHR) in fact re-used the metallurgy of the McKee±Farrar total hip replacement. The BHR components are as-cast highcarbide cobalt±chrome alloy. It is thought that the relative success of the BHR has resulted from improvements in engineering that allow smoothness and roundness to be manufactured reproducibly to a level of accuracy at least one order of magnitude better than the McKee±Farrar implants. At the same time, ceramic bearings have been evolving as a parallel hard bearing technology. Hard bearings promise lower linear wear rates. The tribology of hard bearings suggests that larger heads are preferable for achieving lower wear by allowing true fluid-film lubrication. Coupled with the awareness that stability is aided by larger heads, this has meant that head sizes have tended to increase from those of former designs such as the average 22.5 mm diameter Charnley or 26 mm diameter Exeter femoral heads. Linear wear remains an issue, but the new bearings have brought new problems. Osteolysis around a polyethylene implant is well recognised and in a sense predictable. Implant designs vary greatly in their susceptibility to particle generation, but the pathophysiology is largely understood. Ceramic particles are believed to be relatively inert, but metal particles derived from metal on metal hips are small in total volume yet high in number because of their nanometre size, as shown from ultrastructural studies.42,43 Nanoparticles of metal may have considerable biological effects, though there is a need for basic and clinical research studies to clarify the situation with respect to metal-on-metal implants. The current level of knowledge is also considered in more detail in the section on biological responses to implants (Chapter 15).

19.3.8 Fracture Intra-operative periprosthetic fracture is a feature of uncemented designs that is not seen in cemented joint replacements. The incidence is approximately 0.5%. In the majority of cases prophylactic wiring and protected weight bearing are sufficient to achieve a good clinical result. For hip resurfacing, femoral fracture is a major consideration as it accounts for a high percentage of early failures. Risk factors are thought to include: female sex, patient weight,18 low bone density, intra-operative femoral neck notching and varus positioning of the component,44 and learning curve effects for the surgeon.45 None of these has a simple relationship to femoral neck fracture. Vascularity of the femoral head is much discussed in relation to hip resurfacing and has been suggested as a risk factor for fracture.46

© 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

19.4

471

Current solutions

19.4.1 Implant designs and materials Hip resurfacing in its current incarnation of large metal-on-metal bearings, using a hybrid of uncemented acetabular cup and cemented femoral component has galvanised hip arthroplasty and implant design. Tribology is now a science taught to clinicians, and arthroplasty surgeons must be aware of the principles of friction, lubrication and wear in order to make informed choices with their patients. Large heads appear to have a lower dislocation risk and to be associated with better functional scores. Range of movement probably relates to head size relative to the femoral neck width, rather than head size per se. Conservation of bone, albeit usually necessitating a full capsulotomy, is a primary and unique advantage of the resurfacing concept. With large numbers of hard bearings being used, it is inevitable that there has also evolved some concept of how large metal-on-metal bearings can have problems. It is believed that poor alignment of the couple, particularly an `open' cup position (that is one with an excessive abduction angle), leads to point loading and excessive generation of metal wear particles. These particles might have systemic or local effects on cells, aspects of which are considered briefly in the next section.

19.4.2 Implant biology Cobalt and chromium ions have been recorded in the blood of individuals with metal-on-metal implants at levels approaching those found in high occupational exposure, but it is not certain how long levels remain at this ceiling or how much harm is caused. Carcinogenesis is not known to occur more frequently with joint replacement compared with the general population according to the latest analyses. Primary hip arthroplasty, taking all-comers, has a mortality of approximately 1 in 200. Revision surgery is more hazardous than that. It therefore seems reasonable to counsel patients with this perspective in mind when discussing the possibility of rare neoplasia. Vigilant observation is, however, also required and should include the prospective collection of data from large populations of individuals with hip arthroplasties, such as national registers not only of joint replacement but also of cancer. Only then can meaningful information be obtained and, from this, knowledge increased. Local hypersensitivity and pain are recognised, including a syndrome known as ALVAL (aseptic lymphocyte-dominated vasculitis associated lesion). The effect of implants on the local environment and on the body as a whole are increasingly appreciated. In addition, the complex but very relevant relationship between mechanical and biological environment is starting to be understood. This has pertinence not only with respect to implant failure as in the case of ALVAL, but also to implant fixation. Porous coatings have a range of tolerances © 2008, Woodhead Publishing Limited

472

Joint replacement technology

for the pore size but also for micromovement. Outside the optimal conditions, osteo-integration fails to occur. As well as bone-like chemical structures, implants now mimic bone architecture. Biocompatible metals such as tantulum and titanium can now be fashioned in such a way as to so closely resemble the trabecular structure of bone that osteo-integration is now feasible to an extent greatly exceeding previous expectations. In particular, interest has centred on how to avoid early femoral failure following hip resurfacing. At the time of writing, it is generally agreed that a small amount of valgus inclination relative to the native neck is advantageous.47,48 Avoidance of femoral notching, particularly superiorly, is also important, with biomechanical and finite element analysis indicating that even a small degree of notching predisposes to fracture.49 The detrimental biomechanical effect of leaving exposed cancellous bone at the component bone interface has also been demostrated.50 Total hip replacement design is also evolving. Primary uncemented hip replacements may work on a cortical fit principle (e.g., Synergy, Smith and Nephew) or the more conservative cancellous impaction technique (e.g., Corail Hip, De Puy).

19.4.3 Surgical approaches Minimising incision size and muscle dissection have been important research areas over the past 10±15 years. Many techniques have been described and some novel prostheses developed or revisited to make implantation feasible with these less-invasive surgical approaches. Concerns have been raised about deep infection rates and nerve palsy.51 However, it is likely that this interest has led to a general reduction in standard incision lengths and to an improvement in instrumentation.52 The mini-posterior approach has the advantage that it can be introduced incrementally by those already familiar with the posterior approach; this might reduce risk of complications while the technique is mastered.

19.5

Computer navigation

The use of computer navigation in total knee replacement is well established.53±56 However, it has been slower to gain widespread acceptance in total hip replacement. This almost certainly centres on the difficulty in registering pelvic landmarks and uncertainty as to the optimum acetabular component orientation. The potential benefits of navigation are: (a) improved accuracy in the restoration of leg length, (b) precise data on femoral and acetabular offset, (c) improved acetabular and femoral component angular placement, and (d) the availability of precise range of movement and impingement data. In hip resurfacing, the use of navigation has proved attractive in providing an accurate method of aligning the femoral component57 and potentially avoiding © 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

473

subsequent femoral neck fracture. The additional advantage of aligning both components in the couple optimally may also be possible, thereby reducing wear.

19.5.1 Types of navigation system The navigation systems fall into three main categories: 1. Systems based on computed tomography (CT). These rely on a pre-operative CT scan, which is then correlated intra-operatively using pre-defined registration points. This technique has the advantage of providing the surgeon with a precise model of the intra-operative field from the preoperative scan. The disadvantage of this system is the added step of acquiring a CT scan. The use of low-dose protocols has reduced the radiation exposure from the scans.58 However, their lack of availability in general orthopaedic practice remains a hurdle. 2. Fluoroscopic computer navigation. This provides a reference from intraoperative fluoroscopic images. Uptake has been limited due to concerns over the added intra-operative time acquiring images, increased radiation exposure to the patient and staff and worries about maintaining sterility during the fluoroscopic image acquisition. 3. Imageless computer navigation. This has proved the most popular modality. Improved algorithms and computing power can now provide the surgeon with the required accuracy without the need for additional imaging. The acquisition of defined anatomical points and `clouds' of points provides the systems with detail sufficient to develop a morphed anatomical model of the patient.

19.5.2 Potential benefits of computer navigation Leg length restoration The use of navigation can provide precise leg length measurement, with the surgeon able to delineate accurately the difference between the preoperative and postoperative leg lengths. However, this then leaves the surgeon with the dilemma of deciding to what length the leg should be restored. Leg length perception to the patient is governed by many issues, including preoperative habitation, pelvic obliquity, hip contractures in both the sagittal and coronal planes and pathology above and below the hip (such as a severe knee deformity or a scoliosis). The use of accurately measured antero-posterior full leg radiographs plus standing and sitting pelvic radiographs is recommended. The sitting pelvic radiograph in particular can help in defining whether a pelvic obliquity is fixed or correctable. Navigation technology then gives the surgeon precise information, thereby avoiding leaving the patient with a clinically perceivable leg length inequality and the associated morbidity. © 2008, Woodhead Publishing Limited

474

Joint replacement technology

Offset restoration/optimisation Computer navigation gives the surgeon precise information in correctly restoring acetabular and femoral offset. The data allow intra-operative adjustment on either the femoral or acetabular side if a modular system is used. Femoral offset is the distance from the centre of rotation of the femoral head to a line bisecting the long axis of the femur. It can be altered by using high offset femoral stem components or head components incorporating different neck lengths. Acetabular offset is the distance between the centre of rotation of the hip and the midline of the body. It can be modified by adjusting the depth of acetabular reaming or the use of a lateralised liner. Component orientation The optimum position of the femoral and acetabular components during total hip replacement still remains an enigma. The work by Lewinnek et al.59 suggested a `safe zone' in which the acetabular component should be placed. They suggest an abduction angle of 40ë  10ë and an ante-version angle of 15ë  10ë. The accuracy of placement of the acetabular component without computer navigation is poor, with one multi-centre study demonstrating that only 26% of acetabular components were within the `safe zone'.60 Even when mechanical alignment guides are used there is still a considerable percentage (42%) outside the `safe zone'.61 The inaccuracy is due to variations in patient anatomy and variations in the reference planes used for component measurement. Murray62 demonstrated the differences in measurement of acetabular orientation, depending on whether observations are defined on an anatomical, radiographical and operative basis. He provided nomograms to inter-relate these different measurements with reference to the modality being used; an operative abduction angle of 40ë and anteversion angle of 15ë would translate into an abduction angle of 41ë and anteversion angle of 11ë on radiographical measurement. These discrepancies compound the confusion as to the `perfect' acetabular position. Further discomposure is encountered when establishing an origin of reference for acetabular orientation. DeGioia et al.63 demonstrated that there is a diversity of pelvic flexion and extension (5±70ë), when moving from a sitting to standing position. Variability in pelvic movement in both the sagittal and coronal plane suggests that registering an absolute position from the pelvic anatomy may not represent the `functional' requirements of different individuals, so that consideration of the whole sagittal and coronal `balance' of the patient may be required.64 The factors effecting the acetabular component orientation with respect to the whole body can be divided into spinal, pelvic and acetabular variances. It may be that until all these are accounted for, the acetabular component will not be in a position to achieve the `perfect' orientation in all patients. The `safe zone' of Lewinnek et al.59 is likely therefore to be an oversimplification of the exact acetabular position and further studies are required. © 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

475

At present, pelvic orientation is commonly registered using the anterior pelvic plane. This requires registration of both anterior superior iliac spines (ASIS) and the pubic tubercles or the symphysis pubis. The technical difficulties of registering boney landmarks at the time of surgery can lead to inaccuracy, particularly in the presence of considerable adipose tissue. Wolf et al.,65 examined pelvic registration errors, showing that a 4 mm error in ASIS localisation leads to a 7ë error in acetabular anteversion. Registration is classically done with the patient supine; the patient then has to be turned into a lateral position while maintaining sterility of the pelvic reference frame. The technique can be modified to register the patient in the lateral position or a semi-lateral position. However, the boney landmarks can be even more difficult to palpate in these positions than in the supine position, leading to inaccuracy.66 These adversities have led others to explore alternative concepts of acetabular positioning. Archbold et al.67 have suggested utilising the transverse acetabular ligament and acetabular labrum as a reference for acetabular positioning. This has the advantage of not having to register the pelvic planes. However, this still remains to be fully validated and concerns remain about whether the native acetabular position should be recreated in hips that have developed secondary osteoarthritis due to an unfavourable biomechanical alignment, such as in developmental dysplasia. Ranawat has popularised the idea of `combined anteversion' when positioning the acetabular component.68 The idea of first positioning the femoral component and then adjusting the position of the acetabular component relative to this, does have considerable theoretical merit. The algorithms by Yoshimine and Ginbayashi24 can be used to computer navigate the acetabular component into the most advantageous position and maximise the range of movement without impingement. These concepts require further development, as the present algorithms provide the range of movement data with respect to prosthesis on prosthesis impingement, without consideration of bone or soft tissue impingement. The increasing use of hard-on-hard bearings in total hip replacement in the form of hip resurfacing and of large metal on metal total hip replacements has provided a greater insight into the importance of the bearing couple. Elevated blood metal ions seen after metal-on-metal joint replacement arise when the joint is edge loading or in an unfavourable couple during episodes of greatest load. Computer navigation should allow the surgeon to place the components in the optimal position to achieve the most effective bearing couple for the common activities of daily life in a way that might not be feasible without rapid calculations and carefully controlled algorithms. The potential benefits are unlikely to be limited to metal-on-metal bearings. Data from in vivo studies have reproduced the forces generated in the hip during different physiological activities.69 The incorporation of these data into algorithms, describing the contact areas of the bearing would provide computer navigation having another level of sophistication. © 2008, Woodhead Publishing Limited

476

19.6

Joint replacement technology

Conclusions

Technological advances in hip replacement have reduced many of the problems that faced the arthroplasty surgeon. Modularity, larger femoral heads and bone conservation are the present trends in implant design as the mechanics and engineering have advanced. It is likely that the next generation of implants will move towards a deeper appreciation of the biological environment in which the implant performs. It is also likely that the main battlegrounds in the prevention of implant failure will remain similar; technical surgical issues, loosening and wear. Finally it is likely that advances will be matched by an escalation of patient expectation.

19.7

References

1. McCarthy, J. C.; Bono, J. V.; and O'Donnell, P. J.: Custom and modular components in primary total hip replacement. Clin Orthop Relat Res, 344: 162±71, 1997. 2. Bauman, S.; Williams, D.; Petruccelli, D.; Elliott, W.; and de Beer, J.: Physical activity after total joint replacement: a cross-sectional survey. Clin J Sport Med, 17(2): 104±8, 2007. 3. Mobasheri, R.; Gidwani, S.; and Rosson, J. W.: The effect of total hip replacement on the employment status of patients under the age of 60 years. Ann R Coll Surg Engl, 88(2): 131±3, 2006. 4. Levy, R. N.; Levy, C. M.; Snyder, J.; and Digiovanni, J.: Outcome and long-term results following total hip replacement in elderly patients. Clin Orthop Relat Res, 316: 25±30, 1995. 5. Aderinto, J., and Brenkel, I. J.: Pre-operative predictors of the requirement for blood transfusion following total hip replacement. J Bone Joint Surg Br, 86(7): 970±3, 2004. 6. Bowditch, M. G. and Villar, R. N.: Do obese patients bleed more? A prospective study of blood loss at total hip replacement. Ann R Coll Surg Engl, 81(3): 198±200, 1999. 7. Flugsrud, G. B.; Nordsletten, L.; Espehaug, B.; Havelin, L. I.; and Meyer, H. E.: Risk factors for total hip replacement due to primary osteoarthritis: a cohort study in 50,034 persons. Arthritis Rheum, 46(3): 675±82, 2002. 8. Jacobsen, S., and Sonne-Holm, S.: Increased body mass index is a predisposition for treatment by total hip replacement. Int Orthop, 29(4): 229±34, 2005. 9. Karlson, E. W.; Mandl, L. A.; Aweh, G. N.; Sangha, O.; Liang, M. H.; and Grodstein, F.: Total hip replacement due to osteoarthritis: the importance of age, obesity, and other modifiable risk factors. Am J Med, 114(2): 93±8, 2003. 10. McLaughlin, J. R., and Lee, K. R.: The outcome of total hip replacement in obese and non-obese patients at 10- to 18-years. J Bone Joint Surg Br, 88(10): 1286±92, 2006. 11. Patel, A. D., and Albrizio, M.: Relationship of body mass index to early complications in hip replacement surgery: study performed at Hinchingbrooke Hospital, Orthopaedic Directorate, Huntingdon, Cambridgeshire. Int Orthop, 31(4): 439±43, 2007. 12. Sadr Azodi, O.; Bellocco, R.; Eriksson, K.; and Adami, J.: The impact of tobacco use and body mass index on the length of stay in hospital and the risk of post-operative © 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

477

complications among patients undergoing total hip replacement. J Bone Joint Surg Br, 88(10): 1316±20, 2006. Heisel, C.; Silva, M.; and Schmalzried, T. P.: Bearing surface options for total hip replacement in young patients. Instr Course Lect, 53: 49±65, 2004. Kim, Y. H.; Kook, H. K.; and Kim, J. S.: Total hip replacement with a cementless acetabular component and a cemented femoral component in patients younger than fifty years of age. J Bone Joint Surg Am, 84-A(5): 770±4, 2002. Sedel, L.; Nizard, R. S.; Kerboull, L.; and Witvoet, J.: Alumina±alumina hip replacement in patients younger than 50 years old. Clin Orthop Relat Res, 298: 175± 83, 1994. Evans, B. G., and Salvati, E. A.: Total hip replacement in the elderly: cost-effective alternatives. Instr Course Lect, 43: 359±65, 1994. Garellick, G.; Malchau, H.; Herberts, P.; Hansson, E.; Axelsson, H.; and Hansson, T.: Life expectancy and cost utility after total hip replacement. Clin Orthop Relat Res, 346: 141±51, 1998. Marker, D. R.; Seyler, T. M.; Jinnah, R. H.; Delanois, R. E.; Ulrich, S. D.; and Mont, M. A.: Femoral neck fractures after metal-on-metal total hip resurfacing: a prospective cohort study. J Arthroplasty, 22(7 Suppl 3): 66±71, 2007. McMinn, D. J.; Daniel, J.; Pynsent, P. B.; and Pradhan, C.: Mini-incision resurfacing arthroplasty of hip through the posterior approach. Clin Orthop Relat Res, 441: 91± 8, 2005. Siebel, T.; Maubach, S.; and Morlock, M. M.: Lessons learned from early clinical experience and results of 300 ASR hip resurfacing implantations. Proc Inst Mech Eng [H], 220(2): 345±53, 2006. Cobb, J. P.; Kannan, V.; Brust, K.; and Thevendran, G.: Navigation reduces the learning curve in resurfacing total hip arthroplasty. Clin Orthop Relat Res, 463: 90± 7, 2007. Burroughs, B. R.; Hallstrom, B.; Golladay, G. J.; Hoeffel, D.; and Harris, W. H.: Range of motion and stability in total hip arthroplasty with 28-, 32-, 38-, and 44-mm femoral head sizes. J Arthroplasty, 20(1): 11±19, 2005. Fricka, K. B.; Marshall, A.; and Paprosky, W. G.: Constrained liners in revision total hip arthroplasty: an overuse syndrome: in the affirmative. J Arthroplasty, 21(4 Suppl 1): 121±5, 2006. Yoshimine, F., and Ginbayashi, K.: A mathematical formula to calculate the theoretical range of motion for total hip replacement. J Biomech, 35(7): 989±93, 2002. Sugano, N.; Nishii, T.; Miki, H.; Yoshikawa, H.; Sato, Y.; and Tamura, S.: Mid-term results of cementless total hip replacement using a ceramic-on-ceramic bearing with and without computer navigation. J Bone Joint Surg Br, 89(4): 455±60, 2007. Hedley, A. K.; Gruen, T. A.; Borden, L. S.; Hungerford, D. S.; Habermann, E.; and Kenna, R. V.: Two-year follow-up of the PCA noncemented total hip replacement. Hip: 225±50, 1987. Butt, A. J.; McCarthy, T.; Kelly, I. P.; Glynn, T.; and McCoy, G.: Sciatic nerve palsy secondary to postoperative haematoma in primary total hip replacement. J Bone Joint Surg Br, 87(11): 1465±7, 2005. Crawford, J. R.; Van Rensburg, L.; and Marx, C.: Compression of the sciatic nerve by wear debris following total hip replacement: a report of three cases. J Bone Joint Surg Br, 85(8): 1178±80, 2003. Jerosch, J.: Femoral nerve palsy in hip replacement due to pelvic cement extrusion. Arch Orthop Trauma Surg, 120(9): 499±501, 2000.

© 2008, Woodhead Publishing Limited

478

Joint replacement technology

30. Oleksak, M., and Edge, A. J.: Compression of the sciatic nerve by methylmethacrylate cement after total hip replacement. J Bone Joint Surg Br, 74(5): 729±30, 1992. 31. Pritchett, J. W.: Nerve injury and limb lengthening after hip replacement: treatment by shortening. Clin Orthop Relat Res, 418: 168±71, 2004. 32. Schmalzried, T. P.; Amstutz, H. C.; and Dorey, F. J.: Nerve palsy associated with total hip replacement. Risk factors and prognosis. J Bone Joint Surg Am, 73(7): 1074±80, 1991. 33. Sosna, A.; Pokorny, D.; and Jahoda, D.: Sciatic nerve palsy after total hip replacement. J Bone Joint Surg Br, 87(8): 1140±1, 2005. 34. Bal, B. S.; Haltom, D.; Aleto, T.; and Barrett, M.: Early complications of primary total hip replacement performed with a two-incision minimally invasive technique. Surgical technique. J Bone Joint Surg Am, 88 Suppl 1 Pt 2: 221±33, 2006. 35. Jablonski, M.; Gorzelak, M.; Kuczynski, P.; Piasecki, J.; and Turzanska, K.: Less invasive posterior surgical approach for hip joint replacement ± complications and limitations. Orthop Traumatol Rehabil, 9(1): 25±30, 2007. 36. Lidwell, O. M.; Elson, R. A.; Lowbury, E. J.; Whyte, W.; Blowers, R.; Stanley, S. J.; and Lowe, D.: Ultraclean air and antibiotics for prevention of postoperative infection. A multicenter study of 8,052 joint replacement operations. Acta Orthop Scand, 58(1): 4±13, 1987. 37. Ringberg, H.; Sanzen, L.; Thoren, A.; and Walder, M.: Bacteriologic evidence of infection caused by coagulase-negative staphylococci in total hip replacement. J Arthroplasty, 13(8): 935±8, 1998. 38. Buchholz, H. W. and Engelbrecht, H.: Uber die Depotwirkung einiger Antibiotica bei Wermischung dem Kunstharz Palacos. Chirurg, 41: 511±15, 1970. 39. McNally, M. A.: Insufficient duration of venous thromboembolism prophylaxis after total hip or knee replacement when compared with the time course of thromboembolic events. J Bone Joint Surg Br, 89(10): 1409, 2007. 40. Skedgel, C.; Goeree, R.; Pleasance, S.; Thompson, K.; O'Brien, B.; and Anderson, D.: The cost-effectiveness of extended-duration antithrombotic prophylaxis after total hip arthroplasty. J Bone Joint Surg Am, 89(4): 819±28, 2007. 41. Karachalios, T.; Tsatsaronis, C.; Efraimis, G.; Papadelis, P.; Lyritis, G.; and Diakoumopoulos, G.: The long-term clinical relevance of calcar atrophy caused by stress shielding in total hip arthroplasty: a 10-year, prospective, randomized study. J Arthroplasty, 19(4): 469±75, 2004. 42. Case, C. P.; Langkamer, V. G.; James, C.; Palmer, M. R.; Kemp, A. J.; Heap, P. F.; and Solomon, L.: Widespread dissemination of metal debris from implants. J Bone Joint Surg Br, 76(5): 701±12, 1994. 43. Doom, P. F.; Campbell, P. A.; Worrall, J.; Benya, P. D.; McKellop, H. A.; and Amstutz, H. C.: 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±11, 1998. 44. Long, J. P., and Bartel, D. L.: Surgical variables affect the mechanics of a hip resurfacing system. Clin Orthop Relat Res, 453: 115±22, 2006. 45. Morlock, M. M.; Bishop, N.; Ruther, W.; Delling, G.; and Hahn, M.: Biomechanical, morphological, and histological analysis of early failures in hip resurfacing arthroplasty. Proc Inst Mech Eng [H], 220(2): 333±44, 2006. 46. Beaule, P. E.; Campbell, P. A.; Hoke, R.; and Dorey, F.: Notching of the femoral neck during resurfacing arthroplasty of the hip: a vascular study. J Bone Joint Surg Br, 88(1): 35±9, 2006. © 2008, Woodhead Publishing Limited

Hip replacement: clinical perspectives

479

47. Anglin, C.; Masri, B. A.; Tonetti, J.; Hodgson, A. J.; and Greidanus, N. V.: Hip resurfacing femoral neck fracture influenced by valgus placement. Clin Orthop Relat Res, 465: 71±9, 2007. 48. Davis, E. T.; Olsen, M.; Zdero, R.; Waddell, J.; and Schemitsch, E. H.: Femoral component alignment and the risk of femoral neck fracture following hip resurfacing. Presentation at the American Academy of Orthopaedic Surgeons, San Diego, 2007. 49. Davis, E. T.; Olsen, M.; Papini, M.; Zdero, R.; Waddell, J.; and Schemitsch, E. H.: A biomechanical and finite element analysis of femoral neck notching during hip resurfacing. Presentation at the American Academy of Orthopaedic Surgeons, San Diego, 2007. 50. Olsen, M.; Davis, E. T.; and Schemitsch, E. H.: The biomechanical effect of exposed cancellous bone in hip resurfacing arthroplasty. Presentation at the American Academy of Orthopaedic Surgeons, San Francisco, 2008. 51. Rosenberg, A. G.: The ugly underbelly of the MIS movement: in opposition. J Arthroplasty, 22(4 Suppl 1): 102±5, 2007. 52. Paillard, P.: Hip replacement by a minimal anterior approach. Int Orthop, 31 Suppl 1: S13±5, 2007. 53. Decking, R.; Markmann, Y.; Fuchs, J.; Puhl, W.; and Scharf, H. P.: Leg axis after computer-navigated total knee arthroplasty: a prospective randomized trial comparing computer-navigated and manual implantation. J Arthroplasty, 20(3): 282±8, 2005. 54. Haaker, R. G.; Stockheim, M.; Kamp, M.; Proff, G.; Breitenfelder, J.; and Ottersbach, A.: Computer-assisted navigation increases precision of component placement in total knee arthroplasty. Clin Orthop Relat Res, 433: 152±9, 2005. 55. Rosenberger, R. E.; Hoser, C.; Quirbach, S.; Attal, R.; Hennerbichler, A.; and Fink, C.: Improved accuracy of component alignment with the implementation of imagefree navigation in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc, 2008 Mar, 16(3):249±57. 56. Sparmann, M.; Wolke, B.; Czupalla, H.; Banzer, D.; and Zink, A.: Positioning of total knee arthroplasty with and without navigation support. A prospective, randomised study. J Bone Joint Surg Br, 85(6): 830±5, 2003. 57. Davis, E. T.; Gallie, P.; Macgroarty, K.; Waddell, J. P.; and Schemitsch, E.: The accuracy of image-free computer navigation in the placement of the femoral component of the Birmingham Hip Resurfacing: a cadaver study. J Bone Joint Surg Br, 89(4): 557±60, 2007. 58. Lattanzi, R.; Baruffaldi, F.; Zannoni, C.; and Viceconti, M.: Specialised CT scan protocols for 3-D pre-operative planning of total hip replacement. Med Eng Phys, 26(3): 237±45, 2004. 59. Lewinnek, G. E.; Lewis, J. L.; Tarr, R.; Compere, C. L.; and Zimmerman, J. R.: Dislocations after total hip-replacement arthroplasties. J Bone Joint Surg Am, 60(2): 217±20, 1978. 60. Saxler, G. et al.: The accuracy of free-hand cup positioning ± a CT based measurement of cup placement in 105 total hip arthroplasties. Int Orthop, 28(4): 198±201, 2004. 61. Hassan, D. M.; Johnston, G. H.; Dust, W. N.; Watson, G.; and Dolovich, A. T.: Accuracy of intraoperative assessment of acetabular prosthesis placement. J Arthroplasty, 13(1): 80±4, 1998. 62. Murray, D. W.: The definition and measurement of acetabular orientation. J Bone Joint Surg Br, 75(2): 228±32, 1993. © 2008, Woodhead Publishing Limited

480

Joint replacement technology

63. DiGioia, A. M.; Hafez, M. A.; Jaramaz, B.; Levison, T. J.; and Moody, J. E.: Functional pelvic orientation measured from lateral standing and sitting radiographs. Clin Orthop Relat Res, 453: 272±6, 2006. 64. Vaz, G.; Roussouly, P.; Berthonnaud, E.; and Dimnet, J.: Sagittal morphology and equilibrium of pelvis and spine. Eur Spine J, 11(1): 80±7, 2002. 65. Wolf, A.; Digioia, A. M., 3rd; Mor, A. B.; and Jaramaz, B.: Cup alignment error model for total hip arthroplasty. Clin Orthop Relat Res, 437: 132±7, 2005. 66. Spencer, J. M.; Day, R. E.; Sloan, K. E.; and Beaver, R. J.: Computer navigation of the acetabular component: a cadaver reliability study. J Bone Joint Surg Br, 88(7): 972±5, 2006. 67. Archbold, H. A.; Mockford, B.; Molloy, D.; McConway, J.; Ogonda, L.; and Beverland, D.: The transverse acetabular ligament: an aid to orientation of the acetabular component during primary total hip replacement: a preliminary study of 1000 cases investigating postoperative stability. J Bone Joint Surg Br, 88(7): 883±6, 2006. 68. Lucas, D. H., and Scott, R. D.: The Ranawat Sign. A specific maneuver to assess component positioning in total hip arthroplasty. J Orthop Tech, 2: 59±61, 1994. 69. Bergmann, G.; Deuretzbacher, G.; Heller, M.; Graichen, F.; Rohlmann, A.; Strauss, J.; and Duda, G. N.: Hip contact forces and gait patterns from routine activities. J Biomech, 34(7): 859±71, 2001.

© 2008, Woodhead Publishing Limited

20

Knee replacement: clinical perspectives

J D B L A H A , University of Michigan Medical School, USA

20.1

Introduction

It is difficult to overstate the enormous impact that total joint replacement in general, and total knee replacement in particular, has had on the treatment of arthritis. The significant pain relief afforded by this surgery is accepted universally. The popular press and other media, along with the recent phenomenon of direct-to-patient marketing, drive the increasing demand for this elective procedure. As the demand continues to increase, implant companies, hospitals, and orthopaedic surgeons have attempted to differentiate themselves from one another by advertising in classical media and on the Internet with claims of better pain relief and function, and quicker return to full activity. Even with the numbers of joint replacements performed each year amounting to the hundreds of thousands, there is still no consensus on the best implant design or even the design features that are most important for knee replacement. Clinical research has been hampered by a lack of validated and generally accepted measures of outcome. Patient-reported outcome measures have varied considerably based on age and expected activity level, with older patients being much more satisfied than younger ones in spite of lower scores in validated assessment methods. The disconnect seems to be based on patients' expectation and ability to resume activities (i.e. function) rather than the surgeons' expectation of pain relief or the measurements of alignment, stability and range of motion (ROM) that dominate most knee evaluation scores. Tools such as gait and motion analysis exist but are expensive and time consuming to implement. Thus total knee designs continue to vary considerably, and information is inadequate to make scientifically valid judgments about which knee replacement design is better. Design changes have been abundant but, if outcome measures determine success, few advances really have been made based on design characteristics alone. A significant problem arises when assessing the results of knee arthroplasty through a patient's subjective response. Patients usually present to the surgeon with a primary problem of pain and secondary decrease in function because of © 2008, Woodhead Publishing Limited

482

Joint replacement technology

the pain. When the pain is relieved, a deficit in function may be acceptable in the patient's mind as simply the best that can be achieved with a total knee arthroplasty. In the interest of affirming a good clinical outcome, most surgeons do not probe deeply to discover functional deficits. A patient with one type of knee replacement prosthesis has no way of knowing if a different prosthesis might have afforded more (or less) functional improvement.

20.2

Kinematics and knee joint prosthesis design

20.2.1 Kinematics It stands to reason that a knee joint prosthesis should mimic the motion of the normal knee as closely as possible. Significant technological advancements have been made in the understanding of the kinematics of the knee since the 1960s when total knee joint prostheses initially were designed. However, these advances appear to be simply a rediscovery of what already was known from the anatomic dissections and measurements made in the 19th century.1 In the late 1960s, the prevailing model of knee joint motion was the four-barlink model, more correctly termed a crossed-four-bar-link. This model was conceived from measurements made on some of the first radiographs of the knee after the discovery of the Roentgen ray in the early 1900s (Fig. 20.1). The evolution and varying success of clinical results of total knee arthroplasty cannot be understood without first understanding this model, which implant designers were using when designing total knee replacement prostheses. The work of O'Connor and Goodfellow is central to the early development of knee replacement and the design implication of the four-bar-link.2 Rollback is one property of the four-bar-link that has been essential to implant design. The segments of the four-bar-link (femur, tibia, and two cruciate ligaments) were modeled to be rigid links. In such a linkage the point of zero velocity (i.e. the point around which one segment is moving relative to another) is always at the crossing point of these two links. Because the crossing point moves as the segments move, the point of zero velocity must move. When applied to the knee this means the femur must move backward on the tibia as the knee joint flexes. It also could be said that the tibia moves forward as the knee flexes, but the classic description was made as though the tibia were fixed and the femur moved around it. This phenomenon was termed rollback, which really is a misnomer because the condition of rolling requires that the point of zero velocity be at the contact point between the two surfaces, in this case the femur and tibia. If the femur were to be strictly rolling, the entire femur would roll off the back of the tibia before the range of knee motion was complete. The early total knee implants had a significant failure rate. These failed implants attempted to preserve the cruciate ligaments while still having conforming femoral±tibial articulations. The failure analysis pointed at least in part © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

483

20.1 The crossed four-bar-link model of the knee consisting of the anterior cruciate ligament AB, the posterior cruciate ligament CD, the femur CB and the tibia AD. If the links are modeled as rigid, then the point of zero velocity must always be at the crossing point of the ligaments I. As a consequence of this crossed, rigid four-bar-link model the femur must move backward on the tibia with flexion (rollback) and forward on the tibia with extension (initially published JJ O'Connor, TL Shercliff and E Biden et al. The geometry of the knee in the saggital plane. J Engng Med Proc Inst Mech Eng Part H 1989; 203: 223± 223, with permission).

to `kinematic conflict' (Fig. 20.2).3±7 It was thought that the articular conformity impeded the knee's obligatory antero-posterior motion and that the components loosened because of the stresses incurred by this conflict. Thus, virtually every total knee joint design stressed its respect for the four-bar-link and the lack of surface conformity. More recent studies of knee kinematics have shown conclusively that the knee joint does not function as a crossed four-bar-link.8±11 No obligatory posterior motion of the femoral condyle occurs with flexion. Instead, in open-chain experiments the medial side of the knee remains in a nearly constant position with the spherical portion of the medial condyle articulating in a shallow socket, which is deepened by the meniscus, on the tibial surface. The constraint of the medial © 2008, Woodhead Publishing Limited

484

Joint replacement technology

20.2 A failed Geomedic prosthesis (a) radiographic exam and (b) removed components. This prosthesis had highly conforming articular surfaces while preserving both cruciate ligaments. The crossed-four-bar-link model held that the necessity of rollback, which would be constrained by conforming surfaces, was a `kinematic conflict' and the reason for failure (courtesy of Herbert Kaufer, MD). © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

485

ball (femoral condyle) and socket (concave medial tibia) is not rigid. The ball can move out of the socket in response to external loads on the tibia or femur as is seen in closed-chain models. Contraction of the musculature around the knee stabilizes the ball near to the socket and increased tone of the muscles around the knee decreases the compliance, especially on the medial side.12 The lateral side articulation, which is like two discs rather than a ball-in-socket, is much more free to move in an antero-posterior direction around the more stable medial side. This allows the tibia to rotate internally or externally depending on the functional demands placed on the lower extremity. Generally the tibia rotates internally with flexion, moving the femur toward the posterior surface. (This sometimes is incorrectly termed lateral rollback.) Volitionally or with external loads the tibia can be externally rotated so posterior motion of the condyle is not obligatory. In full flexion the lateral condyle does roll (with the point of zero velocity located at the contact point between the femur and the posterior tibia) around the back of the tibia, pushing it forward and allowing full flexion.13 As the knee extends, a phenomenon known as `screw home' occurs, which is thought to be an obligatory internal rotation at terminal extension (between approximately 20ë and 0ë flexion). As the knee flexes the reverse motion is thought to be obligatory as flexion is initiated. This phenomenon is not well understood and may be absent from the knee in certain conditions. A recent study of the knee suggests that screw home is dependent on the distal `extension' radius.14 As this portion of the femur engages the extension facet on the tibia, the femur is lifted away from the tibia, and the lateral side, because it has little constraint, follows, making the screw home motion. Anatomic measurements of the femur have suggested that the extension facet is not present in all knees and this could explain why the phenomenon has not been found in all knees. Thus, the knee joint can be thought of as a universal joint that can transmit torque in multiple positions by remaining stable on the medial side and allowing rotation of approximately 20ë by change in the anterior-to-posterior contact position on the lateral side (Fig. 20.3). The knee joint moves around three orthogonal axes fixed to the femur: the flexion±extension axis that passes through the center of the medial and lateral condyles, the internal±external rotation axis that passes through the center of the condyle to the depth of the shallow tibial socket in a superior-to-inferior direction, and an axis of adduction that passes orthogonally front-to-back through the medial condyle (Fig. 20.4). The femur with the fixed axes can translate out of the resting position (i.e., the ball in the depth of the socket on the medial side). These combined motions can account for all kinematics that have been found at the knee. The cruciate ligaments have been studied extensively and the classic fourbar-link model has been refined to allow elastic elements and differential forcedeformation so as not to be at odds with the `more stable medial' model of knee joint motion.15 The ligaments likely act first as stretch receptors sensing translation of the medial side out of the `ball-in-socket' kinematics and abduc© 2008, Woodhead Publishing Limited

486

Joint replacement technology

20.3 The knee joint can be thought of as a universal joint that can transmit torque in multiple positions by remaining stable on the medial side while allowing axial rotation of the tibia relative to the femur by anterior±posterior motion on the lateral side. In this picture the climber has flexed and internally rotated her right knee to gain a foothold on the rock in preparation for generation of torque through quadriceps contraction to move up the rock face. While the knee generally rotates internally with flexion (sometimes incorrectly called lateral rollback) this can be overcome with muscular effort as is seen in this figure. (Unlisted Images, Inc.)

tion±adduction and antero-posterior positioning on the lateral side.12 These changes in the four-bar-link model are leading to a convergence of opinion about knee joint motion, namely that the knee acts differently in its kinematics on the medial and lateral sides. This convergence has led to changes in knee joint prosthesis design and represents an advance in joint replacement technology. The evolution of knee joint design in the light of these changing views of function is described in the next sections of this chapter.

20.2.2 Total knee joint prosthesis design In the popular patient-oriented media and on many Internet websites, the terms `total knee' and `partial knee' have been used to identify tricompartmental and © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

487

20.4 A simplified model of knee kinematics suggests that the knee joint can be modeled as being more stable on the medial side than the lateral. In this model, the knee joint moves around three orthogonal axes fixed to the femur, intersecting in the middle of the medial condyle. The flexion axis passes roughly through the middle of the medial and lateral condyles, the rotation axis passes vertically through the femoral condyle into the depth of the shallow socket on the medial tibia, and the ab-adduction axis passes from anterior to posterior through the femur roughly parallel to the plateau of the tibia (courtesy of R Obert and P Stemniski).

unicompartmental prostheses. In the strict sense of the nomenclature, a total joint prosthesis is one that provides resurfacing to both sides of an articulation (e.g., femur±tibia or femur±patella at the knee) without regard to how many compartments are involved. The term total joint prosthesis is used in this nomenclature in contrast to `hemi-prosthesis' which applies resurfacing to only one-half of an articulation. Notwithstanding the popular press, the strictly correct nomenclature will be used in this chapter: tricompartmental, bicompartmental, and unicompartmental total knee prostheses. Little significant technological advancement has been made in joint replacement prostheses since their inception. A few problems have been discovered and remedied, but the remedies have done little to the actual kinematic design of the knee implant. For example, problems with modularity led to improvements in the rigidity of tibial polyethylene components assembled intra-operatively. This has helped to lessen wear debris generated on the backside of tibial inserts.16±18 Problems of polyethylene quality were addressed through changes in manufacturing and sterilization.19,20 On the articulating surfaces, however, the © 2008, Woodhead Publishing Limited

488

Joint replacement technology

changes have been subtle and often done for marketing advantage only. It is useful to enumerate the types of prostheses available and to view which data are available regarding patient outcomes. Considering the large number of prostheses implanted, published studies represent only a very small fraction of the total. The studies often have been done by the design surgeons and bias in the publications certainly exists because of the tendency to publish only good results. There is an obvious need for randomized controlled trials of total knee prostheses but these trials will probably be hampered by the lack of validated instruments of assessment that could further address the function of the arthroplasty from the patient's perspective. Tricompartmental total knee arthroplasty Early in the development of total knee prostheses, three types of prosthetic designs (usually tricompartmental) were available based on the intrinsic stability of the prosthesis and the amount of stability (e.g. ligaments) left in the intact knee. These were the hinge, semi-constrained, and unconstrained prostheses. The hinge total knee prosthesis eliminated the ligaments at the knee and substituted mechanical stability through prosthetic constraint. The semiconstrained prosthesis left some knee ligaments intact and had correspondingly less intrinsic stability, consistent with the prevailing idea that kinematic conflict should be avoided. Unconstrained prostheses had the least intrinsic stability and left the most stability in the ligamentous structures of the knee. The single-axis hinge prosthesis initially had some excellent results, but had a high incidence of loosening and mechanical failure (Fig. 20.5).21±24 Recognition that the knee functions as a three-dimensional structure and not a simple hinge led to development of the tri-axial hinge. In cases where no stability exists within the knee itself (e.g., tumor resection or revision of failed arthroplasty), triaxial hinges have had satisfactory results,25 but concerns remain that the constraint in the articulation will lead to implant loosening and mechanical failure. The descriptors of `semi-constrained' and `unconstrained' implants have been abandoned in favor of descriptions of prostheses based on whether the posterior cruciate ligament is still present after surgery. (Some attempts have been made to retain both the anterior and posterior cruciate ligaments, but these arthroplasty designs represent a small percentage of total knee replacements performed and will not be mentioned further.) A posterior cruciate retaining (commonly abbreviated PCR) prosthesis (Fig. 20.6) removes the anterior cruciate ligament but retains all or part of the posterior cruciate ligament. In some cases the insertion of the cruciate into the top of the tibia is left intact and in other cases only the part of the ligament that inserts onto the posterior tibia is retained. As a class, however, posterior cruciate © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

489

20.5 The fixed axis hinge, although it represented a step forward from what was previously available and had some good results, had a high incidence of loosening and mechanical failure (courtesy of Andrew G. Urquhart, MD).

20.6 The posterior cruciate retaining (PCR) types of implants remove the anterior cruciate ligament but retain all or most of the posterior cruciate ligament. As a class, PCR implants lack conformity between the femur and tibia and rely on the posterior cruciate strength and appropriate balancing to maintain stability. In full extension there is a slight upward slope to the tibial component to limit anterior translation of the femur. In flexion the posterior cruciate ligament must be competent and appropriately tensioned to prevent anterior translation of the femur (courtesy of Biomet). © 2008, Woodhead Publishing Limited

490

Joint replacement technology

retaining prostheses lack conformity between the articulating surfaces of the femur and the tibia to accommodate what is still termed rollback, which is in theory caused by the increasing tightness of the posterior cruciate ligament as flexion increases.26 The posterior cruciate ligament is `balanced' during the procedure so that it is not too tight in extension and will not tighten excessively as the knee flexes, to prevent the femur from sliding forward on the tibia. Conceptually, appropriate balancing of the posterior cruciate ligament allows the ligament to pull the femur backwards relative to the tibia with increasing flexion and provides for antero-posterior stability in extension. In its pure state, this type of prosthesis has little antero-posterior stability from the contribution of the components, but depends entirely on the posterior cruciate ligament for that stability. Many designs of posterior cruciate retaining prostheses have been marketed with slightly different conformity between the femur and tibia and thus differing amounts of stability. It is therefore quite difficult to generalize the results for the use of these components without knowing the exact design features of each one. A posterior cruciate sacrificing implant requires removal of the cruciate ligament. (Although `sacrificing' is the common parlance, it might better be termed `removing'.) In most, but not all cases, prosthetic constraint is added to the implant to substitute for the posterior cruciate ligament. The use of the posterior cruciate substituting (abbreviated PS, for posterior stabilized) implant involves the removal of the posterior cruciate ligament and a constraint imposed by the prosthetic design substitutes for the expected function of that ligament. The most common design of PS knee has a cam on the femur and a post on the tibia (Fig. 20.7). The condyles of the femoral component can have variable constraint based on conformity with the tibial component in the medial and lateral compartments. The cam and post can have different shapes, sizes, and positions. This significant variability means that, just as for posterior cruciate retaining implants, it is likely not appropriate to generalize results from one prosthetic type to another. A more constrained type of posterior stabilized implant, often referred to as a constrained condylar design (abbreviated CCK) (Fig. 20.8) has been designed for revision circumstances and for cases in which the ligaments are not lax enough to consider a hinge prosthesis, but are more lax than would be appropriate for the use of a standard posterior stabilized implant. Implant selection is a judgment on the part of the surgeon, and no agreement has been reached on the exact indications. These CCK implants are quite similar, though certainly not identical, across various implant designs, so results could be generalized cautiously. A number of systems are available that provide high congruency between the tibial insert and a standard non-posterior stabilized femoral component (Fig. 20.9). These components were introduced to address the issue of instability in posterior cruciate retaining arthroplasties in which the posterior cruciate liga© 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

491

20.7 Two different versions of the posterior stabilized (PS) total knee prosthesis. Differences in the position of the spine on the tibia, cam on the femur and condylar conformity between the femur and tibia can markedly change the kinematics and stability of this type of prosthesis (courtesy of Biomet and Zimmer).

ment failed and patients presented with symptoms of instability. These types of components have been used by some as a posterior-cruciate-stabilized prosthesis, calling into question the belief that a posterior-stabilized knee needs a cam and post mechanism. Reports of the use of these implants have not been widespread but probably could be generalized from one system to another in terms of the analysis of outcome (M Goossens and C van der Straeten, personal communication).28 The mobile bearing tricompartmental total knee prosthesis deserves mention as a special case of prosthesis design (Fig. 20.10). The meniscal bearing type of implant was designed for both anterior and posterior cruciate preservation while the rotating platform version has been available for either posterior cruciate preservation or substitution. Cruciate substituting rotating platform knees have either high conformity (deep dish) or the more standard cam post design.3,29,30 The meniscal bearing design for tricompartmental replacement proved somewhat unreliable and was difficult to insert correctly, so that a single bearing encompassing both the medial and lateral side was designed with the © 2008, Woodhead Publishing Limited

492

Joint replacement technology

20.8 The most constrained condylar design (often designated the constrained condylar knee or CCK) is shown. This design is most frequently used for revisions and difficult primary cases because of the need to improve fixation with stems and kinematics and stability with metal augments (courtesy of Wright Medical Technology). © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

493

20.9 The prosthesis shown has nearly full conformity between the femoral component and the tibial thus providing the maximum resistance to anterior± posterior (AP) displacement. The conformity has not led to increased loosening or wear at seven years of follow-up (M Goossens and C van der Straeten, personal communication) and a patient satisfaction study has found increased patient satisfaction with prostheses that have greater AP stability27 (courtesy of Wright Medical Technology).

platform rotating around a central post (either on the polyethylene component inserted into the tibial metal back or with a metal detail engaging the tibial mobile insert). One iteration of this type of implant allowed the tibia to move freely in any direction on the tibial tray and has been referred to as a hockey puck insert (Fig. 20.10).31,32 The mobile-bearing implants were presented to the joint replacement community as having an advantage in providing more normal kinematics, and improved function with this type of implant was expected.2 Patent protection limited the number of available designs, but when the patent expired, many different designs were marketed around the world. The concept of the rotating platform mobile bearing implant is similar enough for the different individual devices that, for the most part, results can be generalized for the class in terms of outcome assessment. Unicompartmental total knee arthroplasty When total knee replacement was introduced, surgeons recognized that joint degeneration frequently was severe in one compartment only (usually the medial compartment but sometimes the lateral), so that the other compartments could look nearly normal (Fig. 20.11). Studies of cartilage indicated that what looked normal was normal in terms of morphology and biochemistry,33 so the idea of replacing just one compartment with a femoral and a tibial component gained © 2008, Woodhead Publishing Limited

494

Joint replacement technology

20.10 The tricompartmental, meniscal bearing implant (a, b) proved difficult to align correctly and had problems with dislodgment of the mobile polyethylene pieces. The one piece `rotating platform' type components (c, d) have proved much more robust and available from almost every manufacturer (images courtesy of DePuy Orthopaedics, Inc (a, b) and Wright Medical Technology (c, d).

© 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

495

20.11 In the belief that a total joint replacement should only remove and resurface damaged cartilage, unicompartmental prostheses have been developed and used for many years starting in the 1970s. Early enthusiasm was dampened by a significant failure rate. With the knowledge that irradiated polyethylene could have been the problem leading to many failures, there has been a resurgence in interest in unicompartmental total knee replacement (photos courtesy of David Waxman, MD).

favor. In these cases the anterior and posterior cruciate ligaments most often were intact to provide stability, and it was reasoned that very little conformity in the articulation was necessary. In fact, respecting the need to avoid kinematic conflict, conformity was thought to be a severe drawback to function and stability of a unicompartmental joint replacement design. Most components were so similar as to be nearly indistinguishable (with the exception of the meniscal bearing component described next), so results can be more appropriately generalized. After some initial enthusiasm, unicompartmental replacement was almost abandoned in the 1980s because of severe prosthetic wear, loosening, and failure. With the understanding of the deleterious effects of irradiation of polyethylene and manufacturing improvements in this material, unicompartmental arthroplasty had a renaissance of sorts in the late 1990s. Results are variable among the reports available, possibly because of operative technique and patient selection, but the design of the implants is similar enough that it is difficult to relate failures to design.34±36 The Oxford unicompartmental knee deserves its own discussion (Fig. 20.12). Based on the four-bar-link model, this prosthesis was designed with a polyethylene component that could move anteriorly and posteriorly on a polished tibial surface while the femoral component moved against the proximal surface. © 2008, Woodhead Publishing Limited

496

Joint replacement technology

20.12 The Oxford mobile bearing unicompartmental prosthesis has been very successful over a long clinical history (photo courtesy of Biomet).

This allowed a large articular area of contact so that the femur could seek its own position on the tibia. Excellent long-term results have been reported by the designers and by others.37 Attempts to reproduce their results in other designs have not been as successful. Less successful clinical results with some meniscal bearing prostheses such as the LCS meniscal bearing (not rotating platform) knee have been reported at orthopedics meetings but have not been followed by publications in the orthopedic literature. This demonstrates the problem of publication bias against negative results. The `bi-uni' is used by some surgeons for replacement of both medial and lateral compartments of the knee without replacement of the patellofemoral joint.38 A recently introduced prosthesis (the `Deuce') is designed to replace the medial and patellofemoral compartment of the joint (Fig. 20.13).39 No long-term reports are available to attest to the success of this technique, but early reports presented at orthopedic meetings indicate success. © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

497

20.13 Again attempting to minimize the amount of potentially healthy bone and cartilage removed, the `Deuce' prosthesis has been introduced to resurface the patellofemoral and medial tibiofemoral joints. Early reports are encouraging but the mid- and long-term success are yet to be reported (photo courtesy of Gerry Engh, MD).

Designs based on `new' kinematics The kinematic work that has been referenced previously in this chapter regarding the `new' kinematics of the knee (i.e., not the four-bar-link model) has led to several designs that attempt to keep the medial side of the knee from moving in the antero-posterior direction, while allowing the lateral side of the knee to move freely from anterior to posterior (Fig. 20.14).40±42 This relatively small change in articular geometry may significantly impact patient function and thus would be considered a technological advance. One of these designs uses ball-insocket geometry on the medial side, one a rotating platform that is more stable on the medial side than on the lateral, and one a combination of cams, posts, and articular conformity to guide motion. Ten-year results may soon be available for two of these designs, whereas the third design has only recently been introduced. One design of total knee has extrapolated from kinematic studies of arthritic knees to conclude that the lateral side of the joint should remain more stable than the medial. Clinical results for this design have not yet been published (WA Hodge et al., www.encoremed.com/products/knee/3Dknee/index.htm). © 2008, Woodhead Publishing Limited

498

Joint replacement technology

20.14 Three recent total knee prosthesis designs that attempt to reproduce the more stable medial side while allowing more freedom on the lateral. The Advance Medial Pivot (a) (Wright Medical Technology) uses a medial ballin-socket geometry, the Journey (b) (Smith & Nephew) asymmetric condyles and both anterior and posterior cams to interact with the tibial post and the MBK (c) (Zimmer) a mobile bearing with an interlocking post on the tibia (photos courtesy of Wright Medical Technology, Smith & Nephew and Zimmer).

These designs based on `new' kinematics stress antero-posterior stability rather than respect for `rollback' and avoidance of `kinematic conflict'. One very interesting report compared patients' preference in the circumstance in which both knees had a total knee arthroplasty but these were of different design.27 These results suggest that patients prefer a knee that has better antero© 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

499

posterior stability to one that has less stability.27 This supports a widely held, but as yet unproved, clinical impression that function is directly related to stability. High flexion designs At the outset of total knee replacement in the 1960s, most surgeons were satisfied with knee motion from near full extension to about 90ë flexion. In those days the operation was applied primarily to older, debilitated patients in the Western hemisphere who found this amount of flexion adequate for their activities of daily living. As total knee joint replacement has become available around the globe it has become clear that 90ë flexion is not adequate to perform culturally important activities. Now in the West, the generation of patients to whom joint replacement is being offered is similarly satisfied only with obtaining the most flexion possible.43 Knee designs have been introduced to achieve higher flexion, and these changes represent a technological advancement. The designs have increased the amount of flexion possible before impingement of the tibial component against the femur by making the posterior condyle of the prosthesis thicker, to increase offset in this region of flexion and by using various articular surface changes in geometry to prevent posterior impingement.44±46 The early clinical results have shown modest improvements in average flexion with some cases showing impressive full flexion.47 Long-term clinical results are not yet available for these patients and the effect of the articular geometry changes on long-term wear and wear-induced loosening or osteolysis is not known. Patellofemoral joint arthroplasty In tricompartmental total knee arthroplasty the anterior flange of the femoral component provides at least a hemiarthroplasty for the patellofemoral joint. In the United States most patients have a polyethylene component placed on the posterior surface of the patella, providing a total resurfacing of that joint, whereas in Europe most patients do not have a patellar replacement. The comparable results between the European and US series leave the impression that the hemiarthroplasty functions as well as a total arthroplasty at this joint. Most US surgeons share the opinion that patients with a hemiarthroplasty report more anterior knee pain.48,49 European and some North American surgeons have detailed a surgical technique that, in their experience, minimizes the potential for anterior knee pain. They describe a patella-friendly groove on the anterior flange of the tricompartmental component to assure pain-free function with a hemiarthroplasty at the patellofemoral joint.50±52 Patients who have arthrosis of the patellofemoral joint only are candidates for a unicompartmental hemi- or total arthroplasty of that joint. Prostheses for the total and hemi-replacement of this joint were available in the past, but few were © 2008, Woodhead Publishing Limited

500

Joint replacement technology

implanted compared with the more standard tricompartmental joint because of problems with the prostheses. Interest has been renewed with the advent of several prostheses and the emergence of opinion leaders who speak about the various devices, but no consensus of opinion has been reached about the use of these devices.53 Is there a clinical difference? Based on the studies that have been published in peer-reviewed literature, it is difficult to ascertain a clinical difference based on the type of prosthesis implanted. A great deal of information about the outcome of prostheses is reported at various orthopedic meetings, and in such circumstances opinions often are stronger than the data that are reviewed and published. Manufacturers want to differentiate their products from others but also want to make only small changes because `minor modifications' of an implant design do not require the additional regulatory burden of proving safety and efficacy. An example of such a change is the `gender-specific' implant introduced and marketed by one manufacturing company. While the human knee comes in an infinite number of sizes and configurations, total knee replacement components must be finite. It has long been recognized that some femora have a smaller medial-to-lateral measurement compared with a larger anterior-to-posterior measurement.54±56 When examining data on sizing from studies published in the literature it is reasonable to conclude that the dimensions of the femur vary with the stature of an individual but, because women are generally smaller than men, it is a potentially spurious conclusion that femoral shape varies with sex.57,58 Size `mismatch' has been recognized from the beginnings of total knee arthroplasty, occasionally leading to a component that is slightly too large (overhang) or slightly too small (underhang). In spite of recognition of size mismatch, no reports have been published about any clinical problem caused by this slight size mismatch. The Q angle is a measurement made from the anterior superior spine of the pelvis to the center of the patella and then to the tibial tubercle.59 The name Q angle is meant to imply the angle of pull of the quadriceps muscle (Fig. 20.15). However, no part of the quadriceps originates from the anterior superior spine from which the measurement is taken. In fact, the vector origin of the quadriceps is suggested to be the mid-portion of the femoral neck.58,60±62 Data on the size of the femur and the Q angle have been used in a redesign of total knee femoral implants. The observation that patients with a wider pelvis compared with their height (i.e. women) have a larger Q angle has been used to suggest that a change in the angle on the trochlear groove of the anterior flange of the femoral component and a change in the size of the implants is necessary to accommodate differences. The differences in size are minimal and the Q-angle measurement spurious. Nonetheless, in a brilliant marketing move, but one of © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

501

20.15 The `Q angle' has been defined as the angle made between a line drawn from the anterior superior iliac spine (ASIS) to the midpoint of the patella and a line from the midpoint of the patella to the tibial tubercle. This line is frequently referenced as the line of action of the quadriceps muscle. However, no part of the quadriceps muscle originates from the ASIS. Observations of anatomic dissections make it reasonable to assume that the vector origin of the quadriceps is, instead, the middle part of the femoral neck. There is little difference between males and females in the angles formed by the shaft and hip-to-knee line.62 This leads to the logical conclusion that the vector of quadriceps pull is very nearly the same in males and females (modified from Grants Atlas, used with permission).

dubious technological advancement, changes in sizing and trochlear groove geometry have been incorporated into a gender-specific design and the components marketed directly to women as the first knee made for a woman. Such is the problem with identifying and documenting true technological advances in a maze of marketing hype. The only means of demonstrating superiority of one prosthesis over another ± the prospective, randomized trial ± is difficult and expensive to conduct and patients are loth to volunteer for such a © 2008, Woodhead Publishing Limited

502

Joint replacement technology

trial when they are bombarded with information that they must have the latest and greatest technological advance. The future looks bright for incremental changes in implant design, but dismal for the science to prove that the changes actually make a difference in clinical outcome.

20.3

Analysis of the kinematics of total joint prostheses

Prior to marketing, total knee prostheses are subjected to some testing to determine the motion characteristics of the components. Implantation in cadaver knees with attempt to mimic open chain or closed chain exercise has been reported.8,63,64 Robotics have been used to determine compliance of joints at various ranges of motion and kinematics have been inferred from these data.65,66 It has been difficult to observe the motion of the knee joint in vivo. The technique of video motion analysis has been used in living patients to evaluate the knee in use.67±70 These types of study offer experimental methods to design articular surfaces that can mimic the normal knee, based on the assumption that a knee that moves more like the normal will function more like the normal. The problems of hype and the difficulty in performing scientifically valid clinical trials make proof of this assertion difficult. The advancing basic techniques of kinematic analysis represent a significant technological advancement for total knee replacement. As discussed earlier in this chapter, these types of analysis demonstrate that in the cadaver and in the living human, the rigid four-bar-link model of knee kinematics is not correct. In most cases the medial side of the knee remains stable while the lateral side rotates around it.71 A different type of kinematics termed lateral pivot has been seen in some patients and has been suggested to be part of the pathology of osteoarthritis at the knee. Analysis of cruciate ligament deficiency and reconstruction has been conducted and the results have added to our knowledge of the pathology of these conditions (WA Hodge et al., www.encoremed.com/ products/knee/3Dknee/index.htm).

20.3.1 Paradoxical motion The technique of video motion analysis, applied to knees with a total knee joint prosthesis implanted, has demonstrated a phenomenon that has been termed paradoxical motion (Fig. 20.16).69 Because of the prevailing four-bar-link model of knee kinematics, it was expected that, when the posterior cruciate ligament was intact and the total knee was apparently functioning well, the femur would move backwards on the tibia with flexion. Video motion analysis demonstrated anterior motion of the femur with flexion. This motion is paradoxical to the fourbar-link model theory and hence the name. Many prostheses have been evaluated using video motion analysis and it can be concluded that paradoxical motion is a function of the stability imparted by © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

503

20.16 A technique of matching an image projected by fluoroscopy with threedimensional models of total knee components has allowed detection of threedimensional movements of implants in use. When this technique was applied to posterior cruciate retaining knees investigators expected to see `rollback'. Instead, the femur slid forward on the tibia. Because they expected rollback and the femur went the opposite direction this type of motion was termed `paradoxical' motion.

the articular conformity of a total knee prosthesis. The finding of paradoxical motion also has implications for the function of total joint prostheses. There are significant posterior to anteriorly directed loads on the knee during function. If the knee cannot be stabilized to prevent this type of motion, patients change the way in which they move to accommodate the antero-posterior instability. This type of instability can help explain the quadriceps-avoidance gait, difficulty ascending and descending stairs, and anterior knee pain, all of which have been seen in total knee joint replacement patients.41

20.3.2 Range of motion Total knee joint designs have been introduced to gain greater amounts of flexion. Many design features of these implants came from kinematic analysis of normal and replaced knees. Although the current designs have not achieved full flexion, this is an area of significant research and many iterations of the highflexion designs can be expected in the near future. © 2008, Woodhead Publishing Limited

504

Joint replacement technology

20.3.3 Surgical technique Taken in its simplest form, the technique of total knee arthroplasty has not changed in many years. Five bone cuts are made on the femur ± distal, anterior, posterior, and two chamfer cuts. One cut is made on the tibia and one on the patella, if it is resurfaced. Contracted ligaments either are released or removed and the components are inserted. Each system has alignment jigs, cutting blocks, and trial devices that often add up to an almost overwhelming number of instruments to be sterilized and made ready for a case. But the essence of the operation remains six or seven cuts and an implantation.

20.3.4 Alignment The originators of total knee replacement each had their specific technique for aligning the replaced joint (Fig. 20.17). The mechanical axis is a line drawn on an AP radiograph (representing the coronal plane) of the knee from the center of the femoral head to the middle of the tibial plateau and on to the middle of the ankle. Some time in the late 1970s or early 1980s the concept of the relationship of the mechanical axis to the replaced knee appeared. Without a specific published experimental study, the concept that total knee arthroplasty was to be perpendicular to the mechanical axis became dogma.72 This relationship was called classical alignment by a group that favored attempting to reproduce the position of the joint surfaces relative to the overall limb ± so-called anatomic alignment.73 These paradigms for alignment co-existed during the 1980s, a period during which excessive wear of irradiated polyethylene and patellar subluxation problems were significant.74±76 The author and a group of coworkers produced a total knee prosthesis that has nearly total conformity on the medial side to mimic normal kinematics and stability and to reduce the contact stress on polyethylene.41 It became clear that the usual paradigms for alignment led to some significant potential problems with kinematics, so we sought to test these assumptions. In the laboratory, motion of the cadaver knee was induced in an open-chain model and the axis of flexion of the knee was calculated. (Recall that the `new' kinematics of the knee suggests that there is a single axis of flexion, rotation and adduction, all of which meet in the center of the medial side.) The flexion axis calculated is not perpendicular to the mechanical axis but rather a plane perpendicular to the flexion axis through the trochlear groove of the femur passing through the neck of the femur proximally and through the tibial tubercle and the neck of the talus distally. This line is parallel to a line from the middle of the femoral head to the middle of the medial condyle where the flexion, rotation, and adduction axes meet. We have thus suggested that kinematically correct alignment would place the flexion axis of the replaced knee in the same position of the flexion axis of the normal knee. For a component that has the lateral and medial condyles that © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

505

20.17 (a) Alignment in total knee arthroplasty is known to be critical to longterm success. Failure to achieve proper alignment has been associated with wear and loosening of components. The mechanical axis of the lower extremity is a line drawn on a long anterior-posterior radiograph of the limb from the middle of the femoral head through the middle of the knee to the middle of the ankle. A properly aligned limb will show a nearly straight mechanical axis. (b) The flexion axis of the knee approximates the `clinical' epicondylar axis (see Fig. 20.4). If the axis of flexion in the replaced knee is to approximate that of the normal knee, all resections on the bones should be parallel to the flexion axis. (c) The `anatomic' alignment scheme seeks to remove equal amounts of bone from the medial and lateral aspects of the femur and tibia. Using this alignment the mechanical axis can be restored but the resections are not equidistant from the flexion axis. To restore the flexion axis a component with differing radii on the medial and lateral sides would be necessary. © 2008, Woodhead Publishing Limited

506

Joint replacement technology

20.17 continued (d) In the `classical' alignment technique, all resections are intended to be perpendicular to the mechanical axis. The technique also makes resections that are not parallel to the flexion axis. (e) A line drawn perpendicular to the flexion axis through the trochlear groove (functional axis) progresses proximally through the femoral neck lateral to the head and medial to the greater trochanter. (This position is approximately where the vector origin of the quadriceps has been reasoned to be. See Fig. 20.15.) (f) Resections made perpendicular to the functional axis are parallel to the flexion axis. Components with equal radii on the medial and lateral condyles will still reproduce the correct flexion axis (courtesy of R Obert and P Stemniski). © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

507

are symmetric (i.e. of equal proportions) this would mean that all cuts on the bone ± femur and tibia ± would be parallel to the flexion axis. The effect of this paradigm of alignment is to place the total knee prosthesis at a slight (about 3ë) varus inclination instead of perpendicular to the mechanical axis. This change in alignment paradigm is not generally accepted, but has worked very well in our hands and represents a potential advancement in the surgical technique of arthroplasty.77,78

20.3.5 Navigation Considerable time and money have been devoted to the interest in using computer-assisted instrumentation to decrease surgical variability in total knee replacement. Large, expensive, and difficult to implement systems have given way to smaller units that can be implemented in the operating theater (Fig. 20.18). The systems work by locating the three-dimensional position of instruments in the calibrated space in the operating room (OR) using infrared, lightemitting diodes, or magnetic tracking. With registration of the patient's anatomy in the same space, the position of bone cuts can be judged by the computer and the surgeon can be advised as to whether a given cut is appropriate for that knee joint. To date most orthopaedists have not adopted this extensive and evolving technology. Those who have adopted it can demonstrate fewer large errors (i.e. fewer outliers) when using these systems, but as yet computer-assisted surgery has not led to an improvement in the clinical results for patients.79±83

20.3.6 Minimally invasive surgery Arguably the most significant change in technique for total knee arthroplasty has been the trend toward smaller incisions and potentially less damaging approaches to the surgical procedure. Using the example of arthroscopic techniques for intraarticular surgery, which markedly decreased the morbidity of knee procedures, surgeons began to perform knee replacement surgery with smaller exposures. The techniques first were applied to unicompartmental surgery but rapidly expanded to tricompartmental surgery. No one technique has emerged as superior to another, and indeed it is difficult to separate the effects of postoperative pain control, rehabilitation, patient selection, and education from the effects of the lesser surgical procedure. The elements of the minimally invasive revolution (as it is referred to in Internet hype) are skin incision, approach to opening the retinaculum and quadriceps, handling of the patella to allow exposure of the articular surfaces, use of special smaller instruments for aligning bone cuts, the pre- and postoperative treatment of pain and inflammation, and the rehabilitation protocols. The area is changing rapidly and any written text is likely to be outdated before it is printed. It is sufficient to state that the trend to lesser surgery in total knee arthroplasty is a technological advancement that has, taken as a © 2008, Woodhead Publishing Limited

508

Joint replacement technology

20.18 Computer navigation uses three-dimensional positioning of instruments referenced to a patient's skeleton to position resections. These systems have demonstrated their precision although generally with some modest increase in operating time and expense (photo courtesy of BrainLAB AG).

whole, improved the early clinical outcome of total knee arthroplasty. Little data have been published to suggest that long-term results are better or worse than more conventional operations, but early work has shown a slightly higher number of incorrectly placed components when a limited approach is used.84±86 Clearly there is great potential to combine minimally invasive techniques with navigation to allow proper placement of components with the least surgical trauma.

20.3.7 Pain control and limited postoperative range of motion One of the biggest problems facing the joint replacement surgeon is the patient with what appears to be aggressive scar formation often coupled with severe pain that limits motion and compromises the postoperative result. In spite of the considerable clinical problem this causes, little published work exists to establish the cause or the treatment of this vexing problem. This condition has been © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

509

associated with chronic depression which is itself associated with an increased perception of pain.87±89 Patients with fibromyalgia, who frequently have a problem with chronic depression, seem to be more at risk to experience this problem. Preoperative cyclooxygenase inhibitors (COX-2), gabypentum and amytriptylene, indwelling peripheral nerve catheters, regional anesthesia, and injections of various combinations of drugs around the knee at wound closure have all been tried to reduce pain in the postoperative period in an attempt to decrease the occurrence of this problem. The changes in anesthesia that have come with minimally invasive surgery to control this problem represent a technological advance, but the problem remains. This is an area ripe for investigation and one where a technological advance would be welcome.

20.4

Summary

Total knee joint arthroplasty continues to be a remarkably successful operation for pain relief, and has been well documented in the literature although the level of evidence is less than ideal. From the data that have been published in peerreviewed journals it is not possible to determine that one type of replacement prosthesis is better than another. Older and less active patients seem to function well with a total knee replacement, whereas the younger, more active patients frequently identify functional deficits. Tricompartmental total knees with either a total or hemi-replacement of the patellofemoral joint, unicompartmental tibiofemoral total joints, bi-unicompartmental and patellofemoral total joints continue to be used, but the application of these various prostheses based on peer-reviewed publications is not clear. Direct-to-consumer marketing and the widely available hype from the media and Internet websites make it difficult to do the randomized controlled trials that would clearly establish one type of prosthesis or procedure to be superior.90 A change in ideas about the kinematics and alignment of the knee has led to a change in knee joint prosthesis design that may address at least some of these functional issues. The classical, rigid four-bar-link kinematic model has been supplanted by a model that is stable on the medial side and mobile on the lateral side to form a universal joint. Prostheses that conform to this model are now approaching 10-year patient outcomes with reported (but not published) satisfactory results. One patient preference study suggests that prostheses that are stable to antero-posterior motion are preferred by patients. Computer-assisted surgery may allow more accurate placement of implants, but data are inconclusive as of this writing. Minimally invasive surgery has introduced a number of new technologies, especially in the areas of pain control and postoperative rehabilitation, that may lead to better and quicker recovery from arthroplasty procedures. Patients' response to surgery varies greatly and investigations to try to understand this variability and treat it are an area of emerging technology with great potential. © 2008, Woodhead Publishing Limited

510

20.5

Joint replacement technology

References

1. Pinskerova V, Maquet P, Freeman MA. Writings on the knee between 1836 and 1917. J Bone Joint Surg Br 2000; 82(8): 1100±1102. 2. Goodfellow J, O'Connor J. The mechanics of the knee and prosthesis design. J Bone Joint Surg Br 1978; 60: 358±369. 3. Lotke PA, Ecker ML, McCloskey J, Steinberg ME. Early experience with total knee arthroplasty. JAMA 1976; 236(21): 2403±2406. 4. Matthews LS, Goldstein SA, Kaufer H. Experiences with three distinct types of total knee joint arthroplasty. Clin Orthop Relat Res 1985; 192: 97±107. 5. Murray DG. Total knee arthroplasty. Clin Orthop Relat Res 1985; 192: 59±68. 6. Townley CO. The anatomic total knee resurfacing arthroplasty. Clin Orthop Relat Res 1985; 192: 82±96. 7. Vince KG. Principles of condylar knee arthroplasty: issues evolving. Instr Course Lect 1993; 42: 315. 8. Blaha JD, Mancinelli CA, Simons WH, Kish VL, Thyagarajan G. Kinematics of the human knee using an open chain cadaver model. Clin Orthop Relat Res 2003; 410: 25±34. 9. Hill PF, Vedi V, Williams A, Iwaki H, Pinskerova V, Freeman MA. Tibiofemoral movement 2: the loaded and unloaded living knee studied by MRI. J Bone Joint Surg Br 2000; 82(8): 1196±1198. 10. Iwaki H, Pinskerova V, Freeman MA. Tibiofemoral movement 1: the shapes and relative movements of the femur and tibia in the unloaded cadaver knee. J Bone Joint Surg Br 2000; 82(8): 1189±1195. 11. Komistek RD, Dennis DA, Mafouz M. In vivo fluoroscopic analysis of the normal human knee. Clin Orthop Relat Res 2003; 410: 69. 12. Blaha JD, Wojtys EM. Motion and Stability of the Normal Knee. In: Scott RD, editor. Surgery of the Knee, 4th edn. Philadelphia: Elsevier; 2007, 227±239. 13. Nakagawa S, Kadoya Y, Todo S, Kobayashi A, Sakamoto H, Freeman MA et al. Tibiofemoral movement 3: full flexion in the living knee studied by MRI. J Bone Joint Surg Br 2000; 82(8): 1199±1200. 14. Blaha JD, DeBoer D, Barnes CL et al. Kinematics of the cadaver knee under openchain and closed-chain conditions. In preparation, 2008. 15. O'Connor J, Shercliff T, Fitzpatrick D et al. Mechanics of the knee. In: Daniel D, Akeson W, O'Connor J, editors. Knee Ligaments: Structure, Function, Injury and Repair. New York: Raven Press; 1990, 201±237. 16. Engh GA, Koralewicz LM, Pereles TR. Clinical results of modular polyethylene insert exchange with retention of total knee arthroplasty components. J Bone Joint Surg Am 2000; 82(4): 516±523. 17. Engh GA, Lounici S, Rao AR, Collier MB. In vivo deterioration of tibial baseplate locking mechanisms in contemporary modular total knee components. J Bone Joint Surg Am 2001; 83-A(11): 1660±1665. 18. Parks NL, Engh GA, Topoleski LD, Emperado J. The Coventry Award. Modular tibial insert micromotion. A concern with contemporary knee implants. Clin Orthop Relat Res 1998; 356: 10±15. 19. Tanner MG, Whiteside LA, White SE. Effect of polyethylene quality on wear in total knee arthroplasty. Clin Orthop Relat Res 1995; 317: 83±88. 20. White SE, Paxson RD, Tanner MG, Whiteside LA. Effects of sterilization on wear in total knee arthroplasty. Clin Orthop Relat Res 1996; 331: 164±171. 21. Gibbs AN, Green GA, Taylor JG. A comparison of the Freeman-Swanson (ICLH) © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

22. 23. 24. 25. 26. 27. 28. 29. 30. 31.

32. 33.

34. 35. 36. 37. 38. 39.

511

and Walldius prostheses in total knee replacement. J Bone Joint Surg Br 1979; 61B(3): 358±361. Jones EC, Insall JN, Inglis AE, Ranawat CS. GUEPAR knee arthroplasty results and late complications. Clin Orthop Relat Res 1979; 140: 145±152. leNobel J, Patterson FP. Guepar total knee prosthesis. Experience at the Vancouver General Hospital. J Bone Joint Surg Br 1981; 63-B(2): 257±260. Woodburn KR, Braidwood AS. GUEPAR total knee prosthesis. Results and outcome of seven years' use in a district general hospital. J R Coll Surg Edinb 1990; 35(1): 56±60. Jones RE, Barrack RL, Skedros J. Modular, mobile-bearing hinge total knee arthroplasty. Clin Orthop Relat Res 2001; 392: 306±314. Rosenberg AG, Knapke DM. Posterior cruciate-retaining total knee arthroplasty. In: Scott WN, editor. Surgery of the Knee, 4th edn. Philadelphia: Churchill Livingstone; 2006, 1522±1530. Pritchett JW. Patient preferences in knee prostheses. J Bone Joint Surg Br 2004; 86(7): 979±982. Hofmann AA, Tkach TK, Evanich CJ, Camargo MP. Posterior stabilization in total knee arthroplasty with use of an ultracongruent polyethylene insert. J Arthroplasty 2000; 15(5): 576±583. Buechel FF, Sr. Long-term followup after mobile-bearing total knee replacement. Clin Orthop Relat Res 2002; 404: 40±50. Sorrells RB, Voorhorst PE, Murphy JA, Bauschka MP, Greenwald AS. Uncemented rotating-platform total knee replacement: a five to twelve-year follow-up study. J Bone Joint Surg Am 2004; 86-A(10): 2156±2162. Aigner C, Windhager R, Pechmann M, Rehak P, Engeleke K. The influence of an anterior-posterior gliding mobile bearing on range of motion after total knee arthroplasty. A prospective, randomized, double-blinded study. J Bone Joint Surg Am 2004; 86-A(10): 2257±2262. Kim YH, Kim JS. Comparison of anterior-posterior-glide and rotating-platform low contact stress mobile-bearing total knee arthroplasties. J Bone Joint Surg Am 2004; 86-A (6): 1239±1247. Brocklehurst R, Bayliss MT, Maroudas A, Coysh HL, Freeman MA, Revell PA et al. The composition of normal and osteoarthritic articular cartilage from human knee joints. With special reference to unicompartmental replacement and osteotomy of the knee. J Bone Joint Surg Am 1984; 66(1): 95±106. Emerson RH, Jr., Higgins LL. A comparison of highly instrumented and minimally instrumented unicompartmental knee prostheses. Clin Orthop Relat Res 2004; 428: 153±157. Marmor L. Unicompartmental knee arthroplasty. Ten- to 13-year follow-up study. Clin Orthop Relat Res 1988; 226: 14±20. Scott RD, Cobb AG, McQueary FG, Thornhill TS. Unicompartmental knee arthroplasty. Eight- to 12-year follow-up evaluation with survivorship analysis. Clin Orthop Relat Res 1991; 271: 96±100. Murray DW. Mobile bearing unicompartmental knee replacement. Orthopedics 2007; 30(9): 768±769. Banks SA, Fregly BJ, Boniforti F, Reinschmidt C, Romagnoli S. Comparing in vivo kinematics of unicondylar and bi-unicondylar knee replacements. Knee Surg Sports Traumatol Arthrosc 2005; 13(7): 551±556. Engh GH. Presented at Current Concepts in Joint Replacement, Orlando, Florida, Dec. 2006.

© 2008, Woodhead Publishing Limited

512

Joint replacement technology

40. Aglietti P, Baldini A, Buzzi R, Lup D, De LL. Comparison of mobile-bearing and fixed-bearing total knee arthroplasty: a prospective randomized study. J Arthroplasty 2005; 20(2): 145±153. 41. Blaha JD. The rationale for a total knee implant that confers anteroposterior stability throughout range of motion. J Arthroplasty 2004; 19(4 Suppl 1): 22±26. 42. Victor J. Presented at International Society of Arthroposcopy, Knee Surgery and Orthopaedic Sports Medicine, Florence, Italy, May 2007. 43. Weiss JM, Noble PC, Conditt MA, Kohl HW, Roberts S, Cook KF et al. What functional activities are important to patients with knee replacements? Clin Orthop Relat Res 2002; 404: 172±188. 44. Bellemans J, Banks S, Victor J, Vandenneucker H, Moemans A. Fluoroscopic analysis of the kinematics of deep flexion in total knee arthroplasty. Influence of posterior condylar offset. J Bone Joint Surg Br 2002; 84(1): 50±53. 45. Han HS, Kang SB, Yoon KS. High incidence of loosening of the femoral component in legacy posterior stabilised-flex total knee replacement. J Bone Joint Surg Br 2007; 89(11): 1457±1461. 46. Li G, Most E, Sultan PG, Schule S, Zayontz S, Park SE et al. Knee kinematics with a high-flexion posterior stabilized total knee prosthesis: an in vitro robotic experimental investigation. J Bone Joint Surg Am 2004; 86-A(8): 1721±1729. 47. Kim YH, Sohn KS, Kim JS. Range of motion of standard and high-flexion posterior stabilized total knee prostheses. A prospective, randomized study. J Bone Joint Surg Am 2005; 87(7): 1470±1475. 48. Burnett RS, Boone JL, McCarthy KP, Rosenzweig S, Barrack RL. A prospective randomized clinical trial of patellar resurfacing and nonresurfacing in bilateral TKA. Clin Orthop Relat Res 2007; 464: 65±72. 49. Forster MC. Patellar resurfacing in total knee arthroplasty for osteoarthritis: a systematic review. Knee 2004; 11(6): 427±430. 50. Burnett RS, Haydon CM, Rorabeck CH, Bourne RB. Patella resurfacing versus nonresurfacing in total knee arthroplasty: results of a randomized controlled clinical trial at a minimum of 10 years' followup. Clin Orthop Relat Res 2004; 428: 12±25. 51. Mayman D, Bourne RB, Rorabeck CH, Vaz M, Kramer J. Resurfacing versus not resurfacing the patella in total knee arthroplasty: 8- to 10-year results. J Arthroplasty 2003; 18(5): 541±545. 52. Whiteside LA, Nakamura T. Effect of femoral component design on unresurfaced patellas in knee arthroplasty. Clin Orthop Relat Res 2003; 410: 189±198. 53. Argenson JN, Flecher X, Parratte S, Aubaniac JM. Patellofemoral arthroplasty: an update. Clin Orthop Relat Res 2005; 440: 50±53. 54. Hitt K, Shurman JR, Greene K, McCarthy J, Moskal J, Hoeman T et al. Anthropometric measurements of the human knee: correlation to the sizing of current knee arthroplasty systems. J Bone Joint Surg Am 2003; 85-A Suppl 4: 115±122. 55. Mensch JS, Amstutz HC. Knee morphology as a guide to knee replacement. Clin Orthop Relat Res 1975; 112: 231±241. 56. Seedhom D, Longton E, Wright V, Dowson D. Dimensions of the knee. Radiographic and autopsy study of sizes required by a knee prosthesis. Ann Rheum Dis 1972; 31: 54±58. 57. Rankin EA, Bostrom M, Hozack W, Jacobs JJ, McCarthy JC, O'Connor MI et al. Gender-specific knee replacements: a technology overview. J Am Acad Orthop Surg 2008; 16: 63±67. 58. Blaha JD, Mancinelli CA. Using the Transepicondylar Axes to Define the Sagittal Morphology of the Distal Femur. American Academy of Orthopaedic Surgeons, 2002. © 2008, Woodhead Publishing Limited

Knee replacement: clinical perspectives

513

59. Insall J, Falvo KA, Wise DW. Chondromalacia patellae. A prospective study. J Bone Joint Surg Am 1976; 58(1): 1±8. 60. Eckhoff D, Hogan C, DiMatteo L, Robinson M, Bach J. Difference between the epicondylar and cylindrical axis of the knee. Clin Orthop Relat Res 2007; 461: 238± 244. 61. Eckhoff DG, Dwyer TF, Bach JM, Spitzer VM, Reinig KD. Three-dimensional morphology of the distal part of the femur viewed in virtual reality. J Bone Joint Surg Am 2001; 83-A Suppl 2(Pt 1): 43±50. 62. Yoshioka Y, Siu D, Cooke TD. The anatomy and functional axes of the femur. J Bone Joint Surg Am 1987; 69(6): 873±880. 63. Blankevoort L, Huiskes R, de LA. Helical axes of passive knee joint motions. J Biomech 1990; 23(12): 1219±1229. 64. Churchill DL, Incavo SJ, Johnson CC, Beynnon BD. The transepicondylar axis approximates the optimal flexion axis of the knee. Clin Orthop Relat Res 1998; 356: 111±118. 65. Most E. Development of a 6-DOF Robotic Test System for Studying the Biomechanics of Total Knee Replacement. Cambridge, MA: Massachusetts Institute of Technology; 2000. 66. Most E, Li G, Sultan PG, Park SE, Rubash HE. Kinematic analysis of conventional and high-flexion cruciate-retaining total knee arthroplasties: an in vitro investigation. J Arthroplasty 2005; 20(4): 529±535. 67. Banks SA, Hodge WA. Accurate 3D measurement of dynamic knee replacement motions using single-plane fluoroscopy: validation and in vivo application. Trans Am Soc Biomech 1995; 19: 163. 68. Banks SA, Markovich GD, Hodge WA. In vivo kinematics of cruciate-retaining and -substituting knee arthroplasties. J Arthroplasty 1997; 12(3): 297±304. 69. Dennis DA, Komistek RD, Hoff WA, Gabriel SM. In vivo knee kinematics derived using an inverse perspective technique. Clin Orthop Relat Res 1996; 331: 107±117. 70. Stiehl JB, Dennis DA, Komistek RD, Keblish PA. In vivo kinematic comparison of posterior cruciate ligament retention or sacrifice with a mobile bearing total knee arthroplasty. Am J Knee Surg 2000; 13(1): 13±18. 71. Freeman MA, Pinskerova V. The movement of the normal tibio-femoral joint. J Biomech 2005; 38(2): 197±208. 72. Rohr W. Primary total knee arthroplasty. In: Chapman M, Madison M, editors. Operative Orthopaedics. Philadelphia: JB Lippincott; 1998, 715±726. 73. Krackow K. The Technique of Total Knee Arthroplasty. St. Louis: Mosby; 1990. 74. Cadambi A, Engh GA, Dwyer KA, Vinh TN. Osteolysis of the distal femur after total knee arthroplasty. J Arthroplasty 1994; 9(6): 579±594. 75. Kobayashi A, Freeman MA, Bonfield W, Kadoya Y, Yamac T, Al-Saffar N et al. Number of polyethylene particles and osteolysis in total joint replacements. A quantitative study using a tissue-digestion method. J Bone Joint Surg Br 1997; 79(5): 844±848. 76. Revell PA, Weightman B, Freeman MA, Roberts BV. The production and biology of polyethylene wear debris. Arch Orthop Trauma Surg 1978; 91(3): 167±181. 77. Simons WS, Blaha JD. Defining the Functional Axis for Total Knee Alignment. American Academy of Orthopaedic Surgeons, 2002. 78. Blaha JD, KS Int Meetings, Jackson Hole, WY, 2006. 79. Bathis H, Perlick L, Tingart M, Luring C, Zurakowski D, Grifka J. Alignment in total knee arthroplasty. A comparison of computer-assisted surgery with the conventional technique. J Bone Joint Surg Br 2004; 86(5): 682±687. © 2008, Woodhead Publishing Limited

514

Joint replacement technology

80. Bolognesi M, Hofmann A. Computer navigation versus standard instrumentation for TKA: a single-surgeon experience. Clin Orthop Relat Res 2005; 440: 162±169. 81. Decking R, Markmann Y, Fuchs J, Puhl W, Scharf HP. Leg axis after computernavigated total knee arthroplasty: a prospective randomized trial comparing computer-navigated and manual implantation. J Arthroplasty 2005; 20(3): 282±288. 82. Krackow KA, Bayers-Thering M, Phillips MJ, Bayers-Thering M, Mihalko WM. A new technique for determining proper mechanical axis alignment during total knee arthroplasty: progress toward computer-assisted TKA. Orthopedics 1999; 22(7): 698±702. 83. Stulberg SD, Loan P, Sarin V. Computer-assisted navigation in total knee replacement: results of an initial experience in thirty-five patients. J Bone Joint Surg Am 2002; 84-A Suppl 2: 90±98. 84. Dalury DF, Dennis DA. Mini-incision total knee arthroplasty can increase risk of component malalignment. Clin Orthop Relat Res 2005; 440: 77±81. 85. Scuderi GR, Tenholder M, Capeci C. Surgical approaches in mini-incision total knee arthroplasty. Clin Orthop Relat Res 2004; 428: 61±67. 86. Tria AJ, Jr, Coon TM. Minimal incision total knee arthroplasty: early experience. Clin Orthop Relat Res 2003; 416: 185±190. 87. Brander V, Gondek S, Martin E, Stulberg SD. Pain and depression influence outcome 5 years after knee replacement surgery. Clin Orthop Relat Res 2007; 464: 21±26. 88. Dorr LD, Chao L. The emotional state of the patient after total hip and knee arthroplasty. Clin Orthop Relat Res 2007; 463: 7±12. 89. Eckman E, Koman A. Acute pain following musculoskeletal injury and orthopaedic surgery. J Bone Joint Surg Am 2004; 86: 1316±1327. 90. Bozic KJ, Smith AR, Hariri S, Adeoye S, Gourville J, Maloney WJ et al. The 2007 ABJS Marshall Urist Award: The impact of direct-to-consumer advertising in orthopaedics. Clin Orthop Relat Res 2007; 458: 202±219.

© 2008, Woodhead Publishing Limited

21

Intervertebral disc joint replacement technology

N H A L L A B , Rush University Medical Center, USA

21.1

Introduction

The use of spinal fixation devices is steadily increasing; from 1993 to 2003 the number of cervical and lumber fusions increased 111% to roughly 105 fusions per 100 000 people in the United States or approximately 305 000 fusions per year (2003) (Cowan et al., 2006). This represents a large market force driving the development of improved intervertebral disc replacement technologies, such as total disc arthroplasties (or intervertebral disc replacements (IDR)). The ultimate goal of IDR technology is to replace fusion as a treatment modality by both pain elimination and structural restoration with preservation of spinal mobility (flexion±extension, lateral-bending and rotation). The first step in this pursuit is to exceed the clinical outcome and performance of spinal fusions in those people where operative intervention (e.g. fusion) is clearly indicated. Current designs of IDRs incorporate all three major classifications of materials that have been used in artificial joints for over half a century, i.e. metals, polymers, and ceramics. Of the three basic types of materials, metals have provided an unbeatable combination of high strength with ductility, fracture toughness, hardness, corrosion resistance, cost-effective formability and biocompatibility, such that they have been the central material used in total joint replacements (TJR) over the past 50 years. The other two types of materials (polymers and ceramics) are currently used as components in contemporary TJR devices. Polymers generally have provided low-friction surfaces for articulating bearing surfaces with some degree of shock absorption in joint replacement applications. Disc replacement technologies are broadening the field of polymers employed for use in loadbearing orthopedic applications, the majority of which are employed for shock absorption and structural support (e.g., poly(etheretherketone) (PEEK), polyurethane and silicone). Ceramics have only recently been used in the United States on a wide scale in joint replacement applications in general, and the high wear resistance of ceramic surfaces. Although ideal for articulating surfaces, ceramics have not been used in disc arthroplasty devices to date. © 2008, Woodhead Publishing Limited

516

Joint replacement technology

Current intervertebral disc replacement technologies generally fall into one of three design categories: (1) total disc replacement (TDR) or total disc arthroplasty (TDA), (2) dynamic-posterior stabilization devices and (3) disc/ nucleus replacement/augmentation. Other spine implants designed to eliminate fusions include: · · · · · · ·

annular and intra-discal treatment for degenerative disc disease; facet arthroplasty; lumbar posterior motion sparing technology; cervical disc arthroplasty; lumbar disc arthroplasty; lumbar nucleus pulposus replacement; and biologic disc replacement.

Central to the success of any implant is its biocompatibility, i.e. host response to implant materials and debris is minimized. Biocompatibility is best defined as the ability of a biomaterial to demonstrate host and material response appropriate to its intended application, in this case as a disc replacement (Bogduk, 1999). The determination of biocompatibility has been dominated historically by the characterization of candidate materials based upon the observation of adverse host responses during long-term use. There are likely be no exceptions to this with the varied designs of current IDRs, and clinical follow-up studies will be required to determine which designs/materials will be the most beneficial over the long term. The following chapter will cover modern designs of intervertebral disc replacements focusing on materials, designs, and remaining clinical concerns.

21.1.1 Criteria for intervertebral disc replacement design There are a number of general criteria, that are currently used in IDR design. Motion (kinematic) preservation The biomaterials and the device design should preserve natural joint kinematics. Above all, intervertebral disc prostheses should, to the greatest degree possible, replicate the normal range of motion in all planes and reproduce physiologic stiffness in all planes of motion including axial compression. There are four basic types of movements that occur between articular surfaces in the normal movement of most joints: spinning (rotation), sliding, rolling and axial compression. Many di-arthrodial joint replacements featuring a variable instantaneous axis of rotation (IAR) (e.g., disc, knee, and ankle joint arthroplasties) use a freefloating intermediate component typically comprising ultra-high molecular weight polyethylene (UHMWPE), a so-called mobile bearing (Fig. 21.1). Theoretically, a mobile bearing mimics a variable instantaneous axis of rotation © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

517

21.1 FernstrÎm Ball implants (stainless steel metal spheres), among the first disc arthroplasties to be implanted (courtesy of Dr J Hall).

which more closely approximates the movement of an intact intervertebral disc. However, mobile bearing prostheses raise concerns of wear and particle induced osteolysis. Spacing preservation Biomaterials within a TDR device should maintain the proper intervertebral spacing and provide stability which may include a need to restore collapsed disc space height. All current intervertebral disc prostheses are contained within the disc space; therefore, disc implants must accommodate variations in patient size, level, and height. Biomechanical loading preservation The artificial disc must accurately transmit physiologic stress. The reduction of stress concentration is an important attribute in IDR, to minimize implant movement (e.g. subluxation) and adverse loading of important spinal elements such as adjacent vertebra, facets and ligaments. White and Panjabi, Bogduk and others regard the degree of lumbar lordosis (and consequently any vertebral body forward displacement) as the source of the major forces exerted on the facet (zygapophysial or L-z) joint (Bogduk, 1999). In other words, the forces © 2008, Woodhead Publishing Limited

518

Joint replacement technology

causing anterior subluxation (front compressive displacement) or spinal translation are resisted by the facet joints. Currently all presently marketed lumbar artificial discs, except pure nucleus replacements, have to be implanted from anterior (front) by a retroperitoneal or transperitoneal approach (through the abdominal cavity) (Bertagnoli and Kumar, 2002; BuÈttner-Janz et al., 2002) where the anterior longitudinal ligament is severed, the annulus is removed and the posterior longitudinal ligament is stretched or sometimes resected. This makes the adjacent lumbar vertebral bodies of an affected motion segment prone to forward displacement. This is important because it increases the risk of `overloading' zygapophysial joint integrity as well as affecting the durability and wear resistance of any artificial disc arthroplasty. Performance life: 50-year life expectancy The biomaterials used for an artificial disc must endure repeated high cyclic loading over a time frame of 50 years. This design criterion is conservatively approximated using the average time that a 35-year-old patient will require a well-functioning lumbar disc replacement. With an average of 2 million strides per year and 125 000 significant bends, the number of cycles over a 50-year lumbar implant life expectancy would likely be over 100 million. Not included in this estimation is the subtle articulation and cyclic loading of the disc joints when taking approximately 6 million breaths annually. Thus the fatigue strength and wear resistance of chosen materials must be optimized to minimize wear debris. Biocompatibility TDR replacement materials must be non-toxic and last (i.e. they must be both corrosion and wear resistant). All implants degrade to some degree in vivo and degradation products must, to the greatest degree possible, not incite any excessive pathologic tissue responses (i.e. inflammatory response or neurological damage). Fixation TDR materials must be able to maintain both immediate and long-term bone fixation. Immediate fixation may be accomplished with screws, staples, porous or macrotexture surfaces, or macro-mechanical geometries that are incorporated into the implant design. Failsafe TDR implants must be designed and constructed such that failure of any individual component may not result in life-threatening catastrophe. For IDRs, in the event of an accident or unexpected loading neural, vascular, and spinal tissues must be protected to the greatest degree possible. © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

519

Revisable Revision-friendly materials in TDR are required to minimize the impact of eventual revision or arthrodesis. In addition to the risk of mid- or long-term biomaterial failure inherent to new TJA designs are other factors such as patient activity, iatrogenic-surgeon error, patient non-compliance, or infection that may cause implant failure. In disc arthroplasty, revisions may be life threatening, thus proactive steps to design ease of revision should not be underestimated as an important implant characteristic. Monitorable TDR materials must be able to be monitored in vivo (e.g. radiographically tracked). When the performance of the spine implant is intimately linked to the level of osteolysis and bone loss, as is typical of total joint arthroplasty (TJA), a case for more aggressive follow-up and the necessity of implant tracking in vivo can be made to prevent substantial bone loss and the need for an allograft.

21.2

Orthopedic materials and methodology available for use in intervertebral disc replacements

Metals and polymers remain the central material components of all total joint arthroplasties including modern disc arthroplasty designs (Black, 1996; Park, 1984; Silver and Christiansen, 1999). Polymers provide low-friction surfaces for articulating bearings and shock absorption. Metals provide appropriate material properties such as high strength, ductility, fracture toughness, hardness, corrosion resistance, formability and biocompatibility necessary for use in loadbearing roles required in fracture fixation and TJA. Implant alloys were originally developed for maritime and aviation uses where mechanical properties such as high strength and corrosion resistance are paramount. There are three principal metal alloys used in orthopedics and particularly in total joint replacement: (1) titanium-based alloys, (2) cobalt-based alloys, and (3) stainless steel alloys (Table 12.1) (Black, 1992). Alloy-specific differences in strength, ductility, and hardness generally determine which of these three alloys is used for a particular application or implant component (Table 21.2). However, it is the high corrosion resistance of all three alloys, more than anything, which has lead to their widespread use as load-bearing implant materials (Black, 1988).

21.2.1 Stainless steel alloys: formability and ductility The form of stainless steel most commonly used in orthopedic practice is designated 316LV (American Society for Testing and Materials F138, ASTM F138). Molybdenum is added to enhance the corrosion resistance of the grain © 2008, Woodhead Publishing Limited

Table 21.1 Approximate weight percent of different elements within popular orthopedic alloys (Black and Hastings, 1998) Alloy Stainless steel ASTM F138 CoCrMo alloys ASTM F75 ASTM F90 Ti alloys cp-Ti ± (ASTM F67) Ti±6Al±4V (ASTM F136)

Ni

N

10±15.5 <0.5 <2.0 9±11

* *

* *

* *

Co

Cr

Ti

Mo

Al

Fe

Mn

Cu

W

C

Si

V

*

17±19

*

2±4

*

61-68

*

<0.5

<2.0

<0.06

<1.0

*

61±66 27±30 46±51 19±20

* *

4.5±7.0 *

* *

<1.5 <3.0

<1.0 <2.5

* *

* <0.35 14±16 <0.15

<1.0 <1.0

* *

99 89±91

* *

* *

* *

* *

* 3.5±4.5

* *

* *

* 0.2±0.5 5.5±6.5 *

* Indicates less than 0.05%. Note: Alloy compositions are standardized by the American Society forTesting and Materials (ASTM vol.13.01). © 2008, Woodhead Publishing Limited

WPTF3007

* *

<0.1 <0.08

Table 21.2 Mechanical properties of dominant orthopedic biomaterials (Black and Hastings, 1998) Orthopedic biomaterial

ASTM designation

Elastic modulus (Young's modulus) (GPa)

Yield strength (elastic limit) (MPa)

Ultimate strength

0.5±1.3

20±30

0.0018± 0.009

28±40

30±40t 30±40c 28±40t 33±50c

ASTM F138

190

792

930t

241±820

130±180

43±45

ASTM F75 ASTM F90 ASTM F562 ASTM 1537

210±253 210 200±230 200±300

448±841 448±1606 300±2000 960

655±1277t 1896t 800±2068t 1,300t

207±950 586±1220 340±520 200±300

300±400 300±400 8±50 (RC) 41 (RC)

4±14 10±22 10-40 20

ASTM F67 ASTM 136

110 116

485 897±1034

760t 965±1103t

300 620±689

120±200 310

14±18 8

Polymers UHMWPE Polyurethane Metals Stainless steel Co±Cr alloys

Ti alloys cp-Ti Ti±6Al±4V

c = compression. t = tension. * = No current ASTM standard. RC = Rockwell Hardness Scale. HVN= Vickers hardness number, kg/mm. © 2008, Woodhead Publishing Limited

WPTF3007

(MPa)

Fatigue strength (endurance limit) (MPa)

Hardness

Elongation at fracture

(HVN)

(%)

13±20

60±90 (MPa) 50±120 (MPa)

130±500

21±30

600±720

522

Joint replacement technology

boundaries, while chromium dissipated evenly within the microstructure allows the formation of chromium oxide (Cr2O3) on the surface of the metal. Stainless steels, although less corrosion resistant than Ti or Co±Cr±Mo alloys, possess the greatest ductility of the aforementioned implant metals indicated quantitatively by a three-fold greater `percentage of elongation at fracture' when compared with other implant metals (Table 21.2). This aspect, as well as low cost and ease of formability, presumably led to its use in the first metal-on-metal TDAs and has allowed it to remain as a viable material for TJA.

21.2.2 Cobalt±chromium alloys: wear and corrosion resistance The two basic constituents of all Co-Cr alloys are Co (approximately 65%) and Cr (approximately 35%). Molybdenum is added to decrease the grain size and thus improve mechanical properties, where cobalt±chromium±molybdenum (Co±Cr±Mo) is designated ASTM F-75 and F-76 (Table 21.1). The dominant implant alloy used for total joint components is CoCrMo (ASTM F-75). Although Co±Cr±Mo alloys are among the strongest, hardest and most fatigue resistant of the metals used for joint replacement components, care must be taken to maintain these properties because the use of finishing treatments can also function to reduce these same properties (Table 21.2) (McKellop et al., 1992).

21.2.3 Titanium alloys: corrosion resistance/biocompatibility Two post-World War II alloys, commercially pure titanium (cp-Ti) and titanium alloy (6% aluminum and 4% vanadium, Ti±6Al±4V), remain the two dominant titanium alloys used in implants. Cp-Ti (ASTM F67) is 98±99.6% pure titanium. Titanium alloys are particularly good implant materials because of their high corrosion resistance compared with stainless steel and Co±Cr±Mo alloys. Generally, Ti±6Al±4V (ASTM F-136) is used for joint replacement components because of its superior mechanical properties in comparison to cp-Ti (Table 21.2). A stable passive oxide film (primarily of TiO2) protects both Ti±6Al±4V and cp-Ti components. Thus, Ti alloys are seldom used as materials where resistance to wear is the primary concern (Black, 1988, 1992, 1996; Breme and Biehl, 2001; Mears, 1979; Park, 1984; Silver and Christiansen, 1999). Today's trend towards titanium implants is largely due to clinical history of enhanced biocompatibility in TJA applications. Additionally titanium produces fewer artifacts in magnetic resonance imaging (MRI) and computed tomography (CT) imaging (Bertagnoli and Kumar, 2002).

21.2.4 Ultra-high molecular weight polyethylene (UHMWPE) UHMWPE artificial discs require high wear resistance as well as resistance to cold flow. Although loading is similar to other joint prostheses (e.g., total knee © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

523

arthroplasty, TKA) the relative motion of bearing surfaces is approximately an order of magnitude less. Past reports (Lee and Pienkowski, 1998) indicate that creep occurs in the amorphous regions of the UHMWPE (rather than in the crystalline regions), and recovery occurs within the same regions. It is therefore advantageous to retain a high degree of crystallinity in that of non-highly crosslinked UHMWPEs. This is because PE crystallinity can be reduced by ~15% (Wright and Goodman, 2000) by crosslinking. Consequently, extra crosslinking may not be an advantage for UHMWPE when used as bearing material in artificial discs if creep is a major or limiting concern. The cold flow limit of UHMWPE is 22 N/mm. This is low enough such that an inadequate thickness of UHMWPE components can adversely affect mechanical properties. Total knee replacements have demonstrated that polyethylene components should have a minimal thickness of 6±8 mm. This advice is listed in a European Standard (EN 12564, 19998) and a US Food and Drug Administration (FDA) Guidance document (Draft Guidance for the Preparation of Premarket Notifications, for cemented and semi-constrained total knee prosthesis, 1993). Therefore new artificial discs are careful to maintain component thicknesses in compliance with these standards. This is particularly important when small implants or non-parallel components for large lordotic angles are used where the load limit per mm2 may be exceeded, resulting in creep.

21.2.5 Ceramics The theoretical advantage of hard-on-hard TJA articulating surfaces is low wear. Alumina (Al2O3) and zirconia (ZrO2) ceramics have been used in total hip arthroplasty (THA) for the past 30 years. Ceramics are chemically stable because of their ionic bonds and thus have relatively biocompatible wear debris that cannot be easily degraded into chemically bioreactive forms. Their very low wear rates combined with steadily decreasing rates of fracture (now estimated to occur in 1 in 2000 cases over 10 years) have resulted in the growing popularity of all ceramic THA bearings. The questionable suitability and difficulty in adapting ceramic components for use in TDA during hard-on-hard surface articulation are reflected by the lack of ceramic components in any of the current TDA designs in clinical use. Generally, ceramic application to disc prosthesis is currently restricted to osteophilic surface coatings with hydroxyapatite (HA).

21.2.6 Surface coatings Surface designs and coatings to direct and improve short-term bone ingrowth and fixation are gaining popularity. These surface enhancements include more traditional surfaces such as roughened titanium, porous coatings made of cobalt chromium or titanium beads, titanium wire mesh, plasma-sprayed titanium, and © 2008, Woodhead Publishing Limited

524

Joint replacement technology

newer bioactive non-metallic materials such as hydroxyapatite or other calcium phosphate (CaP) compositions (Berry, 2000; Zeggel, 2000). Over the long term, the potential debonding of the ceramic surface coating from the underlying metal implant may prove to be a significant weakness of these coatings because of the resulting increase in particulate debris. In addition third body wear due to surface coating particles may cause accelerated wear of the bearing surface. The presence of HA in the bearing surfaces of total hip replacements has been reported, as has a case of severe osteolysis caused by third-body wear in an implant with HA coating. The extent to which this thirdbody wear issue will become a clinical problem is unknown (Bostrom and Lane, 2000). The efficacy of using biologic coatings on TDR has not been well established in joint arthroplasty and remains questionable given the critical importance of nearby neural structures.

21.3

Early intervertebral disk replacement designs

The first disc arthroplasties to be used were metal spheres placed in between vertebrae with the disc annulus in place. Cobalt±alloy spheres were implanted as early as 1957 (Harmon, 1963). A decade later in a larger series of patients, stainless steel metal spheres termed FernstroÈm balls, after Dr FernstroÈm (Fig. 21.1) were also being implanted (155 in 103 patients starting in 1969) (FernstroÈm, 1966; McKenzie Alvin, 1995). In 1964 Hjalmar Reitz reported a total of 75 cervical disc arthroplasties with the spherical prosthesis on 32 patients and 19 lumbar discs in 12 patients, for `discogenic backache and sciatica' (Reitz and Mauritius, 1964). Others clinicians around the same time tried using polymethyl methacrylate for the replacement of spinal intervertebral discs with little success (Hamby and Glaser, 1959). From a modern perspective the first disc replacements, FernstroÈm steel balls, seem like an overly simplistic solution for disc replacement, but they were intended to: (1) maintain disc height and (2) preserve lateral bending, flexion± extension, and axial rotation, which they managed to accomplish over for a short time (<5 years). However, clinical performance diminished over time due in large part to subsidence of bone over the ball as a result of the stress concentrations placed over the high pin-point stresses that developed at the relatively small initial contact areas. This small `footprint' and resulting high initial stresses resulted in subsidence and loss of disc height in 88% of cases at 4±7 years follow-up. The modern era of disc arthroplasty began in 1982, when Schellnack and BuÈttner-Janz implanted the first functional artificial intervertebral disc, the SB ChariteÂ, at the Charite hospital in Berlin (Fig. 21.2) (BuÈttner-Janz, 1992). This was based on John Charnley's low-friction arthroplasty principle, which by the 1980s had proved successful in total hip replacement for more than 20 years. The Charite design consisted of an UHMWPE sliding core, which articulated © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

525

21.2 The first functional artificial disc was used as early as 1982, when Schellnack and BÏttner-Janz developed the SB Charite¨ artificial disc which consisted of an UHMWPE sliding core, which articulated unconstrained between two highly polished stainless steel metal endplates (courtesy of Depuy Spine).

unconstrained between two highly polished metal end plates, reproducing the movement of the nucleus within its annular containment. The end plates of this model were made of 1 mm thick steel (URX2CrNiMoN 18.12) and the sliding core was produced from UHMWPE (ChirulenTM). The artificial disc had multiple teeth like projections for fixation to the vertebral end plates (Fig. 21.2). Problems reported with the first series of SB Charite prostheses were attributed to the stress concentrations of the small end plate size, causing subsidence into the bone (BuÈttner-Janz et al., 2002). Additionally, there was a lack of appropriate instrumentation needed to help achieve near physiologic disk range of motion by careful placement along the midline on the frontal plane and 2 mm posterior on the lateral plane. However, even with these shortcomings, these early designs have also demonstrated long-term success as evidenced by first person to receive this early TDR who continues to enjoy an active lifestyle after 25 years of TDR service (Fig. 21.3). The following sections detail the materials selection used in contemporary TDR design for both lumbar and cervical total disc replacements. The similarities in these designs demonstrate a general consensus of which materials may be best suited for TDR and the differences demonstrate the new technologies that may potentially work better.

21.4

Current designs

In addition to what is now the classical total disc arthroplasty with articulating artificial joint implants, there are several disc sparing or motion preservation devices that while not technically TDAs, are new spine/disc implants and are discussed in the following sections. These other types of spine implants include lumbar annular and intra-discal treatment for degenerative disc disease, facet arthroplasty, posterior motion sparing technology, and nucleus replacement © 2008, Woodhead Publishing Limited

526

Joint replacement technology

21.3 The SB Charite¨ artificial disc has demonstrated dramatic success as evidenced by the first person to receive this early TDR who continues to enjoy a return to an active lifestyle after 18 years of TDR service. The end plates of this model were made of 1 mm thick steel (URX2CrNiMoN 18.12) and the sliding core was produced from Chirulen (UHMWPE). The artificial disc had multiple teeth-like projections for fixation to the vertebral end plates (courtesy of Depuy Spine).

technologies. The following list summarizes seven types of motion-preserving spine implants and some of the specific types of implants (in use or in clinical trials) within each category, their articulation couple and primary material constituents: 1. Annular and intra-discal treatment for degenerative disc disease (Fig. 21.4) · Disc/annular repair: Barricaid-Intrinsic Therapeutic (all polymer and metal: cobalt alloy, PTFE) 2. Facet arthroplasty (Fig. 21.5) · Anatomic Facet Replacement System (AFRSTM) Facet Solutions (metalon-metal: cobalt alloy-on-cobalt alloy · Total Facet Arthroplasty SystemÕ (TFASÕ) Archus Orthopedics (metal and polymer) 3. Lumbar posterior motion sparing technology (Fig. 21.6) · Stabilimax NZÕ-Applied Spine (metal-on-metal: articulation: Cobalt alloy-on-Cobalt alloy, Titanium) · TOPSÕ-Impliant (metal±polymer±metal with elastic core: cobalt alloy± polyurethane±cobalt alloy) © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

527

21.4 Example of a new type of spine implant used for annular and intra-discal treatment of degenerative disc disease, Barricaid-Intrinsic Therapeutic (courtesy of Dr J Hall).

21.5 The Anatomic Facet Replacement System AFRS (Facet Solutions) is a facet joint reconstruction device comprising a metal-on-metal (Co alloy on Co alloy) bearing `glide' which uses conventional pedicle screw-type fixation (courtesy of Dr J Hall).

· DynesisÕ-Zimmer (metal±polymer±metal with elastic core-like structures: cobalt alloy±polyurethane±cobalt alloy) · DIAM Medronics (all polymer, polyester, silicone) 4. Cervical disc arthroplasty (Fig. 21.7) · BryanÕ Cervical Disc-Medtronic (metal±polymer±metal: elastic core: cobalt alloy±polyurethane±cobalt alloy) · PCM-Cervitech (metal-on-polymer: articulation: cobalt alloy-onUWMWPE) © 2008, Woodhead Publishing Limited

528

Joint replacement technology

21.6 Dynamic spine stabilization implants: motion preservation implants are some of the more innovative spinal implants currently under investigational use in the lumbar spine in patients receiving decompression surgery for the treatment of clinically symptomatic central or lateral spinal stenosis: (a) Stabilimax (Applied Spine Technology), (b) TOPS (Impliant Inc), (c) Dynesis (Zimmer Inc), all of which use a combination of articulation and spring or elastomeric interior components to provide both articulation and resistance force back to a neutral position (courtesy of Dr J Hall).

· PRESTIGEÕ Cervical Disc-Medtronic (metal-on-metal: articulation: Stainless Steel-on-stainless steel) · PRODISC-CÕ-Synthes (metal-on-polymer: articulation: cobalt alloy-onUWMWPE) · SecureÕ-C-Globus Medical (metal-on-polymer: articulation: cobalt alloyon-UWMWPE) 5. Lumbar disc arthroplasty (Fig. 21.8) · NUBAC-Pioneeer Surgical (polymer-on-polymer: articulation: PEEK-onPEEK) · Charite-DePuy Spine (metal-on-polymer: articulation: cobalt alloy-onUWMWPE) · Prodisc II-Synthes (metal-on-polymer: articulation: cobalt alloy-onUWMWPE) · Maverick-Medtronic (metal-on-metal: articulation: cobalt alloy-on-cobalt alloy) · eDisc-Theken (metal±polymer±metal: elastic core: cobalt alloy±polyurethane±cobalt alloy) · Freedom Lumbar Disc-Axiomed (metal±polymer±metal: elastic core: cobalt alloy±polyurethane±cobalt alloy) · Activ L-Asculap (metal-on-polymer: articulation: cobalt alloy-onUWMWPE) 6. Lumbar nucleus pulposus replacement (Fig. 21.9) · P-plus/C-Plus-Pioneer Surgical, PDN-Ray Medica, Neudisc-Replication Medica, Nucore-Spinewave, Newcleus-Nautilus, DASCOR-Disc Dynamics, Sinitec-Depuy Spine Sinux, Biomet-EBI, Biodisc-Cryolife, Aquarelle-Stryker 7. Biologic disc replacement remains experimental.

There are many different kinds of current IDR designs that are in pre-clinical © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

529

21.7 CERVICAL IDRs: There are a number of cervical total disc replacements in use which employ the three primary types of disc joint articulation. (1) Metalon-polymer articulation: (a) the Prodisc-C (Synthes) and (b) PCM (Cervitech) use Co-alloy endplates that articulate on a polymeric (UHMWPE) core that is mechanically fixed to one of the end plates and articulates in a ball-and-socket type manner. (2) Metal-on-metal articulation: (c) the Prestige Disc (Medtronic) represents a departure from the metal-on-poly articulation and instead uses metal-on-metal articulation where the end plates are constructed of stainless steel or titanium. (3) Elastic core articulation: (d) the Bryan Cervical Disc System (Medtronic) is axially symmetric and incorporates cobalt±chrome alloy clamshell-shaped end plates, which flex upon on a load-bearing polymeric (polyurethane-based) nucleus core. A significant feature in the design of this component is the polyurethane flexible membrane that surrounds the entire articulation and forms a sealed space containing a saline lubricant to reduce friction and prevent migration of any wear and corrosion debris (pictures provided courtesy of DePuy Spine Inc, Medtronic, Spine Solutions Inc. and Dr J Hall).

and clinical trials and that will be competing in the general population soon. These generally fall into one of five types: 1. 2. 3. 4. 5.

Metal-on-polymer bearing surfaces. Metal-on-metal bearing surfaces. Polymer-on-polymer. Elastic core. Biologics.

© 2008, Woodhead Publishing Limited

530

Joint replacement technology

It is important to point out that any disc arthroplasty implant that is not mentioned in this chapter in no way relates to the promise or performance of these omitted implants. Additionally, the rapid rate of company buyouts, product repackaging, and new market entrants, in 2008 will likely make outdated some of the names for IDR implants covered in the next section, by the time this text is in print. Generally current IDR designs have a primary articulation (or motion) using polymer-on-metal, polymer-on-polymer, metal-on-metal, or some form of all-elastic core technology. © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

531

21.8 (opposite) LUMBAR IDRs: there are a number of lumbar total disc replacements in clinical trials or accepted for use, encompassing all available types of joint articulation including metal-on-polymer, metal-on-metal, polymer-on-polymer and flexible core technologies. (a) Metal-on-polymer articulation: the LINKÕ SB Charite¨ III (DePuy Spine). Thick cobalt chrome alloy end plates articulate on a mobile bearing UHMWPE core. The end plates are covered with an osteoconductive surface of titanium/CaP double coating under the tradename `TiCaPÕ'. (b) Metal-on-polymer articulation: the Prodisc (Synthes) lumbar TDR is composed of cobalt±chrome±molybdenum alloy and covered with a porous titanium alloy and articulates on a central core of UHMWPE. (c) Metal-on-polymer articulation: ActivL (Aesculap) uses a polymeric center core intended to allow both translation and rotation and to more closely approximate physiological motion. (d) Metal-on-polymer articulation: Dynardi (Zimmer) is a disc replacement implant with two opposing Co-alloy (Co±Cr±Mo) end plates coated with porous pure titanium for bone ingrowth, that articulate on a semi-constrained UHMWPE core. (e) Polymeron-polymer articulation: NUBACTM (Pioneer Surgical) is a polymer-onpolymer disc arthroplasty device and the first PEEK-on-PEEK articulated disc arthroplasty device. (f) Metal-on-metal articulation: The Maverick Disc (Medtronic) uses metal-on-metal articulation where the end plates are constructed of Co-alloy. (g) Elastic core articulation: the Theken eDisc (Theken Disc) represents another step in the evolution orthopedic implant devices in that, as well as containing an elastic polyurethane-based core, it provides measured in vivo load information to the surgeon and patient via electronic sensors and transmitters. (h) Elastic core articulation: the Freedom Lumbar Disc (Axiomed) uses a viscoelastic polymer (like polyurethane) to replicate the native function of a natural disc. The elastic core in combination with the implant design provides a three-dimensional motion that functions within the natural biomechanics of the spine (pictures provided courtesy of DePuy Spine Inc, Medtronic, Spine Solutions Inc. and Dr J Hall).

21.9 Elastomeric nucleus/disc replacement implants: (a) The DASCORÕ (Disc Dynamcis) device consists of a polyurethane balloon which is inserted into the disc nucleus space after the nucleus has been removed and then filled with curable polyurethane. (b) PDN (Raymedica) is designed to re-establish disc space and replace the nucleus while retaining the annulus fibrosis. The PDN implant consists of a hydrogel core encased in a woven UHMWPE jacket. © 2008, Woodhead Publishing Limited

532

Joint replacement technology

21.4.1 Metal-on-polymer articulation Metal-on-polymer articulating IDR devices are used for both lumbar and cervical disc replacements. SB Charite III artificial disc (Depuy Spine Inc.) The current design of the Charite III includes thick cobalt chrome end plates which replaced the original stainless steel end plates (Fig. 21.8). The metal end plates are coated with an osteoconductive surface of porous titanium beads and CaP double coating under the trade name `TiCaPÕ'. This coating has been tested extensively in primates (McAfee et al., 2003) and in non-cemented joint replacements such as femoral stems, acetabular cups, ankle joint prostheses and dental implants (Liefeith et al., 2003). The end plate coating on the ChariteÁ III consists of three layers. The first two layers are of commercially pure titanium (Ti); the first layer provides a sintered metallic bond between the cobalt±chrome end plate and the coating, and the second layer of plasma-sprayed Ti provides the desired pore size of 75±300 m (Pillar, 1983; Schliephake et al., 1991). The third layer of coating consists of electroplated CaP. The purpose of this bioactive coating is to aid initial fixation resulting in shorter recovery times. Prodisc (Synthes Inc.) The lumbar TDR Prodisc is similar to that of the Charite in material composition and reflects how hip and knee designs have shaped the opinion of which biomaterials may perform optimally for TDR (Fig. 21.8). The two end plates comprise cobalt±chrome±molybdenum alloy and covered with a porous titanium alloy and they articulate on a central core of UHMWPE. The locking polyethylene insert provides a ball and socket type movement which differentiates this design from a mobile bearing design, e.g. the ChariteÂ. Although the ability of the polymer insert to be locked into place on the end plate is touted as a means to provide shock absorption, it is unclear if this represents shock absorption properties superior to mobile bearing inserts. The cobalt±chrome alloy end plates are coated with a pure titanium (Plasmapore) which in similar fashion to the Charite aims to optimize osseo-integration while using materials with well-established orthopedic clinical performance.

21.4.2 Metal-on-metal articulation Metal-on-metal articulating IDR devices are currently put forward for use in both lumbar and cervical disc replacement applications as well as dynamic stabilization devices of the lumbar spine.

© 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

533

Dynamic stabilization: Stabilimax NZ (Applied Spine Technologies Inc.) The Stabilimax NZÕ dynamic spine system is intended to provide stabilization of the lumbar spine in patients receiving decompression surgery for the treatment of clinically symptomatic central or lateral spinal stenosis (Fig. 21.6). It is composed of two ball and socket joints with Co alloy on Co alloy (ASTM F-75) articulation, two titanium alloy (Ti±6Al±4V) screws and an interposed spring of Co alloy in between the two ball and socket joints (Fig. 21.7). This device utilizes a dual-spring mechanism designed to provide maximum stabilization to the spine to address pain from impingement and misalignment, while preserving normal motion. These types of device offer an alternative to IDR because they promise the advantage of a less invasive surgical procedure (posterior approach) than fusion or disc replacement. The successful implementation and outcome of this technology will likely minimize primary and any revision surgical trauma and reduce morbidity and recovery time when compared with TDA implants. There is the evolving potential of these implants to be used in concert with more traditional and less constrained disc arthroplasty designs adding mechanical forces to more accurately reproduce normal in vivo kinematics. Total disc arthroplasty: Prestige Disc The Prestige Disc (Medtronic) represents a departure from the metal-on-polymer articulation and instead uses a metal-on-metal articulation where the end plates are constructed of stainless steel or titanium (Fig. 21.7). The articulation is similar to a ball and socket construct with a more constrained center of rotation and joint kinematics than that of mobile bearing. Because the wear resistance of stainless steel is superior to polyethylene, the particulate debris generated by metal-on-metal couples is an order of magnitude less than that produced by metal-on-polymer couples. However, lack of shock absorption aside, the amount of metallic debris generated is several orders of magnitude higher in this type of design than in that produced by metal-on-poly articulation. And although there is limited experience with stainless steel and titanium spinal TDR implants, metallic debris is purported to cause greater reactivity than polymeric debris (discussed later). The proximity of this implant to vital tissues raises concerns associated with excessive release of metallic debris. However, the relative importance of these issues will be resolved primarily through careful clinical follow-up and peer-reviewed analysis.

21.4.3 Polymer-on-polymer articulation The relatively low loads of the spine compared with knees and hips and the use of new, more wear-resistant polymers have enabled polymer-on-polymer articulation designs of TDA devices. © 2008, Woodhead Publishing Limited

534

Joint replacement technology

NUBACTM (Pioneer Surgical Inc) is a polymer-on-polymer disc arthroplasty device and the first PEEK-on-PEEK articulated disc arthroplasty device. This implant aims to maintain or restore the disc height and mechanical function of the natural nucleus using a less invasive procedure than other total disc arthroplasty implant, i.e. it is inserted into a partially resected disc. The design of the NUBAC seeks to include the major benefits of both total disc arthroplasty and nucleus arthroplasty with a two-piece design comprising an inner ball/socket articulation. It does not restrict any physiological rotational motions, leaving constraint and stability to the retained surrounding annulus and ligaments. The risk of implant extrusion is low because as the segment bends opposite to the annular insertion `window' in the disc there is an increase in the dimension of the NUBAC near the annular window, which makes extrusion of the device more difficult.

21.4.4 Elastic intervertebral replacement devices The advent of elastic core TDR implants have pioneered elastic core TJR technology because the relatively small range of motion required by a disc arthroplasty implants compared to a hip or knee, facilitates the potential of this type of design. There are generally two different kinds of IDR designs with elastic components as a central design feature: (1) those with elastic interposed between two end plates for total disc arthroplasty, and (2) all elastic devices for nucleus replacement. Total disc arthroplasty: Bryan Cervical Disc Prosthesis (Medtronic) The Bryan Cervical Disc System was recently approved for use in July 2007 (Fig. 21.7). Attempted preservation of joint kinematics is accomplished by the internal articular geometry of the prosthesis that is axially symmetric and therefore representative of flexion/extension motions and lateral bending motions. This device also incorporates cobalt±chrome alloy clamshell-shaped end plates, which `articulate' (or more accurately `flex') on a load-bearing polymeric (polyurethane-based) nucleus core. The end plates are covered with a porous coating comprising 250 m titanium beads sintered to the cobalt alloy end plates (Fig. 21.7). A significant feature in the design of this component is the polyurethane flexible membrane that surrounds the entire articulation and forms a sealed space containing a saline lubricant to reduce friction and prevent migration of any wear and corrosion debris. Additionally, this shealth aims to prevent the intrusion of connective tissue and functions as a barrier, preventing soft tissue impingement on the articulating components, and may act as a scaffold for creation of a pseudo-capsule over time. The inclusion of a sheath component to address the clinical concern of wear and corrosion debris (discussed later) is a significant development in the design of orthopedic joint arthroplasty com© 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

535

ponents. However there are risks associated with this type of component feature, such as in vivo rupture and bolus release of particle build-up and/or wear of the polyethylene component due to accelerated polymer degradation via a three-body wear mechanism of trapped debris. Thus the potential utility of this design element will shown by carefull monitoring over time. eDisc (Theken Disc Inc.) The eDisc represents another step in the evolution of orthopedic implant devices (Fig. 21.8) in that it provides measured in vivo load information to the surgeon and patient via electronic sensors and transmitters. While it resembles other implants that seek to restore the elastic load-bearing characteristics of the natural disc using a flexible interpositioned elastic polymer (polycarbonate polyurethane TH200) between two titanium endplates, it also houses a miniaturized microelectronic module that monitors and transmits force data to surgeons. These load data are targeted at better managing the patient's postoperative course, e.g. detecting auto-fusion and facilitating a return to work. Nucleus/disc replacement New devices are coming on the market with the aim of replacing the nucleus or the entire disc with an elastic substitute. These are not traditional disc arthroplasty joint replacements but they are designed to re-establish normal motion and disc space and replace the nucleus while retaining the annulus fibrosis (Fig. 21.9). DASCORÕ (Disc Dynamics Inc.) This device consists of a polyurethane balloon which is inserted into the disc nucleus space after the nucleus has been removed and then filled with curable polyurethane. The balloon is injected with the flowable polymer which is intended to create a patient-specific implant geometry that conforms to the shape and size of the disc space. The flowable polymer cures, creating a firm but pliable implant. This process reportedly requires short operating times. PDN Prosthetic Disc Nucleus (Raymedica Inc.) The PDN implant is a nucleus replacement, where a hydrogel core is encased in a woven polyethylene jacket. Prior to implantation the nucleus replacement device is compressed and dehydrated. After implantation, the hydrogel begins to absorb fluid and expand, because the tightly woven UHMWPE permits the passage of fluid and it takes approximately 4±5 days for the hydrogel to reach maximum expansion. Two side-by-side PDN implants are placed within the disc © 2008, Woodhead Publishing Limited

536

Joint replacement technology

space to provide the necessary disc space height. Mechanical testing has shown the ability of the PDN implant to withstand in vivo loads applied to discs. The PDN is undergoing clinical evaluation in Europe, South Africa, and the United States. While there are advantages of this device in terms of cartilage and tissue sparing, there is limited historical precedent with regard to in vitro testing and in vivo longevity/clinical performance of this kind of implant. It is unclear what amount of implant debris is generated from this implant and how, over the long term, these types of implant will react with the surrounding tissues. However, this device represents a step towards the future of total disc and joint replacement when (if cost permits) tissue engineered replacement biological tissue will be grown in situ or in vitro and replace damaged cartilage and bone. Currently tissue engineered products for the disc replacement are limited to a bioengineered bone morphogenetic protein INFUSETM (Medtronic Sofamor Danek, Memhpis, TN). The FDA approved INFUSETM for use with titanium interbody implants in the lumbar spine to help encourage bone growth.

21.4.5 Biologic intervertebral replacement devices Replacement of degenerated or injured discs with viable biologic tissue that will maintain itself over the long term remains the ultimate goal of tissue engineering in spinal orthopedics. To a large extent this goal remains highly experimental; however, progress is being made. For example, in a recent study by Revell et al. (2007) a tissue engineered disc substitute used two hyaluronan-derived polymeric substitute materials (HYAFFÕ 120, an ester, and HYADDÕ 3, an amide) that were seeded with previously harvested autologous bone marrow stem cells and injected into the nucleus pulposis of the lumbar spine in female pigs. They found that nucleotomy alone resulted in loss of normal intervertebrate disc (IVD) structure and the injected discs had a central nucleus pulposis-like region similar to that of a normal disc. This study and other similar tissue engineering efforts currently under way hold great promise for the future but have yet to be used in human clinical trials.

21.5

Clinical concerns

Elements used in modern orthopedic implant alloys including disc arthroplasty implants (e.g., titanium, aluminum, vanadium, cobalt, chromium, and nickel) are potentially toxic through (i) metabolic alterations, (ii) alterations in host/ parasite/symbiont interactions, (iii) immunologic interactions of metal moieties by virtue of their ability to act as haptens (specific immunologic activation) or anti-chemotactic agents (non-specific immunological suppression), and (iv) chemical carcinogenesis. Co, Cr, V and possibly Ni are essential trace metals in that they are required for normal homeostasis. In excessive amounts, however, cobalt has been © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

537

reported to lead to polycythemia, hypothyroidism, cardiomyopathy, and carcinogenesis; chromium can lead to nephropathy, hypersensitivity, and carcinogenesis; nickel can lead to eczematous dermatitis, hypersensitivity, and carcinogenesis; and vanadium can lead to cardiac and renal dysfunction, and has been associated with hypertension and manic-depressive psychosis. Other non-essential metallic elements can also be toxic. Titanium has been associated with pulmonary disease in patients with occupational exposure and with platelet suppression in animal models. Aluminum has been associated with renal failure, anemia, osteomalacia, and neurological dysfunction, possibly including Alzheimer's disease. All these aforementioned toxicities generally apply to soluble forms of these elements or particulate aerosols and may not apply to the chemical species that result from prosthetic implant degradation. Toxicity associated with any non-traditional load-bearing implant materials is not currently known. It is important to note that despite the potential toxicologic possibilities the association of metal release from orthopedic implants with any metabolic, bacteriologic, immunologic, or carcinogenic toxicity remains speculative: cause and effect have not been well established in humans with knee and hip TJAs (Michel et al., 1991).

21.5.1 Wear The generation of wear debris is the primary source of TJA degradation and the subsequent tissue reaction to such debris is currently extolled as the primary factor limiting the longevity of joint replacement prostheses. Particulate debris generated by wear, fretting or fragmentation induces the formation of an inflammatory reaction, which at a certain point promotes a foreign-body granulation tissue response that has the ability to invade the bone±implant interface. This commonly results in progressive, local bone loss that threatens the fixation of cemented and cementless devices alike. Mechanisms of wear debris generation Wear involves the loss of material in particulate form as a consequence of relative motion between two surfaces. Two materials placed together under load will only contact over the small area of the higher peaks or asperities. After an initial `wearing in' period, wear rates decrease and eventually become linearly dependent on the contact force and sliding distance represented by the steadystate wear equation: V ˆ KFx

21:1 3

where V is volumetric wear (mm /year), K is a material constant of the material couple, F is the contact force (N) and x is the distance of relative travel (mm). In general, the harder of two bearing materials will wear less rapidly. In a © 2008, Woodhead Publishing Limited

538

Joint replacement technology

metal-on-polymer pair, the polymer wears almost exclusively, while in a metalon-ceramic pair, the metal wears to a greater extent. Volumetric wear can be directly related to the number of wear particles released into periprosthetic fluids (typically in the order of billions of particles per year). The wear rates of THA metal±polymer couples are generally in the order of 0.1 mm/year, with particulate generation as high as 1  106 particles per step or per cycle. There is a great deal of variability associated with in vivo wear rates of orthopedic biomaterials, which are generally measured by radiographic follow-up studies. Radiographic wear measurements are expressed as linear wear rates whereas in vitro studies generally report volumetric wear. Clinical wear rates of THA and presumably disc arthroplasty will increase with the following: (1) physical activity, (2) weight of the patient, (3) size of the articulating interfaces, (4) roughness of the counterfaces, and (5) age or chemical degradation of implant materials (e.g., oxidation of the polyethylene). While in general there is limited information regarding the wear behavior of TDRs, a few recent clinical studies have begun to shed some light on the performance of early entries into the market of TDA. Wear and metal-on-polymer IDR In vitro analysis of wear has demonstrated wear rates of metal-on-polymer bearing lumbar TDA devices such as the Charite and the Prodisc (metal-onpoly-lumbar implants), as high as 20.8 and 17.8 mm3/million cycles of wear, respectively, under multi-axis motion and cyclic loading (using flexion± extension 3  6ë, lateral bending 2ë, axial rotation 1:5ë, cyclic loading 200±1750 N in 30 g/l protein) (Pare et al., 2007). However, because of the decreased loading (compared to hips and knees) and the decreased ranges of motions, TDA implants with metal-on-polymer articulation generally demonstrate ten times less wear than hip or knee arthroplasties (Fig. 21.10). However, this decrease does not seem to occur with metal-on-metal bearing disc arthroplasties (Fig. 21.10). Although it is important to note that the amounts of circulating metal in metal-on-metal THA and the first few reports in TDAs are several order of magnitude lower than that known to cause toxicity. Thus the physiologic effects of these elevated amounts of circulating metal remains unknown. An in vitro study by Popoola et al. (2007) of the Dynardi and Prodisc lumbar disc arthroplasties indicated that they had 2 mm3/million cyles and 4 mm3/ million cycles of volumetric wear, respectively (using flexion±extension ÿ3ë to ‡6ë, lateral bending 2ë axial rotation 2ë cyclic loading 600±2000 N in undiluted calf serum at 1±1.5 Hz), and that the addition of abrasive wear particles increased the wear rates of both implants more than three-fold. During this testing the particulate debris generated was described as approximately 1 m in diameter UHMWPE debris from both implants (Popoola et al., 2007). © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

539

21.10 A graphical comparison of data showing the amount of wear debris generated from different types of total joint arthroplasties, showing that there is relatively less (10) polymeric debris generated by a total disc arthroplasty with a metal-on-polymer articulation. However, this difference is not apparent with metal-on-metal articulating implants. Metal±polymer THA ± Callaghan et al., 2007; ceramic±polymer THA ± Minoda et al., 2005; metal-crosslinked polymer ± Kurtz et al., 2005; Heisel et al., 2004; Wroblewski et al., 1996; Saikko et al., 2002; metal±metal THA ± Catelas et al., 2004; ceramic±ceramic THA ± Tipper et al., 2002; metal-UHMWPE TDA ± Popoola et al., 2007; metal±metal TDA ± Pare et al., 2007.

In a study by Anderson et al. (2003) using in vitro wear testing of the Bryan Cervical Disc (metal-on-poly) prosthesis performed in a cervical spine simulator with loads and motions associated with activities of daily living, wear particles were produced at a rate of 1.2 mg per million cycles (particle size 3.9 m diameter, number-based analysis). This resulted in a device height decrease of 0.02 mm per million cycles where approximately 77% of this decrease was postulated to be due to gradual creep of the nucleus under the constant compressive load. In a recent in vivo study by van Ooij et al. (2007) a small case series of four patients undergoing revision of a failed metal on UHMWPE disc arthroplasty demonstrated wear patterns that included adhesive/abrasive wear of the central domed region of the polyethylene core, as well as rim impingement on the articulating polyethylene core of the device. However, there were no calculations of mass loss and thus attributing poor implant performance to wear debris was only speculative, despite the authors' claim that their study `demonstrates the clinical significance of polyethylene wear debris and the potential for osteolysis with total disc replacements.' © 2008, Woodhead Publishing Limited

540

Joint replacement technology

A similar but larger study by Kurtz et al. (2007) of 21 TDR components retrieved from 18 patients undergoing revision TDR surgery and conversion to fusion demonstrated that at an average implant time of 7.8 years the polymer (UHMWPE) articulating middle insert of the device showed an average of 0.3 mm of total wear or 38 m/year of linear wear. Given the effective diameter of this type of disc arthroplasty, this is an amount of wear that is approximately ten times less than that of highly crosslinked UHMWPE used in hip arthroplasty prostheses on either a mm/year or mm3/year basis (Greenwald and Garino, 2001; Kurtz et al., 2005). Again the dominant TDR wear mechanism was adhesive/ abrasive wear at both the dome and rim and the implants `displayed surface damage observed previously in both hip and knee replacements.' There is a strong consensus among researchers in this field that regular long-term followup of patients undergoing TDA is required to assess how intimately wear will correlate with failure and poor implant performance. Wear and metal-on-metal IDR There are relatively few published reports on the wear rates of metal-on-metal disc arthroplasty prostheses. A recent in vitro study of the Maverick implant by Pare et al. (2007) using a spine simulator (flexion±extension 10ë at a constant load of 1200 N in 11.5 g/l protein up to 10 million cycles) found a wear rate of 1.26 mm3 per million cycles. This amount of metal debris, although more bioreactive, compares favorably in volumetric loss to the Charite and the Prodisc (metal-on-poly-lumbar implants), which in the same study demonstrated an average of 20.8 and 17.8 mm3/million cycles of wear, respectively under multiaxis motion and cyclic loading (flexion±extension ÿ3ë to ‡6ë, lateral bending 2ë, axial rotation 1:5ë, cyclic loading 200±1750 N in 30 g/l protein). Another study by Hellier et al. (1992) of estimated total wear volume of intervertebral disc prosthesis with all titanium±6%Al±4%V alloy was reported to be 2.9 mm3/ million cycles. However in general the wear of a metal-on-metal cobalt alloy prosthesis is well below that of metal on polymer as has been well established in the hip arthroplasty arena (Fig. 21.10) (Callaghan et al., 2007; Catelas et al., 2004; Heisel et al., 2004; Kurtz et al., 2005; Minoda et al., 2005; Saikko et al., 2002; Tipper et al., 2002; Wroblewski et al., 1996).

21.5.2 Corrosion All metal alloy implants corrode in vivo. When severe, the degradative process may reduce structural integrity of the implant, and the release of corrosion products is potentially toxic to the host. The corrosion resistance of implant alloys is primarily due to the formation of surface barriers that limit implant corrosion. Kinetic barriers prevent corrosion by physically limiting the rate at which oxidation and reduction processes can take place. Alloys used in orthopedic © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

541

implants rely on the formation of passive oxide films to prevent significant electrochemical dissolution from taking place. Stainless steel alloys generally corrode to a greater extent than either cobalt or titanium alloys (Black, 1988, 1992, 1996; Breme and Biehl, 2001; Jacobs et al., 1994a, 1998a; Pourbaix, 1984).

21.5.3 Metal ion release There has always been concern regarding the release of chemically active metal ions from implants into surrounding tissues and the bloodstream. Normal human serum levels of prominent implant metals are approximately: 1±10 ng/ml Al, 0.15 ng/ml Cr, <0.01 ng/ml V, 0.1±0.2 ng/ml Co and <4.1 ng/ml Ti. Following TJA, levels of circulating metal have been shown to increase (Table 21.3). The values in this table show that following successful primary total joint replacement there are measurable elevations in serum and urine Co, Cr, and Ti. Transient elevations of urine and serum Ni have been noted immediately following surgery. However, urine and serum Al and V concentrations have not been found to be greatly elevated in patients with TJA. In a recent cross-sectional study of 10 subjects with metal-on-metal TDA, Zeh et al. (2007) found that the serum levels of cobalt and chromium after TDA showed concentrations of 4.75 ng/ml or parts per billion (ppb) for cobalt and 1.10 ng/ml or ppb for chromium, which were significantly elevated over control values. Thus surprisingly the concentrations of circulating Co and Cr measured in the serum of people with a metal-on-metal TDA are as high as levels measured in well-functioning total hip arthroplasties. The long-term effects of presumably much higher elevation of metal proximal to the spinal implant remain unknown and are under careful surveillance by implant companies, the FDA, and orthopedic researchers. Although not yet possible for people with disc replacements, postmortem analyses of tissues obtained from subjects with total joint replacement components have indicated that significant increases in Ti, Al, V, Co, and Cr concentrations occur in such tissues as the heart, liver, kidney, spleen, and lymph (Dorr et al., 1990; Jacobs et al., 1998b, 1999b; Stulberg et al., 1994). The response to metallic (and polymeric) debris in lymph nodes includes immune activation of macrophages and associated production of inflammatory cytokines. Metallic and polyethylene wear particles in the liver or spleen are more prevalent in patients who have had a previously failed reconstruction when compared with patients with primary hip or knee arthroplasties (Jacobs et al., 1994b, 1995a; Jasty et al., 1997; Stulberg et al., 1994; Urban et al., 1997, 2000).

21.5.4 Metal hypersensitivity The incidence of metal sensitivity among patients with TJA implants is approximately 25% (roughly twice as high as that of the general population), © 2008, Woodhead Publishing Limited

Table 21.3 Approximate concentrations of metal in human body fluids (10ÿ3 mM or 10 ppb) and in human tissue with and without total joint replacements (Dorr et al., 1990; Jacobs et al., 1994b, 1998a; Stulberg et al., 1987, 1994; Sunderman et al., 1989) Ti

Al 0.08 0.09

Serum

Normal TJA

0.06 0.09

Synovial fluid

Normal TJA

0.27 11.5

Whole blood

Normal TJA

0.35 1.4

4.0 24 0.48 8.1

V

Co

Cr

Mo

<0.02 0.03

0.003 0.007

0.001 0.006

* *

0.10 1.2

0.085 10

0.058 7.4

0.219 0.604

0.086 0.55

0.12 0.45

0.002 0.33

0.058 2.1

0.009 0.104

0.078 0.50

Normal: Subjects without any metallic prosthesis (not including dental). TJA: Subjects with total joint arthroplasty. *Data not available. © 2008, Woodhead Publishing Limited

WPTF3007

Ni 0.007 <0.16

Intervertebral disc joint replacement technology

543

and the average incidence of metal sensitivity among patients with a `failed' implant (in need of revision surgery) is approximately 50±60%. Released metal can activate the immune system by forming complexes with native proteins. These metal±protein complexes somehow trigger an immune response. Metals accepted as sensitizers include Be, Ni, Co and Cr, while occasional responses have been reported to Ta, Ti, and V. Nickel is the most common metal sensitizer in humans followed by Co and Cr. The temporal and physical evidence associated with reactions of severe dermatitis, urticaria, and/or vasculitis to the implantation of orthopedic devices leaves little doubt that the phenomenon of metal-induced hypersensitivity does occur in some cases, currently accepted to be <1% of patients. The degree to which these types of phenomenon will occur with IDR remains unknown and is an area that should be further investigated, especially given the introduction of metal-on-metal articulating disc replacement devices in the cervical and lumbar spine.

21.5.5 Carcinogenesis The carcinogenic potential of metals used in TDA remains an area of concern. The carcinogenic potential of orthopedic implant materials has been documented in a few animal studies. Rat sarcomas were noted to correlate with high serum cobalt, chromium, or nickel content released from metal implants. Recent studies of human TJA populations have found no significant increase in leukemia or lymphoma in patients with implants although these studies did not include large proportions of subjects with metal-on-metal prostheses (Sunderman, 1989). Continued surveillance and longer-term epidemiological studies are required to fully address these issues (Gillespie et al., 1988; Jacobs et al., 1995b, 1999a; Matheisen et al., 1995; Sinibaldi et al., 1976).

21.5.6 Infection Infection is a clinical reality in about 1% of people receiving orthopedic implants. This results in as many as 6000 people per year in the United States requiring intervention and in most cases reoperation for an implanted and infected total joint prosthesis. The degree to which infections will occur with IDR implants is not known, but it is likely that they will occur at roughly the same rate as other total joint replacement implants. However, because of the proximal location of the implant to the spinal cord and the difficulties associated with anterior approach and revision surgery, efforts to avoid infection both preand post-surgery would seem to be more important to IDR than in TJA of the knee or hip. All these clinical concerns highlight the importance of designing an implant that can be easily revised to either a fusion or another implant, if need be. The © 2008, Woodhead Publishing Limited

544

Joint replacement technology

revisability of an implant is not an easily quantified characteristic and thus this important factor falls into what remains the art of surgery and the ultimately strong market force of physician preference.

21.6

Conclusions

With over 30 years of clinical outcome assessment of spine fusion, it is clear that in an aging population and in younger patients that motion-preserving options are desirable and represent the future of degenerative disc disease-related spine surgery. However, the same 30 years of fusion have also provided a bar of clinical performance that must be clearly improved by all the aforementioned IDR and spine arthroplasty devices covered in this chapter. These clinical trials and follow-up studies of the few approved implants are only beginning in the United States and will need to continue for many years before this issue can be resolved. The local systemic and biologic effects of long-term use of metal alloys and polymeric materials remain largely uncharacterized. However, it is important to note that when evaluating the biocompatibility (defined as the ability of a material to demonstrate host and material response appropriate to its intended application) of a particular metal component, the results do not necessarily apply to all implants made of the same material. What works well in the lumbar region as an IDR may not work in a scaled down version in the cervical spine. Reasons for poor implant performance can be attributed to many factors, including manufacturing errors, mechanical design errors, surgical errors and inappropriate choice of material for a given application. Wise material selection cannot compensate for poor implant design or surgical error. It must be emphasized that currently there is no universal `best' material for all implant applications (in contrast to the claims touted by manufacturers' marketing departments). Terminology describing biologic performance such as `biocompatible' must be qualified with the type of application. Biomaterial optimization in TDR design has benefited from the clinical history of TJA materials in the hip and knee. The result of this knowledge is a convergence of materials and designs used for current TDA designs, which generally can be titanium-coated Co-alloy end plates which articulate on a relatively soft polymeric core. It may well be that differences in individual biology (such as predisposition of metal allergy) will dictate which of the various designs of TDA will function best and that keeping a selection of these implants commercially available will be of utility to patients requiring motion preserving disc replacement surgery.

21.7

References

Anderson PA, Rouleau JP, Bryan VE, Carlson CS 2003, `Wear analysis of the Bryan Cervical Disc prosthesis', Spine, vol. 28, no. 20, p. S186±S194.

© 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

545

Berry DJ 2000, Evolution of Uncemented Femoral Component Design, American Academy of Orthopaedic Surgeons, Chicago. Bertagnoli R, Kumar S 2002, `Indications for full prosthetic disc arthroplasty: a correlation of clinical outcome against a variety of indications', Eur Spine J, no. 11, pp. 131±136. Black J 1988, Orthopaedic Biomaterials in Research and Practice, Churchill Livingstone, New York. Black J 1992, Biomaterials, 2nd edn, Marcel Dekker, Inc., New York. Black J 1996, Prosthetic Materials VCH Publishers, Inc., New York. Black J and Hastings G 1998, Handbook of Biomaterial Properties, 1st edn, Chapman & Hall, London. Bogduk N 1999, Clinical Anatomy of the Lumbar Spine and Sacrum, Churchill Livingstone, New York, pp. 55±58. Bostrom PG, Lane JM 2000, Application of Bone Inductive and Conductive Agents to Hip and Knee Reconstruction, American Academy of Orthopaedic Surgeons, Chicago. Breme J, Biehl V 2001, `Metallic biomaterials,' in Handbook of Biomaterial Properties, J. Black & G. Hastings, eds, Chapman and Hall, London, pp. 135±214. BuÈttner-Janz K 1992, The Development of the Artificial Disc SB ChariteÂ, Hundley & Associates, London. BuÈttner-Janz K, Hahn S, Schikora K 2002, `Principles for successful application of the LinkÕ SB Charite artificial disc', OrthopaÈde, vol. 31, pp. 441±453. Callaghan JJ, Rosenberg A, Rubash H 2007, The Adult Hip, Lippincott Willaims & Wilkins, New York. Catelas I, Medley JB, Campbell PA, Huk OL, Bobyn JD. 2004, `Comparison of in vitro with in vivo characteristics of wear particles from metal-metal hip implants', J. Biomed. Mater. Res. B Appl. Biomater., vol. 70, no. 2, pp. 167±178. Cowan JA Jr, Dimick JB, Wainess R, Upchurch GR Jr, Chandler WF, La MF 2006, `Changes in the utilization of spinal fusion in the United States', Neurosurgery, vol. 59, no. 1, pp. 15±20. Dorr LD, Bloebaum R, Emmanual J, Meldrum R 1990, `Histologic, biochemical and ion analysis of tissue and fluids retrieved during total hip arthroplasty', Clinical Orthop Rel Res, vol. 261, pp. 82±95. FernstroÈm U 1966, `Arthroplasty with intercorporal endoprothesis in herniated disc and in painful disc', Acta Chir. Scand. Suppl, vol. 357, pp. 154±159. Gillespie WJ, Frampton CM, Henderson RJ, Ryan, PM 1988, `The incidence of cancer following total hip replacement', J. Bone Joint Surg. [Br.], vol. 70, no. 4, pp. 539± 542. Greenwald AS, Garino JP 2001, `Alternative bearing surfaces: the good, the bad, and the ugly', J. Bone Joint Surg. Am., vol. 83-A, Suppl 2, Pt 2, pp. 68±72. Hamby WB, Glaser HT 1959, `Replacement of spinal intervertebral discs with locally polymerizing methyl methacrylate', J. Neurosurg, vol. 16, pp. 311±313. Harmon PH 1963, `Anterior excision and vertebral body fusion operation for intervertebral disc syndromes of the lower lumbar spine', Clin. Orthop, vol. 26, pp. 107±111. Heisel C, Silva M, la Rosa MA, Schmalzried TP 2004, `Short-term in vivo wear of crosslinked polyethylene', J. Bone Joint Surg. Am., vol. 86-A, no. 4, pp. 748±751. Hellier WG, Hedman TP, Kostuik JP 1992, `Wear studies for development of an intervertebral disc prosthesis', Spine, vol. 17, no. 6 Suppl, pp. S86±S96. Jacobs JJ, Gilbert JL, Urban RM 1994a, `Corrosion of metallic implants,' in Advances in Orthopaedic Surgery, Vol. 2, Stauffer RN, ed., Mosby, St. Louis, pp. 279±319. © 2008, Woodhead Publishing Limited

546

Joint replacement technology

Jacobs JJ, Skipor AK, Urban RM, Black J, Manion LM, Starr A, Talbert LF, Galante JO 1994b, `Systemic distribution of metal degradation products from titanium alloy total hip replacements: an autopsy study', Trans. Orthop. Res. Soc., vol. 19, New Orleans, p. 838. Jacobs JJ, Urban RM, Gilbert JL, Skipor AK, Black J, Jasty M, Galante JO 1995a, `Local and distant products from modularity', Clin. Orthop. Rel. Res., vol. 319, pp. 94± 105. Jacobs JJ, Urban RM, Wall J, Black J, Reid JD, Veneman L 1995b, `Unusual foreignbody reaction to a failed total knee replacement: simulation of a sarcoma clinically and a sarcoid histologically. A case report', J. Bone Joint Surg. [Am.], vol. 77, no. 3, pp. 444±451. Jacobs JJ, Gilbert JL, Urban RM 1998a, `Corrosion of metal orthopaedic implants', J. Bone Joint Surg. [Am.], vol. 80, no. 2, pp. 268±282. Jacobs JJ, Skipor AK, Patterson LM, Hallab NJ, Paprosky WG, Black J, Galante JO 1998b, `Metal release in patients who have had a primary total hip arthroplasty. A prospective, controlled, longitudinal study', J. Bone Joint Surg. [Am.], vol. 80, no. 10, pp. 1447±1458. Jacobs J, Goodman S, Sumner DR, Hallab N 1999a, `Biologic response to orthopedic implants,' in Orthopedic Basic Science, American Academy of Orthopedic Surgeons, Chicago, pp. 402±426. Jacobs JJ, Silverton C, Hallab NJ, Skipor AK, Patterson L, Black J, Galante JO 1999b, `Metal release and excretion from cementless titanium alloy total knee replacements', Clin. Orthop., vol. 358, pp. 173±180. Jasty M, Goetz DD, Bragdon CR, Lee KR, Hanson AE, Elder JR, Harris WH 1997, `Wear of polyethylene acetabular components in total hip arthroplasty. An analysis of one hundred and twenty-eight components retrieved at autopsy or revision operations', J Bone Joint Surg Am, vol. 79, no. 3, pp. 349±358. Kurtz SM, Hozack W, Turner J, Purtill J, MacDonald D, Sharkey P, Parvizi J, Manley M, Rothman R 2005, `Mechanical properties of retrieved highly cross-linked crossfire liners after short-term implantation', J.Arthroplasty, vol. 20, no. 7, pp. 840±849. Kurtz SM, van Ooji A, Ross R, de Waal MJ, Peloza J, Ciccarelli L, Villarraga ML 2007, `Polyethylene wear and rim fracture in total disc arthroplasty', Spine J., vol. 7, no. 1, pp. 12±21. Lee KY, Pienkowski D 1998, `Viscoelastic recovery of creep deformed ultra-high molecular weight polyethylene (UHMWPE) characterization and properties of ultra-high molecular weight polyethylene', ASTM STP, no. 1307. Liefeith K, Holdebrand G, Schade R 2003, `In vitro and in vivo evaluation of a fully resorbable calcium phosphate coatin deposited on TPS coated implants', Third International Essen Symposium on the Working Group on Biomaterials and Tissue Compatibility, Oct 8, Essen, Germany. Matheisen EB, Ahlbom A, Bermann G, Lindgren JU 1995, `Total hip replacement and cancer', J. Bone Joint Surg. [Br.], vol. 77-B, no. 3, pp. 345±350. McAfee PM, Cunningham BW, Shimamoto N 2003, `SB ChariteÁ Disc Replacement: Biologic ingrowth using a non-human primate model', The Lumbar Artificial Disc, edited by Hochschuler SH, McAfee PM, BuÈttner-Janz K, Springer-Verlag, 6, 53± 71. McKellop HA, Sarmiento A, Brien W, Park SH 1992, `Interface corrosion of a modular head total hip prosthesis', J Arthroplasty, vol. 7, no. 3, pp. 291±294. McKenzie Alvin H 1995, `FernstroÈm intervertebral disc arthroplasty: a long-term evaluation', Orthop. Int. Ed., no. 3B, pp. 313±324. © 2008, Woodhead Publishing Limited

Intervertebral disc joint replacement technology

547

Mears DC 1979, `The biologic response to implanted materials,' in Materials and Orthopaedic Surgery, Williams & Wilkins, pp. 196±257. Michel R, Nolte M, Reich M, Loer F 1991, `Systemic effects of implanted prostheses made of cobalt-chromium alloys', Arch. Orthop. Trauma Surg., vol. 110, pp. 61±74. Minoda Y, Kobayashi A, Iwaki H, Miyaguchi M, Kadoya Y, Ohashi H, Takaoka K 2005, `Polyethylene wear particle generation in vivo in an alumina medial pivot total knee prosthesis', Biomaterials, vol. 26, no. 30, pp. 6034±6040. Pare PE, Chan F, Powell ML, Mathews HH 2007, `Wear characterization of the MAVERICK total disc replacment', Trans 7th Annual Meeting Spine Arthroplasty Society, May14, p. 59. Park JB 1984, Biomaterials Science and Engineering, Plenum Press, New York. Pillar RM 1983, `Powder metal-made orthopaedic implants with porous surface for fixation by tissue ingrowth', Clin.Orthop, vol. 176, pp. 42±51. Popoola OO, Shen M, Heller M, Seebeck J 2007, `In vitro wear of UHMWPE inlays in Dynardi and Prodisc spine disc replacment implants', Trans 7th Annual Meeting Spine Arthroplasty Society, May 14, Berlin, p. 49. Pourbaix M 1984, `Electrochemical corrosion of metallic biomaterials', Biomaterials, vol. 5, no. 3, pp. 122±134. Reitz H, Mauritius J 1964, `Intractable headache and cervio-Brachialgia treated by complete replacement of cervical intervertebral discs with a metal prosthesis', S.A. Medical J., pp. 881±884. Revell PA, Damien E, Di SL, Gurav N, Longinotti C, Ambrosio L 2007, `Tissue engineered intervertebral disc repair in the pig using injectable polymers', J. Mater. Sci. Mater. Med., vol. 18, no. 2, pp. 303±308. Saikko V, Calonius O, Keranen J 2002, `Wear of conventional and cross-linked ultrahigh-molecular-weight polyethylene acetabular cups against polished and roughened CoCr femoral heads in a biaxial hip simulator', J. Biomed. Mater. Res., vol. 63, no. 6, pp. 848±853. Schliephake H, Neukam FW, Klossa D 1991, `Influence of pore dimensions on bone ingrowth into porous hydroxylapatite blocks used as bone graft substitutes: a histomorphometric study', Int. J. Oral Maxillofac. Surg., vol. 20, pp. 53±58. Silver FH, Christiansen DL 1999, Biomaterials Science and Biocompatibility, Springer, New York. Sinibaldi K, Rosen H, Liu SK, DeAngelis M 1976, `Tumors associated with metallic implants in animals', Clin. Orthop. Rel. Res., no. 118, pp. 257±266. Stulberg BN, Warson JT, Shull SL, Merritt K, Crow TD, Stulberg SD, Guidry CH 1987, `Metal ion release after porous total knee arthroplasty: a prospective controlled assessment', Trans. Orthopedic Research Society, vol. 12, p. 314, Stulberg BN, Merritt K, Bauer T 1994, `Metallic wear debris in metal-backed patellar failure', J. Biomed. Mater. Res. Appl. Biomater., vol. 5, pp. 9±16. Sunderman FW 1989, `Carcinogenicity of metal alloys in orthopedic prosthesis: clinical and experimental studies', Fundam. Appl. Toxicol., vol. 13, pp. 205±216. Tipper JL, Hatton A, Nevelos JE, Ingham E, Doyle C, Streicher R, Nevelos AB, Fisher J 2002, `Alumina-alumina artificial hip joints. Part II: Characterisation of the wear debris from in vitro hip joint simulations', Biomaterials, vol. 23, no. 16, pp. 3441±3448. Urban RM, Jacobs J, Gilbert JL, Rice SB, Jasty M, Bragdon CR, Galante GO 1997, `Characterization of solid products of corrosion generated by modular-head femoral stems of different designs and materials', in STP 1301 Modularity of Orthopedic Implants, D. E. Marlowe, J. E. Parr, & M. B. Mayor, eds., ASTM, Philadelphia, pp. 33±44. © 2008, Woodhead Publishing Limited

548

Joint replacement technology

Urban RM, Jacobs JJ, Tomlinson MJ, Gavrilovic J, Black J, Peoc'h M 2000, `Dissemination of wear particles to the liver, spleen, and abdominal lymph nodes of patients with hip or knee replacement', J. Bone Joint Surg. [Am.], vol. 82, no. 4, pp. 457±476. van Ooij A, Kurtz SM, Stessels F, Noten H, van Rhijn L 2007, `Polyethylene wear debris and long-term clinical failure of the Charite disc prosthesis: a study of 4 patients', Spine, vol. 32, no. 2, pp. 223±229. Wright TM, Goodman SB 2000, Biomaterials, American Academy of Orthopaedic Surgeons, pp. 181±215. Wroblewski BM, Siney PD, Dowson D, Collins SN 1996, `Prospective clinical and joint simulator studies of a new total hip arthroplasty using alumina ceramic heads and cross-linked polyethylene cups', J. Bone Joint Surg. Br., vol. 78, no. 2, pp. 280± 285. Zeggel P 2000, `Bioactive calcium phosphate coatings for dental implants', Intern. Mag. Oral Impl. no. 1, pp. 52±57. Zeh A, Planert M, Siegert G, Lattke P, Held A, Hein W 2007, `Release of cobalt and chromium ions into the serum following implantation of the metal-on-metal Maverick-type artificial lumbar disc (Medtronic Sofamor Danek)', Spine, vol. 32, no. 3, pp. 348±352.

© 2008, Woodhead Publishing Limited

22

Replacing temporomandibular joints J - P V A N L O O N and L G M D E B O N T , University of Groningen, The Netherlands and G J V E R K E R K E , University of Groningen and University of Twente, The Netherlands

22.1

Introduction

The masticatory system plays an important role during biting, chewing, swallowing, speech, singing and other functions, all directly affecting quality of life. For proper functioning, both temporomandibular joints and the connecting mandible (Fig. 22.1), together with the masticatory muscles and contiguous tissue components, play a major role. In a healthy situation, the masticatory muscles supply the mandible with the required movements and biting and chewing forces, while the left and right mandibular condyles slide smoothly along their articular eminences (Fig. 22.1). Disturbances of the masticatory system can lead to a wide range of both muscular and temporomandibular joint (TMJ) conditions and pathology, resulting in pain, limited mouth opening, headaches, clicking or popping sounds in the TMJ, and impaired masticatory functioning. Looking more specifically at the TMJ, the conditions affecting this joint most frequently are osteoarthritis, condylar fractures and ankylosis. In addition, TMJ disturbances and muscle problems influence each other and may lead to chronic pain and functional impairment.1,2 The vast majority of TMJ patients are women in their third and fourth decades.2 It has been shown by the Groningen TMJ Research Group and other research groups that TMJ degenerative diseases have a considerable self-limiting behaviour, and non-surgical therapy will reduce the presented signs and symptoms in the majority of TMJ patients.3±7 Arthroscopic intervention can be considered when non-surgical efforts have failed. Open joint surgery is indicated only when all other methods were unsuccessful, and pain and limitation of movement mean that the patient's quality of life is affected significantly. In clinical practice a small group of TMJ patients with severe pain and TMJ destruction causing a strong limitation of function remains therapy resistant. For these patients, the available treatment modalities do not offer a proper solution, and alloplastic reconstruction of these mutilated TMJs may be the only remaining treatment option. © 2008, Woodhead Publishing Limited

550

Joint replacement technology

22.1 Lateral view of the temporomandibular joint (TMJ). Left: the condyle at the mandibular side, the articular fossa and eminence at the skull side, and the intervening articular disc. Middle: natural mouth opening, with the condyle translating smoothly along the eminence. Right: imitated condylar translation as a result of the centre of rotation located approximately 15 mm inferiorly of the centre of the condyle. Note that the condyle is removed when a TMJ prosthesis is placed.

The 1980s brought the first commercially available TMJ replacement, the Vitek-Kent prosthesis. This so-called `anatomical replica' consisted of a metal mandibular (jaw side) part, articulating directly against a poly(tetrafluoroethylene) (PTFE) skull part, thus without an intervening artificial disc. As most previous and later designs, the mandibular part resembled a bone plate with a head, fixed by screws to the lateral side of the mandible. The skull part covered the inferior side of the fossa and articular eminence and was fixated by screws to the lateral side of the articular tubercle. The PTFE skull part was less than 1 mm thick, and was coated at the cranial side with Proplast for tissue ingrowth. The same firm also marketed a `disc implant' to replace the articular disc after discectomy, or to resurface the temporal bone of the skull. The PTFE skull parts and disc implants showed a disastrously high wear rate in combination with deteroriation of the Proplast coating. The resulting particles caused severe soft-tissue irritation and bone resorption.8,9 As a major consequence of this disaster, several thousands of these prostheses were removed, with the result that in the United States a corresponding number of patients are now waiting for a solution for their mutilated TMJs.9 Fortunately, in Europe a much more conservative management of TMJ problems was followed, and only few patients received Vitek-Kent products. Since the Vitek-Kent prosthesis, a small number of new TMJ prostheses have been realised.10 The major causes of the small numbers are the complex shape and wide range of movements of the TMJ and the small market, in combination with high financial risks in the case of failing devices. Currently, there are three © 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

551

22.2 The three TMJ prostheses currently available on the market: Left: the TMJ Implants, with a metal±metal articulation; Middle: TMJ Concepts, with a metal head against a metal-backed polyethylene surface; Right: W. Lorentz Surgical, with a metal head against a polyethylene skull part.

TMJ prostheses on the market, namely those from TMJ Implants Inc. (Golden, Colorado), from TMJ Concepts (Ventura, California) and from Walter Lorenz Surgical Inc. (Jacksonville, Florida) (Fig. 22.2). The TMJ Implants prosthesis was designed in the early 1960s. At that time, the mandibular part had a poly(methylmethacrylate) (PMMA) head, functioning against a chrome±cobalt skull part. At the end of the 1990s, the PMMA head was replaced by a chrome±cobalt head, resulting in a metal-on-metal articulation with a small contact area. For the skull part, a large (33) number of different shapes are available, per side, while for the mandibular side four sizes are available. Interestingly, the commercialisation of this design started only in 1988, after the Vitek-Kent disaster. At that time, the device could be marketed in the United States without any clinical study because it had been used, on a very small scale, prior to 1976. From the start of its development, the TMJ Concepts prosthesis has been a fully custom-made product. The skull part is made from titanium mesh, with an ultra-high molecular weight polyethylene (UHMWPE) lining underneath. The mandibular part is made of titanium alloy with a cobalt±chrome head. Just recently, after 15 years of development, the Walter Lorenz (Biomet) prosthesis entered the US market. The design looks similar to the TMJ Concepts, with two important differences. Firstly, the prosthesis consists of a set of standard shapes, both for the skull part as well as for the mandibular part. Secondly, the skull part is made completely out of UHMWPE, without a metal backing. To get a good fit, the articular eminence is flattened and the skull part is placed directly against the bone. Because the articular eminence extends inferiorly, a space remains at the posterior side, between the fossa and the UHMWPE skull part. According to the website of the company, this space may be filled with PMMA bone cement, but there should not be any load on the cement. With fewer than 1000 patients per year, the number of patient applications of all three of these prostheses is very low compared with hip and knee prostheses. © 2008, Woodhead Publishing Limited

552

Joint replacement technology

At the moment only a small number of patients with severe TMJ problems are considered for such a total TMJ replacement. It may be expected, however, that when a clinically proven and properly functioning TMJ prosthesis is available, the indications will broaden considerably. Athough the Vitek-Kent prosthesis was poorly designed for long-term functioning, it nevertheless showed that a total TMJ replacement can significantly decrease TMJ pain and restriction of movement. Therefore, a project was started at the Department of Oral and Maxillofacial Surgery of the Groningen University Hospital, aiming at the realisation of a TMJ prosthesis that meets all necessary requirements. During the development of the TMJ prosthesis, three major problems were faced. The first problem was to imitate the large translatory movements of a healthy TMJ. During rest of the mandible, the unloaded condyle and disc are located in the glenoid fossa, on the dorsal slope of the articular eminence (Fig. 22.1). During all mandibular movements, the (loaded) condyle-disc complex is always located more anteriorly, with anterior movements that can exceed 15 mm for maximal opened mouth position (Fig. 22.1).11 All existing TMJ prostheses lack this anterior condylar movement. Imitation of the anterior movement is especially important because restricted movements of one TMJ cause abnormal deviations of the contralateral TMJ. Although many TMJ patients have symptoms of both TMJs, it is believed that in most cases the problems start on one side, after which malfunctioning affects the contralateral TMJ as well. It is expected that if the symptomatic joint is replaced by a prosthesis that has the ability to imitate condylar translation, the contralateral TMJ can be protected against further damage. Second, the considerable variation in size and shape of the cranial part of degenerated TMJs complicates the fitting of the prosthesis to the skull. Because the small volume of the cranial part of the TMJ leaves very limited possibilities to adapt the bony structures, it is the TMJ prosthesis that must be adaptable to all variations in size and shape of these structures. In addition, the subtle shape of these bony structures also limits the possibilities for stable fixation to the skull. Third, the long remaining lifetime of TMJ patients makes that it is essential that a TMJ prosthesis has a similar long lifetime. To be able to guarantee long lifetime, a TMJ prosthesis should be extensively tested prior to patient application. This evaluation should include the expected wear properties of the prosthesis, because wear particles can cause severe adverse tissue reactions. For hip and knee joint prostheses, the formation of large amounts of wear particles is the major cause of long-term failure.12,13 The results of the Vitek-Kent prosthesis have shown that the TMJ area is also sensitive to wear particles. The necessity of a long lifetime is stressed by the knowledge that the lifetime of hip and knee joint prostheses decreases with every revision surgery.14 It has been indicated that this will be the case for TMJ prostheses as well.2 © 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

553

In this chapter, the requirements for a safe, biocompatible and proper functioning TMJ prosthesis are given. A design that meets these requirements is shown, including the test results and the results of the first clinical application of the developed TMJ prosthesis.

22.2

Temporomandibular joint prosthesis criteria

The three major problems mentioned that should be solved during development of a TMJ prosthesis were elaborated to a complete list of requirements (Table 22.1). A TMJ prosthesis should meet all 11 requirements. Imitation of functional movement is especially important in case of unilateral TMJ replacement because the movements of the prosthetic side influence the movements of the opposite non-replaced TMJ. While in healthy persons condylar translation exceeds 15 mm during maximal mouth opening, prosthetic condylar translation is known to be less than 2 mm.15±17 It has been shown that such a limited condylar translation causes unnatural large lateral deviations of the mandible towards the prosthetic side,18 which will probably adversely affect the non-replaced TMJ. Therefore, a TMJ prosthesis should imitate the condylar translation during mouth opening (requirement 1), without restricting the nonreplaced TMJ (requirement 2). Realisation of a close fit to the skull is complicated by the fact that there is a considerable variation in shape of the skull side of the TMJ, which must fit correctly (requirement 3). In addition, the prosthesis should fit all possible shapes of the mandible (requirement 4), all parts should be of sufficient strength (requirement 5), and the TMJ prosthesis should be stably fixed to the bony structures (requirement 6). A long lifetime is a normal requirement for a joint prosthesis, but for the TMJ this is even more important because most TMJ patients are relatively young, with a life expectancy of 30±60 years. Because, in general, the expected life of a joint replacement decreases with the number of revision surgeries19,20 the aim Table 22.1 Requirements for a TMJ prosthesis 1 2 3 4 5 6 7 8 9 10 11

Imitation of condylar translation Unrestricted mandibular movements Correct fit to the skull Correct fit to the mandible Sufficient mechanical strength Stable fixation to the bony structures Expected lifetime of more than 20 years Low wear rate Wear particles tolerated by the body Biocompatible Simple and reliable implantation procedures

© 2008, Woodhead Publishing Limited

554

Joint replacement technology

should be to limit the number of revision surgeries to one, leading to a required lifetime of the prosthesis of a minimum 20 years (requirement 7). The lifetime of a joint prosthesis, as it is known from hip and knee joint prostheses, is strongly related to the wear rate. It is generally accepted that the constant formation of wear particles leads to bone resorption, which is a major reason for long-term failure of hip- and knee joint prostheses.13,21 It was assumed that the formation of relatively large amounts of wear particles will also adversely affect the TMJ. Therefore, any TMJ implant should be carefully evaluated with regard to this aspect prior to patient application (requirements 8 and 9). Two additional requirements count for permanent implants in general. The prosthesis should be well tolerated by the patient's body (requirement 10), and the prescribed implantation procedure should allow the surgeon to achieve the intended position of all prosthesis parts (requirement 11).

22.3

Design

In the first part of the research in Groningen, it was found that the natural sliding movement of the condyle can be imitated with a fixed, `inferiorly located' centre of rotation (Fig. 22.1).11,18,22,23 The study showed that when the centre of rotation is positioned in the area of the (former) natural condyle, the movements of the non-replaced TMJ exceed the natural movements, while when it is located 15 mm (or more) inferiorly, the movements of the non-replaced TMJ remain within the natural limits.18 This location was therefore considered the optimal centre of rotation. It has also been demonstrated that an inferiorly located centre of rotation does not increase the loading of the non-replaced TMJ, while the maximum load on the TMJ prosthesis itself will be approximately 100 N.24 This inferiorly located centre of rotation had a second advantage, namely that it automatically created space to design a proper, low-wear articulation. For the articulating surfaces of joint prostheses, UHMWPE opposed by a hard counterface seemed a good choice because it is the most frequently applied material, with a history of successful long-term application.14,25 Although there is also long-term experience with hip joint prostheses that use a ceramic± ceramic or a metal±metal articulation, these combinations are more difficult to apply because they wear slowly only if both parts match each other closely.26,27 In general, low contact stresses are advantageous for achieving a low wear rate28,29 and therefore the load was divided over a large contact area by the application of a ball and socket joint. Because the stress in the UHMWPE decreases with increasing thickness,30,31 and a decrease of the stress results in a decrease of the wear rate,28,29 the minimum disc thickness was set to 5 mm. This ball and socket joint agreed well with the idea of one centre of rotation, but restricted any horizontal movement of the mandible. Therefore, the UHMWPE disc was given freedom to make small sliding movements against © 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

555

22.3 The basic design of the developed TMJ prosthesis, consisting of a metal± ceramic skull part, a metal±ceramic mandibular part, and an intervening polyethylene disc. The spherical head of the mandibular part rotates in the disc, while the mandibular part together with the disc has freedom to translate against the skull part.

a flat inferior surface of the skull part, resulting in a `double articulation' with movements at the inferior and superior side of the disc (Fig. 22.3).32 The sliding movements of the disc were assumed to be small. The centre of the spherical head then became the point of rotation of the prosthesis, located at the optimal centre of rotation. The shape of the articulating surfaces could be realised only by a total TMJ prosthesis, replacing all components of the joint. This resulted in a primary design, consisting of a skull part, a mandibular part and, similar to the natural TMJ, an intervening disc (Fig. 22.3). To fit the prosthesis to the skull, three fitting methods were considered, poly(methylmethacrylate) (PMMA) bone-cement, custom-made parts and stock parts. Bone-cement was rejected because of the high temperatures during polymerisation of the cement which may cause necrosis of the supporting bone. Custom-made techniques were rejected because of high costs and doubts about the accuracy of custom-made prostheses. Thus, stock parts were the remaining option. Looking at the natural, healthy TMJ, it was noticed that the articular eminence is the load-bearing area on the skull side, and therefore the prosthesis is loading this part of the skull only. Because of the considerable differences between patients, the articular eminence and the lateral side of the TMJ were fitted separately. This resulted in a skull part composed of two connected parts, a basic part fixed to the articular tubercle and a cylindrically shaped fitting member facing the articular eminence (Fig. 22.4).33,34 During implantation, the © 2008, Woodhead Publishing Limited

556

Joint replacement technology

22.4 The skull part consists of a basic part and a fitting member. The basic part is fixed with bone screws to the lateral side of the TMJ, in the region of the articular tubercle. The fitting member fits the articular eminence and can rotate relatively to the basic part, around a vertical axis.

fitting member has the freedom to rotate around a vertical axis relative to the basic part, allowing the fitting member to follow the shape of the articular eminence while the basic part keeps in contact with the articular tubercle. This `self-adjustment' should result in the best fitting position of both parts. Regarding the mandibular part, the majority of TMJ prostheses uses a flat, non-adaptable mandibular part, positioned against the lateral side of the mandibular ramus.10 No problems have been reported except when using a small number of screws.10,35 The mandibular part therefore consists of a flat plate with a spherical head on top. To ensure an immediate stable fixation, bone screws with sharp thread were selected, similar to self-tapping osteosynthesis screws. The dimensions of the bony structures, especially at the skull side, limited the maximum screw diameter to 2 mm. For extra stability, bicortical fixation was preferred. In addition, to further increase stability, possible movements between screw and prosthesis were eliminated. This `rigid connection' was achieved by a screwthread at the inner side of the screw holes (Fig. 22.5).36 For the insertion of the screws, dental implant techniques were adopted. For the mandibular part, conventional screw fixation is thought to be sufficient, but to increase the stability the rigid connection was also applied. © 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

557

22.5 The rigid connection between screw and prosthesis is achieved by screwthread of the screw gripping in contra screwthread at the inside of the hole in the prosthesis. During the insertion (a) and in final position (b).

The final design of the TMJ prosthesis (Figs 22.6 and 22.7) was based on the design considerations mentioned above, and the selected materials. For the surfaces opposing the UHMWPE disc, zirconium oxide ceramic was chosen because of its excellent biocompatibility and scratch-resistance, in combination

22.6 The designed TMJ prosthesis, consisting of a skull part, a mandibular part, and an intervening artificial disc, in three views: in latero-caudal view (left), in ventro-cranial view (middle) and in medio-cranial view (right). The skull part consists of a basic part and a fitting member. © 2008, Woodhead Publishing Limited

558

Joint replacement technology

22.7 Lateral view of the Groningen TMJ prosthesis on the stereo-lithographic model of the first patient.

with a high strength.37,38 The other parts, except for the screws, were made from commercially pure titanium (cp-Ti), because of its proven biocompatibility. The screws were made from the stronger Ti alloy (Ti6Al4V), for greater safety. For composing the `self-adjusting' skull part, four different basic parts and three different fitting members were designed. The basic parts differed with regard to the caudal-cranial position of the screw holes, the fitting members with regard to the radius of the cylindrical surface facing the articular eminence. The basic part was fixed by three bone screws with the outer two screws rigidly connected to it. Its inferior side was covered by a zirconium oxide inlay. The fitting member could rotate around a pin at its caudal side, while a dovetail joint at its posterior side ensured attachment to the basic part. The mandibular part came in four versions, differing with regard to the overall length and the latero-medial position of the head. The zirconium oxide head had a diameter of 8 mm. Fixation was achieved with five bone screws, three of them being rigidly connected to the mandibular part. The cylindrically shaped UHMWPE disc had an outer diameter of 12 mm. For initial attachment, the disc was given a `snap' connection on the spherical head of the mandibular part. A circular cp-Ti wire around the disc provided X-ray visualisation. Templates were provided with the permanent prosthesis parts, to determine the correct position of these parts and to guide the drill in the correct direction. Small canals led the cooling water directly to the drilling site.39

© 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

22.4

559

Development and test procedures

Prior to the first clinical application, the above described different parts of the design were evaluated and tested and finally animal tests were performed with the complete TMJ prosthesis. The designed articulation resembles the functional shape of the natural TMJ, with rotation at the inferior side and sliding movements at the superior side of the disc. As previously stated, the translatory movements of the disc will be small and, in contrast to the natural situation, rotation will be the major movement. Condylar translation is therefore imitated by the inferiorly located centre of rotation, located in the middle of the spherical head of the prosthesis, approximately 13 mm below the peak of the articular eminence. It appears that this location agrees well with what is considered the optimal centre of rotation, 15 mm below the centre of the natural condyle. The first requirement, imitation of condylar translation, was therefore met. The remaining translatory capacities of the articulation, although small, are advantageous for following the cusps of the molars during masticatory movements. Furthermore, they provide freedom of movement in the contralateral nonreplaced TMJ. The articulation therefore met the second requirement. Fitting the skull is a major problem in TMJ reconstruction patients, because of the irregular shape of their TMJs.10 Therefore, there is a tendency towards custom-made skull parts, which are expensive owing to the complicated and time-consuming procedure. For these reasons a stock-part design was developed. The major invention is the application of two separate parts which fit separate sides of the skull part of the TMJ. This leaves the surgeon more options to achieve a correct fit than with other stock-part designs. To determine the fit of the designed skull part, prototypes of the basic part (four sizes) and the fitting member (three sizes) were tested on 20 dry skulls.34 For every skull, the best fitting basic part and fitting member were selected, and the maximum gap between the fitting member and the articular eminence was subsequently determined. All skulls could be fitted well. The average maximum gap between the fitting member and the articular eminence was 0.2 mm, with a range of 0.11±0.43 mm.34 These results were even better than the accuracy of stereo lithographic models which is in the order of 0.5 mm.40±42 However, the tested skulls had relatively naturally shaped TMJs and the development of the set of fitting members is not finished yet. The final set should be kept small and needs to be further judged with regard to the third requirement. The designed mandibular part resembles existing mandibular parts. Therefore, it is expected that there will be no problems with this part, and thus the fourth requirement seemed to be met. To determine the strength of the skull part and the mandibular part, threedimensional finite element models were developed. The maximum Von Mises stress in the prosthesis parts and the corresponding loads on the screws were calculated. The load on the TMJ prosthesis was set at 100 N, the expected © 2008, Woodhead Publishing Limited

560

Joint replacement technology

maximum load on a TMJ prosthesis.24 For the skull part, the load was applied at the inferior side by the disc, in a cranial direction. A small contact area (1 mm2) was assumed between fitting member and skull. All possible positions of the disc, as well as of the contact area between fitting member and skull, were included in the calculations. For the mandibular part, the direction of the load vector was varied to simulate all possible positions of the mandibular part relative to the skull part. Non-rigidly connected screws were assumed to carry no load. The calculations showed that the stresses in the skull part and the mandibular part remained well below the maximum allowed stress for cp-Ti (i.e. 180 N/mm2), so the fifth requirement was met. For the skull part, the maximum loads on the screws were 150 N in the radial direction and 3 N in the axial direction. For the mandibular part, the maximum loads on the screws were 155 and 11 N in radial and axial directions, respectively. The in vitro strength and stability of the designed rigid connection were tested by static and dynamic loading of rigidly connected screws.36 In the axial direction, the rigidly connected screws could resist static loads of over 500 N. In a radial direction, the screws could resist static loads of over 200 N when loaded at a distance of 2.5 mm from the prosthesis, increasing to over 500 N when loaded close to the prosthesis. During the dynamic tests, the screws were loaded in the radial direction at a distance of 2.5 mm from the prosthesis, with a load varying between ÿ70 and 70 N, for 5 million cycles. This dynamic shear loading did not induce movements between screw and plate.36 Therefore the sixth requirement seemed to be met. The field of TMJ reconstruction has a history of failing TMJ devices, which, among other problems, induced severe bone resorption and degeneration as a result of extreme high numbers of wear particles.8,43,44 Therefore, the expected wear rate of the UHMWPE disc was determined, using a wear testing machine especially developed for this purpose.32,45 The testing machine simulated the movements of the mandibular head against the disc, using a maximal mouth opening of 28ë and a lateral deviation of 2ë. The (constant) load was 200 N, almost twice the expected maximum in vivo loading of the prosthesis.24 The tests ran for 7 million cycles, corresponding to 10 years in vivo functioning.46 From the test results, the expected yearly in vivo wear rate was calculated, for both sides of the disc. The expected total disc wear rate was 0.65 mm3 per year, equivalent to a decrease of thickness of the disc of less than 0.01 mm per year, or 0.2 mm in the required lifetime of 20 years.32 Although the tests showed a low wear rate, no data are available on the acceptable amount of UHMWPE wear particles in the TMJ. Therefore, the results were compared with the wear rate of hip joint prostheses. The experiences with hip joint prostheses have shown that the body can tolerate huge amounts of UHMWPE wear particles, in the range of 25±75 mm3 per year,47,48 the majority of the particles being of the submicrometre size.49 Still, the life expectancy of a hip joint prosthesis is 10±15 years.25 Compared with the yearly © 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

561

wear volumes for hip joint prostheses, 0.65 mm3 per year for our TMJ prosthesis is thought to be a sufficiently small amount to ensure a long lifetime. Furthermore, compared with the initial disc thickness of 5 mm, the expected decrease in thickness of 0.2 mm after 20 years of functioning is minimal. From these results it has been concluded that the seventh and eighth requirements were met. As a final test, the designed TMJ prosthesis was tested in 12 sheep.50 The shape of the prosthesis was slightly adapted for application in sheep, but the general shape and dimensions differed marginally from the human design. The follow-up period ranged from two weeks to four months. One sheep was excluded because we could not achieve a correct position of the prosthesis parts. All sheep functioned with their TMJ prosthesis until they were killed. In the majority of the animals, the prosthesis parts were stable with favourable tissue reactions.50 There was one mechanical failure of a posterior screw of the skull part. This problem could be attributed to an incorrect position of the skull part, resulting in insufficient support of the fitting member and increased loading of the screws. One disc dislocated shortly after the implantation because of an incorrectly positioned skull part in combination with the absence of pre-stress on the disc. A second disc dislocated at a later stage, most likely accidentally. Although 2 out of 11 discs dislocated, these dislocations could be considered accidents and be attributed to lack of surgical experience. Thus, dislocation may be prevented by correct implantation procedures and careful postoperative handling. Furthermore, some weeks postoperatively, fibrous encapsulation will make dislocation virtually impossible. Therefore, there will be little risk of disc dislocation. The in vivo stability of the screws was studied in detail, as this gives an indication of the overall stability. For this purpose, harvested samples were histologically examined and the shear stress between the screws and the bone was calculated from removal torque measurements of the screws. For all mandibular parts, the screws were well incorporated in the bone, with no fibrous tissue layer between the screw and the bone. The average shear stress between the screw and the bone was 3.4 N/mm2 (standard deviation 1.3 N/mm2). For the skull part, most screws were well incorporated in the bone, with shear stresses slightly below those for the mandibular part (2.3 N/mm2, standard deviation 1.2 N/mm2). For long-term stability, the screws must be well integrated by the bone, in the same way as dental screw-implants are securely osseo-integrated in the mandible. However, for dental implants an unloaded time period is preferred to allow the implants to heal into the bone, while, in contrast, the screws of the TMJ prosthesis are immediately loaded. In patients who have not been chewing firmly for a long time, the loading will be limited, and can be further reduced by prescribing a soft diet. The animal tests are considered a rough test because the sheep had no initial restrictions and could not be instructed to unload their prosthesis. Even under these severe loading conditions, the removal shear © 2008, Woodhead Publishing Limited

562

Joint replacement technology

stresses of the screws were in the same range as has been reported for wellintegrated dental implants.12,51,52 This indicated a good integration of the screws in the bone, and long-term stability can probably be achieved. In combination with the in vitro results, the sixth requirement seems to be met. A prosthesis will function properly only when implanted correctly. Therefore, the implantation procedure was included in the development process and refined during the animal tests. Important for the procedure is the rigid connection between screws and prosthesis. After insertion of one rigidly connected screw, the position of the prosthesis parts cannot be changed. This problem was solved by first correctly positioning the prosthesis parts with non-rigidly connected screws, avoiding the need for subsequent positional changes. The usual tightening of the rigidly connected screws does not change the correct position of the prosthesis parts or the position of the screw in the bone, thereby avoiding unfavourable initial stress concentrations in the bone. We therefore found that the rigid screw± prosthesis connection positively influences the implantation procedure. A second important point is the prosthesis articulation, which allows the surgeon positional freedom in all six degrees of freedom and thus facilitates achieving a correct relationship between the mandibular part and the disc. Furthermore, the designed articulation allows the patient to postoperatively reposition the mandibular part in the position that is optimal for the non-replaced TMJ. From the in vitro and the in vivo tests, it was concluded that the TMJ prosthesis met the ninth, tenth and eleventh requirements, regarding reactions to wear particles, biocompatible materials and simple and reliable implantation methods. The pre-clinical phase was therefore closed and the first clinical application prepared.

22.5

First clinical application

Application of the device in patients was allowed after approval of the study design by the Medical Ethical Committee, and after written informed consent of the patients. The first patient was a woman, age 43 years. She was referred to our clinic after multiple TMJ surgeries including discectomy, arthroplasties and joint reconstruction of the right TMJ with a rib-graft (Fig. 22.8). At referral the patient had a persistent one-sided (right) ankylosis, with a maximum mouth opening of 5 mm. Because of the severe joint pain the patient was at a high level of pain medication, including morphine 30 mg three times daily (3td), diclophenac 100 mg two times daily (2td), diazepam 5 mg. Owing to the previous surgeries there was a left-sided open-bite. The patient was dentate. It was decided to reconstruct her right TMJ with a TMJ prosthesis. A stereo-lithographic model, based on three-dimensional computedtomography (CT) data, was used to plan this first operation. The best fitting parts were determined after model surgery. Regarding the skull part, the size of the basic part was determined by the height of the articular tubercle, while the © 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

563

22.8 Preoperative panoramic radiograph of first patient showing a rib graft at the right side and an eroded mandibular condyle at the left side.

shape of the articular eminence determined the curvature of the fitting member. For the mandibular part, its length was determined by the height of the mandibular ramus. The position of the disc, preferably in the middle of the inferior side of the basic part, determined the choice between the mandibular part with medially or with laterally positioned head. The implantation procedure was based on routine open joint surgery principles, following a pre-auricular approach for the skull part and a retromandibular approach for the mandibular part (Fig. 22.9). First a gap-osteotomy was performed at the level of the fossa and the remnants of the rib graft were

22.9 Preoperative view of the Groningen TMJ prosthesis. A pre-auricular approach is followed for positioning the skull part while the mandibular part is placed following a retro-mandibular approach. © 2008, Woodhead Publishing Limited

564

Joint replacement technology

22.10 Postoperative panoramic radiograph of first patient showing right-sided prosthesis in place.

removed to create space for the prosthesis. During surgery intermaxillary fixation was applied. The skull part template was positioned parallel to the cranial edge of the articular tubercle, while it was gently pushed against the articular eminence to make the fitting member rotate into its best fitting position. The screw holes were drilled using a low rotational speed (1500 rpm) and firm pressure, to convert the energy of the drill to cutting and not to frictional heat.39 To ensure proper distances between the screw holes, the template was kept in position by placing a pin in the drilled hole, for the first two drilled holes. The remaining hole could be drilled without further manipulation. The skull part template was then left in place and the mandibular part template was positioned together with the disc template, with some pre-stress on the disc. The screw holes were drilled following the same procedure as used for the skull part. After the correct position of all parts had been achieved the templates were replaced by the permanent prosthesis parts, starting at the skull side. For the permanent skull part, the non-rigidly connected middle screw was inserted first, after which the two rigidly connected screws were inserted. For the permanent mandibular part, a similar procedure was followed. The incisions were closed in layers with a mini-redon in place. Peri-operative prophylaxis against infection was performed with cefuroxim 1500 mg i.v. 3td for 24 hours. Post-operative radiographs showed the correct position of all prosthesis parts (Fig. 22.10). The patient's occlusion was preserved by applying intermaxillary fixation on elastics for the first two days. Thereafter this fixation was intermittent, with only active jaw movements by the patient herself. No physical therapy was prescribed. For pain relief the pre-operative pain medication schedule was continued. Patient recovery was uneventful except for some persistent swelling of the operated side. The patient was discharged at 10 days postoperatively. The initial mouth-opening had at that time increased from 5 mm to 18 mm. The pain level was at the pre-operative level. © 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

22.6

565

Conclusions

The presented TMJ prosthesis design is a mixture of well-known and accepted techniques, and new inventions. Among the well-known techniques are screw fixation and the use of proven biocompatible materials. The main new developments are a double articulation, including an inferiorly located centre of rotation, a self-adjusting skull part which is built from stock parts, and a rigid screw±prosthesis connection. The Groningen TMJ prosthesis successfully made the difficult step from prototype to first patient application. The developed TMJ prosthesis appears to meet 8 of the 11 requirements shown in Table 22.1. The other three requirements, i.e. the fit to the skull (requirement 3), the expected lifetime of the device (requirement 7), and the reliability of the implantation procedure (requirement 11), require further evaluation.

22.7

Sources of further information and advice: useful websites

www.tmj.org www.tmj.com www.tmjconcepts.com www.lorenzsurgical.com

22.8

References

1. Mercuri LG, Wolford LM, Sanders B, White RD, Hurder A, Henderson W. Custom CAD/CAM total temporomandibular joint reconstruction system: preliminary multicenter report. J Oral Maxillofac Surg 1995: 53: 106±15. 2. NIH. National Institute of Health technology assessment conference statement: management of temporomandibular disorders, 29 April 29±1 May, 1996. Oral Surg Oral Med Oral Pathol 1997: 83: 177±83. 3. Boering G. Temporomandibular Joint Osteoarthrosis: A Clinical and Radiographic Investigation. Groningen: University of Groningen, 1966. 4. de Bont LGM. Temporomandibular joint. Articular cartilage structure and function (dissertation). Groningen: University of Groningen, 1984. 5. de Bont LGM, Dijkgraaf LC, Stegenga B. Epidemiology and natural progression of articular temporomandibular disorders. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997: 83: 72±6. 6. De Leeuw R. A 30 year follow-up study of non-surgically treated temporomandibular joint osteoarthrosis and internal derangement (dissertation). Groningen: University of Groningen, 1994. 7. Stegenga B. Temporomandibular joint. Osteoarthrosis and internal derangement: diagnostic and therapeutic outcome assessment (dissertation). Groningen: University of Groningen, 1991. 8. AAOMS. Recommendations for management of patients with temporomandibular joint implants. J Oral Maxillofac Surg 1993: 51: 1164±72.

© 2008, Woodhead Publishing Limited

566

Joint replacement technology

9. Kent JN, Block MS, Halpern J, Fontenot MG. Update on the vitek partial and total temporomandibular joint systems. J Oral Maxillofac Surg 1993: 51: 408±15. 10. van Loon J-P, de Bont LGM, Boering G. Evaluation of temporomandibular joint prostheses. Review of the literature from 1946 to 1994 and implications for future prosthesis designs. J Oral Maxillofac Surg 1995: 53: 984±96. 11. Falkenstrom CH. Biomechanical Design of a Total Temporomandibular Joint Replacement. Enschede: University of Twente, 1993. 12. Ivanoff CJ, Sennerby L, Johansson C, Rangert B, Lekholm U. Influence of implant diameters on the integration of screw implants. An experimental study in rabbits. Int J Oral Maxillofac Surg 1997: 26: 141±8. 13. Willert H-G, Bertram H, Buchhorn GH. Osteolysis in alloarthroplasty of the hip. Clin Orthop Rel Res 1990: 258: 95±107. 14. Harris WH, Sledge CB. Total hip and total knee replacement. Review article. N Engl J Med 1990: 323: 725±31. 15. Kiehn CL, DesPrez JD, Converse CF. Total prosthetic replacement of the temporomandibular joint. Ann Plas Surg 1979: 27: 5±15. 16. Sonnenburg I, Sonnenburg M. Total condylar prosthesis for alloplastic jaw articulation replacement. J Maxillofac Surg 1985: 13: 131±5. 17. Wolford LM. Temporomandibular joint devices: treatment factors and outcomes. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997: 83: 143±9. 18. van Loon J-P, Falkenstrom CH, de Bont LGM, Verkerke GJ, Stegenga B. The optimal center of rotation for a temporomandibular joint prosthesis. A threedimensional kinematical study. J Dent Res 1999: 78: 43±8. 19. Lawrence JM, Engh CA, Macalino GE, Lauro GR. Outcome of revision hip arthroplasty done without cement. J Bone Joint Surg Am 1994: 76: 965±73. 20. Raut VV, Siney PD, Wroblewski BM. Cemented revision for aseptic acetabular loosening. A review of 387 hips. J Bone Joint Surg Br 1995: 77: 357±61. 21. Amstutz H, Campbell P, Kossovsky N, Clarke IC. Mechanism and clinical significance of wear debris-induced osteolysis. Clin Orthop Rel Res 1992: 276: 7± 17. 22. Falkenstrom, C. H. Temporomandibular joint prosthesis. US Patent no. 5,405,393, 1995. 23. van Loon J-P. The Groningen Temporomandibular Joint Prosthesis. Groningen: University of Groningen, 1999. 24. van Loon J-P, Otten E, Falkenstrom CH, de Bont LGM, Verkerke GJ. Loading of a unilateral temporomandibular joint prosthesis. A three-dimensional mathematical study. J Dent Res 1998: 77: 1939±47. 25. Malchau H, Herberts P, Ahnfelt L. Prognosis of total hip replacement in Sweden. Follow-up of 92,675 operations performed 1978±1990. Acta Orthop Scand 1993: 64: 497±506. 26. Clarke IC. Role of ceramic implants. Design and clinical success with total hip prosthetic ceramic-to-ceramic bearings. Clin Orthop Rel Res 1992: 282: 19±30. 27. Streicher RM, Semlitsch M, Schon R, Weber H, Rieker C. Metal-on-metal articulation for artificial hip joints: laboratory study and clinical results. Proc Inst Mech Engrs, Part H 1996: 210: 223±32. 28. Rose RM, Goldfarb HV, Ellis E, Crugnola AM. On the pressure dependence of the wear of UHMWPE. Wear 1983: 92: 99±111. 29. Rostoker W, Galante JO. Contact pressure dependence of wear rates of Ultra high molecular Weight Polyethylene. J Biomed Mater Res 1979: 13: 957±64. 30. Bartel DL, Burstein AH, Toda MD, Edwards DL. The effect of conformity and © 2008, Woodhead Publishing Limited

Replacing temporomandibular joints

31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50.

567

plastic thickness on contact stresses in metal-backed plastic implants. J Biomech Eng 1985: 107: 193±9. Bartel DL, Rawlinson JJ, Burstein AH, Ranawat C, Flynn Jr. WF. Stresses in polyethylene components of contemporary total knee replacements. Clin Orthop Rel Res 1995: 317: 76±82. van Loon J-P, Verkerke GJ, de Vries MP, de Bont LGM. Design and wear testing of a temporomandibular joint prosthesis articulation. J Dent Res 2000: 79: 715±21. van Loon, J-P. Adjustable temporomandibular surgical implant. International patent application no. PCT/NL95/00440, 1995. van Loon J-P, de Bont LGM, Stegenga B, Verkerke GJ. Fitting a temporomandibular joint prosthesis to the skull. J Oral Rehab 2000: 27: 853±9. Wolford LM, Cottrell DA, Henry CH. Temporomandibular joint reconstruction of the complex patient with the Techmedica custom-made total joint prosthesis. J Oral Maxillofac Surg 1994: 52: 2±10. van Loon J-P, de Bont LGM, Verkerke GJ. Comparison of two systems for rigidly connecting 2.0 mm bone screws to an implantable device. In vitro rigidity tests. Br J Oral Maxillofac Surg 2000: 38: 200±4. Shimizu KN, Oka M, Kumar P, Kotoura Y, Yamamuro T, Makinouchi K, Nakamura T. Time-dependent changes in the mechanical properties of zirconia ceramic. J Biomed Mat Res 1993: 27: 729±34. Willmann G, Frueh HJ, Pfaff HG. Wear characteristics of sliding pairs of zirconia (Y-TZP) for hip endoprostheses. Biomaterials 1996: 17: 2157±62. Matthews LS, Hirsch C. Temperatures measured in human cortical bone when drilling. J Bone Joint Surg 1972: 54-A: 297±308. Barker TM, Earwaker WJS, Lisle DA. Accuracy of stereolithographic models of human anatomy. Australas Radiol 1994: 38: 106±11. Lill W, Solar P, Ulm C, Watzek G, Blahout R, Matejka M. Reproducibility of threedimensional CT-assisted model production in the maxillofacial area. Br J Oral Maxillofac Surg 1992: 30: 233±6. Tyndall DA, Renner JB, Phillips C, Matteson SR. Positional changes of the mandibular condyle assessed by three-dimensional computed tomograpy. J Oral Maxillofac Surg 1992: 50: 1164±72. Milam SB. Failed implants and multiple operations. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997: 83: 156±62. Schellhas KP, Wilkes CH, el Deeb M, Lagrotteria L, Omlie MR. Permanent Proplast temporomandibular joint implants: MR imaging of destructive complications. Am J Roentgenol 1988: 151: 731±5. van Loon J-P, Verkerke GJ, de Bont LGM, Liem RSB. Wear testing of a temporomandibular joint prosthesis. UHMWPE and PTFE against a metal ball, in water and in serum. Biomaterials 1999: 20: 1471±8. Gibbs CH, Lundeen HC. Jaw movements and forces during chewing and swallowing and their clinical significance. In: Lundeen HC, Gibbs CH, eds: Advances in Occlusion. Boston: John Wright PSG Inc, 1982: 2±32. Isaac GH, Wroblewski BM, Atkinson JR, Dowson D. A tribological study of retrieved hip prostheses. Clin Orthop Rel Res 1992: 276: 115±25. Kabo JM, Gebhard JS, Loren G, Amstutz H. In vivo wear of polyethylene acetabular components. J Bone Joint Surg 1993: 75B: 254±8. McKellop HA, Campbell P, Park SH, et al. The origin of submicron polyethylene wear debris in total hip arthroplasty. Clin Orthop Rel Res 1995: 311: 3±20. van Loon J-P, de Bont LGM, Spijkervet FKL, Verkerke GJ, Liem RSB. A short-term

© 2008, Woodhead Publishing Limited

568

Joint replacement technology

study in sheep with the Groningen temporomandibular joint prosthesis. Int J Oral Maxillofac Surg 2000: 29: 315±24. 51. Carlsson L, Rostlund T, Albrektsson B, Albrektsson T. Removal torques for polished and rough titanium implants. Int J Oral Maxillofac Implants 1988: 3: 21±4. 52. Tjellstrom A, Jacobsson M, Albrektsson T. Removal torque of osseointegrated craniofacial implants: a clinical study. Int J Oral Maxillofac Implants 1988: 3: 287±9.

© 2008, Woodhead Publishing Limited

23

Replacing ankle joints H K O F O E D , Federiksberg Hospital, Denmark

23.1

Introduction: short history of ankle replacement

The first ankle replacement in modern times was performed in Paris on 16 October 1970 (Lord and Marotte, 1973). It was an inverted hip prosthesis where the cup was placed in the talus. Twenty-five of these cases were performed; the results were mediocre and the prosthesis was abandoned (Lord and Marotte, 1980). In the beginning of the 1970s a number of other designs came to the market. They were all two-part prostheses meant for cemented fixation. While the initial results were claimed to be promising, medium-term results showed high failure rates with loosening, and subsidence of the components, as well as talus necrosis. During the 1980s ankle replacement was generally abandoned by the orthopaedic community. In an editorial it was concluded that without new designs, new fixation methods, and maybe new materials, ankle replacements did not work (Hamblen, 1985). Meanwhile a few still tried to develop better and uncemented prostheses, such as the Agility ankle (Alvine, 1991), the B-P ankle (Buechel et al., 1988), the Scandinavian Total Ankle Replacement (Kofoed and Danborg, 1995) and the TNK ankle (Takakura et al., 1990). The Agility prosthesis is a two-component design, sloppy joint, approved only for cemented use by the US Food and Drug Administration (FDA). It has as far as we know only been used uncemented. It uses an arthrodesis between the fibula and the tibia in order to increase the ingrowth area (Fig. 23.1). The TNK prosthesis has had several developments, but is currently a two-piece design made of ceramics meant for biological ingrowth (Fig. 23.2). The B-P prosthesis (1983) and the STAR* (1984) were the first to use a meniscal bearing design, and they have subsequently been used uncemented. The B-P ankle (Fig. 23.3) is made of titanium. The tibial component is stemmed, and the talus component is placed on top of the talus. The surface towards the bone is titanium beaded. The articulating surfaces toward the polyethylene meniscus are covered with nitrogenous oxide. The STAR prosthesis (Fig. 22.4) has a talus cap covering the medial and the lateral talus facets. It is made of Co±Cr±Mo, and has a polyethylene meniscus. The tibial component is flat with two parallel cylinders © 2008, Woodhead Publishing Limited

570

Joint replacement technology

23.1 Agility Total ankle prosthesis: (a) talus and tibial component; (b) assembled prosthesis; (c) example of implanted prosthesis.

on its back. The surface toward the bone is titanium beaded with electrochemically added calcium phosphate for rapid ingrowth. These four designs are the only ones with long-term results. They have proved to be so effective that once again the last few years has seen a vast number of new prostheses outside the North American market using parts and principles from these original designs. These are nearly all using the meniscal bearing principle and are for uncemented bone fixation. However, because FDA approval is necessary in the United States, the industry has marketed two-component designs for cemented use since 2006. It may be discussed whether this is a step in the wrong direction.

23.2 TNK Total ankle prosthesis: (a) ceramic talus and tibial components; (b) example of implanted prosthesis. © 2008, Woodhead Publishing Limited

Replacing ankle joints

571

23.3 B-P total ankle prosthesis with mobile bearing: (a) the three prosthetic parts, stemmed tibial, meniscal bearing, and talus prosthesis; (b) example of implanted prosthesis.

23.2

Anatomical, biomechanical and biological features of the normal ankle joint

The human ankle joint is unique. It has a free mobile fibula. This is only otherwise found in the brown bear and the elephant in mammals. It is also the most congruent joint in the body. It has long been discussed how the wider anterior part of the talus can pass up into the fork of the ankle still keeping the congruence. The suggestion has been widening of the ankle fork. A more rational explanation is the adaptive rotation of the fibula as shown by stereo-video analysis (Helweg and Kofoed, 1998). During the arc of motion the centre of rotation is shifting depending on the position of the foot (Lundberg, 1998). However, the different axis always passes through the centre of the body of the talus. In the normal ankle joint, extension and flexion are the only movements that take place (Inman, 1976). Inversion and eversion take place in the subtalar joints. The mobility in the tibio-talar junction has been measured radiographically to be 15ë extension and 45ë flexion (Backer and Kofoed, 1989). These movements are controlled not only by the bone © 2008, Woodhead Publishing Limited

572

Joint replacement technology

23.4 STAR total ankle prosthesis with mobile bearing: (a) three prosthetic parts; (b) talus components; (c) tibial components; (d) meniscal bearings; (e) example of implanted prosthesis.

structures but especially by the ligaments, most pronouncedly by the lateral ligaments (Leardini et al., 2001). The muscles in the lower leg are the power machines, and without these not even a normal ankle joint would move. The strength of the bones is related to the subchondral bone of the distal tibia and the upper talus. It has been calculated that only the upper 0.5 cm of the talus and the distal 1 cm of the tibia are strong enough to carry a prosthetic device © 2008, Woodhead Publishing Limited

Replacing ankle joints

573

(Hvid et al., 1985). The distal tibia has a well-developed vascular supply whereas the vascular supply of the talus mainly comes from the tarsal sinus. The arteries radiate towards the talar body surface from this source, and they are end arteries (Crock, 1967). The venous outflow follow the arteries. This means that cutting down in the size of the talus would not only weaken its strength but also put the vascular supply at risk.

23.3

Pathologies leading to degeneration of the ankle joint

Any disease or condition that affects the structures in the ankle joint will lead to cartilage degeneration and subsequently osteoarthritis (OA). The more generalised diseases such as rheumatoid arthritis (RA), psoriatric arthritis and lupus erythromatosis dissiminatus certainly also attack the ankle joint. Haemachromatosis and repeated joint bleedings (haemophilia) are other diseases that destroy the joint. Most common, however, is the late results of trauma, that being either malleolar fractures or Pilon fractures, as well as ligament injuries, causing instability in the ankle joint. Deep infections either haematogenous or caused by surgery, will lead to joint degeneration. Primary osteoarthrosis is very uncommon, and there may be other, more rare, diseases that lead to degeneration of the joint. Whether osteoarthritis is caused by one or the other, it ends up by being painful to a degree that cannot be treated with medicine, that is, with pharmaceuticals. Depending on the primary cause of degeneration there may also be instability or distorsion of the anatomic features ± nature trying to stabilise the joint by typically antero-lateral and postero-medial osteophytes. Both of these will have to be dealt with when treating the painful ankle surgically. Even if the joint is replaced or fused the disease will continue, i.e. neighbouring joints will become involved, and that will have an impact on the entire foot±ankle situation. This will have to be solved either at primary surgery, or later on. Any surgical attempt to give a pain-free situation in the ankle joint will be at the cost of lost function. Fusion certainly does result in this, whereas ankle replacement intends to keep function, which may not be normal but is present to an extent that makes the functions of daily life uncomplicated.

23.4

Contraindications for ankle replacement

Not all degenerated ankle joints can be replaced. The lack of malleoli will not allow an unconstrained prosthesis ± and constrained prostheses have been shown to carry a greater risk for loosening. The lack of ligaments, especially the lateral ones, will give an unstable situation with increased and fast wear of the prosthesis parts. The lack of muscle power excludes the use of an ankle replacement. Recent deep infection ± and maybe also, old osteomyelitis ± are clearly contraindications. Osteoporosis could also be a contraindication. There is © 2008, Woodhead Publishing Limited

574

Joint replacement technology

no clear definition but arbitrarily the degree of osteoporosis can be put at 50% of the normal bone mineral content. Arteriosclerosis with distal pressures below 70 mm Hg would also put the replacement procedure at risk, not only from a soft tissue healing point of view, but also from the ingrowth potential of uncemented devices. Talus necrosis ± even to a minor degree such as osteochondritis ± does put a replacement at serious risk, whether the fixation is by bone cement or if an uncemented implant is tried. Some forms of arthritis are so aggressive that they will encapsulate the replacement and quickly lead to ankylosis. Psoriatric arthritis is a good example. Finally, patients who do not understand the nature of the procedure and the protection needed for an artificial joint will wear it out quickly. No known ankle prosthesis, so far, can tolerate repeated high impact from running and jumping.

23.5

Materials used to replace the ankle

Most modern ankle replacements are three-component devices. Metal parts are used to cover the surfaces of the distal tibia and the talus. The metal can be either chrome±cobalt±molybdenum or titanium alloy. The articulating surfaces are highly polished or covered with nitrogenous oxide. Titanium is the weaker material, and breakage of the tibial component has been reported in contrast with Co±Cr±Mo of the same dimensions. Between the metal parts a so-called floating meniscus, congruent toward both metal parts, is inserted. The meniscus in all three component prostheses consists of polyethylene. The metal surfaces for implantation into the bone rely on biological fixation. Cement is no longer generally used. As the only different parameter in two comparable series it has been shown that biological fixation is superior to cement fixation (Kofoed, 2004). The two-component ceramic on ceramic design, used solely in Japan (TNK), is also uncemented. Owing to the inferior fixation modus, stem cells have been used on the ingrowth surfaces to facilitate bone ingrowth. This has increased the fixation strength by 2.5 (Tohma et al., 2006). Recently, the use of biological composite grafts has been tried. A donor ankle joint graft consisting of 1 cm of the distal tibia including the surface of the articulating medial malleolus and 1 cm of the top of the talus dome is taken out en bloc. A similar resection is performed in the recipient, and the donor joint is trimmed to fit, and inserted. Small screws are used for fixation. The technique is used in the United States (Meehan et al., 2005) and in Europe (Giannini, 2007). The results are preliminary, but considered promising. This procedure can probably not be used universally as this would require direct access to a transplantation team. The cartilage cannot be frozen, but can be kept alive for about 72 h in a physiological solution at 4 ëC. Apart from this the technique raises the question about compatibility and transmission of diseases. Biomechanically, there is an unknown limit to the thickness of the composite implant. Revascularisation of the graft would probably not extend 1 cm. © 2008, Woodhead Publishing Limited

Replacing ankle joints

575

Depending on the bone strength 1 cm may not be enough to withstand the pressures ± up to six times body weight ± in the ankle joint.

23.6

Fixation of ankle prostheses

Fixation with bone cement is no longer performed. It was technically difficult to obtain pressurisation, and it was also difficult to remove excess cement, especially from the posterior part of the joint. Surfaces either with titanium beads, or plasma sprayed hydroxyapatite (HA) are still in use, but most prostheses now use a combination of titanium plasma sprayed beads to which calcium phosphate (Ti-CaP) is added electrochemically in a thin layer (15±30 m). It was not technically possible to add almost pure HA (>98%) to uneven surfaces because the spraying technique would leave holes underneath the HA layer, thereby giving rise to cracks, and possibly third particle wear (loose HA particles). Pure HA (100±200 m) is very slowly absorbed, whereas calcium phosphate (15± 30 m) is resorbed within 4±6 weeks, leaving the fixation provided by bone ingrowth into the rough surface between the titanium beads. Follow-up between the fixation of flat surfaces with pure HA (up to 17 years) and Ti-CaP (up to 8 years) have not shown any significant difference in loosening, but the increase in bone mineral density adjacent to the tibial component is significantly greater with HA coating (Zerahn and Kofoed, 2004). Carlsson et al. (2005) showed by radiostereophotogrammatic analysis that the components were stable throughout a five-year follow-up whether OA or RA cases were concerned.

23.7

The interrelationship between the ankle and the hindfoot

While the ankle joint only moves in extension and flexion, with possibly a little gliding, the subtalar joint and Choparts joint add rotation and gliding to the entire hindfoot (the combined joint line between the talus and the navicular and the joint between the calcaneus and the cuboid bone). This close interrelationship (in German: obere Sprunggelenke und untere Sprunggelenke; the upper ankle and the lower ankle joints) must be preserved when the ankle joint is replaced. This means that the entire hindfoot must be brought into a neutral position in a stable way before the ankle is replaced. Anything else will lead to malalignment and uneven pressures to both bone and prosthetic parts with the risk of component loosening or subsidence. Prostheses that formerly tried to incorporate all of the hindfoot function into the ankle joint did not fare well. The alignment procedure could involve supramalleolar osteotomy, calcaneal osteotomy, Achilles tendon lengthening or posterior tibial tendon reconstruction. Furthermore, simultanous arthrodesis and/or osteotomies may be added in order to keep the foot plantigrade. Resurfacing the ankle is just the final step of such procedures. © 2008, Woodhead Publishing Limited

576

23.8

Joint replacement technology

Long-term results of uncemented current designs

Only four uncemented prostheses can demonstrate long-term results: Agility, BP, STAR and TNK. These long-term results are understandably delivered by the inventors of the prostheses. While the TNK results are solely from the inventor, the three other prostheses have had other users. Most literature references concern the STAR where several articles have discussed medium-term results. The results of the Agility prosthesis from the author's experience (Knecht et al., 2004) with a mean follow-up of 9 years showed 8% non-union of the tibio-fibular arthrodesis, 11% had revision of the prosthesis or arthrodesis, and 76% showed some periprosthetic radiolucency. Independent investigators (Spirt et al., 2004) in another series with a follow-up of mean 33 months found 10% component revisions, and the estimated survival rate at 5 years to be 80%. The B-P prosthesis with deep sulcus after mean 5 years' follow-up showed a 92% survival rate at 12 years (Buechel et al., 2004). Others found a survival rate at 8 years to be 84% (Doets et al., 2006). The STAR prosthesis after a mean 9.5 years follow-up showed a 95% survival rate at 12 years (Kofoed, 2004). Others have found, at mean follow-up of 5 years, a survival rate at 5 years of 92.7% (Wood and Deakin, 2003), Andersson et al. (2003) had a survival rate of 70% with the same followup. In a second series the survival rate was 92% at 5 years (Carlsson, 2006). Schill et al. (1998) found a 94% survival rate at 6 years. The TNK prosthesis in its latest version was followed for a mean of 5 years (2±11 years) with 4.5% having revision of the prosthesis and one third showing radiographic loosening (Takakura et al., 2004). A number of revisions have been encountered in all series unrelated to the prosthesis fixation. These could be removal of heterogenic bone formation, impingement syndromes, meniscal fractures or correction of alignment. Most common, however, are primary wound problems or intra-operative malleolar fractures. These last complications have been shown to be significantly reduced by the surgeon's experience (Andersson et al., 2003; Haskell and Mann, 2004; Carlsson, 2006). Recently, a prospective randomised study of 200 ankle replacements with an average follow-up of four years using the B-P prosthesis or the STAR prosthesis found equally good clinical results, but the B-P replacements had a higher failure rate (Mishra and Wood, 2007).

23.9

Future trends

Ankle replacement has come a long way since its start in 1970. It can no longer be called an experimental procedure, and patients' awareness of the possibility for having the ankle replaced has reduced the need for primary arthrodesis of the ankle. There are still unsolved problems. Ankle replacement is far from as predictable as hip and knee replacement. The procedure is much more © 2008, Woodhead Publishing Limited

Replacing ankle joints

577

demanding, and one of the main problems seems to be the shortage of patients with disease requiring replacement surgery. This leads to small series and too little experience and routine gained by the single surgeon. Operating on fewer than two cases per month is questionable. Another problem seems to be the understanding of correcting the entire hindfoot in order to obtain the correct alignment of the foot±ankle complex. Still another problem is to stabilise the unconstrained resurfaced ankle joint. Solving these problems will certainly give better long-term results. The advantage of a failed ankle replacement is that it may be revised if there is sufficient bone stock, otherwise it can still be fused. The result of the last option is certainly better than if a hip or a knee replacement cannot be revised and requires arthrodesis. Wear problems in ankle replacements can be expected to continue with the available current materials, especially in malaligned and unstable ankles. It has been calculated that the amount of wear is the same as for a knee prosthesis (Kobayashi et al., 2004). New materials and surfaces may be foreseen, but with the complexity of the combined movements in the ankle joint, metal against metal is hardly one of them. Forgiving materials with no wear that can tolerate the impacts to the ankle joint are what we should look for. Creating a whole new surface replacement from stem cells may be an option, but as this is a politically hot potato in Western countries it will probably first be seen in Asia.

23.10 References Alvine FG. Total ankle arthroplasty: New concepts and approaches. Contemp Orthop 1991; 22(4): 397±403. Anderson T, Montgomery F, Carlsson A. Uncemented STAR total ankle prosthesis. Three to eight-year follow-up of fifty-one consecutive cases. J Bone Joint Surg Am 2003; 85A(7): 1321±1329. Backer M, Kofoed H. Passive ankle mobility. Clinical measurements compared with radiography. J Bone Joint Surg Br 1989; 71: 696±698. Buechel FF, Pappas MJ, Iorio LJ. New Jersey low contact stress total ankle replacement; biomechanical rationale and review of 23 cases. Foot Ankle 1988; 8(6): 279±290. Buechel FF Sr, Buechel FF Jr, Pappas MJ. Twenty years evaluation of cementless mobile-bearing total ankle replacements. Clin Orthop 2004; 494: 19±26. Carlsson A. Single-and double coated STAR total ankle replacements: a clinical and radiographic follow-up study of 109 cases. OrthopaÈde 2006; 35(5): 527±532. Carlsson A, Markusson P, Sundberg M. Radiostereometric analysis of the double-coated STAR total ankle prosthesis: a 3±5 years follow-up of 5 cases with rheumatoid arthritis and 5 cases with osteoarthrosis. Acta Orthop Scand 2005; 76(4): 573±579. Crock HV. The bones of the foot. In: The Blood Supply of the Lower Limb Bones in Man (Ed. HV Crock), E and S Livingstone, 1967, pp 72±79. Doets HC, Brand R, Nielsen RG. Total ankle arthroplasty in inflammatory joint diseases with use of two mobile-bearing designs. J Bone Joint Surg Am 2006; 88(6): 1272± 1284. Giannini S. Fresh osteochondral allograft ankle transplantation for ankle arthritis. Lecture 8th EFORT Congress. Firenze, 11±15 June, 2007. © 2008, Woodhead Publishing Limited

578

Joint replacement technology

Hamblen DL. Can the ankle joint be replaced? J Bone Joint Surg Br 1985; 67(5): 689±690. Haskell A, Mann RA. Perioperative complication rate of total ankle replacement is reduced by surgeons experience. Foot Ankle Int 2004; 25(5): 283±289. Helweg J, Kofoed H. The fibula rotates during motion in the ankle joint. In: Current Status of Ankle Arthroplasty (Ed. H Kofoed), Springer 1998, pp. 59±63. Hvid I, Rasmussen O, Jensen NC, Nielsen S. Trabecular bone strength profiles at the ankle joint. Clin Orthop 1985; 199: 306±312. Inman VT. The Joints of the Ankle. Williams and Wilkins 1976. Knecht SI, Estin M, Callaghan JJ, Zimmerman MB, Alliman KJ, Alvine FG, Saltzman CL. The Agility total ankle arthroplasty. Seven to sixteen-year follow-up. J Bone Joint Surg Am 2004; 86-A(6): 1161±1171. Kobayashi A, Minoda Y, Kadoya Y, Ohashi H, Takoka K, Saltzman C. Ankle arthroplasties generate wear particles similar to knee arthroplasty. Clin Orthop 2004; 424: 69±72. Kofoed H. Scandinavian total ankle replacement (STAR). Clin Orthop 2004; 424: 73±79. Kofoed H, Danborg L. Biological fixation of ankle arthroplasty. Foot 1995; 5: 27±31. Leardini A, Catani F, Giannini S, O'Connor JJ. Computer-assisted design of the sagital shapes of a ligament-compatible total ankle replacement. Med Biol Eng Comput 2001; 39(2): 168±175. Lord G, Marotte JH. Total ankle prosthesis. Technique and first results. A propos of 12 cases. Rev Chir Reparatrice Appar Mat 1973; 59(2): 139±151. Lord G, Marotte JH. Total ankle replacement. Rev Chir Reparatrive Appar Mat 1980; 66(8): 527±530. Lundberg A. Kinematics of the normal ankle joint. In: Current Status of Ankle Arthroplasty (Ed. H Kofoed), Springer 1998, pp. 3±9. Meehan R, McFarlin S, Bugbee W, Brage M. Fresh ankle osteochondral allograft transplantation for tibiotalar joint arthritis. Foot Ankle Int 2005; 26(10): 793±802. Mishra V, Wood PLR. A prospective randomized trial of two mobile bearing total ankle replacement implant designs. # 68. American Academy of Orthopaedic Surgeons 74th Annual Meeting. 14±18 Feb. 2007. San Diego. Schill S, Biehl C, Thabe H. Ankle prostheses. Midterm results of Thompson±Richards and STAR. OrthopaÈde 1998; 27(3): 183±187. Spirt AA, Assal M, Hansen ST Jr. Complications and failure after total ankle arthroplasty. J Bone Joint Surg Am 2004; 86-A(6): 1172±1178. Takakura Y, Tanaka Y, Sugimoto K, Tamai S, Masuhara K. Ankle arthroplasty. A comparative study of cemented metal and uncemented ceramic prostheses. Clin Orthop 1990; 252: 209±216. Takakura Y, Tanaka Y, Kumai T, Sugimoto K, Ohgashi H. Ankle arthroplasty using three generations of metal and ceramic prostheses. Clin Orthop 2004; 424: 130±136. Tohma Y, Tanaka Y, Ohgushi H, Kawate K, Taniquchi A, Hayashi K, Isomoto S, Takakura Y. Early bone in-growth ability of alumina ceramic implants loaded with tissue-engineered bone. J Orthop Res 2006, 24(4): 595±603. Wood PLR, Deakin S. Total ankle replacement. J Bone Joint Surg 2003; 85-B: 334±341. Zerahn B, Kofoed H. Ankle arthroplasty. Bone mineral density, gait analysis, and patient satisfaction before and after ankle arthroplasty. Foot Ankle Int 2004; 25(4): 208± 214.

© 2008, Woodhead Publishing Limited

24

Replacing shoulder joints L D E W I L D E , University Hospital of Ghent, Belgium

24.1

Introduction

The first successful shoulder replacement was performed in 1892 by Jules E. Pean. The indication was a painful shoulder destroyed by tuberculous arthritis.1 The implant consisted of an iridescent platinum tube, a hardened rubber ball and two metal loops that attached the ball to the scapula and platinum tube.1,2 The implant had to be removed due to infection two years later. It was the first successful attempt at human joint replacement antedating the first hip replacement by 26 years. Since the early 1950s, shoulder arthroplasty has been modernised by Neer et al. The original humeral head replacement of Charles Neer II, was designed for the treatment of three and four-part proximal humeral fractures,3 but later on it proved to be also clinically effective for the treatment of osteoarthritis, rheumatoid arthritis and avascular necrosis.4±6 In an attempt to further improve results Neer developed a total shoulder arthroplasty with an all-polyethylene cemented glenoid component for glenoid resurfacing.7 Neer's arthroplasty proved to be effective and relatively durable for the treatment of glenohumeral arthritis8 and it became the model for the design of almost all current modern unconstrained shoulder arthroplasties. Since the 1970s, constrained and semi-constrained shoulder prostheses, such as the so-called reverse ball-and-socket design, have been employed as a solution to the vexing problem of glenohumeral arthrosis associated with glenohumeral instability secondary to a functionally impaired or an anatomically deficient rotator cuff.9±17 Nowadays, despite a number of complications, the reverse prosthesis,18 designed by Grammont,19 is felt to be the treatment of choice if anatomical reconstruction of the soft tissues (rotator cuff tendon tears with muscular fatty degeneration) and the bone (severe malunion of the proximal humerus) is no longer feasible20 and if the deltoid muscle is still functioning. Basically, the shoulder prosthesis is composed of a humeral stem with a humeral ball proximally and a polyethylene cup at the glenoid side. At the humeral side the first-generation Neer prosthesis has a fixed head inclination, © 2008, Woodhead Publishing Limited

580

Joint replacement technology

one radius of curvature (44 mm), and two head thicknesses. The secondgeneration total shoulder design developed after the Neer prosthesis is modular, allowing the stem to accommodate heads of variable diameter and neck length, which theoretically achieves better soft-tissue tensioning. The third-generation prosthetic design also is modular, but the option to offset the head from the centre of the stem and sometimes with a variable inclination, has been introduced to match the anatomical dimensions of a human proximal humerus more appropriately. Recently, the stemless so-called resurfacing humeral head prosthesis offering the same advantages as the third-generation prosthesis has become more popular because of its simplicity.21 At the glenoid side prosthetic glenoid design parameters such as implant material, implant shape, method of fixation and use of metal backing have been evaluated by means of experimental and clinical outcome studies.22±29 Despite an abundance of these studies, there is no consensus on the ideal glenoid component design and it remains unclear which type of component (cemented or uncemented, keeled or pegged, convex back or flat back) has the lowest loosening rate and hence the best potential for good long-term results.

24.2

Biomechanics of total shoulder arthroplasty

The shoulder joint is a subtle equilibrium between stability and mobility. To combine good mobility with proper stability, the shoulder joint can be theoretically considered as a two-mating spheres joint (Fig. 24.1).30 The spherical proximal humeral convexity is formed by the tuberosities and the superficial surface of the rotator cuff tendons. It has a radius, R, and articulates with the undersurface of the coracoacromial arch. The spherical humeral articular surface has a radius, r, and articulates with the glenoid. The

24.1 Concentric spheres concept of Matsen: the subacromial sphere and the humeral head sphere have the same centre of rotation. © 2008, Woodhead Publishing Limited

Replacing shoulder joints

581

space between the two radii is filled by the height of the tuberosities and the thickness of the cuff tendons. The centres of the two spheres and the concavities with which they articulate must be superimposed in normal shoulder function. In anatomical shoulder replacement, the surgeon must reproduce an anatomical centre of rotation by a perfect placement of the prosthetic components. Besides this anatomical restoration, normal gliding between the glenoid and the humeral head and between the rotator cuff tendons and the coracoacromial undersurface (in the subacromial space) is necessary to optimise prosthetic shoulder function. The biomechanical movement of the shoulder joint is a spinning rotation. Because this spinning rotation needs to be optimal both at the subacromial space to prevent subacromial impingement and at the glenoid side to prevent a rocking-horse phenomenon (the compression of the humeral head on the opposite edge creates a source of tensile loading), selection of properly designed and anatomically sized components and meticulous placement are essential.31 Modularity in head size and neck length alone (second-generation prosthesis) is insufficient for replication of the proximal humeral anatomy. A slight underestimation of the humeral head is still advised because too large a head prejudices the biomechanics by overtensioning the joint, thereby limiting mobility and creating more pressure on the glenoid cartilage. This results in glenoid wear with a hemi-arthroplasty or in polyethylene wear if the glenoid is resurfaced. A smaller head size will also prevent early rupture of the subscapularis repair due to excessive tension in the rotator cuff, finally leading to early anterior instability and subsequent supraspinatus stretching or tearing. If the head size is too large and the humeral implant too high, abductors (subscapularis and infraspinatus) can be converted into adductors because these features displace the centre of the head more superiorly.32 Inadequate retroversion may lead to eccentric loading at the periphery of the glenoid, which may increase glenoid wear with subsequent loosening.33 Moreover, because the articular surface of the head is mainly with a posterior and medial offset to the medullary axis,34±37 the need for a prosthesis with the option to adjust these offsets is becoming more and more stringent (Fig. 24.2). These third-generation shoulder prostheses have the theoretical advantage that the glenohumeral center of rotation can be reconstructed more accurately. Hertel et al.38 confirmed that the anatomy of the proximal humerus showed a wide range for variables such as medial offset and greater tuberosity offset, but was surprisingly constant for inclination and relative dimensions of the head. The implications for the third-generation prosthetic shoulder design are as follows: stem design and insertion should respect the insertion facet of the supraspinatus (note: the lateral fin can damage the tendon), head inclination should be constant at 130ë (adaptability of the varus/valgus is less important), only one head height per radius is required (less risk of overstuffing the joint), and the medial and posterior offset should be adjustable (if medial and posterior © 2008, Woodhead Publishing Limited

582

Joint replacement technology

24.2 Third-generation prostheses have the ability to offset the humeral head from the axis of the humeral stem in a posterior and medial direction. This allows better coverage of the humeral osteotomy.

offset need to be independent of each other, two different eccentric adjustments are necessary ± Fig. 24.3). The stemless humeral resurfacing prosthesis39 obviates the need for these adjustments and offers the additional advantage of minimal removal of bone, which facilitates revision surgery and eliminates the potential risk of fracture of the humerus because of the avoidance of stress risers. Irrespective of a stemmed or stemless design, the selection of an appropriately sized head is important since biomechanical experiments have shown that a 5±10 mm change in the centre of rotation results in a significant reduction of the lever arms of the deltoid and rotator-cuff muscles during abduction.31,40 A perfect visualisation of the anatomical neck of the humeral head, after resection of the osteophytes, is mandatory to perform a correct bone cut in the case of a third-generation prosthesis, or reaming in the case of resurfacing arthroplasty. The anatomical neck is actually suggested as the humeral landmark for reconstruction of the humeral head. This is followed by meticulous measurement of the humeral head size by means of a calliper so that the proper size can be selected. Although not yet clinically proven, the third-generation prosthesis and the resurfacing prosthesis are felt to be valuable tools for the orthopaedic surgeon to reconstruct the humeral side more anatomically. At the glenoid side no clear guidelines are available yet. Resurfacing of the glenoid is technically demanding surgery. Many alterations in component design, surgical techniques and cementing techniques have already provided significant improvements. From a biomechanical point of view, glenoid resurfacing is crucial because every degree of glenoid retro-/anteversion, varus/valgus angulation will displace the centre of the humeral component posteriorly by 0.5 mm, and vice versa (Fig. 24.4).41 © 2008, Woodhead Publishing Limited

Replacing shoulder joints

583

24.3 Third-generation prostheses require an adjustable posterior and medial offset. If both offsets need to be independent, a double eccentricity is necessary.

Today's gold standard for primary glenoid replacement is a cemented, allpolyethylene component of 2.5 mm thickness. There is still controversy about the size and shape, and whether the component should be curved or flat, keeled or pegged, metal backed or full polyethylene, cemented or uncemented, conforming or non-conforming.7,42±46 Besides that, the surgical technique of glenoid implantation is subject to a high variability.47 Recent studies have defined surgical anatomical points of reference which can guide the surgeon in optimising the glenoid prosthetic position.47±49 The surgeon should balance the © 2008, Woodhead Publishing Limited

584

Joint replacement technology

24.4 The glenoid retroversion must be very accurate because a 0.5 mm displacement of the centre of rotation is created per degree of version.

expected clinical short- and medium-term benefits and the increased risk of preoperative and postoperative morbidity against each other if the glenoid component is implanted. Because placing the glenoid component is technically more difficult, this matter becomes a highly individual surgically determined parameter.50,51 In cases of severe instability or massive rotator cuff tears where soft-tissue reconstruction is no longer feasible, it seems wise not to try to restore an anatomical configuration. In this non-anatomical setting a reverse shoulder prosthesis seems to be indicated20,52 because of its different biomechanical behaviour.52,53 Grammont et al.54 introduced two major innovations in the reverse prosthesis, notably a large glenoid hemisphere with no neck and a small almost horizontally inclined (155ë) humeral component covering less than half of the hemisphere (Fig. 24.5). These non-anatomical features are essential to obtain stability and to restore active elevation over 90ë in patients with a cuffdeficient shoulder. An important limitation of this non-anatomical prosthesis is its inability to restore active internal and external rotation; this is caused mainly by design limitations of the prosthesis,55,56 producing mechanical impingement (Figs 24.6 and 24.7), and a malfunctioning teres minor, respectively.12,14,15,53 This can also be explained by the slackening of the remaining external rotators © 2008, Woodhead Publishing Limited

Replacing shoulder joints

585

24.5 The reverse prosthesis has introduced two major innovations: a large glenoid hemisphere with no neck and a small humeral component inclined almost horizontally (155ë) and covering less than half of the hemisphere.

due to the medialisation of the centre of rotation.53 Whereas the anatomical prosthesis has a spinning rotation requiring minimal room for movement, the reverse prosthesis has a hinged rotation where additional room in three dimensions is required to allow movement (Fig. 24.7). The best known expression of this requirement is scapular notching,53,57 where a mechanical impingement exists between the medial border of the humeral implant and the inferior rim of the glenoid53,58,59 as a consequence of lack of space inferiorly. The mechanical impingement also exists in the transverse plane of the body, thereby limiting the range of external and internal rotation and possibly leading to mechanical prosthetic failure.55 The additional space around the prosthesis is believed to be responsible for the higher rate of infection in reverse total shoulder arthroplasty.14,18,53 A positive effect is the lengthening of the arm;53 this enhances the strength of the deltoid muscle,60 which has been shortened by longstanding cuff tear arthropathy. The strength of the deltoid is also increased by the medialisation of the centre of rotation, thereby recruiting more fibres of the anterior and posterior deltoid to act as abductors. The Delta 3.1 reverse prosthesis (DePuy International Ltd) has a medialisation of the joint rotation centre by 28.5 mm as the major reason for improvement of the deltoid moment arm. An additional lateralisation (from 5 mm preoperatively to a mean of 15 mm postoperatively) of the humeral shaft to enhance the moment arm61 is of no major importance. Although the descend© 2008, Woodhead Publishing Limited

586

Joint replacement technology

24.6 The reverse prosthesis requires extra room for movement. If not present due to malunion and/or scar tissue, motion will be blocked.

ing forces of the deltoid muscle are increased by the cable-pulling effect induced by increased lateralisation of the humeral shaft,62 this prosthesis seems to function well with a significant decrease of the wrapping angle (from 52.5ë preoperatively to 12.4ë postoperatively). Elongation of the deltoid muscle (116.4%) contributes to enhancement of the maximal deltoid moment.63 Thus, the reverse prosthesis creates a new biomechanical environment, in which the deltoid muscle, with more than a doubled moment (Fig. 24.8),60 can become the motor unit of the shoulder despite the lack of rotator cuff muscles. Because of the hinged rotation, the lengthening of the remaining rotator cuff is highly dependent on the spatial positioning of the arm as well as on the positioning of the prosthetic components. Clear guidelines for optimal function have not yet been established yet.64 © 2008, Woodhead Publishing Limited

Replacing shoulder joints

587

24.7 The reverse prosthesis requires extra room for movement. Internal rotation will be limited due to a conflict between the anterior wall of the scapula and the proximal humeral prosthesis (gs, glenosphere; he, humeral epiphysis; hd, humeral diaphysis).

24.8 Moment of the deltoid muscle of the reverse total shoulder prosthesis versus a normal shoulder for the different degrees of abduction in the scapular plane of the body. © 2008, Woodhead Publishing Limited

588

24.3

Joint replacement technology

Indications for total shoulder arthroplasty

24.3.1 Osteoarthritis Symptomatic osteoarthritis of the glenohumeral joint is relatively uncommon; about one shoulder arthroplasty is performed for every ten hip arthroplasties. The exact cause and mechanism of primary osteoarthritis, resulting from a breakdown of articular cartilage, are unknown. Patients develop cartilage erosions, flattening of the joint surface, osteophytes and asymmetric wear of the posterior glenoid. Soft-tissue changes include contractures of the anterior capsule and subscapularis, and a patulous, attenuated posterior capsule. Rotator cuff tears are uncommon except for cuff tear arthropathy, which is defined as a separate clinical entity and will be discussed later. According to the work of Walch,65 defining the amount and the type of posterior glenoid wear has therapeutic consequences. Three types with different degrees of posterior wear and subluxation are distinguished (Fig. 24.9). Moderate and severe posterior glenoid wear (types B2 and C) can compromise anatomical reconstruction with a total shoulder arthroplasty. A hemi-arthroplasty or even non-anatomical surgery are more reasonable options in these cases. Prospective studies are necessary to define the irreparability of the anatomical configuration.

24.3.2 Cuff tear arthropathy Longstanding and large (at least two irreparable retracted tendons) tears of the rotator cuff can result in superior migration of the humeral head. Subsequent erosion of the glenohumeral joint, acromioclavicular joint and undersurface of the acromion, as well as an acetabularisation of the humeral head can be seen on radiographs, indicating cuff tear arthropathy. This destruction can be explained by a chronic loss of the centring effect of the rotator cuff on the humeral head in the glenoid vault resulting in a superior subluxation. Similar to the posterior subluxation of the rotator cuff-intact osteoarthritic shoulder, a classification of the glenohumeral wear with therapeutic consequences can be made (Fig. 24.10).66 A hemishoulder arthroplasty with a `femoralised' extension (global

24.9 Diagram showing the classification of glenoid arthritis according to Walch: type A, concave (A1, moderate; A2, severe), type B, biconcave (B1, moderate; B2, severe), and type C, dysplastic (more than 25ë of retroversion). © 2008, Woodhead Publishing Limited

24.10 Seebauer's classification of cuff tear arthropathy: type IA, acetabularisation; type IB, acetabularisation and medialisation; type IIA, ascension with intact coracoacromial arch (no instability); type IIB, ascension with broken coracoacromial arch (superior instability). © 2008, Woodhead Publishing Limited

WPTF3007

590

Joint replacement technology

advantage cuff tear arthroplasty (CTA)) instead of a total shoulder arthroplasty, to overcome early glenoid loosening, is proposed to relieve pain and to achieve modest functional improvement in this patient group. A stemless resurfacing arthroplasty in `valgus' can be performed to treat type IA, type IB and probably type IIA as well. When the superior glenoid wear worsens (types IIA and B) and for the pseudoparalytic shoulder anatomical reconstruction is no longer feasible and a reverse total arthroplasty is indicated. Further studies are necessary to define if the superior instability is beyond repair.

24.3.3 Rheumatoid arthritis Rheumatoid arthritis can affect the shoulder joint as part of a systemic synovialbased polyarticular disorder. The disease can cause severe pain and disability. Modern medical treatment has reduced the progression of the disease. In rheumatoid arthritis patients undergoing shoulder arthroplasty the most symptomatic joint needs to be treated first. If multiple joints are involved, a staged plan has to be developed. In the arm the proximal joint must be addressed first. In the case of a shoulder arthroplasty determination of the length of the stem must take into account that an elbow arthroplasty will probably be performed at a later date. Therefore it is probably wise to consider a resurfacing arthroplasty of the shoulder, which yields equally favourable results67 as the stemmed design and is technically less invasive (Fig. 24.11). The humeral head size should be anatomically correct because an oversized head results in improper soft tissue tension with limitation of motion and pain. The results of hemi-arthroplasty are satisfactory, but in an appropriately selected patient with rheumatoid arthritis a total shoulder arthroplasty may be considered since the rate of glenoid loosening and failure is no different from that of a patient with osteoarthritis.68 For preoperative evaluation of the glenoid bone stock a computed tomography (CT) scan is performed. If glenoid wear, which is mainly central and superior, is too excessive, resurfacing of the glenoid is not an option. A careful preoperative examination of the rotator cuff is mandatory. The incidence of rotator cuff tears in a rheumatoid patient with advanced arthritis of the glenohumeral joint is 20%.68±70 If the rotator cuff is irreparable, a reverse prosthesis can be considered but the surgeon must be able to address the complications, which are more frequent in this population.13

24.3.4 Osteonecrosis Chronic systemic corticosteroid use and trauma are the main causes of osteonecrosis of the humeral head.71 Total or hemi-shoulder arthroplasty for osteonecrosis perform poorly compared with results in osteoarthritic indications.72,73 The worst results are achieved in post-traumatic osteonecrosis with a history of previous fixation attempts.72 No difference was noted between hemi- and total © 2008, Woodhead Publishing Limited

Replacing shoulder joints

591

24.11 Rheumatoid patients are probably better treated with a stemless prosthesis because it facilitate revision surgery, eliminates the potential risk of fracture of the humerus by avoiding of stress risers in the presence of poor bone quality, and permits elbow arthroplasty at a later date.

shoulder arthroplasty.72,73 Because glenoid wear predictably progresses over time and produces pain, primary replacement of the glenoid is preferred to hemiarthroplasty, especially in older patients. Because managing severe glenoid articular wear in young patients is challenging (loosening of the glenoid, polyethylene disease) it is probably wise to treat these patients early with a HemicapÕ (Fig. 24.12) partial resurfacing prosthesis before glenoid wear occurs (Davidson PA, personal communication).

© 2008, Woodhead Publishing Limited

592

Joint replacement technology

24.12 HemicapÕ implanted for avascular necrosis before the occurrence of degenerative changes in the glenohumeral joint.

24.3.5 Acute proximal humeral fractures Since Neer reported his results in 197074 prosthetic replacement has become a well-accepted method of treatment for selected three- and four-part fractures of the proximal humerus. Although a favourable outcome in terms of pain relief can be expected, function is often not satisfactory. The key to success of a hemiarthroplasty for proximal humeral fractures is a functional rotator cuff mechanism. Traditionally the humeral shaft is built up with an articular surface positioned at an adapted height and version. This converts a four-part fracture into a three-part fracture with the tuberosities still left to be fixed around the humeral head. This step appears to be the most difficult to accomplish. In the literature the main complication responsible for poor results is loss of fixation of the greater tuberosity.75±81 Several factors can lead to this complication, including malpositioning of the prosthetic component, inadequate positioning and/or fixation of the tuberosity, no bone grafting, misjudgement of an `anatomical reduction' and overaggressive rehabilitation.51,75,76 To achieve better and more predictable results, new prosthetic devices have a less bulky prosthetic neck, which permits generous bone grafting. Height and retroversion can be estimated more accurately with advanced implant instrumentation. The prosthesis should © 2008, Woodhead Publishing Limited

Replacing shoulder joints

593

be positioned in approximately 20ë to 30ë of retroversion (relative to the humeral epicondylar axis), and the head height should be 0.5 cm above the level of the greater tuberosity. An essential step in securing the tuberosities around the neck of a prosthesis for trauma is the use of circular cerclage wires. Biomechanical studies have shown that cerclage sutures around the shaft of the prosthesis (Fig. 24.13) improve stability, especially if the reduction is anatomical.82,83 Obviously, a prerequisite is that the prosthesis has a smooth medial border on the metaphysis of the humeral component, since a slightly rough surface can result in breakage of cerclage sutures. A new and promising concept to treat proximal humerus fractures is the DuocentricÕ fracture system. This system differs from the traditional trauma prostheses in that it secures the bone±tendon junction of the rotator cuff, which often remains the only useful landmark, to the holes in the enlarged rim of the

24.13 Securing the tuberosities around the neck of a (trauma) prosthesis with circular cerclage wires is an essential step to improve the fixation of the tuberosities (TM, greater tuberosity; tm, minor tuberosity; phh, prosthetic humeral head; pd, prosthetic diaphysis; IS, infraspinatus; subsc, subscapularis). © 2008, Woodhead Publishing Limited

594

Joint replacement technology

prosthetic head.84 By creating an anatomical soft-tissue repair to the prosthetic head, the reconstruction is built up from the glenoid, rather than from the humeral diaphysis. This allows an anatomical reconstruction of the glenohumeral unit. Restoring the glenohumeral unit first reduces a complex four-part fracture to a two-part fracture, overcoming the difficulties with accurate version and offset. The prelimary clinical results seem to be very promising (Berghs B, personal communication) (Fig. 24.14).

24.14 The DuocentricÕ trauma prosthesis is a new system that fixes the bone± tendon junction of the rotator cuff, which often remains the only useful landmark in fractures of the proximal humerus, to the holes in the enlarged rim of the prosthetic head. It allows rapid rehabilitation: clinical function at three weeks postoperatively. © 2008, Woodhead Publishing Limited

Replacing shoulder joints

595

24.15 Boileau's classification of sequelae of proximal humerus fractures with therapeutic consequences: types 1 and 2 can be treated with a third-generation total shoulder prosthesis or a resurfacing prosthesis because no osteotomy of the greater tuberosity is required, type 3 is best treated with centromedullary grafting and a plate fixation and type 4 is best treated with a reverse shoulder arthroplasty.

24.3.6 Sequelae of proximal humerus fractures Boileau et al.85 identified four types of fracture (Fig. 24.15). Type 1 are sequelae of impacted fractures with humeral head collapse or necrosis, type 2 are irreducible dislocations or fracture-dislocations, type 3 are non-unions of the surgical neck, and type 4 are severe tuberosity malunions. The results of unconstrained shoulder arthroplasty for types 1 and 2 sequelae were predictably good because no greater tuberosity osteotomy was performed. The distorted anatomy was accepted, and both the prosthesis and technique were modified accordingly. Total shoulder arthroplasty yielded better results than hemi-arthroplasty. Patients with type 3 or type 4 sequelae had poor functional results with unconstrained arthroplasty because a greater tuberosity osteotomy was required. These authors85 discourage the use of unconstrained prostheses for types 3 and 4 fracture sequelae. They suggest peg bone grafting86 or low-profile fracture prosthesis for patients with type 3 sequelae and reverse arthroplasty for those with type 4 sequelae.

24.4

Surgical technique

24.4.1 Humeral side Arthroplasty is performed in the beach-chair position and most commonly through a deltopectoral approach. All the adhesions in the subdeltoideal space need to be released. The nerve-to-nerve release as described by Matsen and Lippitt87 is very useful, especially if previous shoulder surgery has been © 2008, Woodhead Publishing Limited

596

Joint replacement technology

performed. The surgeon can then evaluate the subacromial status of the rotator cuff. Magnetic resonance imaging (MRI) or arthro-CT scan is used to evaluate the extent of the tear and the degree of fatty atrophy of the rotator cuff muscles, allowing some preoperative prediction of tear repair feasibility. If irreparable tears are present, or if a repair is tenuous and unlikely to remain intact, the glenoid component should not be implanted. The location of the long head of the biceps tendon is determined. Because this tendon is abnormal in at least 30% of patients with primary osteoarthritis, showing delamination, pre-rupture or hypertrophy,88 a biceps tenodesis at the pectoralis tendon or in the intertubercular groove offers better pain relief and it is advised to routinely perform this procedure as part of a hemi- or total shoulder arthroplasty. If the rotator cuff is intact, the joint is best accessed by means of an osteotomy of the lesser tuberosity.89 This technique provides an easy anterior approach for total shoulder replacement and is followed by consistent bone-tobone healing which can be monitored radiologically, as well as resulting in good clinical subscapularis function. The postoperative appearance of fatty degeneration in the muscle remains unexplained (possibly due to the subscapularis release). The diseased synovium and the anterior, thickened capsule must and can be excised without a risk of instability. The inferior capsule, including the attachment of the long head of the triceps tendon, should be released to allow adequate exposure of the glenoid and to ensure sufficient mobility of the subscapularis muscle. If a hemi-arthroplasty is considered, the labral structures are better left in place to enhance stability of the shoulder joint. With the subscapularis tendon detached, an anterior dislocation of the humeral head and an extreme external rotation of the upper arm can be performed. This allows a clear visualisation of the humeral osteophytes, which are almost always located inferiorly. All osteophytes should be removed so that the capsular insertion is visualised all around the anatomical neck, providing an important surgical landmark (Fig. 24.16). Then, the size of the humeral head should be measured with the use of callipers. If a stemmed humeral prosthesis is implanted, the proximal humeral osteotomy should be performed at the anatomical neck with or without a normal humeral retroversion of 20±30ë. It has not yet been established whether the anatomical neck or a mean humeral retroversion is the best option for humeral osteotomy.90 If a resurfacing prosthesis is considered, the anatomical neck serves as the anatomical landmark for reaming with the selected size. Humeral stem reaming is done according to the instructions provided by the manufacturer of the third-generation implants and taking care not to damage the insertion of the supraspinatus tendon. Then, the humeral head is replaced in the most anatomical position with an adapted medial and posterior offset. Finally the humeral head can be implanted cemented or cementless at the appropriate height, after which the lesser tuberosity osteotomy is closed with double-row suturing91 or a circular cerclage around the humeral stem.83,92 © 2008, Woodhead Publishing Limited

Replacing shoulder joints

597

24.16 All osteophytes should be removed so that the capsular insertion is visualised all around the anatomical neck, providing an important surgical landmark.

24.4.2 Glenoid side If a total shoulder arthroplasty is considered93 it is advisable to resect the labrum instead of sparing it, all around the glenoid rim. The origin of the long head of the biceps tendon is resected as well, after the humeral head has been osteotomised. Then, the remaining cartilage can be scratched off so that the bony glenoid becomes clearly visible. In doing so the surgeon will find that the peripheral rim of the inferior quadrants of the articular surface is located on a circle.92,94,95 Defining the centre of this circle (M) appears to be more reliable (ICC 0.98) than determining the midpoint of the longitudinal axis (0,0) between the most cranial point and the most caudal point, defined as Saller's line (AB) (ICC 0.89) (Fig. 24.17).92 This centre can then be used as a landmark for reaming if glenoid reaming is guided by a K-wire or peg. If a slot is preferred, less accuracy is needed to prepare the implantation of the keeled glenoid. Mechanical reaming of the glenoid is biomechanically better than manual preparation.24 The glenoid must be prepared according to the manufacturer's technique. As the goal of glenoid resurfacing is to provide better stability, © 2008, Woodhead Publishing Limited

598

Joint replacement technology

24.17 Defining the centre of the circle (M) formed by the rim of the inferior quadrants of the glenoid is more accurate than defining the centre of the glenoid (crossing point (O,O) of greatest longitudinal (AB) and anteroposterior (CD) axis.

motion and pain relief, this must be weighed against the complications of clinical glenoid loosening and the possibility of revision surgery. These complications are more frequent with less surgical experience and poor surgical exposure. Moreover, the surgery at the glenoid becomes more difficult with excessive glenoid retroversion (types B and C according to Walch et al.96). The surgeon should plan to correct the version of the glenoid to neutral or risk postoperative posterior instability. Adjustment of humeral retroversion does not correct the tendency of the prosthesis to slide posteriorly. The surgeon needs to decide whether a correction of the version by eccentric reaming or glenoid reconstruction with a bone graft is appropriate and feasible (Fig. 24.18). Most patients have only mild (<1 cm) posterior wear, and the glenoid version can be corrected by preferentially removing the relatively higher anterior bone. Reconstruction is more difficult and yields more inconsistent results than eccentric reaming.97,98

24.4.3 Reverse total shoulder arthroplasty When the humeral head is no longer stabilised in the shoulder socket due to severe irreparable cuff tearing or when normal anatomy of the shoulder cannot be restored without an osteotomy of the greater tuberosity,85 the reverse © 2008, Woodhead Publishing Limited

Replacing shoulder joints

599

24.18 Correction of the glenoid version by reconstruction (B) or by eccentrical reaming (C) is necessary in type B and C glenoids according to Walch.

shoulder prosthesis can help restore comfort and some shoulder function.9±17 In milder cases (types Ia and Ib of Seebauer's classification66) of rotator cuff tear arthropathy, a hemi-shoulder prosthesis with a cranial extension (Cuff Tear Arthropathy, CTA, head, Fig. 24.19) may be indicated. The latter is implanted similarly to the anatomical prosthesis, without an osteotomy of the lesser tuberosity but with an extra osteotomy of the greater tubercle, according to the manufacturer's technique. It creates a smooth gliding underneath the remaining subacromial arch (Fig. 24.18) and the clinical results are similar to those of tuberoplasty.99,100 In severe rotator cuff tear arthropathy (types IIa and IIb of Seebauer's classification66) a reverse prosthesis is indicated.12,14,15 There is some debate as to which approach is the best for implantation of a reverse shoulder prosthesis. The deltopectoral approach has the advantage of being well known and offering a nice view of the inferior part of the glenoid, but if the glenoid is retroverted, glenoid placement becomes difficult and damage to the deltoid muscle can occur. Moreover, as the rotator cuff lesion mainly involves the superoposterior side in cuff tear arthropathy, a (partial) desinsertion (taking of the tendon from the bony insertion) of the remaining subscapularis is © 2008, Woodhead Publishing Limited

600

Joint replacement technology

24.19 Extended humeral prosthetic head for the treatment of mild cuff tear arthropathy (CTA type IA and IB according to Seebauer and coworkers66).

often required through this approach. Therefore the antero-superior MacKenzie approach101 is preferred to the classical deltopectoral approach in those cases. The deltoid is split along the direction of its fibres, taking care to avoid the axillary nerve that runs on its inferior surface (Fig. 24.20) at the turning fold of the subacromial bursa. Opening the shoulder reveals the irregular joint surface of the humeral head. This surface is resected using a cutting guide. Instead of duplicating the normal 30ë retroverted direction of the humeral head joint surface, the head is cut so that it faces directly medially (0±10ë of retroversion).64 The shaft of the humerus is prepared to accept the stem of the

24.20 The antero-superior MacKenzie approach101 to the shoulder joint is indicated in postero-superior cuff tear arthropathy. © 2008, Woodhead Publishing Limited

Replacing shoulder joints

601

humeral component according to the technical instructions of the manufacturer. Then, the glenoid must be exposed completely, especially the inferior part. A surgical desinsertion of the inferior capsule including the long head of the triceps tendon is recommended. Again, the midpoint of the circle formed by the rim of the inferior quadrants of the glenoid is used as the landmark for a pin driven into its centre to act as a guide. This technique ensures the most distal placing of the base plate in an effort to decrease the incidence of inferior scapular notching.58,59 The glenoid surface is drilled and reamed to a flat surface. The base of the base plate is inserted on the prepared surface and fixed into position using four screws. The inferior and superior screws lock into the base plate (metaglene), the anterior and posterior screws are non-locking. An improvement has recently been made with the locking mechanism of the superior and inferior screws allowing an adaptable angulation of the screws, thereby preventing screw placement outside the scapular bone. A trial ball (glenosphere) is placed on the metaglene. Then, the humeral trial component is inserted (this step facilitates glenoid preparation and surgical dissection of the inferior glenoid) and a trial cup is placed on the humeral component allowing the surgeon to examine the shoulder for stability. If softtissue tension is insufficient for stability, an extended neck can be added to the humeral prosthesis. Once the optimal trial components and the positions of the humeral and glenoid components have been verified, the final glenosphere is screwed onto the metaglene and the final humeral socket is fixed to the humeral prosthesis. At the completion of this procedure, if it has been performed through a deltopectoral approach, the subscapularis tendon, if present, is securely repaired to the bone. Any remnants of the posterior cuff can be sutured to the greater tuberosity to improve active external rotational power or a SeverL'Episcopo tendon transfer102 can be performed.

24.5

Complications

Total shoulder arthroplasty can be associated with a multitude of complications, the most common of which include prosthetic loosening, glenohumeral instability, periprosthetic fracture, rotator cuff tears, infection, neural injury and deltoid muscle dysfunction. Glenoid component loosening , the commonest complication of glenoid replacement, has continued to be an unresolved problem.18,103 Recent advances in glenoid component design and fixation, and alternatives such as biological resurfacing104,105 with meniscal allograft tissue105 have yielded satisfactory short- to medium-term results and should be considered for the young arthritic shoulder patient (post-instability). Press-fit humeral stem prostheses have a higher rate of periprosthetic fracture than cemented prostheses.106 Overaggressive rehabilitation may lead to reflex sympathetic dystrophy and stiffness, or to rupture of subscapularis and anterior instability.51 © 2008, Woodhead Publishing Limited

602

Joint replacement technology

The probability of anatomical implant survival is reported to be 93% after 10 years and 87% after 15 years. The glenoid revision rate remains low, 5.6% at a mean 12.2 years' follow-up.107 The survival rate of the reverse prosthesis with replacement of the prosthesis and glenoid loosening as the end points was 91% and 84%, respectively, at 120 months. Shoulders that had an arthropathy with a massive rotator cuff tear showed a significantly better result than those that had a disorder with a different aetiology such as rheumatoid arthritis, trauma or revision arthropathy.108 Two recently published studies have found relationships between the number of surgeries performed (surgical volume) and shoulder replacement (both total and partial shoulder replacement). A significantly higher risk of complications, a longer hospital stay and increased total hospital charges have been reported after hemi-shoulder and total shoulder replacement performed by low-volume surgeons compared with high-volume surgeons.109,110

24.6

Prognostic factors for clinical outcome

Shoulder arthroplasty provides good to excellent pain relief and functional improvement. A number of factors have been found to affect the outcome of unconstrained shoulder arthroplasty for osteoarthritis.85,86,111,112 The postoperative results are dependent on the status of the rotator cuff tendon. Small degenerative tears of the supraspinatus did not have any significant effect on the outcome.111,113 It is recommended to tenodese (attached to the bone or surrounding structure so it cannot move or glide like a normal tendon) the long head of the biceps tendon because it is mostly partially torn and thickened, as well as being painful if not treated.88 The degree of fatty infiltration of the cuff muscles has a direct impact on the functional outcome. Severe fatty infiltration of the infraspinatus (Goutallier stage 3 or 4) is associated with less favourable functional results. These patients had a mean Constant score of only 54 points with a mean active elevation of 104ë, as opposed to 72 points and 147ë when the infraspinatus was normal (stage 1).51,88 Glenoid deformation significantly determines function. Functional results are significantly poorer in a dysplastic glenoid with only active elevation to 130ë in a type-C configuration, and the complication rate is significantly higher with a bioconcave shape.88 The results of unconstrained shoulder arthroplasty vary according to the underlying pathology (Fig. 24.21). Glenoid implantation produces better pain relief in patients with glenoid cartilage wear. The results of unconstrained shoulder arthroplasty depend on the quality of the anatomical reconstruction. Recent reports have suggested that inferior results may be due to component malpositioning or the use of nonanatomical prostheses.113 Third-generation implants or resurfacing humeral head prostheses allow more accurate recreation of proximal humeral anatomy and deliver higher performance.51 Total shoulder replacement was superior in terms of pain relief, active range of movement, activity scores and patient © 2008, Woodhead Publishing Limited

Replacing shoulder joints

603

24.21 Constant score versus ethiopathology. The Constant score is a measuring scale for the surgeon to evaluate in an objective manner the function of the shoulder pre- and postoperatively.

satisfaction, and had comparable complication rates.114 The experience of the surgeon appears to be a major determinant of functional outcome, complication rate and duration of hospital stay.109,110 Humeral prosthetic implantation for fractures remains difficult because the reattachment of the tuberosities may not be permanent, especially in elderly patients.75±79 For that reason some surgeons opt for a reverse total shoulder arthroplasty in these patients.115,116 At a mean follow-up of 86 months, the Constant score was 60 points (contralateral shoulder 83 points). Patients with painful arthritis due to a complete, irreparable rotator cuff tear can achieve good pain relief and modest improvements in motion after an anatomically sized hemishoulder arthroplasty. Those patients who have a pseudoparalytic shoulder due to massive rotator cuff tearing or those with irreparable anatomy of the shoulder can obtain good pain relief and regain a functional shoulder with the reverse shoulder arthroplasty.52,117,118 The Grammont reverse prosthesis can improve function and restore active elevation in patients with incongruent cuff-deficient shoulders; active rotation is usually unchanged. The results are less predictable and the complication and revision rates are higher in patients undergoing revision surgery than in patients with cuff tear arthropathy. The results of the reverse prosthesis depend on the diagnosis and on the remaining cuff muscles, specifically the teres minor and the development of an inferior scapular notching.119 For all indications (massive and irreparable cuff tear associated with arthrosis, cuff tear arthropathy, fracture sequelae with arthritis and failed revision arthroplasty) there was a significant increase in active elevation (from 55ë preoperatively to 121ë postoperatively) and Constant score (from 17 to 58 points) but no significant change in active external or internal rotation. © 2008, Woodhead Publishing Limited

604

24.7

Joint replacement technology

Summary

One century after PeÂan's shoulder arthroplasty and 50 years after Neer's innovations, the French surgeons Walch and Boileau conducted fundamental biomechanical research and clinical multicentre studies to upgrade the humeral prosthesis to the nowadays widely accepted third-generation shoulder arthroplasty. Another Frenchman, Grammont, introduced a reverse shoulder prosthesis concept that improves shoulder biomechanics if normal shoulder anatomy cannot be restored and if the deltoid muscle is still functioning well. New biomechanical studies and higher-volume shoulder surgeons will further improve the postoperative functional results of the established concepts.

24.8

References

1. Lugli T. Artificial shoulder joint by PeÂan (1893): the facts of an exceptional intervention and the prosthetic method. Clin Orthop Relat Res 1978; 133: 215±18. 2. Pean JE, Bick EM. Con prosthetic methods intended to repair bone fragments. Clin Orthop 1973; 94: 4±7. 3. Neer CS II, Brown TH Jr, McLaughlin HL. Fracture of the neck of the humerus with dislocation of the head fragment. Am J Surg 1953; 85: 252±8. 4. Neer CS II. Follow-up notes on articles previously published in the journal: articular replacement for the humeral head. J Bone Joint Surg Am 1964; 46: 1607± 10. 5. Neer CS II. Prosthetic replacement of the humeral head: indications and operative technique. Surg Clin North Am 1963; 43: 1581±97. 6. Neer CS II. Replacement arthroplasty for glenohumeral osteoarthritis. J Bone Joint Surg Am 1974; 56: 1±13. 7. Neer CS II, Watson KC, Stanton FJ. Recent experience in total shoulder replacement. J Bone Joint Surg Am 1982; 64: 319±37. 8. Wilde AH. Shoulder arthroplasty: what it is good for and how good is it. In: Matsen FA, Fu FH, Hawkins RJ, editors. The Shoulder: A Balance of Mobility and Stability. Rosemont (IL): American Academy of Orthopaedic Surgeons; 1992, p. 459±81. 9. Wirth MA, Rockwood CA Jr. Complications of total shoulder-replacement arthroplasty. J Bone Joint Surg Am 1996; 78: 603±16. 10. Wirth MA, Rockwood CA Jr. Complications of shoulder arthroplasty. Clin Orthop Relat Res 1994; 307: 47±69. 11. Frankle M, Siegal S, Pupello D, Saleem A, Mighell M, Vasey M. The Reverse Shoulder Prosthesis for glenohumeral arthritis associated with severe rotator cuff deficiency. A minimum two-year follow-up study of sixty patients. J Bone Joint Surg Am 2005; 87: 1697±705. 12. Seebauer L, Walter W, Keyl W. Reverse total shoulder arthroplasty for the treatment of defect arthropathy. Op Orthop Traumatol 2005; 17: 1±24. 13. Rittmeister M, Kerschbaumer F. Grammont reverse total shoulder arthroplasty in patients with rheumatoid arthritis and nonreconstructible rotator cuff lesions. J Shoulder Elbow Surg 2001; 10: 17±22. 14. Werner CM, Steinmann PA, Gilbart M, Gerber C. Treatment of painful pseudoparesis due to irreparable rotator cuff dysfunction with the Delta III reverse-balland-socket total shoulder prosthesis. J Bone Joint Surg Am 2005; 87: 1476±86. © 2008, Woodhead Publishing Limited

Replacing shoulder joints

605

15. Sirveaux F, Favard L, Oudet D, Huquet D, Walch G, Mole D. Grammont inverted total shoulder arthroplasty in the treatment of glenohumeral osteoarthritis with massive rupture of the cuff. J Bone Joint Surg Br 2004; 86: 388±95. 16. Nwakama AC, Cofield RH, Kavanagh BF, Loehr JF. Semiconstrained total shoulder arthroplasty for glenohumeral arthritis and massive rotator cuff tearing. J Shoulder Elbow Surg 2000; 9: 302±7. 17. Kalandiak S, Wirth MA, Rockwood CA Jr. Complications of shoulder arthroplasty. In: Williams GR, Yamaguchi K, Ramsey ML, Galatz LM, editors. Shoulder and Elbow Arthroplasty. Philadelphia: Lippincott Williams and Wilkins; 2005, pp. 229± 49. 18. Bohsali KI, Wirth MA, Rockwood CA, Jr. Complications of total shoulder arthroplasty. J Bone Joint Surg Am 2006; 88: 2279±92. 19. Grammont PM, Baulot E. Delta shoulder prosthesis for rotator cuff rupture. Orthopaedics 1993; 16: 65±8. 20. Dines JS, Fealy S, Strauss EJ, Allen A, Craig EV, Warren RF, Dines DM. Outcomes analysis of revision total shoulder replacement. J Bone Joint Surg Am 2006; 88: 1494±500. 21. Levy O, Copeland SA. Cementless surface replacement arthroplasty of the shoulder. 5- to 10-year results with the Copeland mark-2 prosthesis. J Bone Joint Surg Br 2001; 83: 213±21. 22. Szabo I, Buscayret F, Edwards TB, Nemoz C, Boileau P, Walch G. Radiographic comparison of flat-back and convex-back glenoid components in total shoulder arthroplasty. J Shoulder Elbow Surg 2005; 14: 636±42. 23. Anglin C, Wyss UP, Pichora DR. Mechanical testing of shoulder prostheses and recommendations for glenoid design. J Shoulder Elbow Surg 2000; 9: 323±31. 24. Collins D, Tencer A, Sidles J, Matsen F 3rd. Edge displacement and deformation of glenoid components in response to eccentric loading. The effect of preparation of the glenoid bone. J Bone Joint Surg Am 1992; 74: 501±7. 25. Walch G, Edwards TB, Boulahia A, Boileau P, Mole D, Adeleine P. The influence of glenohumeral prosthetic mismatch on glenoid radiolucent lines: results of a multicenter study. J Bone Joint Surg Am 2002; 84: 2186±91. 26. Iannotti JP, Gabriel JP, Schneck SL, Evans BG, Misra S. The normal gleno-humeral relationships. An anatomical study of one hundred and forty shoulders. J Bone Joint Surg Am 1992; 74: 491±500. 27. Lazarus MD, Jensen KL, Southworth C, Matsen FA 3rd. The radiographic evaluation of keeled and pegged glenoid component insertion. J Bone Joint Surg Am 2002; 84: 1174±82. 28. Wirth MA, Korvick DL, Basamania CJ, Toro F, Aufdemorte TB, Rockwood CA Jr. Radiologic, mechanical, and histologic evaluation of 2 glenoid prosthesis designs in a canine model. J Shoulder Elbow Surg 2001; 10: 140±8. 29. Gartsman GM, Elkousy HA, Warnock KM, Edwards TB, O'Connor DP. Radiographic comparison of pegged and keeled glenoid components. J Shoulder Elbow Surg 2005; 14: 252±7. 30. Matsen FA III, Lippitt SB (eds): Shoulder Surgery: Principles and Procedures, 1st edn. Philadelphia: WB Saunders; 2004, p. 43. 31. Harryman DT, Sidles JA, Harris SL, Lippitt SB, Matsen FA 3rd. The effect of articular conformity and the size of the humeral head component on laxity and motion after glenohumeral arthroplasty: a study in cadavera. J Bone Joint Surg Am 1995; 77±A: 555±63. 32. Nyffeler RW, Sheikh R, Jacob HA, Gerber C. Influence of humeral prosthesis © 2008, Woodhead Publishing Limited

606

33. 34. 35. 36. 37.

38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50.

Joint replacement technology height on biomechanics of glenohumeral abduction: an in vitro study. J Bone Joint Surg Am 2004; 86-A: 575±80. Pearl ML, Volk AG. Retroversion of the proximal humerus in relationship to prosthetic replacement arthroplasty. J Shoulder Elbow Surg 1995; 4: 286±9. Roberts SN, Foley AP, Swallow HM, Wallace WA, Coughlan DP. The geometry of the humeral head and the design of prostheses. J Bone Joint Surg 1992; 73-B: 647± 50. Boileau P, Walch G. The three-dimensional geometry of the proximal humerus: implications for surgical technique and prosthetic design. J Bone Joint Surg 1997; 79-B: 857±65. Boileau P, Walch G. Anatomical study of the proximal humerus: surgical technique considerations and prosthetic design rationale. In: Walch G, Boileau P, editors. Shoulder Arthroplasty. Heidelberg, Springer, 1998, pp. 69±82. Wallace AL, Phillips RL, MacDougal GA, Walsh WR, Sonnabend DH. Resurfacing of the glenoid in total shoulder arthroplasty: a comparison, at a mean of five years, of prostheses inserted with and without cement. J Bone Joint Surg 1999; 81-A: 510±18. Hertel R, Knothe U, Ballmer FT. Geometry of the proximal humerus and implications for prosthetic design. J Shoulder Elbow Surg 2002; 11(4): 331±8. Copeland SA. Cementless total shoulder replacement. In: Post M, Morrey BF, Hawkins RJ, editors. Surgery of the Shoulder. St. Louis, Mosby Year Book, 1990, pp. 289±93. Takase KR, Yamamoto K, Imakiire A, Burkhead WZ Jr. The radiographic study in the relationship of the glenohumeral joint. J Orthop Res 2004; 22: 298±305. Nyffeler RW, Gerber C. The relevance of anatomical reconstruction. In Walch G, Boileau P, Mole D, editors. 2000 Shoulder Prosthesis: Two to Ten Year Follow-up. Montpellier, France, Sauramps Medical, 2001, pp. 57±59. Bicknell RT, Liew ASL, Danter MR, Patterson SD, King GJW, Chess DG, Johnson JA. Does keel size, the use of screws, and the use of bone cement affect fixation of a metal glenoid implant? J Shoulder Elbow Surg 2003: 12(3): 268±75. Gartsman GM, Elkousy HA, Warnock KM, Edwards TB, O'Connor DP. Radiographic comparison of pegged and keeled glenoid components. J Shoulder Elbow Surg 2005; 3: 252±7. Mileti J, Boardman D, Sperling JW, Cofield RH, Torchia ME, O'Driscoll SW, Rowland CM. Radiographic analysis of polyethylene glenoid components using modern cementing techniques. J Shoulder Elbow Surg 2004; 13: 492±8. Nyffeler RW, Anglin C, Sheikh R, Gerber C. Influence of peg design and cement mantle thickness on pull-out strength of glenoid component pegs. J Bone Joint Surg Br 2003; 5: 748±52. Stone KD, Grabowski JJ, Cofield RH, Morrey BF, An KN. Stress analysis of glenoid components in total shoulder arthroplasty. J Shoulder Elbow Surg 1999; 2: 151±8. Karelse A, Kegels L, De Wilde L. The pillars of the scapula. Clin Anat 2007; 20(4): 392±9 De Wilde LF, Berghs BM, Audenaert E, Sys G, Van Maele GO, Barbaix E. About the variability of the shape of the glenoid cavity. Surg and Rad Anat 2004; 26: 54±9. Hertel R., Lehmann O. Die Schultergelenkpfanne: Anatomische Aspekte und Implikationen fuÈr das Prothesendesign. OrthopaÈde 2001; 30: 363±9. Torchia ME, Cofield RH, Settergren CR. Total shoulder arthroplasty with the Neer prosthesis: long-term results. J Shoulder Elbow Surg 1997; 6: 495±505.

© 2008, Woodhead Publishing Limited

Replacing shoulder joints

607

51. Boileau P, Sinnerton RJ, Chuinard C, Walch G. Review: Arthroplasty of the shoulder. J Bone Joint Surg 2006; 88-B: 562±75. 52. Boileau P, Watkinson D, Hatzidakis AM, Hovorka I. Neer Award 2005: The Grammont reverse shoulder prosthesis: results in cuff tear arthritis, fracture sequelae, and revision arthroplasty. J Shoulder Elbow Surg 2006; 15(5): 527±40. 53. Boileau P, Watkinson DJ, Hatzidakis AM, Balg F. Grammont reverse prosthesis: design, rationale, and biomechanics. J Shoulder Elbow Surg 2005; 14 (1 Suppl S): 147S±61S. 54. Grammont P, Trouilloud P, Laffay JP, Deries X. Etude et reÂalisation d'une nouvelle protheÁse d'eÂpaule. Rhumatologie 1987; 39: 27±38. 55. Nyffeler RW, Werner CML, Simmen BR, Gerber C. Analysis of a retrieved Delta III total shoulder prosthesis. J Bone Joint Surg Br 2004; 86(8): 1187±91. 56. De Wilde L, Walch G. Humeral prosthetic failure of reversed total shoulder arthroplasty. A report of three cases. J Shoulder Elbow Surg 2006; 15(2): 260±4. 57. Delloye C, Joris D, Colette A, Eudier A, Dubuc JE. Mechanical complications of total shoulder inverted prosthesis. Rev Chir Orthop Reparatrice Appar Mot 2002; 88: 410±14 (in French). 58. Nyffeler RW, Werner CML, Gerber C. Biomechanical relevance of glenoid component positioning in the reverse Delta III total shoulder prosthesis. J Shoulder Elbow Surg 2005; 14(5): 524±8. 59. Valenti P, Boutens D, Nerot C. Delta 3 reversed prosthesis for arthritis with massive rotator cuff tear: long term results (> 5 years). In Walch G, Boileau P, Mole D, editors. 2000 Shoulder Prosthesis: Two to Ten Year Follow-up. Montpellier, France, Sauramps Medical, 2001, pp. 253±9. 60. De Wilde F, Plasschaert FS, Audenaert EA, Verdonk RC. Functional recovery after a reverse prosthesis for reconstruction of the proximal humerus in tumor surgery. Clin Orthop Relat Res 2005; 430: 156±62. 61. Rietveld ABM, Daanen HAM, Rozing PM, Obermann WR. The lever arm in glenohumeral abduction after hemiarthroplasty. J Bone Joint Surg 1988; 70B: 561±5. 62. Gagey O, Hue E. Mechanics of the deltoid muscle: a new approach. Clin Orthop 2000; 375: 250±7. 63. De Wilde L, Audenaert E, Barbaix E, Audenaert A, Soudan K. Consequences of deltoid muscle elongation on deltoid muscle performance: a computerised study. J Clin Biomech 2002; 17: 499±505. 64. Karelse ATJ, Bhatia DN, De Wilde LF. Prosthetic component relationship of the reversed delta III total shoulder prosthesis in the transverse plane of the body. J Shoulder Elbow Surg 2007. 65. Walch G, Badet R, Boulahia A, Khoury A. Morphologic study of the glenoid in primary glenohumeral osteoarthritis. J Arthroplasty 1999; 14: 756±60. 66. Visotsky JL, Basamania C, Seebauer L, Rockwood CA, Jensen KL. Cuff tear arthropathy: pathogenesis, classification, and algorithm for treatment. J Bone Joint Surg Am 2004; 86-A Suppl 2: 35±40. 67. Levy O, Funk L, Sforza G, Copeland SA. Copeland surface replacement arthroplasty of the shoulder in rheumatoid arthritis. J Bone Joint Surg 2004; 86A(3): 512±18. 68. Torchia ME, Co?eld RH. Long-term results of Neer total shoulder arthroplasty. In: The American Shoulder and Elbow Surgeons Tenth Open Meeting, New Orleans, LA, 1994, p. 30. 69. Fink B, Singer J, Lamla U, Ruther W. Surface replacement of the humeral head in rheumatoid arthritis. Arch Orthop Trauma Surg 2004; 124(6): 366±73. © 2008, Woodhead Publishing Limited

608

Joint replacement technology

70. Trail IA, Nuttall D. The results of shoulder arthroplasty in patients with rheumatoid arthritis. J Bone Joint Surg 2002; 84-B: 1121±5. 71. Hattrup SJ, Co?eld RH. Osteonecrosis of the humeral head: relationship of disease stage, extent, and cause to natural history. J Shoulder Elbow Surg 1999; 8(6): 559± 64. 72. Hattrup SJ, Co?eld RH. Osteonecrosis of the humeral head: results of replacement. J Shoulder Elbow Surg 2000; 9: 177±82. 73. Mansat P,Huser L, Mansat M, Bellumore Y, Rongieres M, Bonnevialle P. Shoulder arthroplasty for atraumatic avascular necrosis of the humeral head: nineteen shoulders followed up for a mean of seven years. J Shoulder Elbow Surg 2005; 14(2): 114±20. 74. Neer CS 2nd. Displaced proximal humeral fractures: II. Treatment of three-part and four-part displacement. J Bone Joint Surg 1970; 52-B: 1090±103. 75. Boileau P, Trojani C, Walch G, et al. Shoulder arthroplasty for the treatment of the sequelae of fractures of the proximal humerus. J Shoulder Elbow Surg 2001; 10: 299±308. 76. Compito CA, Self EB, Bigliani LU. Arthroplasty and acute shoulder trauma: reasons for success and failure. Clin Orthop 1994; 307: 27±36. 77. Kralinger F, Schwaiger R, Wambacher M, et al. Outcome after primary hemiarthroplasty for fracture of the head of the humerus: a retrospective multicentre study of 167 patients. J Bone Joint Surg 2004; 86-B: 217±19. 78. Tanner MW, Cofield RH. Prosthetic arthroplasty for fractures and fracturedislocations of the proximal humerus. Clin Orthop 1983; 179: 116±28. 79. Green A, Barnard WL, Limbird RS. Humeral head replacement for acute, four-part proximal humerus fractures. J Shoulder Elbow Surg 1993; 2: 249±54. 80. Kay SP, Amstutz HC. Shoulder hemiarthroplasty at UCLA. Clin Orthop 1988; 228: 42±8. 81. Naranja RJ, Iannotti JP. Displaced three- and four-part proximal humerus fractures: evaluation and management. J Am Acad Orthop Surg 2000; 8: 373±82. 82. Frankle MA, Greenwald DP, Markee BA, Ondrovic LE, Lee WE 3rd. Biomechanical effects of malposition of tuberosity fragments on the humeral prosthetic reconstruction for four-part proximal humerus fractures. J Shoulder Elbow Surg 2001; 10: 321±6. 83. Frankle MA, Ondrovic LE, Markee BA, Harris ML, Lee WE 3rd. Stability of tuberosity reattachment in proximal humeral hemiarthroplasty. J Shoulder Elbow Surg 2002; 11: 413±20. 84. De Wilde LF, Berghs BM, Beutler T, Ferguson SJ, Verdonk RC. A new prosthetic design for proximal humeral fractures: reconstructing the glenohumeral unit. J Shoulder Elbow Surg 2004; 13(4): 373±80. 85. Boileau P, Chuinard C, Le Huec JC, Walch G, Trojani C. Proximal humerus fracture sequelae: impact of a new radiographic classification on arthroplasty. Clin Orthop Relat Res 2006; 442: 121±30. 86. Walch G, Badet R, Nove-Josserand L, Levigne C. Nonunions of the surgical neck of the humerus: surgical treatment with an intramedullary bone peg, internal fixation and cancellous bone grafting. J Shoulder Elbow Surg 1996; 5: 161±168. 87. Matsen FA III, Lippitt SB (eds). Shoulder Surgery: Principles and Procedures, 1st edn. Philadelphia, WB Saunders; 2004, p. 334. 88. Kempf JF. Conclusions sur l'omarthrose primitive. In: Walch G, Boileau P, Mole D, editors. 2000 Shoulder Prostheses: Two to Ten Year Follow-up. Montpellier, Sauramps MeÂdical, 2001, pp. 113±18. © 2008, Woodhead Publishing Limited

Replacing shoulder joints

609

89. Gerber C, Pennington SD, Yian EH, Pfirrmann CAW, Werner CML, Zumstein MA. Lesser tuberosity osteotomy for total shoulder arthroplasty. Surgical technique. J Bone Joint Surg Am 2006; 88 Suppl 1 Pt 2: 170±7. 90. De Wilde LF, Berghs BM, VandeVyver F, Schepens A, Verdonk RC. Glenohumeral relationship in the transverse plane of the body. J Shoulder Elbow Surg 2003; 12(3): 260±7. 91. Krishnan S. Oral presentation of a new technique at the Current Solutions in Shoulder and Elbow Surgery Conference, Tampa, FL, 2006. 92. De Wilde LF, Audenaert E, Sys G, Van Maele GO, Barbaix E. About the variability of the shape of the glenoid cavity. Surg Radiol Anat 2004; 26(1): 54±9. 93. Rodosky MW, Bigliani LU. Indications for glenoid resurfacing in shoulder arthroplasty. J Shoulder Elbow Surg 1996; 5: 231±48. 94. Burkhart SS, DeBeer JF, Tehrany AM, Parten PM. Quantifying glenoid bone loss arthroscopically in shoulder instability. Arthroscopy 2002; 18: 488±91. 95. Sugaya H., Moriishi J, Dohi M, Kon Y, Tsuchiya A. Glenoid rim morphology in recurrent anterior glenohumeral instability. J Bone Joint Surg 2003: 85-A: 878±84. 96. Walch G, Boulahia A, Boileau P, Kempf JF. Primary glenohumeral osteoarthritis: clinical and radiographic classification. The Aequalis Group. Acta Orthop Belg 1998; 64 Suppl 2: 46±52. 97. Kelly JD Jr, Norris TR. Decision making in glenohumeral arthroplasty. J Arthroplasty 2003; 18(1): 75±82. 98. Matsen FA III, Lippitt SB (eds). Shoulder Surgery: Principles and Procedures, 1st edn. Philadelphia, WB Saunders; 2004, pp. 481±510. 99. Scheibel M, Lichtenberg S, Habermeyer P. Reversed arthroscopic subacromial decompression for massive rotator cuff tears. J Shoulder Elbow Surg 2004; 13(3): 272±8. 100. Fenlin JM Jr, Chase JM, Rushton SA, Frieman BG. Tuberoplasty: creation of an acromiohumeral articulation ± a treatment option for massive, irreparable rotator cuff tears. J Shoulder Elbow Surg 2002; 11(2): 136±42. 101. Mackenzie DB. The antero superior exposure of a total shoulder replacement. Arthop Traumatol 1993; 2: 71±7. 102. L'Episcopo JB. Tendon transplantation in obstetrical paralysis. Am J Surg 1934; 25: 122±5. 103. Chin PY, Sperling JW, Cofield RH, Schleck C. Complications of total shoulder arthroplasty: are they fewer or different? J Shoulder Elbow Surg 2006; 15(1): 19±22. 104. Burkhead WZ Jr, Hutton KS. Biologic resurfacing of the glenoid with hemiarthroplasty of the shoulder. J Shoulder Elbow Surg 1995; 4: 263. 105. Bhatia DN, van Rooyen KS, du Toit DF, de Beer JF. Arthroscopic technique of interposition arthroplasty of the glenohumeral joint. Arthroscopy 2006; 22(5): 570±5. 106. Iannotti JP, Norris TR. Influence of preoperative factors on outcome of shoulder arthroplasty for glenohumeral osteoarthritis. J Bone Joint Surg Am 2003; 85-A: 251±8. 107. Torchia ME, Cofield RH, Settergren CR. Total shoulder arthroplasty with the Neer prosthesis: long-term results. J Shoulder Elbow Surg 1997; 6(6): 495±505. 108. Guery J, Favard L, Sirveaux F, Oudet D, Mole D, Walch G. Reverse total shoulder arthroplasty. Survivorship analysis of eighty replacements followed for five to ten years. J Bone Joint Surg Am 2006; 88(8): 1742±7. 109. Jain N, Pietrobon R, Hocker S, et al. The relationship between surgeon and hospital volume and outcomes for shoulder arthroplasty. J Bone Joint Surg Am 2004; 86: 496±505. © 2008, Woodhead Publishing Limited

610

Joint replacement technology

110. Hammond JW, Queale WS, Kim TK, McFarland EG. Surgeon experience and clinical and economic outcomes for shoulder arthroplasty. J Bone Joint Surg Am 2003; 85: 2318±24. 111. Edwards TB, Boulahia A, Kempf JF, et al. The influence of rotator cuff disease on the results of shoulder arthroplasty for primary osteoarthritis: results of a multicentre study. J Bone Joint Surg Am 2002; 84-A: 2240±8. 112. Nyffeler RW, Sheik R, Jacob HA, Gerber C. Influence of humeral prosthesis height on biomechanics of glenohumeral abduction: an in vitro study. J Bone Joint Surg Am 2004; 86-A: 575±80. 113. Godeneche A, Boileau P, Favard L, et al. Prosthetic replacement in the treatment of osteoarthritis of the shoulder: early results of 268 cases. J Shoulder Elbow Surg 2002; 11: 11±18. 114. Edwards TB, Kadakia NR, Boulahia A, et al. A comparison of hemiarthroplasty and total shoulder arthroplasty in the treatment of primary glenohumeral osteoarthritis: results of a multicenter study. J Shoulder Elbow Surg 2003; 12: 207±13. 115. Cazeneuve JF, Cristofari DJ. Grammont reversed prosthesis for acute complex fracture of the proximal humerus in an elderly population with 5 to 12 years followup. Rev Chir Orthop Reparatrice Appar Mot 2006; 92(6): 543±8. 116. Van Seymortier P, Stoffelen D, Fortems Y, Reynders P. The reverse shoulder prosthesis (Delta III) in acute shoulder fractures: technical considerations with respect to stability. Acta Orthop Belg 2006; 72(4): 474±7. 117. De Wilde LF, Van Ovost E, Uyttendaele D, Verdonk R. Results of an inverted shoulder prosthesis after resection for tumor of the proximal humerus. Rev Chir Orthop Reparatrice Appar Mot 2002; 88(4): 373±8. 118. Matsen FA 3rd, Boileau P, Walch G, Gerber C, Bicknell RT. The reverse total shoulder arthroplasty. J Bone Joint Surg Am 2007; 89(3): 660±7. 119. Simovitch RW, Zumstein MA, Lohri E, Helmy N, Gerber C. Predictors of scapular notching in patients managed with the Delta III reverse total shoulder replacement. J Bone Joint Surg Am 2007; 89(3): 588±600.

© 2008, Woodhead Publishing Limited

25

Replacing the elbow joints J S A N C H E Z - S O T E L O , Mayo Clinic, USA

25.1

Introduction

Replacement arthroplasty of the elbow is in constant evolution. Although it was initially used mainly in patients with inflammatory arthritis, its indications were expanded to other conditions which place higher demands on the implants. Elbow arthroplasty presents some unique peculiarities. Compared with the hip and knee joints, the elbow is relatively small and its stability depends greatly on ligamentous integrity. Linked semi-constrained elbow arthroplasties became popular in the United States and central Europe; these inherently stable implants raise the concern of increased contact pressures on the already thin polyethylene. Unlinked arthroplasties, popular in the United Kingdom and Asia, may have better tribological properties but are at risk for instability and decreased elbow extension. Elbow arthroplasty is further complicated by the need to violate the extensor mechanism for exposure, the increased risk of infection, the role of the radial head, and potential clinical problems related to the ulnar nerve. Present and future innovations may include the use of linkable implants, alternative bearing surfaces, uncemented fixation, distal humerus hemi-arthroplasties, unicompartmental arthroplasties, and improved revision systems.

25.2

Materials and device design

25.2.1 Implant types There is some confusion regarding the terminology used for implants available to replace the elbow joint. In general, there are two broad categories of implants which differ in the presence or absence of a mechanism linking the humeral and ulnar components (Table 25.1). A common misconception is to equate linking to constraint: some unlinked implants are more constrained than their linked counterparts. Linked/coupled implants The distinguishing feature of this category of implant is the physical linking of the humeral and ulnar components at the time of surgery in order to avoid © 2008, Woodhead Publishing Limited

612

Joint replacement technology

subluxation or dislocation episodes. Early linked implants were constrained hinges which allowed only flexion and extension. These implants were associated with a high failure rate secondary to the transmission of high stresses to the implant±cement±bone interface and other design flaws. Currently, most linked implants are semi-constrained: their linking mechanism behaves as a sloppy hinge, allowing some rotational and varus±valgus play. Semi-constrained implants are believed to transmit less stress to the implant interfaces, which associated with other design improvements has resulted in more reliable longterm fixation. The linked semi-constrained implant most commonly used currently is the Coonrad-Morrey prosthesis (Fig. 25.1). The humeral component is porouscoated distally and presents an anterior flange which increases the rotational stability of the implant and neutralizes the extension forces transmitted to the implant interface. The ulnar component has a plasma-spray metallic coating in its proximal third. Both components are intended to be fixed with polymethylmethacrylate. The components are linked with a cobalt±chrome axis pin which articulates with the polyethylene bushings of the ulnar and humeral components

25.1 Some examples of implants used to replace the elbow joint: (a) Coonrad± Morrey linked semi-constrained elbow arthroplasty, (b) Kudo unlinked minimally constrained elbow arthroplasty, (c) Latitude anatomic linkable prosthesis. © 2008, Woodhead Publishing Limited

Replacing the elbow joints

613

Table 25.1 Main implants used to replace the elbow joint Linked

Unlinked

Linkable

Coonrad-Morrey Discovery GSB III Norway Pritchard Mark II Pritchard-Walker

Capitellocondylar iBP Kudo Norway Pritchard II (ERS) Sorbie Souter-Strathclyde

Acclaim Latitude

and allows approximately 10ë of varus±valgus and rotational laxity. Other linked implants are enumerated in Table 25.1. Unlinked/uncoupled implants In this kind of arthroplasty the components are not mechanically linked. Maintenance of prosthesis articulation depends on the adequate position of each component as well as ligamentous integrity and the dynamic stabilizing effect of the musculature. Most of these implants provide a more or less anatomic resurfacing of the distal humerus and proximal ulna and some incorporate a radial head component. The most popular unlinked implants are the SouterStrathclyde and the Kudo prosthesis (Fig. 25.1). Other unlinked implants are listed in Table 25.1.

25.2.2 Advantages and disadvantages of the different kinds of implants The clinical outcome and long-term survivorship differs from implant to implant, and the results obtained with a given linked or unlinked implant cannot be extrapolated to other members of the same implant family. However, there are a few advantages and disadvantages of each of these two design philosophies (Table 25.2). Linked implants ensure joint stability, even in the presence of severe bone loss or ligamentous insufficiency. These implants not only eliminate one of the main complications of unlinked implants, namely dislocation, but also allow a more aggressive soft-tissue release in patients with preoperative stiffness and deformity, which allows more reliable restoration of elbow motion. On the other hand, the increased constraint associated with implant linking results in increased tension on both the joint surface and the interfaces, which may facilitate polyethylene wear and component loosening. Semi-constrained implants did represent an improvement, but well-fixed semi-constrained implants are at risk for accelerated wear in the presence of ligamentous imbalance. © 2008, Woodhead Publishing Limited

614

Joint replacement technology

Table 25.2 Advantages and disadvantages of linked and unlinked prostheses Linked Advantages · Ensure joint stability · May be used in the presence of ligamentous insufficiency · May be used in the presence of severe bone loss · Better range of motion (soft-tissue release and non-anatomic implantation) Disadvantages · Increased constraint may result in increased tension to the interface and higher risk of mechanical failure secondary to wear and/or loosening · More extensive canal invasion, potentially complicating revision surgery · Cannot be used as hemi-arthroplasty · Component linking may make implantation more difficult · Possible failure of the linking mechanism

Unlinked · Less constrained implants may be associated with a lower risk of wear, loosening and osteolysis · Less bony-invasive, which may be beneficial if revision or resection are required · Some anatomic humeral components may be used as hemiarthroplasty · Most require more accurate component positioning in order to ensure proper articular tracking · It is possible to subluxate or dislocate the joint · Difficult to use when there is the need to compensate for bone loss or ligamentous insufficiency · Limited ability for soft-tissue release or non-anatomic implant positioning in patients with stiffness

Some linked implants also allow replacement in the presence of severe bone loss. Many unlinked designs require the humeral condyles and ulnar notch for component fixation. Bone loss compromises fixation of this kind of component and may render the medial or lateral ligament complexes insufficient if the epicondyles are affected. In addition, patients with severe preoperative stiffness may require non-anatomic implantation of the humeral component to raise the joint line, which makes the use of unlinked implants more complicated. However, linked implants do have substantial disadvantages, especially when they are constrained. In those situations where the remaining bone stock and ligamentous structures are adequate, unlinked implants are at least theoretically at less risk of mechanical failure secondary to wear, osteolysis, and loosening. As a general rule, the stems of unlinked implants are shorter; this is especially beneficial when revision or resection is required. Some anatomic unlinked humeral components may also be used as hemi-arthroplasties. The need for a radial head implant is controversial. On one side, patients with an arthritic radial head or a previous radial head resection may benefit from the use of a radial implant, which may increase stability and result in a greater improvement on the lateral side of the joint. However, from a technical point of view it is difficult to achieve proper alignment and tracking of the radial head © 2008, Woodhead Publishing Limited

Replacing the elbow joints

615

implant, and this component is potentially one more source of wear, osteolysis, and loosening. Currently published data seem to favour the use of linked semi-constrained implants. Little et al.1 recently published a systematic review of the literature on elbow arthroplasty. The overall revision rate has been similar for linked and unlinked implants (11% vs 13%). However, radiographic loosening seems to be higher with unlinked implants (especially the humeral component of the Souter prosthesis). The functional results are similar, with the exception of elbow extension, which seems to be better with linked implants. Recently designed implants have maintained some of the classic features recognized to improve the outcome of elbow arthroplasty (such as the use of a flange), but provide three potential advantages: · The bearing surface design allows the use of thicker polyethylene subjected to less contact pressure. · The instrumentation and design allow a more anatomic reconstruction with more attention being paid to reproduction of the anatomic center of rotation. · The components may be linked after being completely seated. The Latitude system probably is the best example of this new generation of elbow arthroplasty. This modular system is linkable, meaning that the surgeon may choose at the end of the case to leave the implant linked or unlinked depending on his or her intra-operative assessment of stability. In addition, to my knowledge, this is the only system that allows conversion of a distal humerus hemi-arthroplasty to a total elbow arthroplasty without revising the humeral stem.

25.3

Indications and contraindications

Inflammatory arthropathies such as rheumatoid arthritis represent the classic indication of elbow arthroplasty. Those patients with more severe involvement (Mayo Clinic stages III to V) experience great improvements in pain and function. In addition, the polyarticular nature of these conditions may limit the overall activity level of these patients, with a low rate of wear and loosening. In the earlier stages of rheumatoid arthritis, there is usually enough bone stock and ligamentous integrity to allow the use of unlinked implants. The successful outcome of elbow arthroplasty in inflammatory conditions prompted its use for the treatment of other conditions (Table 25.3). Posttraumatic elbow osteoarthritis represents one of the most difficult conditions to treat. Some patients may improve with alternative surgical procedures, such as interposition arthroplasty, but pain relief with elbow interposition is not completely reproducible and some patients may experience postoperative instability. Elbow arthroplasty provides a more reliable outcome, but these younger, more active patients are at risk for early mechanical failure. In general, elbow arthroplasty is best avoided in patients under the age of 60. © 2008, Woodhead Publishing Limited

616

Joint replacement technology Table 25.3 Main indications for elbow arthroplasty · · · · · · · · ·

Chronic inflammatory arthropathies Post-traumatic osteoarthritis Acute distal humerus fractures Distal humerus non-unions Extreme intrinsic stiffness/ankylosis Large post-traumatic bone defects Primary osteoarthritis (rare) Hemophilic arthropathy Reconstruction after tumor resection

Acute comminuted distal humerus fractures in elderly patients or those with previous articular degeneration has emerged as one of the most common indications for elbow arthroplasty in some countries. Stable internal fixation is difficult to obtain in these circumstances, and arthroplasty is used successfully for other fractures (femoral neck, proximal humerus). It is important to emphasize that this is a selective indication, as most patients with distal humerus fractures are best treated with open reduction and internal fixation. Other indications for elbow arthroplasty include the salvage of distal humerus non-union in elderly patients, large post-traumatic defects, as well as elbow reconstruction after tumor resection. Primary osteoarthritis of the elbow usually affects younger patients and is treated successfully in many patients with joint debridement procedures such as osteocapsular arthroplasty.

25.4

Surgical technique overview

25.4.1 Surgical exposure Most of the surgical approaches used for implantation of an elbow arthroplasty require mobilization of the elbow extensor mechanism. Subcutaneous ulnar nerve transposition is routinely performed by most surgeons. The author's preferred exposure is the triceps-reflecting Bryan±Morrey approach;2 other surgeons prefer to split the triceps or use an extended lateral-sided KoÈcher approach. The approach described by Bryan and Morrey (Fig. 25.2) involves detaching the triceps off the olecranon, reflecting it from medial to lateral, maintaining its continuity with the anconeus and the forearm fascia. This approach provides ample exposure of the joint and allows a secure reconstruction of the extensor mechanism, although it is associated with some risk of lateral subluxation of the triceps and weakness in extension. Splitting the triceps in the midline with detachment of its medial and lateral halves from the olecranon also provides a good exposure. The main advantage of this approach is maintenance of the extensor mechanism centralized over the olecranon, but transmuscular approaches are in general less appealing and the repair of the medial half is sometimes unsatisfactory. © 2008, Woodhead Publishing Limited

Replacing the elbow joints

617

25.2 Bryan±Morrey approach.

In some specific circumstances it is possible to perform the replacement by working on both sides of the triceps (so-called bilaterotricipital approach, Fig. 25.3).3 This approach is mostly indicated in the presence of a substantial bone defect at the distal humerus (secondary to trauma or tumor resection), as well as in acute distal humerus fractures and distal humerus non-unions where the distal fragments are resected.

25.4.2 Bony preparation and component insertion The bony preparation is different for each particular system. Most components are stemmed and require preparation of the humeral and ulnar canals with rasps and broaches. The author uses the Coonrad-Morrey system. With this system, the humeral side is prepared first after exposing the joint and releasing the lateral and medial collateral ligaments. The humeral canal is identified and used as a reference to cut a yoke-shaped segment of the distal humerus to accommodate the distal part of the humeral component. Next, the canal is prepared to accept the stem and the anterior cortex of the distal humerus is exposed for future contact with a bone graft placed behind the anterior flange of the humeral component. The ulnar canal is opened at the midportion of the trochlear notch and the canal prepared with right or left broaches. The components are then cemented in place with antibiotic-loaded polymethylmethacrylate, placing a bone graft between the anterior humeral cortex and the humeral flange. The components are then linked together. The ulnar canal is usually relatively narrow, which requires the use of a small flexible cannula to introduce the cement, which should be applied very early. Preoperative stiffness or deformity usually requires extensive soft-tissue balancing and releases. Limited extension may be corrected by anterior capsular release and proximal placement of the humeral component with elevation of the joint line. Limited flexion is corrected by posterior capsular release and occasionally resection of the anterior aspect of the coronoid. © 2008, Woodhead Publishing Limited

618

Joint replacement technology

25.3 Total elbow arthroplasty for the treatment of an acute distal humerus fracture. This procedure may be performed working on both sides of the triceps without violating the extensor mechanism: (a) humeral canal preparation; (b) trial components.

25.4.3 Postoperative management The goal of the early phase of postoperative treatment consists of limited postoperative edema. The elbow is immobilized in extension with an anterior plaster splint and a bulky dressing and the upper extremity is kept elevated. When a linked arthroplasty is used, elbow motion without protection may be initiated in the first few days after surgery depending on the aspect of the wound. Most surgeons keep the elbow immobilized for approximately two weeks after using an unlinked elbow arthroplasty to protect the ligamentous structures and decrease the risk of instability. A nocturnal extension splint is useful for the first few weeks after surgery when there is a marked preoperative flexion © 2008, Woodhead Publishing Limited

Replacing the elbow joints

619

contracture. Elbow extension against resistance should be avoided whenever the extensor mechanism has been violated for exposure. Polyethylene wear is the main limiting factor for the survivorship of current elbow designs. Prior to surgery patients should understand the need to protect their upper extremity. Empirically, patients are recommended to avoid lifting with the involved upper extremity more than 1 kg (2 lb) on a repetitive basis or more than 5 kg (10 lb) on a single event.

25.5

Clinical results

25.5.1 Chronic inflammatory arthritis Several studies have documented the outcome of elbow arthroplasty in rheumatoid arthritis using both linked and unlinked implants. Gill and Morrey4 published the results obtained in 78 consecutive rheumatoid elbows using the Coonrad±Morrey design. At most recent follow-up, 97% of the patients had no or mild pain and the mean arc of motion was from 28ë of extension to 131ë of flexion. The main complications of this series included deep infection (two cases), aseptic loosening (two cases), triceps avulsion (three cases), periprosthetic fractures (two cases), and ulnar component fracture (one case). Survivorship free of revision was 92.4% at 10 years (Fig. 25.4). Gschwend et

25.4 Antero-posterior (a) and lateral (b) radiographs after elbow arthroplasty for rheumatoid arthritis. © 2008, Woodhead Publishing Limited

620

Joint replacement technology

al.5 published the results using the GSB III prosthesis in 65 elbows, 32 of which were rheumatoid, followed for a minimum of 10 years. Overall clinical results were satisfactory and the main complications included infection (6%), loosening (4.6%), and component disengagement (13.6%). Van der Lugt et al.6 published the results obtained in 204 rheumatoid elbows replaced using the Souter-Strathclyde prosthesis and followed for a mean of 6.4 years. At most recent follow-up, only 6 patients complained of pain at rest. Complications included infection (10 cases), humeral loosening (22 cases), and dislocation (4 cases). Kudo et al.7 published the results obtained in 43 elbows replaced with the Kudo prosthesis and followed for a mean of three years; good or excellent results were obtained in approximately 86% of the patients, although some experienced loss of extension. Willems and De Smet published the results of 36 Kudo prostheses in rheumatoid elbows; the main reported complications included infection (one case), instability (two cases), and loosening (six cases).8 In general, most patients with rheumatoid arthritis experience satisfactory pain relief and functional improvement with both linked and unlinked implants. Most patients maintain a good arc of motion and the rate of mechanical failure is small. Some authors believe that the outcome of elbow arthroplasty is similar to the outcome of hip and knee arthroplasty in rheumatoid patients.4,5 Linked arthroplasties allow the treatment of a wider spectrum of pathology, including patients with more extensive involvement, bone defects, and instability.

25.5.2 Trauma Post-traumatic osteoarthritis This is one of the most common conditions affecting the elbow joint. Postoperative pain and stiffness are common sequels of elbow trauma. The first step in the evaluation of these patients is to determine how much the articular surface contributes to the patient's symptoms. Patients with a symptomatic articular surface experience pain with resisted flexion and extension in the mid-arc of motion. The status of the articular surface may be evaluated with radiographs and computer tomography (CT) scan. When the articular surface is responsible for most symptoms, the alternative surgical options are somewhat limited and not totally satisfactory. Arthroscopic debridement is more reliable for impingement pain. Interposition arthroplasty includes placement of a layer of cutis, fascia lata, or Achilles tendon allograft interposed between the humerus and ulna and temporary distraction of the joint with an articulated external fixator for approximately six weeks. This procedure is more reliable for restoration of motion than pain relief.9,10 Other procedures, such as osteoarticular allografts or elbow fusion, have a high rate of complications11 or are poorly accepted by patients.12 Elbow arthroplasty is very attractive © 2008, Woodhead Publishing Limited

Replacing the elbow joints

621

as it provides the best early functional results; however, it is associated with a worrisome rate of mechanical failure, especially in younger patients.13 Schneeberger et al.13 published the results obtained in 41 patients with posttraumatic osteoarthritis using the Coonrad±Morrey prosthesis. The mean age of the patients at the time of surgery was 57 years (range, 32±82 years) and the mean follow-up time was five years. Seventy-three per cent of the patients had no or mild pain and the results were considered satisfactory in 83% of the cases. However, there was a 27% complication rate, including five ulnar component fractures and two revisions for polyethylene wear. These authors concluded that elbow arthroplasty should be relatively contraindicated in patients who are planning to perform substantial physical activities with the involved upper extremity or who are not able to comply with the previously mentioned postoperative restrictions. The relatively high mechanical failure rate of elbow arthroplasty in patients with post-traumatic osteoarthritis has been the main driving force for the development of newer implants with supposedly better wear patterns (Fig. 25.5). There are no published studies on the outcome of these new designs. An alternative strategy in younger patients is to offer them an interposition arthroplasty as their first procedure as long as they understand that pain relief is not reliable; fortunately, the outcome of replacement after failed interposition arthroplasty is equivalent to that of patients without previous interposition.14 Distal humerus fractures Open reduction and internal fixation is the treatment of choice for most distal humerus fractures. However, the outcome of internal fixation may be compromised in a selective group of patients with extensive comminution, osteopenia or previous articular pathology. For elderly patients in this situation, elbow arthroplasty probably represents a better alternative. There are different philosophies for the use of elbow arthroplasty in distal humerus fractures. The author's preferred strategy is to work through a

25.5 Failed elbow components secondary to polyethylene wear in a patient with post-traumatic osteoarthritis. © 2008, Woodhead Publishing Limited

622

Joint replacement technology

bilaterotricipital approach, resect the fractured fragments, and complete the arthroplasty. When the distal fragments are resected, the collateral ligament complexes and the flexor±pronator and extensor±supinator groups are detached. A linked arthroplasty is needed to compensate for the ligamentous insufficiency. The forearm muscular groups are sutured to the triceps to seal the joint; interestingly, resection of the humeral condyles does not seem to affect grip strength or strength in flexion, extension, pronation, or supination.15 An alternative philosophy consists of fixing the condyles to preserve the integrity of the collateral ligaments and replace the articular surface with a distal humerus hemi-arthroplasty or a total elbow replacement. The outcome of total elbow arthroplasty in selected patients with complex distal humerus fractures is quite satisfactory. Kamineni and Morrey recently reviewed the results obtained in a consecutive series of 43 patients followed for a mean of seven years.16 Most patients achieved a satisfactory Mayo Elbow Performance Score and the mean arc of motion was from 24ë of extension to 131ë of flexion (Fig. 25.6). However, nine patients required a reoperation, including five cases of component revision. Other authors have published

25.6 Postoperative radiograph (a) and final range of motion (b and c) after elbow arthroplasty for an acute distal humerus fracture in an elderly female patient. © 2008, Woodhead Publishing Limited

Replacing the elbow joints

623

similar outcomes.17±20 Frankle et al.17 performed an interesting comparative study between internal fixation and arthroplasty in 24 fractures affecting women over 65 years old and obtained a better result in the arthroplasty group. Distal humerus non-union The salvage of distal humerus non-union in selected patients represents a good indication for elbow arthroplasty. Most distal humerus non-unions are treated with internal fixation and bone grafting. However, elderly patients with osteopenia and very limited bone stock may benefit more from elbow arthroplasty. Morrey and Adams published the results obtained in 36 patients with a mean age of 68 years followed for an average time of four years after elbow arthroplasty for distal humerus non-union.21 Results were rated as satisfactory in 86% of the cases; there were two infections and three patients with excessive polyethylene wear in the others. The author and colleagues recently updated the Mayo Clinic experience using elbow arthroplasty for the salvage of 92 distal humerus nonunions. At a mean follow-up of 6.7 years (range, 2±20 years), 79% of the patients had no or mild pain and the mean range of motion was from 22ë of extension to 135ë of flexion. Complications included aseptic loosening in 16 patients, component fracture in 5 patients, deep infection in 5 patients and bushing wear in 1 patient.

25.5.3 Other indications Total elbow arthroplasty has also been successfully used in patients with severe stiffness or ankylosis,22 gross instability secondary to large bony defects,23 haemophilic arthropathy,24 and reconstruction after tumor resection.25

25.6

Complications

25.6.1 Infection Deep periprosthetic infection affects the elbow more commonly than other joints. This is attributed to the thin soft-tissue envelope of the elbow as well as the higher risk of infection in patients with relative immune suppression secondary to inflammatory conditions or failed previous surgical procedures for trauma. Currently, the incidence of infection after elbow arthroplasty is estimated to be between 2% and 4%.1,26 Antibiotic-loaded polymethylmethacrylate is used routinely for implant fixation in an effort to decrease the rate of infection. Acute infections may be treated with irrigation, debridement, polyethylene exchange, and retention of the components. Chronic infections may be treated with two-stage reimplantation or resection depending on the nature of the infection, patient needs, and remaining bone and soft tissues. © 2008, Woodhead Publishing Limited

624

Joint replacement technology

25.6.2 Ulnar neuropathy The overall rate of ulnar neuropathy is difficult to estimate as patients with sensory symptoms are not reported accurately on most published studies about elbow arthroplasty. The incidence of severe ulnar neuropathy probably is around 5%.1 Most surgeons recommend routine subcutaneous ulnar nerve transposition at the time of arthroplasty to prevent postoperative ulnar nerve dysfunction.

25.6.3 The extensor mechanism The rate of extensor mechanism dysfunction is also difficult to analyze in the published literature and is probably underestimated. In Little et al.'s systematic review of the literature the incidence of triceps insufficiency was 3%. Poor softtissue quality as present in many patients with rheumatoid arthritis may affect the quality of the triceps repair at the end of surgery. Patients with symptomatic dysfunction of the extensor mechanism may benefit from surgical reconstruction of the extensor mechanism using either an anconeus rotation flap or an Achilles tendon allograft (Fig. 25.7).27

25.6.4 Instability Unlinked elbow arthroplasty may be complicated by subluxation or dislocation (Fig. 25.8). The rate of dislocation is approximately 5%; the overall rate of instability (dislocation or subluxation) is about 15%.1 There are different treatment options. Dislocation presenting in the first few weeks after surgery may respond to closed reduction and immobilization. However, most patients with instability require revision surgery for ligamentous reconstruction or revision to a linked elbow arthroplasty.28

25.6.5 Mechanical failure The overall rate of aseptic loosening after elbow arthroplasty probably ranges between 5% and 10%, and it is different for different implant designs. According to Little et al., the published aseptic loosening rate is 2% for the Coonrad± Morrey prosthesis, 8% for the Souter prosthesis, and 18% for the Kudo prosthesis.1 Polyethylene wear and osteolysis, component fracture, and component disengagement are additional modes of mechanical failure whose rate is difficult to estimate. Polyethylene wear probably is the limiting factor for the durability of elbow arthroplasty in young active patients.

25.6.6 Periprosthetic fractures Elbow periprosthetic fractures are classified based on the location of the fracture, the fixation of the components and the need to use special reconstructive © 2008, Woodhead Publishing Limited

Replacing the elbow joints

625

25.7 Extensor mechanism insufficiency after elbow arthroplasty may be reconstructed using an anconeus rotation flap (a) or an Achilles tendon allograft (b).

techniques for bone loss (Fig. 25.9).29 Most fractures of the humeral condyles may be treated non-operatively provided they are not associated with instability in the case of unlinked prosthesis. Most periprosthetic fractures require component revision and internal fixation using plates or cortical strut allografts (Fig. 25.10).9,30

25.7

Revision surgery

The increasing use of elbow arthroplasty, especially in younger patients with increased functional demands, has resulted in a substantial increase in the prevalence of revision surgery. Most revision surgeries require the use of a © 2008, Woodhead Publishing Limited

626

Joint replacement technology

25.8 Dislocation after unlinked elbow arthroplasty. (a) antero-posterior view; (b) lateral view.

25.9 Periprosthetic elbow fracture classification. © 2008, Woodhead Publishing Limited

Replacing the elbow joints

627

25.10 Periprosthetic humeral fracture treated with humeral component revision and internal fixation with anterior and posterior cortical strut allografts and wires.

linked prosthesis, as the severity of bone loss and ligamentous insufficiency in the revision setting rarely permits the use of an unlinked implant. A careful preoperative evaluation of the patient prior to revision surgery is critical for success. The physical examination should consider the condition of the skin, location of previous incisions, range of motion, joint stability, muscle function and strength, as well as the location and function of the ulnar nerve. The possibility of infection should always be considered and investigated with baseline laboratory studies including white cell count, sedimentation rate, and C reactive protein. Joint aspiration for cell count and cultures should probably be considered in every patient and is mandatory if there is a high suspicion of infection or the parameters mentioned above are elevated. Preoperative radiographs should also be analyzed carefully to evaluate the fixation of the components and the severity of bone loss. A few basic principles apply to all revision cases. The skin overlying the elbow joint is very fragile; the previous skin incision should be used whenever possible and the soft tissues should be handled with extreme care. The ulnar nerve should be identified and protected in all cases; complex humeral reconstructions also require identification and protection of the radial nerve. In many instances, component revision may be performed working on both sides of the triceps, especially in the presence of severe bone loss. Component and cement removal should be done with extreme care, as intra-operative perforations and fractures can occur easily; the use of high-speed burs and flexible cannulated canal reamers is recommended, and sometimes it is necessary to © 2008, Woodhead Publishing Limited

628

Joint replacement technology

create a controlled osteotomy of the humerus or ulna. In the absence of infection, it is reasonable to preserve well-fixed cement and use cement within cement technique for implant fixation. King et al. reported the initial Mayo Clinic experience in a consecutive series of 41 revision elbow arthroplasties followed for a mean of six years.31 Most patients experience a substantial improvement in pain and function and many were able to resume activities of daily living. However, there was a high incidence of complications, including intra-operative fractures and radial or ulnar nerve dysfunction. More recent studies have documented a high success rate with revision techniques used in the presence of bone loss, including cortical strut allografts, impaction grafting and allograft-prosthetic composites.29,30,32,33 In general terms, allograft-prosthetic composites have provided inferior results compared with other techniques.

25.8

Summary

The field of elbow arthroplasty continues to experience substantial improvements. Currently, elbow replacement represents a successful treatment alternative for patients with inflammatory conditions as well as selected patients with post-traumatic osteoarthritis, elderly patients with low, comminuted distal humerus fractures, the salvage of distal humerus non-union, ankylosis, haemophilic arthropathy, and elbow reconstruction after tumor resection. Some linked arthroplasty designs seem to be associated with a better outcome and allow the management of a wider range of pathology. There is interest in the development of improved designs which will decrease the rate of polyethylene wear and mechanical failure in higher-demand patients and provide increased flexibility in the primary and revision setting. The role of distal humerus hemi-arthroplasty, linkable implants, and components for the radial head need further investigation. The success of elbow arthroplasty depends greatly on the surgeon's familiarity with the anatomy and surgical approaches to the elbow joint, the proper selection and implantation of prosthetic components, and compliance with postoperative recommendations. Although elbow arthroplasty is sometimes the only option to improve pain and function in a wide range of patients, this procedure may be associated with complications which may be difficult to solve, including infection, extensor mechanism dysfunction, periprosthetic fractures, wear, loosening, and osteolysis. Fortunately, revision techniques developed over the last few years allow successful treatment of some of these complications.

25.9

References

1. Little, C. P.; Carr, A. J.; and Graham, A. J.: Total elbow arthroplasty: a systematic review of the literature in the English language until the end of 2003. J Bone Joint Surg Br, 87(4): 437±44, 2005. © 2008, Woodhead Publishing Limited

Replacing the elbow joints

629

2. Bryan, R. S., and Morrey, B. F.: Extensive posterior exposure of the elbow. A triceps-sparing approach. Clin Orthop, 166: 188±92, 1982. 3. Alonso-Llames, M.: Bilaterotricipital approach to the elbow. Its application in the osteosynthesis of supracondylar fractures of the humerus in children. Acta Orthop Scand, 43(6): 479±90, 1972. 4. Gill, D. R., and Morrey, B. F.: The Coonrad-Morrey total elbow arthroplasty in patients who have rheumatoid arthritis. A ten to fifteen-year follow-up study. J Bone Joint Surg Am, 80(9): 1327±35, 1998. 5. Gschwend, N.; Scheier, N. H.; and Baehler, A. R.: Long-term results of the GSB III elbow arthroplasty. J Bone Joint Surg Br, 81(6): 1005±12, 1999. 6. van der Lugt, J. C.; Geskus, R. B.; and Rozing, P. M.: Primary Souter-Strathclyde total elbow prosthesis in rheumatoid arthritis. J Bone Joint Surg Am, 86-A(3): 465± 73, 2004. 7. Kudo, H.; Iwano, K.; and Nishino, J.: Total elbow arthroplasty with use of a nonconstrained humeral component inserted without cement in patients who have rheumatoid arthritis. J Bone Joint Surg Am, 81(9): 1268±80, 1999. 8. Willems, K., and De Smet, L.: The Kudo total elbow arthroplasty in patients with rheumatoid arthritis. J Shoulder Elbow Surg, 13(5): 542±7, 2004. 9. Cheng, S. L., and Morrey, B. F.: Treatment of the mobile, painful arthritic elbow by distraction interposition arthroplasty. J Bone Joint Surg Br, 82(2): 233±8, 2000. 10. Morrey, B. F.: Post-traumatic contracture of the elbow. Operative treatment, including distraction arthroplasty. J Bone Joint Surg Am, 72(4): 601±18, 1990. 11. Dean, G. S.; Holliger, E. H. T.; and Urbaniak, J. R.: Elbow allograft for reconstruction of the elbow with massive bone loss. Long term results. Clin Orthop, 341: 12±22, 1997. 12. McAuliffe, J. A.; Burkhalter, W. E.; Ouellette, E. A.; and Carneiro, R. S.: Compression plate arthrodesis of the elbow. J Bone Joint Surg Br, 74(2): 300±4, 1992. 13. Schneeberger, A. G.; Adams, R.; and Morrey, B. F.: Semiconstrained total elbow replacement for the treatment of post-traumatic osteoarthrosis. J Bone Joint Surg Am, 79(8): 1211±22, 1997. 14. Blaine, T. A.; Adams, R.; and Morrey, B. F.: Total elbow arthroplasty after interposition arthroplasty for elbow arthritis. J Bone Joint Surg Am, 87(2): 286±92, 2005. 15. McKee, M. D.; Pugh, D. M.; Richards, R. R.; Pedersen, E.; Jones, C.; and Schemitsch, E. H.: Effect of humeral condylar resection on strength and functional outcome after semiconstrained total elbow arthroplasty. J Bone Joint Surg Am, 85A(5): 802±7, 2003. 16. Kamineni, S., and Morrey, B. F.: Distal humeral fractures treated with noncustom total elbow replacement. J Bone Joint Surg Am, 86-A(5): 940±7, 2004. 17. Frankle, M. A.; Herscovici, D., Jr.; DiPasquale, T. G.; Vasey, M. B.; and Sanders, R. W.: A comparison of open reduction and internal fixation and primary total elbow arthroplasty in the treatment of intraarticular distal humerus fractures in women older than age 65. J Orthop Trauma, 17(7): 473±80, 2003. 18. Gambirasio, R.; Riand, N.; Stern, R.; and Hoffmeyer, P.: Total elbow replacement for complex fractures of the distal humerus. An option for the elderly patient. J Bone Joint Surg Br, 83(7): 974±8, 2001. 19. Garcia, J. A.; Mykula, R.; and Stanley, D.: Complex fractures of the distal humerus in the elderly. The role of total elbow replacement as primary treatment. J Bone Joint Surg Br, 84(6): 812±16, 2002. © 2008, Woodhead Publishing Limited

630

Joint replacement technology

20. Ray, P. S.; Kakarlapudi, K.; Rajsekhar, C.; and Bhamra, M. S.: Total elbow arthroplasty as primary treatment for distal humeral fractures in elderly patients. Injury, 31(9): 687±92, 2000. 21. Morrey, B. F., and Adams, R. A.: Semiconstrained elbow replacement for distal humeral nonunion. J Bone Joint Surg Br, 77(1): 67±72, 1995. 22. Mansat, P., and Morrey, B. F.: Semiconstrained total elbow arthroplasty for ankylosed and stiff elbows. J Bone Joint Surg Am, 82(9): 1260±8, 2000. 23. Ramsey, M. L.; Adams, R. A.; and Morrey, B. F.: Instability of the elbow treated with semiconstrained total elbow arthroplasty. J Bone Joint Surg Am, 81(1): 38±47, 1999. 24. Kamineni, S.; Adams, R. A.; O'Driscoll, S. W.; and Morrey, B. F.: Hemophilic arthropathy of the elbow treated by total elbow replacement. A case series. J Bone Joint Surg Am, 86-A(3): 584±9, 2004. 25. Sperling, J. W.; Pritchard, D. J.; and Morrey, B. F.: Total elbow arthroplasty after resection of tumors at the elbow. Clin Orthop Relat Res, 367: 256±61, 1999. 26. Yamaguchi, K.; Adams, R. A.; and Morrey, B. F.: Infection after total elbow arthroplasty. J Bone Joint Surg Am, 80(4): 481±91, 1998. 27. Sanchez-Sotelo, J., and Morrey, B. F.: Surgical techniques for reconstruction of chronic insufficiency of the triceps. Rotation flap using anconeus and tendo achillis allograft. J Bone Joint Surg Br, 84(8): 1116±20, 2002. 28. O'Driscoll, S. W., and King, G. J.: Treatment of instability after total elbow arthroplasty. Orthop Clin North Am, 32(4): 679±95, ix, 2001. 29. Sanchez-Sotelo, J.; O'Driscoll, S.; and Morrey, B. F.: Periprosthetic humeral fractures after total elbow arthroplasty: treatment with implant revision and strut allograft augmentation. J Bone Joint Surg Am, 84-A(9): 1642±50, 2002. 30. Kamineni, S., and Morrey, B. F.: Proximal ulnar reconstruction with strut allograft in revision total elbow arthroplasty. J Bone Joint Surg Am, 86-A(6): 1223±9, 2004. 31. King, G. J.; Adams, R. A.; and Morrey, B. F.: Total elbow arthroplasty: revision with use of a non-custom semiconstrained prosthesis. J Bone Joint Surg Am, 79(3): 394± 400, 1997. 32. Loebenberg, M. I.; Adams, R.; O'Driscoll, S. W.; and Morrey, B. F.: Impaction grafting in revision total elbow arthroplasty. J Bone Joint Surg Am, 87(1): 99±106, 2005. 33. Mansat, P.; Adams, R. A.; and Morrey, B. F.: Allograft-prosthesis composite for revision of catastrophic failure of total elbow arthroplasty. J Bone Joint Surg Am, 86-A(4): 724±35, 2004.

© 2008, Woodhead Publishing Limited

26

Replacing joints with pyrolytic carbon

J S T A N L E Y , J K L A W I T T E R and R M O R E , Wrightington Hospital, UK

26.1

Introduction

Upper limb surgeons seek to relieve pain, correct deformity, restore function and improve the appearance of hand, wrist, elbow and shoulder joints damaged by disease or trauma. Joint pain, deformity and dysfunction can result from rheumatoid arthritis, osteoarthritis and post-traumatic conditions. During the past 28 years, pyrolytic carbon (PyC) has been used successfully as a material for the construction of upper limb load-bearing total joint prostheses such as the metacarpophalangeal (MCP) joint and the proximal interphalangeal (PIP) joint. Successful hemi-arthroplasty applications, in which only one component of the joint is replaced, include the MCP joint, the PIP, the carpometacarpal (CMC) joint and radial head as well as inter-positional articulating surface spacers for use in the CMC joint. Advantages of PyC over traditional materials such as polymers and metals that have been realised in this experience are: · · · · · ·

elimination of wear-related failures; absence of osteolytic adverse tissue reactions; excellent fatigue resistance; minimisation of stress-shielding effects and bone resorption; excellent compatibility with joint cartilage and bone tissues; non-cemented fixation via bone apposition.

The intent of this chapter is to introduce the PyC material itself and to survey the successful applications in joint surgery. Further, new applications of PyC as a platform for conservative resurfacing strategies in load-bearing joints are discussed.

26.2

What is pyrolytic carbon?

Pyrolytic carbon (PyC) used for joint replacements is a specific form of elemental carbon that has been tailored for strength, durability and compatibility © 2008, Woodhead Publishing Limited

632

Joint replacement technology

within the often hostile biological environment. The PyC structure is an imperfectly crystalline, turbostratic graphene or graphite-like pure elemental carbon structure. Elemental carbon materials cover a broad spectrum of possible physical and mechanical properties ranging from pure diamond, the hardest material known, to graphite, and on to soot and lamp black, which are some of the softest materials. Diversity in carbon materials arises from the allotropic nature of carbon, meaning that it can exist in multiple pure and partially crystalline forms. Although most pure elemental carbons are biocompatible and biostable, not all carbons have the appropriate mechanical properties for use in long-term loadbearing implant applications. Pyrolytic carbon, as used in joint replacements, is prepared as a coating up to 1 mm in thickness deposited on a graphite substrate. The graphite substrate defines the implant shape and size while the PyC coating provides strength and durability. Typically, the underlying graphite substrate is doped with tungsten to impart radio-opacity because the pure carbon PyC is radio-transparent. Thus, in cases of direct bone apposition, X-ray images of PyC joint replacements show an apparent radiolucent layer around the implant that is actually the radiotransparent PyC layer. In contrast, there exist `carbon' implants that are composite materials consisting of carbon fibres embedded in polymeric materials, such as carbon fibre filled poly(etheretherketone) (PEEK). For these composite carbon-filled materials, the implant properties depend upon the polymer matrix along with the amount, distribution and orientation of the carbon fibre filler. The composite, carbon-filled materials are a completely different class of materials; thus the mechanical properties, durability and clinical experience are not comparable.

26.2.1 Composition The PyC material used in Ascension Orthopedics implants is atomically pure elemental carbon. There are other versions of PyC that are alloyed by the addition of small amounts of silicon or boron, which produces small inclusions of hard silicon carbides or boron carbides within the carbon matrix.

26.2.2 Structure/properties Medical PyC materials have a graphene (graphite-like) turbostratic carbon crystallographic structure with c layer spacing of approximately 0.348 nm. Crystallite (continuous crystalline array regions) sizes are very fine, in the order 2.5±4.0 nm and are randomly oriented. Because of the very small crystallite size and random orientations, bulk PyC physical and mechanical properties are isotropic.1,2 A turbostratic (disordered layers) PyC crystallographic structure is depicted in Fig. 26.1. © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

633

26.1 Allotropic crystalline forms of carbon: diamond, graphite and turbostratic PyC. Turbostratic PyC is a graphene (graphite-like) structure that has order within the graphene sheet layers but disorder between the layers. The disorder between layers greatly strengthens PyC relative to graphite.

A high-resolution transmission electron microscope (TEM) micrograph of pure PyC is given in Fig. 26.2. This is a view through an atomically thin crosssection that shows the turbostratic structure and crystallites.3,4

26.2 High-resolution TEM micrograph of pure PyC turbostratic structure. The wavy lines are the turbostratic graphene layers and the dark spots are crystallites (Ling Ma, Studies on pyrolytic carbons for biomedical applications, Doctoral Dissertation 1997, Univ. California, Los Angeles). © 2008, Woodhead Publishing Limited

634

Joint replacement technology

26.2.3 Production processes Pyrolytic carbons are prepared by heating a hydrocarbon such as propane to approximately 1200±1400 ëC in the absence of oxygen. At high temperature, the hydrocarbon breaks down to carbon free radicals, which then covalently bond to form the solid PyC material. In order to produce isotropic PyC, a fluidised bed reactor is needed. The reactor, illustrated in Fig. 26.3, consists of a vertical tube containing a fine dispersion of refractory material particles, which are levitated using an inert carrier gas, such as nitrogen or helium, fed into the bottom of the tube. Induction coils on the outside of the tube heat the contents. The graphite substrate components are added to the fluidised bed. At temperature, a hydrocarbon gas is added to the mixture and the pyrolysis reaction takes place.5,6 This process is described as a chemical vapour deposition. Implant components are manufactured by placing preformed graphite substrates within the fluidised bed that subsequently become coated with PyC during the reaction. The resulting components are layered structures consisting of an outer shell of PyC encasing and bonded to an inner refractory substrate such as fine-grained graphite. Implants are typically coated to approximate 0.5 mm PyC thickness. The exterior of the PyC coated parts then may be ground and polished for articulating surfaces or left as-coated to enhance bone apposition.7

26.2.4 Beneficial properties for orthopaedic implants Some of the important mechanical and physical properties of the pure and silicon-alloyed PyC materials appropriate for use in long-term implants are given in Table 26.1.8,9 PyC flexural strength, fatigue and wear resistance provide adequate structural integrity for a variety of implant applications. The density is low enough to

26.3 Schematic of reactor used to prepare PyC coatings. © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

635

Table 26.1 Pyrolytic carbon properties Property Flexural strength (MPa) Young's modulus (GPa) Strain-to-failure (%) Hardness (DPH, 500 g load) Density (g/cm3) Poisson's ratio Fracture toughness (MPa m1/2) Fatigue threshold (MPa m1/2 ) Fatigue crack velocity (m/cycle) Coefficient of friction Residual stress (MPa)

Pure PyC

Typical Si-alloyed PyC

493:7  12 29:4  0:4 1:58  0:03 235:9  3:3 1:93  0:01 0:28  0:04 1:68  0:05 1.11 3.98  10ÿ15
K 70.3 0.15 18.2

407:7  14:1 30:5  0:65 1:28  0:03 287  10 2:12  0:01 0:22  0:01 1:17  0:17 0.7 4.15  103
K 88.9 0.15 28.6

allow components to closely approximate the density of the tissues that the implants replace. Relative to orthopaedic applications, Young's modulus is in the range reported for bone7,9 which provides biomechanical compatibility and minimises stress shielding at the prosthesis±bone interface. The beneficial strength and elastic properties for pure PyC are as shown in Fig. 26.4. Fluidised bed isotropic PyCs are remarkably fatigue resistant because crystallographic mechanisms for fatigue crack initiation and damage accumulation are not significant in the PyC at ambient temperatures.10±13 There is strong evidence for the existence of a fatigue threshold that is very nearly the single cycle fracture

26.4 Pure PyC has an excellent match to bone stiffness (wet human femur, L ± longitudinal, T ± transverse) and a higher strength. © 2008, Woodhead Publishing Limited

636

Joint replacement technology

strength.10±15 Paris-Law fatigue crack propagation rate exponents are in the order of 80 and da/dN fatigue crack propagation testing displays clear evidence of a fatigue crack propagation threshold.10±13 In practical terms, fatigue in PyC implants is synonymous with single cycle overload fracture and there have been no clear instances of fatigue failure in a clinical implant during the accumulated 30 year experience.16±18 PyC wear resistance is excellent. Wear testing performed in the 1970s identified titanium alloy, cobalt±chromium alloy and PyC as low wear contact materials for use in contact with PyC.19,20 This study determined that wear in PyC occurred due to an abrasive mechanism and interpreted wear resistance as approximately proportional to the ratio H 2 =2E, where H is the Brinell hardness number and E is Young's modulus. This criterion is related to the amount of elastic energy that can be stored in the wearing surface.20 The greater the amount of stored energy, the greater the wear resistance. Successful lowwearing contact couples used for mechanical heart valves include PyC against itself, cobalt±chromium alloy and ELI titanium alloy. Observed wear in retrieved PyC mechanical heart valve prosthesis implant components utilising PyC coupled with cobalt±chromium alloy is extremely low with PyC wear mark depths of less than 2 m (39.4 micro-inch) at durations of 17 years.21±26 Wear in the cobalt±chromium components was higher, 19 m at 12 years.21±26 Wear depths in valves utilising polymeric components such as the polyacetyl Delrin in contact with cobalt±chromium and titanium alloys are much higher at 267 m at 17 years.22,23 Design of mechanical heart valve prostheses has evolved to bileaflet all PyC devices in which the PyC leaflets articulate directly with the PyC valve housing. In this configuration, the PyC on PyC leafet hinge joint open and closes approximately 40 million times each year. Observed wear depths in all PyC bileaflet valves are less than 3.5 m at 13 years,27,28 which demonstrates that PyC-on-PyC articulation has outstanding wear resistance. Incorporation of PyC in heart valve prostheses has eliminated wear as a failure mode.7,16 PyC is an excellent material for orthopaedic applications because of advantages over metallic alloys and polymers:2,7,9,29,30 · · · · · ·

A modulus of elasticity similar to bone to minimise stress shielding. Excellent wear characteristics. Excellent fatigue endurance. Excellent biocompatibility with bone and hard tissue. Excellent biocompatibility with cartilage. Fixation by bone apposition.

A brief comparison of PyC properties with those of convention orthopaedic implant materials is given in Table 26.2. Pyrolytic carbon (PyC) coatings for orthopaedic implants can reduce wear, wear particle generation, osteolysis and aseptic loosening, thus extending implant useful lifetimes. Furthermore, good © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

637

Table 26.2 Material properties of orthopaedic materials Property

PyC

Al2O3

TZP

CoCrMo

UHMWPE

Density (g/cc) Bend strength (MPa) Young's modulus E (GPa) Hardness H (HV) Fracture toughness K1c (MN/m3/2) Elongation at failure (%) Poisson's ratio H2/2E **

1.93 494 29.4 236* 1.68

3.98 595 400 2400 5

6.05 1000 150 1200 7

8.52 690 (uts) 226 496

0.95 20* 1.17 NA

2 0.28 7.6

0.15 0.2 12.2

0.2

1 0.3 1.8

>300

* The hardness value for PyC is a hybrid definition that represents the indentation length at a 500 g load with a diamond penetrant indentor because PyC elastically completely recovers the microhardness indentation. A replica material such as a cellulose acetate coating, or a thin copper tape is used to `record' the fully recovered indentation length. Although unusual, this operational definition for hardness is a common practice used throughout the PyC heart valve industry. ** Approximate values, there are no exact conversions.

PyC compatibility with the native cartilage and bone joint tissues enables conservative hemi-arthroplasty replacements as an alternative to total joint replacement.

26.3

History of pyrolytic carbon use

Isotropic, fluidised bed pyrolytic carbons (PyC), appropriate for cardiovascular applications originated at General Atomics in the late 1960s as a cooperative effort between an engineer and a surgeon. The engineer, Jack Bokros, was working with pyrolytic carbons as coatings for nuclear fuel particles at General Atomics. The surgeon, Vincent Gott of University of Wisconsin, was searching for thromboresistant materials for cardiovascular applications. Together, they tailored a specific fluidised bed isotropic pyrolytic carbon alloy with the biocompatibility, strength and durability needed for long-term structural implant applications.31

26.3.1 Heart valves In the early 1960s, heart valve prostheses constructed from polymers and metal were prone to early failure from wear, thrombosis and reactions with the biological environment. Prosthesis lifetimes were limited to several years because of wear in one or more of the valve components. Incorporation of PyC as a replacement for the polymeric valve components successfully eliminated wear as an early failure mechanism.27 Subsequently, in most mechanical cardiac valve designs, metallic materials were replaced with PyC by the mid-1970s. © 2008, Woodhead Publishing Limited

638

Joint replacement technology

Since that time, the clinical experience with PyC includes more than 4 million implants and in the order of 20 million patient-years without wear-related device failures.

26.3.2 Finger joints The first use of PyC for joint reconstruction was the development of an implant to replace the metacarpophalangeal (MCP) joint of the hand. Arthroplasty of the MCP joint has for decades been accomplished using flexible silicone rubber interpositional spacers. Silicone rubber implants are effective in reducing joint pain and improving cosmetic appearance; however, they are much less effective in re-establishing functional joint motion and functional pinch and grip strength. The medical literature suggests that silicone rubber MCP spacers are acceptable for use in cases having low strength and limited motion, but are not indicated for use in cases having an outcome potential for high strength and motion (i.e. patients with osteoarthritis, post-traumatic joint conditions or early rheumatoid arthritis). Thus, there exists a clear clinical need for an MCP joint implant that relieves pain and restores normal motion and function. The possibility of PyC as a material for orthopedic applications was recognised and investigated. Preclinical investigations included a series of PyC MCP finger joint replacements implants in baboons.28 Subsequently MCP replacements were implanted in patients between 1979 and 1987.32 A subsequent 10 year follow-up demonstrated excellent performance.33 Thus, PyC was proven to provide excellent performance in small joint orthopedic devices.33

26.4

Review of pyrolytic carbon joint clinical history/ performance

26.4.1 The pyrolytic carbon metacarpophalangeal early experience Cook et al.34 studied hemi-joint implants with a PyC femoral head in the canine hip and observed a greater potential for acetabular cartilage survival in PyC than for cobalt±chromium±molybdenum alloy and titanium alloy femoral heads. There were significantly lower levels of gross acetabular wear, fibrillation, eburnation, glycosaminoglycan loss and subchondral bone change for PyC than the metallic alloys. This study will be addressed in more detail later in this chapter. Tian et al.35 surveyed in vitro and clinical in vivo PyC orthopaedic implant studies conducted during the 1970s through the early 1990s and concluded that PyC demonstrated good biocompatibility and good function in clinical applications. Preclinical evaluations of PyC MCP joint implants were conducted by Cook et al.28 Four non-constrained uncemented prostheses were implanted into long © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

639

finger joints in four baboons, one cemented PyC implant and one cemented polyethylene/metal implant were used as controls. The implants were evaluated by radiography and histology nine months post-insertion. The uncemented PyC implants all exhibited functional fixation with no evidence of bone resorption and no evidence of foreign body reaction or intracellular particles in the soft tissues. Both cemented prostheses evidenced bone resorption and/or gross implant loosening, which matched the clinical experience with cemented prostheses. The study results were interpreted as a demonstration of the potential for biological fixation with PyC and the possibility for PyC stemmed implants to offer significant improvement as a material for joint reconstruction. Based on successful animal use, Dr Beckenbaugh concluded that the PyC MCP joint replacements were suitable for therapeutic use in his patients. During the time between 1979 and 1987, 151 PyC MCP prostheses were implanted in 53 patients by Dr Beckenbaugh and Dr Linscheid.32 A ten-year follow-up of PyC MCP finger joint replacements was published in 1999.33 Twenty-six patients with 71 implants from this group were available for long-term evaluation at an average of 11.7 years after implantation. This follow-up revealed that the use of the PyC MCP prostheses had improved the functional range of motion and position of fingers as compared to preoperative conditions: · · · · ·

No complaints of implant specific pain. A 13ë improvement in the arc of motion. A 16ë decrease in extension lag. Progressive ulnar deviation had been arrested. Patients satisfied with functional and cosmetic results.

There was a high prevalence of joint stability with no adverse remodelling or resorption of bone observed. A high percentage of implants showed evidence of direct bony fixation. Preliminary survivorship analysis demonstrated an average annual revision rate of 2.1%, and a 16-year survival rate of 70.3%. None of the revised implants had any visible changes due to wear or deformity of the surfaces or stems, and no evidence of intracellular particles or particulate synovitis was found. These very favourable long-term results with the PyC MCP implants demonstrated excellent performance. The device was biologically and biomechanically compatible, wear resistant, robust and durable. State-of-the-art CAD-CAM (computer-aided design/manufacture) and manufacturing methods were subsequently used to refine the MCP design in order to simplify the surgical implantation techniques, to make the stem shape more anatomical, resistant to lateral rotation and to expand the range of anatomical sizes. The current MCP design is shown in Fig. 26.5. European CE mark approval was obtained in 1999 and US Food and Drug Administration (FDA) premarket approval was obtained in 2001.

© 2008, Woodhead Publishing Limited

640

Joint replacement technology

26.5 AOI MCP and proximal interphalangeal (PIP) components and X-ray. Components are shown in lateral views relative to the X-ray.

26.4.2 The pyrolytic carbon metacarpophalangeal current experience The promising results shown for the MCP in the 1999 long-term study have been confirmed in the past seven years' experience in worldwide utilisation. Approximately 8000 joint arthroplasties with the refined Ascension Orthopedics MCP replacements have been performed to date. The MCP implant has the advantages of requiring minimal bone removal and approaches a `resurfacing'. Collateral ligaments and soft tissues are preserved, thus anatomical alignments can be restored along with normal kinematics, strength and cosmetic appearance. The non-constrained design allows an anatomical range of motion: 15ë in radial and ulnar abduction, 20ë in hyperextension and 90ë in flexion. With the strength and durability of PyC, no revisions have been needed because of post-surgical fracture (fractures due to traumatic injury are possible) and no wear-related failures have occurred. For the best results, the non-constrained PyC MCP prostheses require intact soft tissues (tendons, ligaments, muscles, etc.) for stability, as occurs in patients with osteoarthritis and post-traumatic injury. Patients with rheumatoid arthritis may benefit in the early stages of disease progression. But, the effectiveness of the non-constrained joint replacements decreases in rheumatoid patients with the loss of soft-tissue integrity.

26.4.3 The pyrolytic carbon proximal interphalangeal joint The success of PyC MCP arthroplasty has been extended to the proximal interphalangeal (PIP) joint. An anatomic PyC bicondylar PIP replacement designed and manufactured by AOI, as an alternative to PIP arthrodesis, obtained European CE mark approval in 2001 and FDA approval as a Humanitarian Device in 2002 (see Fig. 26.5). The PIP offers the same advantages as the MCP along with additional features fine-tuned for the PIP joint. The additional © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

641

features include bicondylar articulating surfaces and anatomic stems for rotational stability, a dorsal notch for extensor, central slip, tracking to increase stability during flexion/extension, a dorsal collar to suppress subluxation and a low profile for minimal bone removal. An anatomical range of motion from 20ë hyperextension to 100ë flexion is provided. Furthermore, the bicondylar design suppresses deviation deformity allowing use in border digits. To date, approximately 8000 joint arthroplasties have been performed with the AOI PIP. Clinical outcomes with the PyC PIP are promising. Clinical results demonstrate pain relief, increased range of motion, improved grip and pinch strength, and improved appearance with general agreement that pyrolytic PIP arthroplasty is superior to arthrodesis in the management of PIP joint arthritis. Furthermore, there is evidence of superior function compared with silicone spacers in terms of grip and pinch strength restoration. Fixation in the PyC stem occurs because of appositional bone growth, not osseous integration.36 The non-cemented press fit stem of the PyC prostheses achieve fixation by a mechanism that involves the formation of a tight fibrous membrane and appositional bone growth according to Wolff's law.37 Final PyC implant fixation occurs 6±24 months after surgery.36 A fixed prosthesis remains firmly seated in the medullary canal and does not `piston' with articulation as do silicone spacers. The press-fit removes the risk of tissue damage due to heat generated during polymerisation, and, if a revision is needed, it is possible because the implant can be extracted. Because of the radiolucency of the pyrolytic carbon coating and the fibrous membrane, there is an apparent lucent line or zone around the implant. In good fixation, the lucent zone is uniform in width. However, variations in the angle of the radiographic beam may distort the lucent zone to give the impression of loosening. Review of the literature describing MCP and PIP arthroplasty using PyC prostheses provides a clear picture of the outcomes surgeons intend to achieve from the procedure and the complications encountered during the course of treatment. The desired outcomes of joint arthroplasty are: (1) relief of pain, (2) improvement in hand function, (3) reduction or correction of finger deformity, (4) improvement in the appearance of the hand, and (5) obtaining a long-term result. The ability of the PyC MCP and PIP prostheses to attain these outcomes is generally assessed as `positive' or `promising'. Longer follow-up periods will help better to determine the efficacy of the devices.36 Relative to traditional arthrodeses, the results of arthroplasty with PyC implants are clearly superior in terms of restored function. Unlike contemporary silicone spacer implants or implants with metal/polymer articulations, the PyC implants do not exhibit measurable wear and thus will not generate wear particles to initiate osteolytic processes and synovitis. PyC implant fractures due to traumatic injury are possible, but PyC fractures due to biomechanical loading and intrinsic wear processes alone are extremely unlikely. © 2008, Woodhead Publishing Limited

642

Joint replacement technology

Patient selection is important because of the necessity of good bone and soft tissue quality. Furthermore, the interaction of the soft tissues and the PyC implant must be carefully considered during surgery and post-operative care.36 Fluoroscopy is recommended as a means to verify fit and alignment and to adjust the joint spaces because misalignment can result in squeaking and can produce off-axis forces that can lead to migration. In addition, complications may arise36 in which it is necessary to perform tenolysis and additional soft tissue reconstruction or implant upsizing or downsizing to correct instability. A significant virtue of the PyC appositional fixation is that revision is always available as an option, in contrast to fixation by cement or osseous integration.

26.5

Design and testing of pyrolytic carbon joint replacement implants

PyC is a brittle material, which means that it does not undergo ductile, plastic deformation as do metals. As a practical consideration, special design methods have been developed for brittle materials, such as Weibull theory38 that must be used in order to produce safe and reliable designs. An example of the procedure used to design and validate the strength of the PyC MCP implants follows. The design of a PyC implant having sufficient strength to ensure mechanical reliability requires knowledge of two critical factors: (1) the biomechanical performance requirement for the device and (2) the failure probability of the device under service loads. To determine the appropriate biomechanical load requirements for the PyC MCP implant, a review of the biomechanics literature was conducted to establish the magnitude and direction of the joint reaction force (JRF) resulting from demanding hand function. Performance parameters of interest were: (1) grasp and pinch strength for normal and diseased hand and (2) the magnitude and direction of the MCP JRF resulting from the isometric functions of grasp and pinch. MCP biomechanical requirement results are summarised in Table 26.3.39 The key parameter was a worst case JRF of 350 N which represents a maximum likely value for activities of daily living. Table 26.3 MCP biomechanical performance requirements Parameter

Grip strength (N)* Pinch strength (N)* Joint flexion angle for grip and pinch (ë) Joint reaction force JRF (N) Angle of JRF relative to long axis

Male dominant hand

Female dominant hand

380 60 60 350 20

228 42 60 350 20

Diseased arthritic hand 5±20

*Non-dominant hand grip and pinch strength is approximately 90% of the dominant hand strength.40 © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

643

Test methods based upon existing standards for orthopaedic components were adapted to evaluate the MCP design performance relative to the biomechanical requirements. Tests were then performed to determine single cycle fracture strength and to evaluate endurance for 10 million cycles. Test results were then evaluated relative to the biomechanical requirements using Weibull theory to determine the risk of failure. For brittle materials, such as PyC, Weibull theory accounts for dispersion in strength by using an empirical fit of component strength data to a statistical distribution function. This statistical Weibull distribution provides estimates of both the probability of failure P…x† and a reliability factor, e.g. …1 ÿ P…x†† as a function of the applied load or stress x. The Weibull distribution has the form: P…x† ˆ 1 ÿ expÿ…x=0 †

m

26:1

where P…x† is the probability of failure for the load or stress x, m is a dimensionless factor that is related to the dispersion in failure strength and 0 is related to the median failure strength in units of load or stress. Analysis of component strength data for the mid-size MCP yielded the following Weibull distribution parameters: m ˆ 11:9 and 0 ˆ 991 N. The cumulative probability of failure Weibull distribution, equation [26.1], is plotted in Fig. 26.6 using these parameters. Test results are summarised in Table 26.4. The strength of the mid-size implant was determined to be in the order of 983 N, which gives a safety factor

26.6 Cumulative Weibull probability of failure distribution for the mid-size MCP (size 30) as a function of load in newtons. Distribution parameters are m ˆ 11:9 and 0 ˆ 991 N. The slope of the ascending part of the curve is proportional to m. Low m values have a broad slow rise (m < 8), while high m values (m > 20) rise steeply. The mid point of the rise is approximately the mean strength (50% probability of failure). © 2008, Woodhead Publishing Limited

644

Joint replacement technology Table 26.4 MCP design performance verified by testing Parameter Assumed JRF (N) Strength measured (N) Safety factor Weibull probability of failure

Value 356 983 2.75 0.000 005 6

of 2.75 relative to the worst case JRF and a Weibull risk of failure of approximately 6 in a million. The interpretation of the test results is as follows. For a mid-size implant placed in the index or long finger of a healthy adult male, with one-third of the stem bone support lost, and subjected to a maximum 350 N JRF, the estimate of fracture rate is very conservative, because all patients will not be healthy adult males and will include females, the elderly and those with hand strength reduced due to disease or trauma. Experience with the PyC MCP joint demonstrates the effectiveness of this approach in achieving implant safety and reliability. From the period October 1999 to September 2007, approximately 8000 Ascension PyC MCP joints have been implanted with no documented case of postoperative implant stem fracture occurring due to normal hand function. Wear testing was performed for 10 million cycles using a Co±Cr alloy bearing on UHMWPE as a control. No measurable wear occurred for PyC specimens and, as expected, measurable penetrating wear occurred in the UHMWPE control. Probabilistic rather than deterministic design methods must be applied when brittle materials are used in a structure. Knowledge of the failure probability and the mechanical performance requirements provide the basis for determining a safety factor sufficient to ensure an implant device having a brittle material component, such as PyC, is safe and reliable.

26.6

Hemi-joint arthroplasty

Joint reconstruction can be either total arthroplasty or hemi-arthroplasty. Total joint arthroplasty is indicated when both sides of the joint are diseased and an implant consisting of two parts, one for each side of the joint, is used to replace the entire joint. Joint hemi-arthroplasty is a more conservative procedure indicated when only one side of the joint is diseased or damaged and a one-part implant is used to replace only that side of the joint. Current hemi-arthroplasty implants typically use a Co±Cr metal alloy as the portion of the device that contacts the tissues (cartilage and or bone) of the opposing native joint. Damage to native joint tissues from contact with Co±Cr alloy implants, resulting in loss of joint space and recurring pain, is a known shortcoming of current hemiarthroplasty procedures. PyC (pyrolytic carbon) has been shown to result in significantly less damage to cartilage and bone than Co±Cr alloy. © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

645

26.6.1 Compatibility of pyrolytic carbon with cartilage Cook et al.34 studied cartilage degradation in 45 canine acetabula after implantation of prostheses with articulating surfaces of pyrolytic carbon, Co±Cr alloy and titanium alloy for a period ranging from 2 weeks to 18 months. Gross specimens and histological sections were compared with the non-operated (control) acetabulum of the same animal. Cartilage articulating with pyrolytic carbon exhibited significantly lower levels of gross wear, fibrillation, eburnation, glycosaminoglycan loss and subchondral bone change than with metallic surfaces. Survivorship analysis showed a 92% probability for cartilage articulating with PyC at 18 months, as compared with only a 20% probability of survival for cartilage with either of the metallic alloys. Cook concluded that although the exact mechanism for the enhanced cartilage survival observed with the pyrolytic carbon implants remains unclear, it could be due to the lower elastic modulus of the PyC or possibly due to the surface chemistry of the PyC. Figure 26.7 illustrates the condition of the canine acetabula following contact with PyC and with CoCr alloy. Cook reported similar results comparing the wear characteristics of aluminum oxide (Al2O3) to PyC.37 The poor cartilage wear characteristics of the metal alloys reported by Cook are consistent with the clinical results for hemi-arthroplasty implants made of metal alloy as summarised by van der Meulen:41 Compared to total joint replacement, hemi-arthroplasty procedures are more conservative; involving shorter surgical time and lower medical and prosthesis costs. Hemi-arthroplasty success, however, is often limited by pain from cartilage erosion and loss of joint space. The severe cartilage degradation resulting from articulation with a metal implant is of great concern if hemi-arthroplasty is to be a more successful surgical procedure.

Kawalec et al.42 investigated PyC and Co±Cr alloy as materials for hemiarthroplasty in an animal model mimicking an arthritic joint. PyC and Co±Cr alloy resurfacing implants were placed in the canine knee joint. The cartilage on the lateral side of the tibial plateau was abraded to create a full-thickness, arthritictype defect which exposed the subchondral bone. PyC and Cr±Cr alloy implants were placed in the lateral femoral condyle in contact with the subchondral bone exposed by the cartilage defect and the joints evaluated after a year. Histogical

26.7 Canine acetabulae following hemi-arthroplasty in contact with femoral heads, PyC (left) and CoCr (right). Acetabulae in contact with PyC exhibited significantly lower levels of gross wear, fibrillation, eburnation and subchondral bone changes than did acetabulae in contact with CoCr.34 © 2008, Woodhead Publishing Limited

646

Joint replacement technology

examination of the tibial defects revealed a smooth bony surface for both implant groups. Microscopic surface cracks in the subchondral bone were present adjacent to the implants being seen in 14% of the PyC implants and in 100% of the Co±Cr alloy implants. Fibrocartilage regeneration was seen in 86% of the PyC implants and in 25% of the Co±Cr implants. Kawalec and colleagues concluded that PyC implants were better tolerated as hemi-arthroplasty implants in the canine arthritic joint model than Co±Cr alloy implants.42 Ascension Orthopedics, located in Austin, Texas, has developed several hemi-arthroplasty and inter-positional arthroplasty PyC implants for the upper extremity. The implants are in some cases designed to articulate with cartilage of the opposing native joint surface and in other cases are intended to articulate with bone tissue. Four PyC implants for reconstruction of the base of the thumb (CMC) joint are shown in Figs 26.8 and 26.9 as examples of hemi-arthroplasty and inter-positional PyC implants.

26.8 Hemi-arthroplasty CMC implants. CMC PyroSaddle and CMC PyroHemi-Sphere. © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

647

26.9 Hemi-arthroplasty CMC implants. CMC PyroSphere and CMC PyroDisk.

The Saddle CMC is a hemi-arthroplasty implant designed to articulate with the saddle-shaped surface of the native trapezium and the PyroHemi-Sphere is a hemi-arthroplasty designed to be a ball and socket articulation where the spherical head of the implant is placed in a mating concave cavity created in the body of the trapezium. The PyroSphere is an inter-positional arthroplasty where the spherical implant is placed in mating concave cavities created in the base of the metacarpal and in the body of the trapezium. The PyroDisk is an interpositional arthroplasty to be inserted into shallow concavities created in the metacarpal and trapezium bones and designed with a central hole through which a tendon graft can be placed to stabilise the implant. Ascension Orthopedics has also developed a carbon modular radial head (CMRH) hemi-arthroplasty prosthesis having a titanium stem and a PyC on cartilage articulating surface, as shown in Fig. 26.10. Radial head arthroplasty is common for young post-traumatic patients where the implant will be subjected to demanding function for many decades. The CMRH has a unique snap-lock mechanism for attaching the PyC head to the metal stem that eliminates tensile © 2008, Woodhead Publishing Limited

648

Joint replacement technology

26.10 Hemi-arthroplasty Carbon Modular Radial Head (CMRH) implant.

residual stresses produced by traditional taper lock attachments. The elimination of tensile residual stresses is important because of potential for extended in vivo durability compared with other devices using taper lock attachments. European CE mark approvals have been obtained for all of these PyC devices, in addition to FDA 510K approvals, with the exception of the PyroDisk, which is pending.

26.6.2 Compatibility of pyrolytic carbon with bone When used in hemi-arthroplasty, the bearing surface of the implant contacts the cartilage of the opposing native joint surface and the compatibility of the implant material with cartilage is of first importance. In the case of metal hemiarthroplasty implants, wear damage to cartilage results in loss of the cartilage covering the joint, loss of joint space and exposure of the underlying bone. Destruction of cartilage and painful contact with the underlying bone tissue is a recognised failure mode of hemi-arthroplasty. Although PyC has been shown to be much less damaging to cartilage than metal alloys, it is possible in the long term that PyC will be in contact with bone tissue when used in a hemiarthroplasty implant. Laboratory wear testing has demonstrated that PyC is also remarkably less damaging to bone tissue than metal alloys or ceramics. Identically shaped and finished PyC and Co±Cr alloy saddle-shaped implants intended for replacement of the base of the metacarpal bone of the thumb were wear tested against bone using a joint function simulator. The tests measured the amount of bone loss after 1 million, 2.5 million and 5 million cycles of joint © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

649

motion. Following 1 million cycles, the Co±Cr alloy specimens had eroded approximately 100 times more bone volume than identical PyC specimens. Continued testing of the Co±Cr alloy specimens was not possible because bone erosion had penetrated the bone specimen to the metal mounting fixture. Wear testing of the PyC was continued to 5 million cycles and even after this time the PyC wear volume was less than one-tenth that observed with the Co±Cr alloy after 1 million cycles. The appearance of the bone surfaces after 1 million cycles and a plot of the wear volume data are shown in Fig. 26.11. The results of this experiment have been reproduced using spherical test specimens of PyC, titanium alloy and zirconia ceramic tested against bone and the PyC modular radial head prosthesis and Co±Cr alloy metal modular radial head prosthesis. In the test comparing bone bearing against the PyC modular radial head and Co±Cr alloy metal modular radial head prostheses. wear damage to bone caused by the PyC modular implant was remarkably less than the control. In fact, the Co±Cr radial head had consumed all of the available bone counter-face by 0.5 million cycles, while the PyC/bone couples endured 5 million cycles. Obvious wear

26.11 Above: appearance of bone counter-face after 1 million cycles in contact with CoCr (left) and PyC (right). Below: plot of bone volume vs. number of cycles. © 2008, Woodhead Publishing Limited

650

Joint replacement technology

26.12 CMRH implants following wear testing in contact with bone. The black PyC implants and bone counter-faces endured 5 million cycles. The CoCr control consumed the bone counter-face at 0.5 million cycles and was removed from the test. The CoCr articulating surface is heavily scored after 0.5 million cycles while the PyC articulating surfaces retain the original bright surface finish after 5 million cycles.

damage occurred on the bearing surface of the Co±Cr control specimen while no visible damage occurred on the bearing surface of the PyC implants as shown in Fig. 26.12. The results of the comparative bone wear testing are even more remarkable than the comparative cartilage wear results reported by Cook et al.28,34 Metal alloys and zirconia ceramic cause massive wear damage against bone when compared with PyC. It is likely that the pain seen clinically following loss of joint space and implant to bone contact with metal hemi-arthroplasty is the result of aggressive wear damage to the underlying bone caused by the metal materials. It now seems possible that even in conditions where PyC contacts bone when used as a hemi-arthroplasty, its favourable wear characteristics contribute to reduced joint pain and acceptable joint function. In addition to the material of construction, other factors such as proper joint kinematics and bearing surface size and shape are necessary in realising a hemi-arthroplasty that will result in minimal wear damage to native joint tissues. While animal and laboratory test results are good indicators, results from clinical use of PyC in hemi-arthroplasty applications are necessary to validate a PyC hemi-arthroplasty bearing surface provides benefits to patients. Although early clinical results are encouraging, more clinical experience will be necessary to truly validate the potential for benefits to patients. Several surgeons aware of the potential for PyC as a material for hemiarthroplasty have used one component PyC MCP implants in trauma cases, with one joint surface remaining intact, for hemi-arthroplasty. The results have been encouraging. Mark Ross and his colleagues have reported their experiences with © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

651

ten PyC hemi-arthroplasties (eight PIP and two MCP) with an average follow up of 13 months (range 3 months to 23 months) in which all of the joints remain in situ.43 Some examples follow. Mark Ross, MD, of Brisbane, Australia, has used one component of a twocomponent PyC total MCP joint implant to treat patients following traumatic damage to only one side of the joint. Initial clinical and functional results have been optimistic. An example is a young woman who suffered a fracture of the base of the first phalanx due to a sports injury. The distal component of a total PyC MCP implant was used as a hemi-arthroplasty to replace the base of the injured phalanx. Twenty-three months following implant surgery a tenolysis was performed to release a stiff joint resulting from to the trauma. A picture of the flexed MCP joint at the time of tissue release surgery shows the native cartilage of the metacarpal head to be healthy and undamaged as shown in Fig. 26.13. The excellent survival of the native joint surface is obvious. Robert Beckenbaugh treated a painful base of the thumb joint with a PyroHemi-Sphere implant in a female patient. The patient had a fall four months following surgery resulting in a fracture of the trapezium as shown in Fig. 26.14. The implant was not damaged. The fracture remained unresolved seven months

26.13 Appearance of native joint tissues in hemi-arthroplasty contact with MCP during a tenolysis procedure 23 months post-implantation. The native cartilage of the metacarpal head, shown in flexion, appears to be healthy and undamaged. The excellent survival of the native joint surface in contact with PyC is obvious. © 2008, Woodhead Publishing Limited

652

Joint replacement technology

26.14 PyroHemi-Sphere in a patient with an unresolved fracture of the trapezium caused by a fall. The trapezium counter-face (arrow) demonstrates a glistening fibrocartilage surface.

post-operatively. The patient was then successfully treated with a suspension arthroplasty; the implant was left in place and the trapezium removed. At time of removal the trapezium appeared to have a glistening fibrocartilage surface on the portion of the trapezium bearing against the PyC implant as shown in the right-hand frame of Fig. 26.14. Calcified histologic sections and microradiographs produced from the trapezium specimen are shown in Fig. 26.15. A dense, avascular, fibrocartilagenous tissue is observed to have formed between the bone and PyC implant bearing surface. The corresponding microradiograph suggests that a continuous bony layer is developing beneath fibrocartilage. The bony layer appears to resemble formation of a native joint subchondral plate. Although anecdotal and speculative, the histological sections and microradiographs suggest the initial PyC to bone-bearing surface has not resulted in continuous penetrating wear of trapezial bone but instead allowed for formation of a stable fibrocartilagenous

26.15 A dense, avascular, fibrocartilagenous tissue (F) is observed to have formed between the bone (B) and PyC implant bearing surface (I). The corresponding microradiograph suggests that a continuous bony layer is developing beneath fibrocartilage. The bony layer appears to resemble formation of a native joint subchondral plate. © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

653

tissue between the implant and bone. The potential benefits resulting from a hemi-arthroplasty that initially bears on bone and in time transforms to a stable pseudo-articular joint structure are enormous.

26.7

Conclusion

To date, PyC has shown great promise for long-term success in total joint arthroplasty in small, non-load-bearing joints in the upper extremities. Several PyC hemi-arthroplasty applications, in non-load-bearing joints, also demonstrate good results and are expected to provide a more durable alternative than metals. However, all of the PyC joint replacements require careful patient selection for good quality bone and soft tissues. Furthermore, prosthesis sizing, alignment and interactions with soft tissue are critical considerations during surgery and postoperative rehabilitation. The expectation is that PyC implants in joint applications will prove more functional, aesthetic, durable and complication free than existing alternatives.

26.8

Forward-looking statement with respect to pyrolytic carbon in orthopaedics

The PyC clinical experience in small joints has demonstrated advantages relative to other traditional materials for joint replacements: excellent durability in total joint applications, enhanced compatibility with cartilage and bone in hemiarthroplasty and potentially significantly extended device lifetimes. With contemporary design and manufacturing techniques, extensions of PyC as a platform for conservative resurfacing strategies in load-bearing joints are in development. Challenges in larger joints involve a modular structure for the implants that joins PyC for articular resurfacing to bend-resistant metal components for stems. Excellent solutions to this problem have been developed. Furthermore, it is possible to selectively treat PyC surfaces with biologically active substances such as hydroxyapatite, or other selective surface treatments as may be desired. Such selective surface treatments enable a `localised' spot or partial resurfacing for isolated defects in large articular surfaces because of PyC compatibility with cartilage and bone. In effect, localised defects within a large articular surface can be repaired using a small PyC insert rather than replacing the entire surface. Prostheses with three differentiated surfaces can be realised, with one surface optimised for contact with articular cartilage, the opposite surface for contact and fixation with the subchondral bone and the third edge surface optimised for fixation with fibrocartilage. Thus, there are a number of possibilities for PyC in joint applications that can be imagined. Realisation of new PyC devices will require interactions between the surgeons' ability to envision and the engineers' ability to enable.

© 2008, Woodhead Publishing Limited

654

26.9

Joint replacement technology

References

1. Bokros J, Carbon biomedical devices, Carbon 1977; 15: 355±371. 2. Haubold AD, More RB, Bokros JC, Carbons, Handbook of Biomaterial Properties, Black J, Hastings G (eds.), Chapman & Hall, London, 1998: 464±477. 3. Ma L, Sines G, High resolution structural studies of a pyrolytic carbon used in medical applications, Carbon 2002; 40: 451±454. 4. Ma L, Studies on pyrolytic carbons for biomedical applications, Doctoral Thesis 1997, Univ. California, Los Angeles. 5. Bokros JC, Deposition, structure and properties of pyrolytic carbon. In Walker PL (ed), Chemistry and Physics of Carbon, Vol. 5. Marcel Dekker, Inc., New York, 1969: 1±118. 6. Kaae JL, The mechanism of deposition of pyrolytic carbon, Carbon 1985; 23(6): 665±667. 7. More RB, Haubold AD, Bokros JC, Pyrolytic carbon for long-term medical implants. In Ratner B, Hoffman A, Schoen F, Lemons J (eds), Biomaterials Science: An Introduction to Materials in Medicine, 2nd edition, Elsevier Academic Press, 2004; 170±180. 8. Haubold AD, Shim HS, Bokros JC, Carbon in medical devices. In Williams DF (ed.), Biocompatibility of Clinical Implant Materials, CRC Press, Boca Raton, FL, 1981; 2: 3±42. 9. More RB, Sines G, Ma L, Bokros JC, Pyrolytic carbon. In Encyclopedia of Biomaterials and Biomedical Engineering, Marcel Dekker, 2004. 10. Gilpin CB, Haubold AD, Ely JL, Fatigue crack growth and fracture of pyrolytic carbon composites. In Ducheyne P, Christiansen D (eds), Bioceramics, Vol. 6, Butterworth-Heinemann Ltd, Oxford, 1993: 217±223. 11. Ma L, Sines G, Fatigue of isotropic pyrolytic carbon used in mechanical heart valves, J Heart Valve Dis 1996; 5 (Suppl. I): S59±S64. 12. Ma L, Sines G, Unalloyed pyrolytic carbon for implanted heart valves, J Heart Valve Dis 1999; 8(5): 578±585. 13. Ma L, Sines G, Fatigue behavior of pyrolytic carbon, J Biomed Mat Res 2000; 51: 61±68. 14. Ritchie RO, Dauskardt RH, Yu W, Brendzel AM, Cyclic fatigue-crack propagation, stress corrosion and fracture toughness behavior in pyrolite carbon coated graphite for prosthetic heart valve applications, J Biomed Mat Res 1990; 24: 189±206. 15. Beavan LA, James DW, Kepner JL, Evaluation of fatigue in pyrolite carbon. In Ducheyne P, Christiansen D (eds), Bioceramics, Vol. 6, Butterworth-Heinemann Ltd, Oxford, 1993: 205±210. 16. Haubold AD, On the durability of pyrolytic carbon in vivo, Med Prog Through Technol 1994; 20: 201±208. 17. Schoen FJ, Carbons in heart valve prostheses: foundations and clinical performance. In Zycher M (ed.) Biocompatible Polymers, Metals and Composites, Technomic, Lancaster, PA, 1983: 240±261. 18. Bokros JC, Haubold AD, Akins RJ, Campbell LA, Griffin CD, Lane E, The durability of mechanical heart valves replacements: past experience and current trends. In Bodnar E, Frater RWM (eds) Replacement Cardiac Valves, Pergamon Press, New York, 1991: 21±47. 19. Shim HS, Schoen FJ, The wear resistance of pure and silicon-alloyed isotropic carbons, Biomat Med Dev Art Org 1974; 2(2): 103±118. 20. Shim HS, The wear of titanium alloy, and UHMW polyethylene caused by LTI © 2008, Woodhead Publishing Limited

Replacing joints with pyrolytic carbon

655

carbon and Stellite 21, J Bioengr 1977; 1: 223±229. 21. Schoen FJ, Titus JL, Lawrie GM, Durability of pyrolytic carbon-containing heart valve prostheses, J Biomed Mater Res 1982; 16: 559±570. 22. More RB, Silver MD, Pyrolytic carbon prosthetic heart valve occluder wear: in vivo vs. in vitro results for the BjoÈrk-Shiley prosthesis, J Appl Biomater 1990; 1: 267± 278. 23. More RB, An examination of two retrieved long-term human implant Bjork-Shiley valves, Med Prog Through Technol 1994; 20: 195±200. 24. More RB, Haubold AD, Silver MD, Pyrolytic carbon wear in retrieved mechanical heart valve prosthesis implants, 25th Annual Meeting of the Society for Biomaterials, 1999: 553. 25. More RB, Chang BC, Hong YS, Cao BK, Butany J, Wear analysis of retrieved mitral bileaflet mechanical heart valve prostheses, Presented to the Society for Heart Valve Disease, 1st Biennial Symposium, London, June 2001. 26. More RB, Haubold AD, Silver MD, Pyrolytic carbon wear in retrieved mechanical heart valve prosthesis implants, 25th Annual Meeting of the Society for Biomaterials, 1999: 553. 27. Ely JL, Emken MR, Accuntius JA, Wilde DS, Haubold AD, More RB, Bokros JC, Pure pyrolytic carbon: preparation and properties of a new material, On-X carbon for mechanical heart valve prostheses, J Heart Valve Dis 1998; 7: 626±632. 28. Cook SD, Beckenbaugh R, Weinstein A, Klawitter J, Pyrolite carbon implants in the metacarpophalangeal joint of baboons, Orthopedics 1983; 6(8): 952±961. 29. Reilly DT, Burstein AH, Frankel VH, The elastic modulus for bone, J Biomech 1974; 7: 271. 30. Reilly DT, Burstein AH, The mechanical properties of bone, J Bone Joint Surg Am 1974; 56: 1001. 31. LaGrange LD, Gott VL, Bokros JC, Ramos MD, Compatibility of carbon and blood. In Hegyeli RJ (ed), Artificial Heart Program Conference Proceedings, US Government Printing Office, Washington, DC, 1969; Chapter 5: 47±58. 32. Beckenbaugh RD, Preliminary experience with a non-cemented nonconstrained total joint arthroplasty for the metacarpalphalangeal joints, Orthopedics 1983; 6(8): 962± 965. 33. Cook SD, Beckenbaugh RD, Redondo J, Popich LS, Klawitter JJ, Linscheid RL, Long term follow-up of pyrolytic carbon metacarpophalangeal implants, J Bone Joint Surg 1999; 81A(5): 635±648. 34. Cook SD, Thomas KA, Kester MA, Wear characteristics of the canine acetabulum against different femoral prostheses, J Bone Joint Surg 1989; 71B: 189±197. 35. Tian CL, Hetherington VJ, Reed S, A review of pyrolytic carbon: application in bone and joint surgery, J Foot and Ankle Surg 1993; 32(5): 490±498. 36. Bravo CJ, Rizzo R, Hormel KB, Beckenbaugh RD, Pyrolytic carbon proximal interphalangeal joint arthroplasty: results with minimum two-year follow-up evaluation, J Hand Surg 2007; 32A: 1±11. 37. Cook SD, Anderson RC, Thomas KA, Kester MA, Haddad RJ, Jr, Articular cartilage response to implant material modulus and method of fixation, 11th Annual Meeting of the Society for Biomaterials, San Diego, CA, 25±28 April 1985. 38. DeSalvo GJ, Theory and Structural Design Applications of Weibull Statics, Westinghouse Electric Co Astronuclear Laboratory report WANL-TME-2688, 1970. 39. 19 Nov 2001PMA # P000057 40. Incel NA, Ceceli E, Durukan PB, Erdem HR, Yorgancioglu R, Grip strength: effect of hand dominance, Singapore Med J 2002; 43(5): 234±237. © 2008, Woodhead Publishing Limited

656

Joint replacement technology

41. van der Meulen M et al., guide.stanford.edu/96reports/96dev5.html 42. Kawalec JS, Hetherington VJ, Melillo TC, Carbin H, Evaluation of fibrcartilage regeneration and bone response at full-thickness cartilage defects in articulation with pyrolytic carbon or cobalt±chromium ally hemiarthroplaties. J Biomed Mater Res 1998; 41: 534±540. 43. Couzens G, Hussain N, Gilpin D, Ross M, Pyrocarbon PIPJ and MCPJ hemiarthroplasty, abstract presented IFSSH, Budapest 2004.

© 2008, Woodhead Publishing Limited