Philippe
Daniel J. Neyret
Mollura
Chris Butcher
Matthew
P. Lungren
Guillaume
Michael
R.B.Demey
Evans
Editors

Clinical Medicine
Covertemplate
Surgery
of the Knee
SubtitleEdition
for
Second
Clinical Medicine Covers T3_HB
Second Edition

1123
3
2

Surgery of the Knee

Philippe Neyret • Chris Butcher
Guillaume Demey
Editors

Surgery of the Knee
Second Edition

Editors
Philippe Neyret
Infirmerie Protestante
Lyon
Caluire
France

Chris Butcher
Healthpoint
Abu Dhabi
UAE

Guillaume Demey
Clinique de la Sauvegarde
Lyon Ortho Clinic
Lyon
France

ISBN 978-3-030-19072-9    ISBN 978-3-030-19073-6
https://doi.org/10.1007/978-3-030-19073-6

(eBook)

© Springer Nature Switzerland AG 2020
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed
to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Foreword

In the early 1970s, Albert Trillat and his team, including Henry Dejour, Gilles Bousquet and
Jean Luc Lerat, fascinated the European world of knee surgery with their didactic “Journées
Lyonnaises de Chirurgie du Genou”. Trillat’s OP Program with “Un petit ménisque pour commencer” regularly attracted many people, among them myself, to watch him perform his art.
There was a progressive snowball effect with an increasing number of Journées attendants and
followers.
An important aspect of this early Lyon School was the controlled follow-up and critical
study of the results, which were presented at the next biannual Journées Lyonnaises de
Chirurgie du Genou. This analysis allowed an evolution of the operative procedures by
improvement, not only of the technique but also the indications. The knowledge and know-­
how amassed by calculated trial and error grew continuously and was the basis for the trust
placed in the recommendations of the Lyon School.
The new second edition continues in this Lyon style and in the progressively wider field of
knee surgery. It is still directed by the same critical scientific principles, now under the guiding
spirit of Philippe Neyret and his team and with an enlarged international group.
Why is this great work of the Lyon School so important? “Do it right the first time” is the goal
for an operation, and to start by learning from one’s own failures is not anymore the aim. Revisions
after failure never bring the potential of the primary procedure. If a surgeon nowadays wishes to
start performing a new procedure, he needs more than just some nice slides from a conference or
from a journal. Better that, he should learn from those who developed the procedure, who have
already dealt with the unanticipated problems that arise during the learning curve.
In this updated work, the step-by-step descriptions and pictures for each operation increase
the possibility of doing it right the first time. One should follow these experienced instructions
without modification at first, but never perform the operation without thinking and trying to
understand the underlying functional and biomechanical principles.
You will find here a full spectrum of procedures covering the big field of knee surgery with
43 chapters. It starts with the general principle for the patient positioning and the preoperative
setup for sports surgery with arthroscopy; meniscus surgery; anterior and posterior cruciate
reconstructions and revisions, including arthroscopic technique for PCL reconstruction; the
lateral ligament and posterolateral corner reconstruction; and followed by the classification of
bicruciate ligament lesions and their treatment options. The chapters on synovectomies and
cartilage lesions guide us to the degenerative problems, with eight chapters on various osteotomies and unicompartmental knee arthroplasty. The steps and strategies for total knee arthroplasty and TKA revision surgery are included in eight further chapters.
There is a large part dedicated to patellar conditions, a corner stone of the strong Lyon Knee
School foundation, and these are again presented in eight chapters, followed by chapters about
the management of the stiff knee. The last part of this great work contains clinical cases which
reveal particular learning principles gleaned from decades of experience.
Werner Muller
Former Head of Ortho-Trauma Department, Kantonsspital Bruderholz
Basel-Landschaft, Switzerland
v

Foreword

“See one, Do One, Teach One” is the principle of learning a surgical technique in a previous
generation! This concept was especially applied to orthopaedics and, more specifically, to what
we refer to today as sports surgery and surgery of the degenerative arthritic knee. Fortunately,
for both patients and students of the arthritic knee, analogue education has been replaced with
digital learning. We now have the ability to “see one” via text, illustrations, excellent colour
pictures and an online presence, a marked improvement over being a member of a casual audience in a somewhat crowded operating theatre.
This second edition of Surgery of the Knee, by Philippe Neyret, Chris Butcher and Guillaume
Demey, affords all readers the opportunity of really “seeing one”. The part in the text, Surgery
for Degenerative Conditions, is inclusive of principles, indications, contraindications and
detailed techniques in rather specific chapters. It is a method of teaching indicative of the Lyon
School of Orthopedics, renowned for surgery of the sports and arthritic knee. The guiding
forces in the Lyon School (1958) were Albert Trillat, succeeded by Henry Dejour (1978) and
now Phillippe Neyret (1998). Phillippe Neyret has surrounded himself with stellar younger
partners, Sebastian Lustig and Elvire Servien.
“Doing one” is another issue! The editors of this book have addressed the complexity of
surgical performance in a very professional manner. The part devoted to surgery for degenerative conditions of the knee, the largest in the book, is well focused on osteotomy, partial
and total knee replacement. Osteotomy as a treatment for a degenerative arthritic knee is a
much more prevalent procedure in the European culture as compared to the American
approach. As a surgeon with significant femoral and tibial osteotomy experience, I am so
pleased by and admiring of the author’s devotion to detail. A procedure based on realigning
the femur and tibia leaves very little room for error. Incorrect alignment or improper fixation impedes healing of fracture or osteotomy. It’s obvious that the detailed surgical explanations of and improvements in osteotomy fixation have enhanced the success of osteotomies
about the knee. The aforementioned previous generation just “saw one (osteotomy)” and
then “did one”.
Partial and total knee replacement chapters are also very detailed with excellent pictures
and illustrations that allow us students to get a better grasp on performing the appropriate
arthroplasty. This descriptive and visual experience so benefits results and the development of
the “happy patient”.
This second edition of Surgery of the Knee is very current, a difficult task for textbooks,
even with an online production. The new chapters on TKR: Steps and Strategies, and
Robotic-­Assisted Knee Arthroplasty are “a must” in today’s teaching environment, and
Surgery of the Knee has captured it well. “Doing one” is now easier, thanks to editors of
Surgery of the Knee.
“Teaching one” used to be applied only to the Professor simply telling the student what he
or she did or did not do correctly; that was all that was required because the Professor said so!
Those days are over! (Thank God for the patients!) Today, the first person who has to be taught
and constantly challenged is the Professor, and his biggest critic must be himself. The authors

vii

viii

Foreword

share their persistent evolution in the development of their own experiences. New isn’t always
correct, but students must teach themselves the pros and cons of any new approach before they
can “teach” their peers and physicians in training. It’s apparent that the ­contributors to the
second edition of Surgery of the Knee understand this approach and have thus advanced our
educational experience.
Congratulations on showing surgeons of the degenerative arthritic knee the new definition
of “See one, Do one, Teach one”.
W. Norman Scott,
NYU Langone Medical Center, New York, USA

Foreword

I am honoured with the kind invitation of “Mon Cher Ami” Prof. Philippe Neyret to write a
preface in the second edition of this excellent book Surgery of the Knee.
Lyon/France has always been one of the central focuses of the knee surgery, since the
Founding Presidency of Prof. Albert Trillat, MD, in the “Lyon School of Knee Surgery”
between the years 1969 and 1978. In this respect, he is one of the real heroes in knee surgery.
In addition, the second President of this specialized school, Prof. Henry Dejour, MD, has
always been one of the leading names until the present day. The contributions of the “Lyon
School of Knee Surgery” towards the advances in knee surgery have always been worthy of
commendation until now and no doubt will continue to be in the future.
Apart from the developments in Europe and America, the advances in the management of
knee disorders in the Asia-Pacific region are also striking in terms of the cultural and social
lifestyles. As sitting on the floor is a traditional habit on the axis from Turkey to Japan, knee
problems have a special focus. Moreover, of the world’s population, currently, approximately
three-fifths are living in the Asia-Pacific region. Globally, the countries that belong to the associations and societies such as EFORT, AAOS, APOA, PAOA, ISAKOS, ESSKA, SICOT, etc.
have played critical roles in intercontinental relationships, in the integration of cultures and
communities, in the formation of the scientific consensus and eventually in the protection of
world peace.
Considering the dramatic increase of young surgeons, we should always keep in mind that
they are the future of our big family. Thus, one of our main duties must be to support their
contributions in all meetings, and in the associations, in order to give them opportunities to
expand their vision, which will in turn expand the vision of orthopaedics and traumatology,
and sports medicine further, like a domino effect. FORTE have had an important mission. In
this respect, the Past President, Dr. G. Huri, and his friends; the General Secretary of the APOA
Young Ambassador Forum, Dr. O. Bilge; and the ESTRO Study Group supported by Prof.
Philippe Neyret, Dr. S. Lustig, Dr. E. Servien and all their colleagues should be strongly supported. Additionally, I sincerely congratulate my friend Prof. Philippe Neyret for his personality, enthusiastically dedicated to education, and his ongoing works, which are always supportive
for these young orthopaedic surgeons. I wanted to share my thoughts with you through this
nice, high-­quality book.
As one of the Past Presidents of ISAKOS, and Lyon’s School of Knee Surgery, Prof.
Philippe Neyret knows the world and read it very well with a realistic and futuristic vision, as
one of the leaders in our field. He has made tremendous efforts for the advancement of knee
surgery, sports medicine and arthroscopy. One of them is indeed this unique book in your
hands, which joins the educational visions of the best knee experts from all around the world
together, covering many of the traditional and new aspects of knee problems in a systematic
and comprehensive fashion, including future prospects. Personally, I share the feeling of both
the difficulty of preparing the second edition of a high-quality book at the beginning of the
process and the pleasure to be able to teach the eager colleagues, in the end.

ix

x

Foreword

In this respect, I recommend strongly to all colleagues Prof. Philippe Neyret and his co-­
author, Dr. Chris Butcher, for the birth of the second edition of this book. I am sure that all the
readers will enjoy and benefit from this state-of-the-art book to a maximum extent and look
forward to the next edition with the new advances on the knee surgery. The success of the editors in this work is praiseworthy, and I wish the continuity of their worldwide achievements.
Mahmut Nedim Doral
Department of Orthopedics and Traumatology
Faculty of Medicine
Ufuk University
Dr. Rıdvan EGE Hospital
Ankara, Turkey
Department of Orthopaedics and Traumatology
Department of Sports Medicine
Faculty of Medicine
Hacettepe University
Ankara, Turkey

Foreword

Knee surgery has evolved rapidly over the past years. The renewed interest in exploring and
restoring the native anatomy has sparked this evolution. With regard to the anterior cruciate
ligament (ACL), for example, its anatomy had been described as early as the 1800s. The Weber
brothers described the ACL as having two bundles. Reconstruction of the ACL started to be
performed in the early 1900s, using an open, single-bundle technique. For young surgeons, it
may be difficult to envision doing a large open surgery for an ACL injury, but the technique
was actually quite anatomic. In 1939, Dr. Palmer even proposed a double-bundle technique in
his now famous thesis, although the importance of this document was not understood at the
time of publication.
The transition to arthroscopic surgery was made in the 1980s. The goal was primarily to
avoid large incisions, long surgical time and the need for prolonged immobilization. However,
in its infancy, the arthroscopic techniques focused primarily on faster, standardized techniques
which did not necessarily restore the anatomy. Although the preliminary results of arthroscopic
ACL surgery were promising, the high rate of osteoarthritis during mid- to long-term follow­up was unsettling. It would take several more years to figure out the problem.
Advanced imaging modalities allowing three-dimensional mapping of the anatomy, precise
measurement of landmarks and biomechanical and kinematic studies were needed to show we
had been placing the new ACL in a nonanatomic position and that this was causing altered
knee biodynamics. So back to anatomy, we went with new arthroscopic techniques individualized to restore the ACL to its native dimensions, collagen orientation and insertion sites. This
change back to restoration of anatomy and protecting long-term knee health has been apparent
in the evolution of various surgeries about the knee.
Professor Philippe Neyret did a phenomenal job putting together the second edition of this
excellent book. It provides surgeons with the most up-to-date surgical methods for various
knee problems ranging from ACL tears to osteoarthritis, all aimed to provide patients with the
best potential for a successful outcome. This book is a must-read for orthopaedic surgeons. As
surgeons, we should always strive to maintain quality of life and long-term knee health for our
patients. This book provides perspectives from all over the world and solutions for both common and rare problems about the knee. This is further enhanced by the use of helpful figures
and illustrations to allow practical application of the teachings of Professor Philippe Neyret
and colleagues. Congratulations on the publication of the new and even better second edition.

Freddie H. Fu
Division of Sports Medicine, Department of Athletics
School of Health and Rehabilitation Sciences
School of Education, Swanson School of Engineering
Pittsburgh, PA, USA

xi

Preface

This knee surgery “Traité” reports the techniques we had developed, used and improved in
Lyon over the past 30 years. This second English edition provides details not only about the
most frequent surgical procedures we perform during our daily practice but also some original
solutions we would propose in unusual situations.
Included are many schemes and more than 1000 pictures, to provide the inexperienced resident or fellow with an extensive description of the techniques. The aim is to provide not only
the technical steps but also some simple and reliable concepts that underlie them.
We have updated all the previous chapters and added several new chapters including one on
the use of robotics in arthroplasty and one entitled “Steps & Strategies in Total Knee
Arthroplasty”. We originally presented this topic with Paul Rivat during the “Journées
Lyonnaises du Genou” in 1999 and have now totally revised it with Chris Butcher. In it, we
focus on some principles of the surgery, independent of instrumentation or implants. Despite
new technologies that allow the surgeon to operate less invasively and with more precision,
during the coming decades, the orthopaedic surgeon will probably still need a comprehensive
approach to knee surgery; this implies a global understanding of all the anatomic and other
factors in the patient’s surgical treatment.
We also report some clinical cases we had operated on, including some original surgical
management. The goal is to share some disputable decisions we have made during our Lyon
practice. By introducing these controversial treatment options, we hope to stimulate some
thought in our readers.
Success of surgery depends on reliable and reproducible technique (the “how”) but also
good indications and timing (the “who” and, even more difficult, the “when”). During the last
30 years as Chief of the Department of Orthopaedic Surgery in Lyon following my mentors, A
Trillat and H Dejour, I travelled in every continent in order to learn, share my experience and
practise knee surgery with my colleagues. These various ways of practice gave me the opportunity to confirm that the techniques described in this edition are also robust; this means they
can be reliably reproduced in countries where specialized training and equipment are scarce.
I would like to thank all the contributors of the previous editions coordinated by Peter
Verdonk and Tarik Aitsiselmi in many different languages. Guillaume Demey coordinated the
first English version and Chris Butcher this second one.
I am particularly grateful to Professors Werner Muller and Mahmut Doral, Freddie Fu and
Norman Scott, who all accepted to write a foreword.
I really hope this new edition of the “Traité” will help the young surgeon and also give also
some ideas for further developments to my colleagues.
Lyon, France

P Neyret

xiii

Contents

1	The General Principals of Patient Positioning and Setup���������������������������������������   1
G Demey, R Magnussen, P Neyret, and C Butcher
Part I Sports Surgery
2	Arthroscopy of the Knee��������������������������������������������������������������������������������������������   5
P Archbold, LN Favarro Francisco, RK Prado, R Magnussen,
P Neyret, and C Butcher
3	Meniscectomy ������������������������������������������������������������������������������������������������������������� 17
P Archbold, LN Favarro Francisco, RK Prado, R Magnussen,
P Neyret, and C Butcher
4	Meniscal Suture����������������������������������������������������������������������������������������������������������� 21
Maad AlSaati, S Thompson, R Desmarchelier, G Demey,
P Neyret, and C Butcher
5	Anterior Cruciate Ligament Reconstruction: Surgical Technique ����������������������� 31
R Magnussen, AM Ozturk, G Demey, P Neyret, and C Butcher
6	Anterior Cruciate Ligament Reconstruction with
Six-Strand Hamstring Tendon Graft ����������������������������������������������������������������������� 57
S Orduna, N Darwich, and C Butcher
7	Revision Anterior Cruciate Ligament Reconstruction ������������������������������������������� 71
R Magnussen, G Demey, P Neyret, and C Butcher
8	Reconstruction of the Posterior Cruciate Ligament����������������������������������������������� 87
E Servien, G Demey, R Magnussen, P Neyret, and C Butcher
9	Posterolateral Corner and Lateral Collateral Ligament Reconstruction������������� 99
E Servien, R Magnussen, P Neyret, and C Butcher
10	Dislocations and Bicruciate Lesions ������������������������������������������������������������������������� 105
S Lustig, R Magnussen, P Neyret, and C Butcher
11	Synovectomies of the Knee����������������������������������������������������������������������������������������� 115
P Archbold, A Pinaroli, and P Neyret
12	Surgical Management of Chondral and Osteochondral Lesions��������������������������� 125
P Archbold, T Aït si selmi, C Bussière, P Neyret, and C Butcher
13	Iliotibial Band Syndrome������������������������������������������������������������������������������������������� 133
P Archbold, G Mezzadri, P Neyret, and C Butcher

xv

xvi

Part II Surgery for Degenerative Conditions
14	Surgical Indications in the Treatment of Osteoarthritis����������������������������������������� 139
P Archbold, JL Paillot, P Neyret, and C Butcher
15	Osteotomy: General Concepts and Indications������������������������������������������������������� 147
P Archbold, JL Paillot, P Neyret, and C Butcher
16	Varus Distal Femoral Osteotomy: Lateral Opening ����������������������������������������������� 153
P Verdonk, R Magnussen, P Neyret, and C Butcher
17	Valgus High Tibial Osteotomy: Lateral Closing and Medial Opening ����������������� 159
R Debarge, F Trouillet, G Demey, R Magnussen,
P Neyret, and C Butcher
18	Varus High Tibial Osteotomy: Medial Closing ������������������������������������������������������� 173
R Debarge, P Archbold, P Neyret, and C Butcher
19	Varus High Tibial Osteotomy: Lateral Opening����������������������������������������������������� 179
R Debarge, P Archbold, P Neyret, and C Butcher
20	Flexion High Tibial Osteotomy: Anterior Opening������������������������������������������������� 187
P Archbold, P Verdonk, E Servien, P Neyret, and C Butcher
21	Double Osteotomy������������������������������������������������������������������������������������������������������� 193
S Lustig, MF AlSaati, R Magnussen, P Neyret, and C Butcher
22	Patellar Femoral Arthritis and the Lateral Partial Patellectomy��������������������������� 199
S Lustig, LN Favarro Francisco, R Magnussen, P Neyret, and C Butcher
23	Unicompartmental Knee Arthroplasty��������������������������������������������������������������������� 205
S Lustig, A Daher, R Magnussen, P Neyret, and C Butcher
24	Unicompartmental Knee Arthroplasty (UKA)
After UKA to the Other Compartment ������������������������������������������������������������������� 219
Maad AlSaati, S Lustig, R Magnussen, P Neyret, and C Butcher
25	Total Knee Arthroplasty: Steps and Strategies ������������������������������������������������������� 227
C Butcher and P Neyret
26	Total Knee Arthroplasty in Medial Arthritis: Surgical Technique������������������������� 287
G Demey, R Magnussen, P Neyret, and C Butcher
27	Total Knee Arthroplasty in Lateral Arthritis: Specifics and
Surgical Techniques ��������������������������������������������������������������������������������������������������� 313
P Archbold, J Pernin, G Demey, P Neyret, and C Butcher
28	Computer-Assisted Total Knee Arthroplasty����������������������������������������������������������� 325
S Lustig, R Badet, Maad AlSaati, P Neyret, and C Butcher
29	Robotic Assisted Unicompartmental Knee Arthroplasty ��������������������������������������� 333
S Lustig, C Batailler, E Servien, and P Neyret
30	Total Knee Arthroplasty After Valgus Osteotomy of the Tibia������������������������������� 341
G Demey, H Hobbs, P Neyret, and C Butcher
31	Revision Unicompartmental Knee Arthroplasty����������������������������������������������������� 349
G Demey, R Magnussen, P Neyret, and C Butcher
32	Revision Total Knee Arthroplasty: Planning and
Technical Considerations������������������������������������������������������������������������������������������� 365
S Lustig, R Magnussen, P Neyret, and C Butcher

Contents

Contents

xvii

Part III Surgery for Patellar Conditions
33	Surgical Management of Episodic Patellar Dislocation ����������������������������������������� 381
E Servien, P Archbold, P Neyret, and C Butcher
34	Deepening Femoral Trochleoplasty��������������������������������������������������������������������������� 401
E Servien, P Archbold, P Neyret, and C Butcher
35	Patellar Tendon Shortening��������������������������������������������������������������������������������������� 405
E Servien, P Archbold, and P Neyret
36	Acute Ruptures of the Quadriceps and Patellar Tendons��������������������������������������� 409
G Demey, R Magnussen, C Fiquet, P Neyret, and C Butcher
37	Chronic Rupture of the Extensor Apparatus����������������������������������������������������������� 419
G Demey, R Magnussen, C Fiquet, S Lustig, P Neyret, and C Butcher
38	Patella Fractures��������������������������������������������������������������������������������������������������������� 435
G Demey, R Magnussen, P Neyret, and C Butcher
39	Surgical Management of the Stiff Knee ������������������������������������������������������������������� 441
R Debarge, P Archbold, P Neyret, and C Butcher
40	Lengthening of the Patella Tendon ��������������������������������������������������������������������������� 451
G Demey, P Archbold, and P Neyret
41	Stiffness of the Knee: Release According to Judet��������������������������������������������������� 457
H Hobbs, J Bruderer, G Demey, and P Neyret
42	Principles of Knee Surgery: Case Examples ����������������������������������������������������������� 463
P Neyret and C Butcher
43	Postoperative Complications������������������������������������������������������������������������������������� 495
G Demey, R Magnussen, and P Neyret
Index������������������������������������������������������������������������������������������������������������������������������������� 499

Introduction

In this manuscript, the authors describe in detail the most frequently performed interventions
in primary knee surgery. It is commonly believed that surgical knowledge is transmitted by
word of mouth. It is the close contact between the student and his teacher that will shape and
tailor the former on a day-to-day basis. These experiences are undeletable, and cannot be
undone. Deliberately or not, the quality and imperfections of his teachers will condition the
future of the young surgeon. This can be a good or a bad thing.
Not everybody has had the chance to be inspired by the enthusiasm of Albert Trillat and
later be moulded by his magic trio: Henri Dejour, Gilles Bousquet and Jean Luc Lerat. The best
advice given by Albert Trillat was as follows: “Take the best of every one of your teachers, but
ignore their imperfections”. It is exactly that advice that we would also like to impart on our
readers—while you will probably find some issues in this book controversial, you will surely
find certain surgical techniques and interventions that will be of value during your surgical
activity.
I would like also to warn the reader about two commonly made mistakes. The first mistake
is to adopt and repeat a surgical technique without evaluating the outcome of his own patients.
The second mistake is to perform a surgical step out of sequence. It has to be made very clear
that each step has a specific reason within the sequence of steps during a surgical procedure. It
was Henri Dejour who advised me: “Do exactly as I do during a period of three years and then
you may evolve in your own way”.
The young surgeon, full of talent and imagination, is tempted to abruptly break with his
teaching history once he leaves his teaching hospitals, but what a waste of time and energy that
would be!
Now more than ever, the orthopaedic surgeon is confronted with time restrictions. He not
only has to schedule his surgical programme but has to schedule the sequence of the different
surgical procedures. Moreover, he has to know when it is suitable to quickly proceed through
the different steps of the surgery and also when it is required to slow down and take his time,
during those critical moments. This knowledge is only acquired through practice, experience
and thoroughness.
Working your way through the surgical list is not just working repeatedly through a single
surgical intervention. It is working through three or four, or even eight to ten, surgical interventions. The surgical programme thus has to be coordinated. One should strictly plan the strategy
beforehand, thus limiting the possibility of surprises. The sequence of the list should be carefully considered. This planning should be done as late as possible the evening before the surgery due to the many variables (side of surgery, type of anaesthesia, day care surgery, the
availability of the surgical instruments, the availability of the surgical personnel, prosthetic
surgery, infected cases, etc.). Most commonly, a surgical list consists of a series of classic,
well-timed interventions, without surprises. But sometimes, your meticulously planned list is
disturbed by a difficult case which demands your attention and energy. In our department, we
have agreed to allow only one such case during a regular surgical list. On occasion, there may
be a case which would benefit from the presence of two surgeons, whose combined experience
and concentration may just provide the best solution in the moment.

xix

xx

It is obvious that surgical techniques change and evolve. But despite modern revolutions in
genetics, biologics and computer-assisted surgery, the presented surgical techniques can serve
as a solid basis upon which future innovations will be built. Some of these innovations will be
discussed in the respective chapters.
In conclusion, the purpose of this book is to transmit some of this knowledge, and as
Talleyrand said, “That which goes without saying, goes even better when it is said”.
P Neyret
This edition is an evolution of previous manuscripts. Like many evolutions, it is formed by
the collective thought of an expanding group of people. The first manuscript, “Précis de chirurgie du genou: Mes 10 opérations” (“Knee Surgery: My Ten Procedures”), was a synthesis of
decades of surgical development at the Lyon School and not only presented a practical guide
to commonly performed procedures but conveyed some of the spirit of the school from the
originators of the movement. The next was an expanded text, “Surgery of the Knee”, which
represented an evolution in two ways. Firstly, it included some less common and more challenging problems, which sought to stimulate ideas as well as give didactic instruction. Secondly,
it was formed with the collaboration of experts from around the globe, whose experience and
influence included, in some shape or form, the Lyon School.
This second edition is a continuation of this theme, as the original lessons and spirit are
adapted, both to the educational and cultural backgrounds of surgeons and patients of the
global community and by the natural evolution of surgical treatment. All education is a balance; one seeks to comprehend what is already known and to cultivate enquiry of what is not.
We hope that the mix of advice and stimulation in this book serves this purpose.
C Butcher

Introduction

1

The General Principals of Patient
Positioning and Setup
G Demey, R Magnussen, P Neyret, and C Butcher

In this book, we describe many surgical techniques, some of
which are technically demanding. Although these operations
differ significantly in many ways, the initial positioning and
setup of the patient generally remains the same.
• The procedure is performed under general or spinal anesthesia. Less commonly the procedure can be performed
with a nerve block and mild sedation.
• The patient is positioned on the operating table in the
supine position. A padded horizontal post is positioned
distally on the table to hold the knee in a 90° flexed position. This has the advantage of allowing the knee to be
held at either 90° of flexion (when the heel rests on the
post) or 110° of flexion (when the toes rest on the post)
without changing the post position (Fig. 1.1a). A second
post can be placed proximally on the table to allow a position of hyperflexion (Fig. 1.1b). The posts do not obstruct
full extension (Fig. 1.2). In the situation of a stiff knee
with limited flexion, the second post can be placed distally in order to maintain the knee maximally flexed during the first step of the arthrolysis.
• A lateral support controls the rotation of the hip, with the
thigh resting on the support. Slight external rotation of the
hip is set prior to inflation of a pneumatic tourniquet
(Figs. 1.1 and 1.3).

• The pneumatic tourniquet is placed as high as possible on
the thigh. Once the lower limb has been prepped, it is
exsanguinated by elevation. Tightly wrapping the leg or
the use of a rubber Esmarch for exsanguination is generally not necessary. The tourniquet is then inflated to
300 mmHg, or adjusted to systolic blood pressure. If the
patient has a history of vascular disease, a pneumatic
tourniquet is positioned as proximally as possible, but
generally not inflated. We consider the use of sterile disposable exsanguinating tourniquets (e.g., HemaClear®) in
hemorrhagic procedures where sterility is paramount,
such as total knee arthroplasty.
• The surgical leg is prepped with a betadine and alcohol
solution. After prepping the foot, it is covered with a size
9 glove. The leg is then elevated and held by the foot
while the rest of the limb is prepped. A stocking is then
rolled up the leg to the level of the tourniquet, and an
arthroscopy drape is used to complete the sterile field.
• The stocking is opened with scissors. The planned surgical incision and any previous surgical scars are marked
with a pen. An adhesive drape with or without antiseptic
(e.g., Opsite™, Ioban™) is applied, always allowing for the
possibility of extending the expected incision proximally
or distally (Fig. 1.4).

G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae
© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_1

1

2

G Demey et al.

a

b

Fig. 1.1 (a) The distal horizontal post allows for 90° and 110° of flexion, (b) a proximal horizontal post maintains hyperflexion

Fig. 1.2 Extension of the knee

Fig. 1.4 Leg prepped and ready for surgery

Fig. 1.3 The lateral support controls hip rotation; note the slight external rotation prior to the inflation of a pneumatic tourniquet

Part I
Sports Surgery

2

Arthroscopy of the Knee
P Archbold, LN Favarro Francisco, RK Prado,
R Magnussen, P Neyret, and C Butcher

General Information

Patient Positioning

It is essential that an orthopedic surgeon acquires the skills to
perform arthroscopy at an early stage of his/her training. It
can be compared to playing golf or driving a car, as when
these skills are acquired at an early age, they appear to come
naturally. The subject of this chapter is not to give a general
overview of the surgical options and situations encountered
but instead to describe some tricks which can facilitate
arthroscopy and provide some general concepts. The specific
techniques and major indications are presented in detail in
the corresponding chapters.

The patient is placed on the operating table in the supine
position. A tourniquet is placed high around the proximal
thigh which facilitates hemostasis and visualization. A horizontal post is positioned distally on the table to hold the knee
in a 90° or at 110° of flexion. A further advantage is that this
setup can be used to combine arthroscopic and open surgery
(e.g., arthroscopic and open surgery on the patella or an osteotomy) (Fig. 2.2a, b). The vertical post which is positioned at
the proximal third of the thigh acts to resist valgus stress of
the knee. The quality and ease of the arthroscopy is facilitated by the ability to place the knee in valgus for improved
visualization of the medial compartment. The post needs to
be adjusted to allow the surgeon to be situated between the
leg and the table and thus able to stabilize the knee with valgus stress without the help of an assistant (Fig. 2.3). To assist
with this, some surgeons also place a sandbag under the contralateral hemipelvis to prevent excessive internal rotation of
the flexed knee.
The lower limb is prepared in a sterile manner using
either alcoholic iodine or alcoholic chlorhexidine
(chlorhexidine in the situation of iodine allergy). The foot
is covered with a size 9 surgical glove, and then the lower
limb is covered with a stocking and subsequently an
extremity sheet. The limb is elevated for several seconds,
and the tourniquet is inflated. The pressure of the tourniquet

Surgery
Patient Preparation
Following arrival at the hospital and prior to the administration of any medications, the operative site is marked by the
surgeon with an indelible marker (Fig. 2.1). General anesthesia is preferred as it offers complete amnesia of the intraoperative period, is quickly administrated, and allows early
mobilization. In the case of older patients, obese patients,
and patients with limited respiratory capacity or a difficult
airway, regional anesthesia can be used. This technique also
allows for optimal pain relief in the operated limb.

P Archbold · LN Favarro Francisco · RK Prado · R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

Fig. 2.1 The operative extremity is marked by the surgeon personally

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_2

5

6

P Archbold et al.

a

b

Fig. 2.2 Patient positioning. A horizontal post is positioned distally on the table to hold the knee in a 90° flexed position (a) or at 110° of flexion
when the toes rest on the post (b)

sidered very important not only to document the intervention
but also to inform the patient as to the surgical findings in a
simple and accurate way.
One needs fully equipped arthroscopic towers available
for every operating room and a sufficiently large number of
arthroscopes and cameras. The arthroscopic infusion fluid
chosen is an isotonic saline solution. If using a pump rather
than gravity, a minimal pressure of 50 mmHg suffices to
obtain an acceptable distension of the knee joint.
The sleeve for the arthroscope has an inlet and an outlet.
We use a 4 mm diameter lens with 30° angulation. In case of
reconstruction of the PCL or arthroscopy in the posterior
region of the knee, a 70° arthroscope can be used.

Fig. 2.3 Valgus stress on the knee using a vertical post laterally positioned at the proximal third of the thigh

should be a minimum of 200 mmHg above the diastolic
blood pressure or in absolute numbers approximately
300 mmHg. It should never be inflated longer than 120 min.
Nowadays many surgeons preferred to avoid a tourniquet
and add local injections before creating portals, in order to
reduce postoperative pain.

Instrumentation
The essential piece of equipment is an arthroscopic tower
including a camera, a monitor, and a light source. An
arthroscopic pump, a shaver, and a recording device are
­useful aids and are routinely used. Image recording is con-

Surgical Instruments
The list of available instrumentation becomes longer each
day as arthroscopic techniques advance. Certain instruments, however, are indispensable and form the “basic toolbox”: the probe, the arthroscopic scissors, a punch, a large
grasper, and a cannula (Fig. 2.4). The punch, which is the
indispensable instrument for performing a meniscectomy,
has a straight tip and may have a slightly angled shaft to
facilitate sliding under the femoral condyle (Fig. 2.5a).
Radiofrequency (RF) ablation (controlled ablation aka
coblation) is also an interesting technology (Fig. 2.5b).
With this we can treat meniscal tears, and we can clean
ligament insertion area and manage some cartilage lesion
(e.g., flap). RF tools are good for reaching the posterior
horn of the meniscus due to the fact that the tip of the device
is typically small and curved and also narrower than a
shaver or a basket.

2

Arthroscopy of the Knee

7

Fig. 2.4 Surgical devices
used in our practice: from left
to right: probe, beaver knife,
punch, biopsy forceps,
grasping forceps (Wolf
forceps), suction cannula

Fig. 2.5 (a) Meniscectomy
using the punch, (b)
radiofrequency probe

a

Surgical Technique
Portals
Triangulation is the most effective arthroscopic technique. It
requires two portals, which can be made in a variety of
positions:
––
––
––
––

The anterolateral portal
Two anteromedial portals
The superolateral portal
The superomedial portal

b

––
––
––
––

The posteromedial portal
The posterolateral portal
The lateral parapatellar portal of Patel
The posteromedial and posterolateral portals as described
by Philippe Beaufils.

The anterolateral and the two anteromedial portals are the
most frequently used. Using these three portals, one can perform 95% of the surgical procedures. The other portals are
considered accessory portals (Fig. 2.6). The choice of the
medial portal depends on the indication and on the
arthroscopic finding. One should never hesitate to make an

8

P Archbold et al.

The Two Anteromedial Portals
The Inferior Anteromedial Portal
This portal provides access to the anteromedial joint space,
which is situated just above the medial meniscus. In order to
avoid damage to the medial meniscus or the infrapatellar
branch of the saphenous nerve, some strict rules must be
adhered to. The knee should be flexed at 90° with the foot positioned on the distal post. Please note that the lower anteromedial portal is situated closer to the femorotibial joint line than
the anterolateral portal. The inferomedial portal is also farther
from the patella tendon. With the scope in the anterolateral portal, one can transilluminate this medial area for guidance.
This skin incision is again made with an 11 blade directed
superiorly (and never inferiorly!). The skin incision is vertical and 5–8 mm in length. Under arthroscopic control, the
blade is visualized as it enters the joint just proximal to the
superior surface of the medial meniscus. One can now turn
the scalpel 90° and widen the capsular incision horizontally
just above the meniscus. This portal, which is just above the
superior surface of the medial meniscus, allows easy access
to the medial compartment for a successful meniscectomy. A
portal placed too centrally will enter into the area of Hoffa’s
fat pad, significantly limiting visualization.

Fig. 2.6 Left knee, anatomical landmarks (anterior tibial tuberosity,
patella, lateral tibial plateau), anterolateral, low anteromedial, and high
anteromedial portals

additional third portal or to reverse the location of the camera
and instrumentation between portals. Attempting to use inadequate portals to avoid an additional incision will surely lead
to mistakes.

 he Anterolateral Portal
T
This portal is used to introduce the camera. It allows good
visualization of the joint. The skin landmarks are: medially, the lateral edge of the patellar tendon; inferiorly, the
lateral tibial plateau; and superiorly, the lateral femoral
condyle. The entry point of the portal is just inferolateral
to the patella. An 11 blade is introduced with the blade
angled proximally (to protect the meniscus), and a vertical
incision is made in the soft spot situated between the lateral tibial plateau, the lateral femoral condyle, and the
inferolateral patella.
If the incision is placed too low, the available space to
position the camera will be reduced, and there is a risk of
damaging or cutting the anterior horn of the lateral
meniscus. This situation is often encountered in cases of
patella baja or when the surgeon opts to reuse a previous
skin incision.

 he Superior Anteromedial Portal
T
Symmetrical to the anterolateral portal, thus more central and
more proximal than the lower anteromedial portal, this portal
gives improved access to the intercondylar notch. In the figureof-four position, this more proximal portal provides perfect
visualization of the lateral compartment and provides optimal
access to treat lateral meniscal lesions. Both the inferior and
superior anteromedial portals can be used in combination.

 he Superolateral Portal
T
This portal, superior and lateral to the patella, gives access to
the patellofemoral compartment, the suprapatellar pouch,
and the lateral condylar gutter. This portal can be used for the
evaluation of patellofemoral cartilage and patellar tracking
as well as for arthroscopic synovectomy and arthrolysis.
 he Superomedial Portal
T
This portal is symmetrical to the superolateral portal but on
the medial side of the patella. The entry point is slightly
more proximal (approximately 2–3 cm above the patella) to
allow easy instrumentation.
 he Posteromedial Portal
T
This portal is used to visualize the posterior compartment
and posterior horn of the medial meniscus. The tibial insertion of the posterior cruciate ligament can also be visualized.
The skin incision should be proximal enough to allow the
entry point on the capsule to be in contact with the posterior

2

Arthroscopy of the Knee

part of the medial condyle. This positioning allows optimal
orientation of the instruments. A skin incision that is situated
too distal will increase the difficulty of the surgery.
In order to facilitate the correct positioning of the skin
incision, a spinal needle can be introduced. The knee should
be positioned in the figure-of-four position and the capsule
of the knee completely distended. The cutaneous and capsular entry point of the needle can be visualized by transillumination with the camera placed through the intercondylar
notch (Fig. 2.7).

9

 he Posterolateral Portal
T
This approach can be used to visualize the posterolateral
compartment and the posterior horn of the lateral meniscus.
In order to prevent any injury of the common peroneal nerve,
the portal must always be located anterior to the biceps tendon. As for the posteromedial approach, a spinal needle can
be used to help determine the exact site of the skin incision.
The needle is introduced just posterior to the lateral femoral
condyle with the knee in a 90° flexed position. The tip of the
needle can be visualized with the arthroscope placed in the
intercondylar notch (Fig. 2.8a). The skin incision is subsequently made with an 11 blade guided by the spinal needle
and oriented toward the condyle. Scissors may be used to
dilate the portal (Fig. 2.8b). The combination of the posteromedial and posterolateral portals has been previously
described in detail by Philippe Beaufils. A long blunt
arthroscopic obturator inserted through the posterior medial
portal and gently across the knee may help locate the correct
posterior lateral portal.
 he Lateral Parapatellar Portal of Patel
T
This portal is situated along the lateral border of the patella
but more laterally and proximally than the classic anterolateral portal. This approach gives an excellent view on the lateral femorotibial compartment and in particular the anterior
horn of the lateral meniscus.

Fig. 2.7 Posteromedial portal, left knee. A spinal needle is inserted
through the capsule distended by arthroscopy fluid

a

All the above-mentioned portals can be used for instrumentation, but the anterolateral and anteromedial portals are the
most frequently used (Fig. 2.9). We have never used the
transpatellar portal described by Gillquist. Some surgeons
b

Fig. 2.8 Posterolateral portal, left knee. (a) The spinal needle aids positioning. (b) Scissors can be used to dilate the portal

10

P Archbold et al.

Fig. 2.10 Placing the arthroscope to examine the patellofemoral joint

Fig. 2.9 Portals, left knee. Anterolateral portal (arthroscope, right) and
low anteromedial portal (instrument, left)

use it in order to have a better visualization of the posterior
aspect of the notch.

Arthroscopic Steps
After anesthetic induction, the knee is re-examined to complete the physical examination. The sleeve and obturator are
introduced through the anterolateral portal in the direction of
the femoral notch with a knee in the 90° flexed position. The
sleeve is then directed to suprapatellar pouch while extending
the knee. The inflow cannula is subsequently connected, and
the joint is distended. The obturator is removed, and the
camera is introduced. The intra-articular examination is
performed in a systematic fashion. The sequence of the
surgical steps is inspection, then palpation, and subsequently
treatment. Beginning with the inspection of the knee, one can
adapt and choose the necessary portals for instrumentation.
The knee cavity is inspected in a systematic sequence:
• The patellar femoral compartment
• The medial and lateral tibiofemoral compartment
• The intercondylar notch

 he Patellofemoral Compartment
T
The lower limb is extended in neutral rotation to rest on the
operating table. The exploration starts in the suprapatellar
pouch. We do not use an outflow sleeve for lavage except in
cases of a significant hemarthrosis. The camera is introduced
through the anterolateral portal, and the lens is angled proximally (Fig. 2.10). The intra-articular space is opened up by
distention.
With the camera lens oriented proximally and in the direction of the patella, one can examine the medial facet, the lateral facet, and the central ridge of the patella (Fig. 2.11).
Excellent visualization of the patellofemoral compartment can also be achieved through a suprapatellar portal (frequently used in the case of episodic dislocation of the
patella). After inspection of the patella and trochlea, the
arthroscope is progressively retracted to the point where adequate visualization of the patellofemoral compartment is
lost. The arthroscope is now rotated into the neutral position
with the camera lens orientated at 90° looking toward the
suprapatellar pouch. This maneuver allows simultaneous
inspection of the patella and trochlea. In particular it allows
inspection of the part of the trochlea just above the notch.
Although generally not inspected, this zone is frequently
damaged (deceleration lesions).
The camera lens remains at the 90° and now slides along
the medial femoral gutter while the hand that is holding the
camera is now directed proximally, bringing the camera distally into the medial tibiofemoral joint (Fig. 2.12a–d).
 he Medial Tibiofemoral Compartment
T
The lower limb is now elevated off the table and held in position by placing it on the contralateral iliac crest of the surgeon
(Fig. 2.13). Subsequently, the medial femoral condyle and the
medial compartment can be visualized with the knee in about
30° of flexion. The knee can now be placed into valgus (the

2

Arthroscopy of the Knee

a

11

b

Fig. 2.11 Patellofemoral compartment: (a) patella, (b) trochlea

patient’s foot is moved laterally while the thigh is held by the
post). It is often helpful for an assistant to place a downward
force on the thigh at the level of or just distal to the tourniquet
to prevent excessive knee flexion (Fig. 2.13). This maneuver
helps open up the medial compartment, allowing the entire
body of the medial meniscus to be easily visualized. The synovial fringes can be easily identified because of the slight pink
color. The free meniscus border is checked as well as the anterior and posterior horns (Fig. 2.14). As yet we have not performed a percutaneous release of the deep fibers of the MCL
to improve visualization, as has been described by H. Paessler.
Pushing in the popliteal fossa with the fingers can help bring
the posterior horn of the meniscus more anteriorly, and external rotation of the tibia may help visualization. Palpating the
medial meniscus should be carried out with a probe introduced
through the anteromedial portal (Fig. 2.15). The peripheral
attachments of the meniscus are evaluated, potential meniscus
lesions investigated, and the quality and texture of the meniscus assessed.
The articular cartilage of the medial femoral condyle and
the medial tibial plateau are palpated (Fig. 2.16). Slowly
flexing the knee allows evaluation of both the integrity of the
whole articular surface and the quality of the cartilage.

Lateral Tibiofemoral Compartment
The arthroscope is not removed from the joint, and the knee
is now positioned in the Cabot position, that is, the knee in
varus and flexed to 90° (Fig. 2.17). The foot rests on the contralateral tibia. The hip is flexed, abducted, and in external
rotation. This maneuver opens up the lateral compartment
(Fig. 2.18). The superior and inferior surface of the anterior,

body, and posterior horn of the lateral meniscus can be visualized. The intra-articular course of the popliteal tendon can
be seen. It runs anteriorly and superiorly from its origin posterior tibia to its insertion on the lateral femoral condyle. It is
necessary to check the tendon and the hiatus since specific
anatomic variations have been observed (Fig. 2.19). The
meniscal wall and the menisco-popliteal attachments can be
well visualized. Anatomical variations of the lateral meniscus can be observed and potentially treated (discoid meniscus, hypermobile meniscus).

 he Intercondylar Notch
T
To visualize the intercondylar notch, the knee is flexed to 90°
with the foot resting on the post (Fig. 2.20). A synovial
extension between the Hoffa fat and the lateral condyle, also
known as the ligamentum mucosum or infrapatellar plica,
can obstruct adequate visualization of this region. If this is
the case, we routinely resect this structure with the shaver at
its attachment superior to the notch, allowing it to fall out of
the way of the camera.
The ligament of Humphrey and the PCL can now be
observed in the upper part of the intercondylar notch
(Fig. 2.21). They occupy the medial 2/3 of the intercondylar
notch while the anterior cruciate ligament has a more horizontal course to the back of the notch where it inserts on the
lateral femoral condyle. The appearance of the intercondylar
notch as an inverted “U” or a capital “A” is noted in the operative report as well as the presence of any osteophytes.
The anterior cruciate ligament can be easily recognized due
to its white color and covering with a thin, vascularized
synovium (Fig. 2.22). The two separate bundles of the ACL can

12

P Archbold et al.

a

b

c

d

Fig. 2.12 Changes in the position of the arthroscope from the patellofemoral compartment to the medial tibiofemoral compartment. The
arthroscope looks at the suprapatellar recess (a), it is switched to the

neutral position parallel to the joint space (b), the arthroscope is rotated
down in the medial tibiofemoral compartment (c) while the leg is positioned in valgus (d)

Fig. 2.13 Position of the surgeon and assistant during a medial
meniscectomy

be frequently discerned. Its origin is very proximal (“deep”) on
the lateral condyle and posterior (“low”) with the knee in flexion. The tension of the ACL can be tested by palpation. The
posterolateral bundle is only under tension near full extension.
The posterior cruciate ligament is covered by the more
horizontally oriented ligament of Humphrey and by synovial
tissue (Fig. 2.23). The ligament of Humphrey, also known as
the anterior meniscofemoral ligament, should not be mistaken for the PCL. It originates from the posterior horn of the
lateral meniscus, then crosses the PCL anteriorly to insert
just in front of the PCL on the medial femoral condyle. The
PCL can be palpated at a level of its insertion on the femoral
condyle. The surface area of the ligament of Humphrey is
less than 30% of the PCL.
Despite the presence of the cruciate ligaments in the intercondylar notch, access to the posterior knee compartment is
possible. This can be done by gently gliding the sleeve and

2

Arthroscopy of the Knee

Fig. 2.14 Normal medial meniscus, body

13

Fig. 2.16 Palpation of tibial plateau cartilage

Fig. 2.15 Palpation of the medial meniscus, showing an unstable
lesion. Note the femoral chondral lesions

the round-tipped obturator along the medial femoral condyle
underneath the PCL with a knee at 90° of flexion.
While pushing the sleeve and obturator, the knee is gently
flexed up to 110°. The obturator is removed. The camera is
introduced, and the posterior compartment is now visualized.
The insertion of the PCL on the tibia can be observed (Fig. 2.24).
At the end of the arthroscopy, we manually apply pressure
to the suprapatellar pouch and flex the knee in order to evacuate the intra-articular fluid.

Fig. 2.17 Cabot (figure-of-four) position to assess the lateral tibiofemoral compartment

14

P Archbold et al.

Fig. 2.20 Position for the examination of the intercondylar notch
Fig. 2.18 Normal lateral meniscus. Note that the meniscus is inclined
horizontally on the screen

Fig. 2.19 Popliteus tendon

Fig. 2.21 The intercondylar notch, left knee. The femoral origins of
the PCL (left) and ACL (right, in front of the PCL) are observed

After a meniscectomy or cartilage procedure, small
meniscal or cartilaginous fragments can sometimes be found
in the portals. As these can result in persistent irritation and
induration of the wound, we pinch the portal between two
fingers to remove them.

Postoperative Care
No drain is necessary. Nonabsorbable skin sutures are removed
on the tenth postoperative day. Thrombo-­prophylaxis is not
necessary except for patients at higher risk for DVT. The use
of routine prophylactic antibiotics is not recommended.

2

Arthroscopy of the Knee

15

Fig. 2.24 The tibial insertion of the PCL (left knee) from the anterior
lateral portal
Fig. 2.22 Anterior cruciate ligament (ACL)

or fourth postoperative day; however, this will often be too
short and needs to be modified depending on the extent of the
surgery, as well as local cultural factors.
Activities of daily life are limited for the first week, and
professional activities are limited for 2–4 weeks, depending on the profession. All patients are reviewed at day 45
to assess postoperative recovery. Sport activities are usually allowed at 4–6 weeks, unless there is a specific
contraindication.
The patients should always be informed before the intervention of the rare but real risk of infection, as well as the
possibility of a longer than usual rehabilitation period. This
information is essential. Arthroscopic surgery should not be
considered or presented to the patient as being harmless.
Treatment failures or persistent lesions are observed in 1% of
our patients.

Complications

Fig. 2.23 The red star shows the ligament of Humphrey (anterior
meniscofemoral ligament) in front of the PCL, left knee

Outpatient treatment is preferred unless specific medial or
social concerns require an inpatient stay.
Mobilization of the knee is performed immediately, and
physiotherapy is prescribed for nine sessions. In some circumstances, we allow the patients to drive a car on the third

Paresthesia and hypoesthesia are not commonly observed
after an arthroscopy. By systematically transilluminating the
skin during creation of the anteromedial portal, one can limit
the frequency of lesions to the sensitive nerve branches in
this area. Dysesthesias have also been reported and can lead
to a complex regional pain syndrome (algodystrophy).
We have not observed skin necrosis. A possible reason
could be that we do not perform arthroscopic lateral
patellar release with an electrocoagulator. We have never
observed tibial or femoral fractures during an arthroscopic

16

procedure. We have observed three cases of a complete
medial collateral ligament tear in our long experience
over 20 years. These lesions have healed uneventfully
with conservative treatment. We have always been able to
extract parts of broken instruments or meniscal fragments. Stopping inflow of the irrigation fluid can facilitate extraction.

P Archbold et al.

Other Complications
Iatrogenic intra-articular injuries can be reduced if not
completely eliminated through careful attention to detail.
Perhaps most importantly, the head of the shaver should
always be visualized before shaving, especially in the posterior compartment in order to avoid an injury to neurovascular structures.

3

Meniscectomy
P Archbold, LN Favarro Francisco, RK Prado,
R Magnussen, P Neyret, and C Butcher

Medial meniscectomy is one of the most frequently performed surgical procedures. The technical difficulties associated with this procedure are not always appreciated.
Sometimes it can be harder to perform a medial meniscectomy than to perform an ACL reconstruction.
When confronted with a meniscal lesion, the following
questions should always be asked:

enhanced imaging techniques. Arthro-CT is useful when
patients have a history of meniscectomy or meniscal suturing
as well as for the evaluation of articular cartilage. Arthro-­
MRI is a more recent technique, the full potential of which
will become apparent in time.

The Medial Meniscectomy

–– Is the knee stable or is there ligamentous laxity?
–– Is it a degenerative tear or a traumatic tear?

A partial medial meniscectomy is a quick and reliable procedure with excellent short- and long-term results. The outBefore surgery, it is critical to document the clinical his- come is better in traumatic meniscal injuries within a stable
tory including patient activity level and to perform a detailed knee joint with no cartilage damage (Fig. 3.1a, b).
physical examination. AP and lateral radiographs of the knee
A medial meniscectomy is typically performed while
are necessary, and a Rosenberg view is required after the age working through an anteromedial portal, keeping the followof 50 years.
ing rules in mind: (1) the meniscus wall should always be
In patients above age sixty, we routinely perform an MRI respected, (2) one has to be “economical” with the resection,
to look for any degenerative change in the underlying sub- and (3) iatrogenic cartilage damage should be avoided.
chondral bone or extrusion of the meniscus. If this is the
In the case of a degenerative tear, the meniscectomy
case, a conservative rather than surgical management is pre- should be more aggressive; however, the integrity of the
ferred. Nevertheless when there is a recent additional trauma, meniscal wall should always be respected. With degeneramechanical symptoms, and sport expectations, there is still a tion, meniscal tissue undergoes structural changes (Fig. 3.2).
place for the medial meniscectomy in degenerative meniscal This deterioration in the quality and mechanical characteristears.
tics of the meniscus means that a flap or tear left in situ can
This detailed patient workup allows us to choose the most result in the failure of the treatment.
appropriate mode of treatment for each patient. In the situaIn this instance (older patients, degenerative lesions), one
tion of symptoms with previous history of meniscal surgery, has to compromise. We perform a more aggressive meniswe also order MRI. Sometimes it is useful to utilize contrast-­ cectomy but always with respect for the meniscus wall
(Fig. 3.3). In many cases, a stable tear such as a horizontal
cleavage tear in the meniscus wall is left in place and not
P Archbold · LN Favarro Francisco · RK Prado · R Magnussen
resected.
Centre Albert Trillat, Lyon, France
A medial meniscectomy is generally performed using the
P Neyret
anteromedial portal while viewing through the anterolateral
Infirmerie Protestante, Lyon, Caluire 69300, France
portal (see Chap. 2). A curved basket is adequate in the
e-mail: Philippe.neyret01@gmail.com
majority of cases in order to follow the shape of the condyle
C Butcher (*)
and reach the upper surface of the meniscus. A 4 mm shaver
Healthpoint, Abu Dhabi, UAE
can be used to debride frayed meniscal tissue as well.
e-mail: c.butcher@healthpoint.ae

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_3

17

18

a

P Archbold et al.

b

Fig. 3.1 Medial meniscal tear (flap) (a) treated by partial meniscectomy (b)

Fig. 3.2 Degenerative tear of the medial meniscus demonstrating
structural changes

Fig. 3.3 Degenerative tear of the medial meniscus and cartilage lesions

Specific Cases
Surgeons should be aware that degenerative meniscal tissue
is softer than normal meniscus, and aggressive use of the
shaver or punch can quickly lead to a loss of the meniscal
wall and effective total meniscectomy in these patients if
care is not taken.

In some cases, access to the posterior horn of the medial
meniscus is difficult. In these cases, it is often helpful for the
assistant to place downward pressure on the thigh just distal
to the tourniquet. This maneuver allows the surgeon to apply
valgus load to the knee more effectively without the knee

3 Meniscectomy

going into flexion. Applying pressure in the posteromedial
crease of the knee joint is often helpful in delivering the posterior horn of the medial meniscus into a position where
debridement can be performed. Sometimes the use of the
coblator can make the procedure easier (see Chap. 2). In
spite of these tricks, a peripheral posterior tear associated
with anterior cruciate ligament laxity remains difficult to
treat. The posterior horn has a frequent tendency to escape
behind the femoral condyle. Fortunately, many of these
lesions can be sutured, improving outcomes and avoiding the
need for partial meniscectomy.
In the case of a bucket handle tear that cannot be sutured,
we usually start by cutting the tear at its anterior root. The
anterior cut should not leave a large stump of torn tissue
behind because this piece of meniscus will be difficult to
excise later in the case. For this procedure, we prefer the
banana knife rather than the punch. Banana knife is a slightly
curved scalpel fixed on a tubular scalpel holder. Cutting of
the anterior part is easier if the bucket handle tear is reduced.
After the anterior resection, the posterior resection of the
bucket handle tear is performed. With a punch, the most posterior part of the tear is addressed by creating a curved progressive resection of the posterior attachment and then
removing the entire torn piece of meniscus in one. Leaving a
very small attachment may aid in preventing the loose meniscus from escaping before being removed with the grasper.
Meniscal roots must be respected, to avoid defunctioning of
the remaining tissue.

19

damage the articular cartilage. An alternative is to use a
back-cutting punch.

The Meniscal Cyst
A meniscal cyst is most frequently found in the lateral meniscus. The general treatment principle is to preserve the meniscus
wall. Additional investigations such as MRI and arthro-CT can
help in the diagnosis of the lesion (Fig. 3.4a–d). In some cases,
the cyst is in continuity with the joint space while in other cases
no clear communication with the joint can be found.
In order to preserve the meniscus, we usually combine an
arthroscopic procedure to debride any intra-articular meniscus
lesions with a direct open approach to resect the cyst and close
the connection to joint with vertical sutures (Fig. 3.5a, b).
In rare cases, the cyst can be treated by arthroscopy
alone—where there is already a defect in the meniscus wall
and with small cysts.
Patients undergoing meniscal cyst resection should
always be informed in advance of the possibility of an additional skin incision, the risk of a residual swelling, and the
possibility of recurrence.
a

Lateral Meniscectomy
The surgical technique is the same for a degenerative lesion
as for a traumatic lesion.
Radial tears in the middle segment are treated with a
meniscectomy in the form of an arc. The popliteal hiatus
should be respected if possible. If the tissue anterior to this
hiatus is resected, the clinical outcome in the mid term will
be worse. The posterior horn and the middle segment are
easily accessible with the punch with a knee in the figure-offour position. In very posterior lesions, access can be limited by the tibial spines. In this case, the viewing and
working portals should be reversed. It can be very difficult
to excise tears of the anterior horn with the punch; however,
they can be addressed more easily using the shaver, taking
care not to resect the insertion of the anterior horn or to

Fig. 3.4 Meniscal cyst of the lateral meniscus. (a) Preoperative MRI,
(b1, b2) arthroscopic views, (c) cyst resection using a direct open
approach, (d) vertical sutures to close the connection to joint

20

P Archbold et al.

b1

b2

c

d

Fig. 3.4 (continued)

Fig. 3.5 Suture of anterior
horn. (a) Rare case:
disinsertion of the anterior
horn of a lateral discoid
meniscus. (b) Absorbable
sutures are used to preserve
the meniscus tissue

a

b

4

Meniscal Suture
Maad AlSaati, S Thompson, R Desmarchelier, G Demey,
P Neyret, and C Butcher

Introduction

Classification

Menisci are intra-articular fibrocartilaginous structures that
have multiple functions, including shock absorption, control
of anterior tibial translation, lubrication of the joint, and possibly proprioception. Owing to their many functions and the
risk of meniscectomy-associated osteoarthritis, the management of meniscal tears should include “meniscus-­preserving”
procedures wherever possible.
Studies on meniscal healing have demonstrated that the
prognosis of meniscal tears is closely related to their location
within the meniscal tissue and meniscal blood supply. The
most peripherally located lesions in the “red-red” zone have
an excellent prognosis owing to the relatively abundant
blood supply. Lesions in the middle third “red-white” zone
are on the edge of the vascular zone and are also likely to
heal based on reasonable blood supply. The most central
lesions, located in the avascular area or “white-white” zone,
have a limited capacity to heal.
Classically, open meniscal repair with vertical sutures via a
posteromedial or posterolateral approach to the knee was considered to be the gold standard of meniscal saving surgery.
However, there are complications of an open approach, including hypoesthesia, neuroma formation, and often the need for a
longer hospital stay than is required for a simple arthroscopy.
Accordingly, we now prefer an all-inside technique for repair
of the body and posterior aspect of the meniscus.

We use the ISAKOS classification, which takes into consideration length, depth, and location of meniscal lesions. The
meniscus can be divided into three zones according to its
width (Zone 1: synovial-meniscal junction or “red-red” zone,
Zone 2: “red-white” zone, and Zone 3: free edge of the
meniscus or “white zone”) and three areas from front to back
(anterior horn, middle segment, and posterior horn). There
are several types of lesions according to their appearance:
vertical, horizontal, radial, flap, complex, and finally discoid
meniscus.

Maad AlSaati · S Thompson · R Desmarchelier
Centre Albert Trillat, Lyon, France
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

Indications
The ideal indication for repair is a tear of the lateral meniscus
that is located peripherally and occurs in a young patient. It
is important to note, however, that tears less than 10 mm in
length in this location are likely to heal spontaneously and
can be left in situ.
General contraindications to meniscal repair include tears
with a radial, flap, or complex configuration, although factors
such as alignment, activity, and age of patient may modify the
decision. These lesions should be treated with partial meniscectomy, leaving as much normal meniscus as possible.
A variety of meniscal suturing techniques exist. The open,
“all-outside” method is considered to be a reference standard. However, owing to the popularity of arthroscopy and
demand for less invasive solutions, several new meniscal
repair methods have been developed.
Whatever the technique, a number of principles should be
followed:
• The lesion should be perforated to create vascular channels that are conducive to healing.
• The junction between the meniscus and capsule should be
debrided using a rasp or a synovial knife.
• The knee must be stable or stabilized.

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_4

21

22

Maad AlSaati et al.

We primarily describe the techniques used in our
experience:
• All-Inside technique (Morgan) using Fast Fix 360®
• Inside-Out technique (Henning)
• Outside-In technique (Warren)
The choice of technique depends on the location of the
lesion:
• Posterior horn lesion: All-Inside or Inside-Out technique
• Middle segment lesion: All-Inside or Outside-In
technique
• Anterior horn lesion: Outside-In technique
• Very peripheral lesion not visible at arthroscopy: Open or
Outside to Inside technique with a posteromedial portal

 ll-Inside Technique Using FastFix 360®
A
Implant
Surgery is performed through the standard anterolateral and
anteromedial arthroscopic portals. The procedure begins
with the identification of the lesion, assessment of its stability, its potential for healing, and technical capability to be
sutured (Fig. 4.1). The lesion is then refreshed using a rasp or
with perforations.
Prior to introduction into the joint, the FastFix 360®
device is prepared by cutting the protective sheath to
ensure passage of the implant beyond the meniscal wall
and is typically set at 16–18 mm. It is then introduced
through the anteromedial portal, with its removable protective sheath still in place (Fig. 4.2). The sheath is
removed and the positioning of the implant is determined
by the type of lesion and suture configuration desired
(vertical or horizontal sutures). The first implant is
inserted until the protective sheath comes into contact
with the meniscus (Fig. 4.3a–d).

Fig. 4.1 Arthroscopic view identifying the lesion

Fig. 4.2 Introducing FastFix 360® with its protective sheath

4

Meniscal Suture

23

a

b

c

d

Fig. 4.3 (a–d) Implementation of the first implant

24

a

Maad AlSaati et al.

b

Fig. 4.4 (a, b) Loading the second implant

Passage of the implant through the outer rim of intact
meniscus is easily felt by the surgeon. The device is
rotated 90° to facilitate the docking of the anchor. The
inserter is retracted back into the joint and the second
implant is loaded (Fig. 4.4a, b). It is introduced into the
meniscus in the same manner as before (Fig. 4.5a–c). The
free end is then tightened carefully by gentle traction. The

knot-pusher is used to adjust the suture and optimize the
tension of the suture construct (Fig. 4.6a–c). The free end
is cut. Additional devices can be placed depending on size
and stability of the tear. The different devices are placed
from the posterior part to the anterior part of the lesion.
The stability of meniscal repair is determined with the
arthroscopic probe.

4

Meniscal Suture

a

c

Fig. 4.5 (a–c) Implementation of the second implant

25

b

26

a

Maad AlSaati et al.

b

c

Fig. 4.6 (a–c) Tightening the knot by gentle traction and then with the knot-pusher

I nside-Out Technique Using Aiming Cannulas
(Henning)
Similar to the previous technique, surgery begins with the
identification of the lesion and debridement before repair. The
aiming cannula (single or double) is introduced through the
appropriate portal and is directed toward the lesion (Fig. 4.7).
Two needles with nonabsorbable sutures are passed through
the cannulas. They are retrieved through an open approach
with the knee flexed to 90°; on the medial side, the incision is
longitudinal posterior to the posterior edge of the medial collateral ligament; on the lateral side, it is posterior to the pos-

terior edge of the lateral collateral ligament (Fig. 4.8). Good
reduction of the lesion is controlled arthroscopically while
pulling on the sutures externally. The knot is tied over the
capsule through the open approach. We did not use this technique for many years, but we describe it because it can be
useful when modern devices are not available.

Outside-In Technique with a Loop (Warren)
A needle is inserted through the capsule and the meniscal
tear. A loop is crafted with a stiff suture and passed into the

4

Meniscal Suture

a

27

b

Fig. 4.7 Suture from inside out with single or double cannulas

a

b

Fig. 4.8 Suture with double-cannula passage of the needles and retrieval of the suture

joint (Fig. 4.9). A second needle is inserted, and a second
suture is passed through the needle and through the loop in
the first suture. Both needles are then removed, leaving the
sutures in place. The looped suture is then pulled from the
joint, bringing the second suture with it. Reduction is
achieved by pulling on both strands. The knot is tied over the
capsule through a small incision (Fig. 4.10).

Rather than using a looped suture to retrieve the repair
suture, we currently use Meniscus Mender II®, an outside-in
suture kit for the passage of suture through an intra-articular
deployable loop. This technique is useful to suture the anterior part of the meniscus, particularly that of the lateral
meniscus.

28

a

Maad AlSaati et al.

b

Fig. 4.9 Suture from outside to inside with needle. Both sutures are retrieved through the anterior approach. A loop is crafted to pull the second suture

Meniscal Root Repair
The effect of a complete root tear is similar to a total meniscectomy (Fig. 4.11). Acute injuries of the roots should
therefore be actively sought and repaired. This is especially
the case with ACL injuries, due to the frequency of association and the importance of the repair for stability. In lesions
of degenerative menisci, the effect of repair is less certain.
The technique involves creating a receptive base for tissue
repair. Proprietary instruments facilitate the creation of a
transosseous tunnel and suture insertion for stabilization
(Fig. 4.12).

Fig. 4.10 The appearance of the tear before tightening the knot.
Traction on the strands confirms good apposition of the edges of the
lesion prior to tying the knot

4

Meniscal Suture

29

Postoperative Care

Fig. 4.11 Meniscal root tear

a

Meniscal repair is typically performed as an outpatient surgery. Knee range of motion is started on the first postoperative day. The aim is to have full extension, but hyperextension
is prohibited. Flexion beyond 120° is delayed up to 6 weeks
postoperatively. The configuration of the tear and stability of
the repair can alter these restrictions. There is no agreement
in the literature regarding weight bearing and range of
motion restrictions. However, it seems logical to allow flexion and axial loading in cases of repair of longitudinal tears
(the compression forces act to push the repaired meniscus
together until hyperflexion is reached). Thus full weight
bearing is permitted immediately in such patients. By contrast, no weight bearing should be allowed in the event of
repair of a radial tear.
Return to sport is not permitted until 4 months postoperatively at the earliest. Return to sports involving pivoting or
contact is not permitted until 6 months postoperatively.

b

Fig. 4.12 (a) Creating a transosseous tunnel for root repair, (b) a completed repair

5

Anterior Cruciate Ligament
Reconstruction: Surgical Technique
R Magnussen, AM Ozturk, G Demey, P Neyret,
and C Butcher

Classification of ACL insufficiencies

Introduction
Isolated

Reconstruction of the anterior cruciate ligament is the
method of choice for the treatment of chronic anterior laxity.
Differences exist in surgical technique, graft choice, and how
surgeons choose graft position.
We prefer to use a bone patellar tendon bone autograft
for this reconstruction. This technique was initially
described by Lambda in 1937 and popularized by Kenneth
Jones. In the procedure described by Kenneth Jones, the
patellar tendon bone graft remained attached at its tibial
insertion. Franke, followed by Dejour and Clancy, promoted
the use of a free graft.
To guide our treatment strategy, we use the classification
as proposed by the Henri Dejour school (Fig. 5.1). This classification takes into account associated lesions and degree of
anterior laxity. These are a function of both the initial acute
injury and the time elapsed. With increasing time, the potential for associated lesions, resulting laxity, and degeneration
increases.
Initially the laxity may be “isolated,” and occasionally “partial” due to either some healing of the fibers to the PCL or some
remaining intact fibers, usually in the posterolateral bundle.
In a minority (<5% of cases), there is an associated
acute posterior-lateral lesion, which may dictate careful
assessment and treatment of both laxity and coronal alignment to prevent failure of the reconstruction (Fig. 5.2).

R Magnussen · AM Ozturk
Centre Albert Trillat, Lyon, France
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

Evolved

ACL Laxity
with
pre-OA

OA
due to
ACL laxity

Posterolateral < 5%
25-35 yrs

Fig. 5.1 Classification of ACL laxities according to Henri Dejour

Fig. 5.2 ACL and posterolateral deficiency often represents a frontal
imbalance that needs to be corrected by combined ligament and bony
surgery. The role of the osteotomy is to protect the grafts, and a small
amount of valgus is enough (2 or 3°). In the absence of LCL graft, an
obvious hypercorrection would be required

C Butcher
Healthpoint, Abu Dhabi, UAE

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_5

31

32

R Magnussen et al.

Fig. 5.4 Lower limb axis assessed with fluoroscopy

Fig. 5.3 Evolved ACL deficiency. Repeated meniscal, capsular and
ligamentous injury, or surgical meniscectomy allows greater sagittal
laxity and chondral injury from abnormal loading patterns. Early
degenerative changes are revealed by x-ray or MRI

In these and the “evolved” ACL laxities, associated lesions
(often posterior medial) develop with time and are followed by ACL laxity with pre-arthritis (Fig. 5.3). In this
condition, repeated meniscal, capsular and ligamentous
injury, or surgical meniscectomy allow greater sagittal laxity and chondral injury from abnormal loading patterns.
Symptomatically instability is still dominant, but radiological investigation reveals early degenerative change.
Attention must be made not only to the deficient soft tissues but to the coronal and sagittal alignment. Combined
ligament reconstruction and osteotomy, in either the frontal or sagittal plane, is a surgical option these cases (see
later in chapter, Figs. 5.4 and 5.5).
After perhaps 20–30 years, true OA secondary to ACL
laxity may develop, with the typical posterior medial tibial
defect, or “cupula.” Symptoms from the degeneration are
dominant, usually without feelings of instability. ACL reconstruction is not indicated. Either osteotomy (preferably closing wedge) or more commonly TKA is appropriate, whereas
UKA is not (Figs. 5.6 and 5.7).

Fig. 5.5 Fixation of the osteotomy with staples is an option in the
sports patient

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

33

Kenneth Jones Surgical Technique
Although we have modified the technique to utilize a free
bone patellar tendon bone graft, we continue to call this
operation a KJ procedure. Our technique was largely inspired
by Pierre Chambat.

Setup and Clinical Examination
The setup we use to perform an ACL reconstruction is the
same that we use for nearly all types of knee surgery
(Fig. 5.8). After the patient is under anesthesia and the
extremity sheet has been applied but prior to inflation of the
tourniquet, the knee is once again tested for anterior laxity
with the Lachman-Trillat test and the pivot shift test.

 arvest of the Bone Patellar Tendon
H
Bone Graft
Fig. 5.6 Medial condyle sitting in the posterior “cupula” during a TKA

We generally harvest the graft prior to arthroscopy in order
to avoid swelling of the soft tissues. The skin incision starts
at the inferior pole of the patella and continues 2 cm distal
to the tibial tubercle (Fig. 5.9). In total the paramedian skin
incision is 6–8 cm in length and situated on the medial border of the patellar tendon. Dissection is performed down to
the tenosynovium, which is vertically incised down the lateral aspect of the patella tendon and carefully elevated from

Fig. 5.8 Routine setup

Fig. 5.7 Posterior polyethylene wear in a UKA implanted for osteoarthritis secondary to ACL deficiency

34

R Magnussen et al.

the anterior aspect of the tendon. The lateral and medial
­borders of the tendon are exposed as well as its insertion on
the tibial tubercle and its origin on the distal pole of the
patella.

 reparation of the Patella Tendon Part
P
Harvest of the graft starts with the tendinous part. We use a
specifically designed double blade scalpel (Fig. 5.10). The
graft width is 10–11 mm. The tendon is incised in the direction of its fibers (Fig. 5.11). The proximal and distal osteotendinous transition zones of the tendon are marked with a
23 blade. The bone blocks are marked by incising the periosteum with the blade.
 erforation of the Bone Blocks
P
Before the bone blocks are cut, three holes are drilled with a
2 mm drill in the future bone blocks (Fig. 5.12). Two holes
are created proximally in the patella and one distally on the
tibial tubercle. We find it easier to drill these holes prior to
graft harvest rather than on the back table.
Fig. 5.9 Incision landmarks

Fig. 5.10 Double-blade
scalpel

Fig. 5.11 Identifying medial and lateral sides of the patellar
tendon

Fig. 5.12 Drilling of bone blocks prior to cutting them

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

35

 arvest of the Bone Blocks
H
In order to facilitate the harvest of the bone blocks, the longitudinal incisions in the patellar tendon are opened up using a
Farabeuf retractor. A small angled blade saw is used for the harvest, with a stop to avoid cutting deeper than 10 mm (Fig. 5.13).
Tibial Bone Block
The tibial bone block is shaped in a specific way. It is trapezoidal in the shape of a champagne cork. The width is 10 mm
proximally, widening gradually to 12 mm in the distal 10 mm of
the block. The overall length of the tibial bone block is at least
25 mm, and it is 10 mm thick (Fig. 5.13). The tibial bone block
is then detached using a curved osteotome starting proximally.
Patellar Bone Block
The patellar bone block is prepared using a small blade saw. The
dimensions of the patellar bone block are 10 mm in width and
15 mm in length. The tibial bone block and distal portion of the
graft are lifted out of the harvest site and pulled proximally. The
adhesions between tendon and Hoffa fat pad are dissected until
the inferior pole of the patellar is clearly visible. A small osteotome of 10 mm in width is used to detach the patella bone block;
its thickness should be between 5 and 8 mm. The osteotome
should be introduced parallel to the anterior cortex of the patella
(Fig. 5.14). One must take care not to fracture the patella when
detaching the bone block. Any effort to detach it by prying with
the osteotome must be avoided. If needed, a second osteotome
is placed anterior to the first, in order to detach the patella bone
block. The free graft is then prepared by the surgeon of the back
table (Figs. 5.15 and 5.16).

Fig. 5.13 Cutting the tibial bone block

Fig. 5.14 Patellar bone block detachment

Fig. 5.15 Patellar tendon graft

36

R Magnussen et al.

Fig. 5.16 Autograft shape
and size

30-35 mm
12 mm

15 mm
9 mm

10 mm

5-8 mm
Baguette tibiale

Baguette rotulienne

Fig. 5.18 Bone block calibration and adjustment

Fig. 5.17 Periosteal suture

The defect in the tendon is closed with interrupted resorbable stitches. The tenosynovium is carefully closed above the
tendon and remaining periosteum closed over the bony
defects (Fig. 5.17).

 reparation of the Bone Patellar Tendon
P
Bone Graft
This step of the procedure can be performed by an assistant while notch preparation and tunnel drilling continue.
The first step in the preparation of the bone patellar tendon bone graft is the sizing of the bone blocks (Fig. 5.18).
The edges and corners of the patellar bone block should
be rounded using Liston scissors and cutting scissors. It
should pass easily through the 9 mm hole of the graftsizing block (Fig. 5.19). The proximal end of the tibial
bone block should engage in the 10 mm hole of the graft-

Fig. 5.19 Graft-sizing block

sizing block but should not completely pass. This illustrates the press fit that will be obtained on the femoral
side. Pull sutures are introduced into the patellar bone
block to aid in graft passage.
Through the two drill holes in the patellar bone block, a
FiberWire suture is introduced in a figure of “8.” This strong
suture allows axial traction on the graft during passage.
A number 5 resorbable suture is placed in the tibial bone

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

37

Fig. 5.20 Bone patellar
tendon bone graft after
preparation

Fig. 5.21 Arthroscopic
exploration

block (Fig. 5.20). This suture will allow the retraction of the
bone block in case of problems with the femoral fixation.
The prepared graft is subsequently kept in a physiological
solution with vancomycin. The graft should not be covered
with gauze as it increases the risk of finding the graft in the
trash together with the gauze.

Arthroscopy
I ntercondylar Notch Preparation
The scope is introduced through the anterolateral portal. The
instruments are introduced through the anteromedial portal

(see chapter on arthroscopy) (Fig. 5.21). Any meniscal or
cartilage lesions are evaluated and treated (Fig. 5.22). If a
meniscal repair is considered necessary, the repair should be
performed prior to the anterior cruciate ligament reconstruction (see chapter on arthroscopy). If a posterior meniscal
lesion is suspected, but the visualization is poor, a posterior
compartment view must be obtained via either the Gillquist
maneuver or creation of a posteromedial portal (our choice)
to evaluate the posterior part of the meniscus.
First, the remnant of the anterior cruciate ligament is visualized as well as the morphology of the intercondylar notch.
Preparation of the intercondylar notch is done in a systematic
way. While in the past we removed the remaining fibers of

38

Fig. 5.22 Arthroscopic view of a medial meniscus tear

the ACL, we now carefully analyze the remaining footprint
to determine the insertion sites of the ACL and try to preserve the remaining fibers unless they obscure visualization.
If there is any impingement of the old fibers on the anterior
part of the notch once the graft is passed, these fibers are
resected. Overzealous clearing of the wall of the notch limits
the blood supply to the healing graft. We generally prefer to
do any resection with a radiofrequency device as it preserves
underlying osseous anatomy better than a shaver. In spite of
the desire to preserve remnant tissue in the notch when possible, obtaining clear visualization is essential to avoid the
most frequent error of malpositioning: a femoral tunnel positioned too anteriorly (Fig. 5.23).

Notch Plasty
In our hands, a notch plasty is rarely needed. We decide to
perform a notch plasty when an impingement of the graft is

R Magnussen et al.

Fig. 5.23 Posterior edge of the intercondylar notch (anterolateral
portal view)

observed in the intercondylar notch. This most frequently
involves the superior part of the intercondylar notch and less
frequently the lateral side.
To perform a notch plasty, the knee is placed in semiflexion. The zone of impingement is removed using a curve
osteotome. This osteotome is positioned on the cartilage
bone transition zone. By gently tapping it with a mallet, the
osteotome easily removes the zone of impingement. The
bony debris should be carefully removed and the remaining
notch smoothed with a shaver, burr, or coblation.

The Femoral Tunnel
Accurate femoral tunnel positioning is key to successful
ACL reconstruction. The appropriate tunnel location is proximal and posterior to the lateral intercondylar (Resident’s)
ridge (Fig. 5.24). For single-bundle reconstructions, we pre-

Fig. 5.24 Femoral insertion sites

AM
PL

Lateral
intercondylar
ridge
Lateral
bifurcate ridge

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

39

Fig. 5.26 Appearance of pin

Fig. 5.25 Femoral guide

fer to place the tunnel proximal to the lateral bifurcate ridge
(when visualized) where the anteromedial bundle inserts and
take care that it is posterior to the resident’s ridge. Once the
appropriate tunnel position has been identified, the femoral
drill guide is introduced through the anteromedial portal to
the desired location (Fig. 5.25). The bullet is subsequently
introduced into the jig. This indicates the position of the lateral skin incision. The skin incision should be situated on the
lateral surface of the lateral condyle anterior and proximal to
the lateral collateral ligament. The incision must be sufficiently lateral to avoid opening the suprapatellar pouch.
The skin and the fascia lata are incised, and the bullet is
introduced through the guide until it is in contact with the
bone. A guide pin is subsequently introduced into the bullet
and driven across the lateral condyle and into the intercondylar
notch. The guide is removed, and the pin is driven 4–5 mm
into the notch (Fig. 5.26). At this point, the position of the
guide pin is checked while viewing through the medial portal.
The surgeon now introduces a curette to prevent inadvertent
guide pin advancement during over-drilling. We have specially
designed curette with a small hole in the middle, which accepts
only the tip of the guide pin. A femoral tunnel 6 mm in diameter is drilled over the guide pin (Fig. 5.27). The direction of

Fig. 5.27 Femoral tunnel drilling (6 mm drill first)

the cannulated drill should be perfectly parallel to the guide
pin. The progression of the cannulated drill should be progressive and smooth. If abnormal resistance is noted, the drill
should be retracted immediately and its direction should be
checked. Using a sharp drill will make it easier to detect resistance due to misdirection early. In case of misdirection, metal
debris is produced. Next a 10 mm cannulated drill is introduced over the guide pin, and the femoral tunnel is enlarged.
The guide pin is retracted and bone debris from the tunnel is
carefully removed (Figs. 5.28, 5.29, and 5.30). Preparation of
the femoral tunnel in two steps has two advantages:

40

R Magnussen et al.

Fig. 5.30 Femoral tunnel (anteromedial portal view)

Fig. 5.28 Femoral tunnel (anterolateral portal view)

inspected with the scope to verify the circumferential presence of cancellous bone (Fig. 5.32). The corner of the femoral tunnel can be rounded with the curette to lower the risk of
graft erosion. Finally a plug is placed in the external aperture
of the tunnel to avoid leakage of the irrigation fluid during
tibial tunnel preparation.

 he Tibial Tunnel
T
The tibial drill guide is introduced through the anteromedial portal (Fig. 5.33). The guide is positioned at a point
located according to the following landmarks, corresponding to the footprint of the original anterior cruciate ligament (Fig. 5.34):
• In front of the posterior cruciate ligament, just lateral to
the cartilage of the medial tibial plateau
• Just behind the anterior horn of the medial meniscus
• Medial to the anterior horn of the lateral meniscus
Fig. 5.29 Well-positioned femoral tunnel assessed on a 3D CT scan
(AM bundle)

• First, smooth progression of the drill without the need to
use excessive force.
• Second, the tunnel position can be adjusted by 2–3 mm if
necessary. This correction is done by moving the guide pin
with the curette (hence the hole in the curette). The direction of the 10 mm drill can thus be adjusted by 2–3 mm in
the previously drilled 6 mm tunnel (Fig. 5.31a–c).
All debris should be removed from the tunnel using suction (if one forgets to perform this step, one will be confronted with the presence of bony debris on the lateral side of
the condyle visible on the postop x-ray). Next, the tunnel is

This position is usually aligned with the two femoral condyles in 90° of flexion. The guide is set at 45°, and the entry
point on the tibial metaphysis is medial to the tibial tubercle.
The bullet is subsequently introduced, the guide pin is driven
into the knee, and its position is checked first in flexion
(Fig. 5.35).
The knee is then extended and the position of the guide
pin is checked to ensure no impingement occurs between the
notch and the guide pin. A 3 mm minimal distance should be
present between the guide pin and the intercondylar notch in
order to avoid any conflict between the notch and the graft.
This concept is called graft clearance and was introduced by
R. Jakob. Once the final position is checked, the curette is
placed over the guide pin, and the 6 and 9 mm cannulated
drills are introduced over the guide pin (Fig. 5.36). It is very

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

a

41

b

c

Fig. 5.31 (a–c) Precise positioning can be obtained with progressive enlargement of the tunnel diameter (6 and 10 mm diameter)

Fig. 5.32 Femoral tunnel inspection (outside view)

Fig. 5.33 Tibial guide

42

Fig. 5.34 Tibial guide positioning

Fig. 5.35 Pin appearance

important to respect this sequence (first 6 followed by 9)
because a 9 mm drill could induce a fracture of the tibial
spine if used straightaway. Moreover, as for the femoral tunnel, the position of the tibial tunnel can be adjusted by
2–3 mm if necessary by changing the position of the guide
pin before using the 9 mm drill. Tunnel debris is aspirated,
and the entry hole of the tunnel is cleared from soft tissues
which could block the entrance of the bone block (Fig. 5.37).
Because the graft will be passed from proximal to distal, this
step is key to ensuring smooth graft passage.

R Magnussen et al.

Fig. 5.36 Tibial tunnel drilling (6 mm drill first)

Fig. 5.37 Tibial tunnel (anterolateral portal view)

I ntroduction of the Bone Patellar Tendon Bone Graft
With the knee flexed to 30°, a pull suture is introduced
through the tibial and then the femoral tunnel in a retrograde fashion with a suture guide. We verify arthroscopically that the suture guide does not perforate the posterior
cruciate ligament. The pull suture is captured in the intercondylar notch with a grasper introduced through the femoral tunnel. The traction sutures from the graft are looped
through the pull suture (Figs. 5.38 and 5.39). This allows
the introduction of the graft in an anterograde fashion first
through the femoral condyle, into the intercondylar notch,

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

43

Fig. 5.40 A Wolff grasper is sometimes used to guide the bone block
migration through the notch

Fig. 5.38 Passage of a passing suture

Fig. 5.39 Introduction of the graft

and then through the tibial tunnel. It is sometimes possible
that the passage of the patella bone block in the notch is
difficult, particularly if the bone block is too long. In these
cases, a Wolff grasper is introduced through the anteromedial portal to guide the bone block through the notch
(Fig. 5.40). Once the graft is introduced in the tibial tunnel,
impaction of the bone block in the femoral tunnel can be

Fig. 5.41 Graft impaction

initiated (Fig. 5.41). The orientation of the bone block in
the femoral tunnel should be controlled. The tendon attachment site should be positioned posteriorly in the femoral
tunnel. During impaction, it is essential to exert some traction on the graft and to be sure that the graft progresses into
the tibial tunnel. This is to prevent invagination of the tendinous portion of the graft between the femoral bone block
and the tunnel wall like an accordion, resulting in a para-

44

R Magnussen et al.

Fig. 5.42 “Accordian paradox”. Impaction without distal tension on
the graft

Fig. 5.43 “Accordian paradox”. The graft starts to invaginate between
the bone block and the wall of the tunnel

Fig. 5.45 Isometry of the graft is tested before tibial fixation by strong
traction on the suture during flexion and extension

• Absence of jamming of the bone block in the tibial tunnel

Fig. 5.44 “Accordian paradox”. Full invagination of the graft

doxical situation: The more the bone block is impacted on
the femoral side, the less the graft progresses into the joint
(Figs. 5.42, 5.43, and 5.44).
The bone block on the femoral side is advanced with an
impactor and mallet while the graft is kept under tension.
The bone block on the femoral side should be well impacted
and should not be palpable outside of the femoral condyle.

Graft Fixation
Prior to fixation, the following should be checked:
• Isometry of the graft during flexion and extension between
5 and 90° (Fig. 5.45)
• Absence of impingement of the graft in the intercondylar
notch

On the anteromedial side of the tibia, a 2 mm drill hole is
made through cortical bone distal to the tibial tunnel, and
connecting to the tibial graft harvest site (Fig. 5.46). The
FiberWire loop from the graft will be tied over this bone
bridge providing the first tibial fixation (Figs. 5.47, 5.48, and
5.49). A guidewire is then introduced through the tibial tunnel inside the knee. The guidewire should be on the
­anterolateral border of the bone block in the tibial tunnel
(Fig. 5.50). If necessary, the position of this guide pin can be
modified. Its position is secured inside the knee with a
grasper. A resorbable interference screw (Biosure, Smith &
Nephew), 25 mm in length and 9 mm in diameter, is introduced as additional fixation over the guidewire (Fig. 5.51).
The interference screw is introduced under arthroscopic
vision until it reaches the level of the joint line. In case of a
long graft, contact between this interference screw and the
bone block is preferred over screw tendon contact. This combination of the suture over the bone bridge and interference
screw provides double fixation on the tibial side, which we
recommend in all cases.
Prior to closure, four variables have to be checked:
A—the position of the screw with respect to the bone block
in the tibial tunnel.
B—the tension of the fixation and the tension of the graft, the
posterior fibers should be tensioned in extension while
the anterior fibers are somewhat slack.

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

45

Fig. 5.46 Drilling of transosseous tunnel (additional tibial fixation)

Fig. 5.48 One limb of the Fiberwire has been passed through the bone
tunnel

Fig. 5.47 One limb of the Fiberwire will be passed through the bone
tunnel, with the aid of another needle and suture

Fig. 5.49 The two limbs of the Fiberwire have been tied over the bony
bridge

C—absence of impingement within the intercondylar notch.
D—a stable Lachman-Trillat test

through the anteromedial portal. The subcutaneous tissues
are closed with a resorbable suture, and the skin is closed
subcuticularly or by using skin staples.
An additional compressive bandage is applied that will be
removed 1 h postoperatively.

At the end of the procedure, the tourniquet is deflated and
hemostasis achieved. An intra-articular drain is introduced

46

R Magnussen et al.

Fig. 5.50 Position of the graft
and interference screw/guidewire

20˚

Fig. 5.51 Position of the
interference screw in the tibial
tunnel

Postoperatively
• The knee is put into a brace with 20° of flexion 48 h, in
order to prevent patella infera.
• AP and lateral plain x-rays are performed immediately
postoperatively.
• Low molecular weight heparins are prescribed for
10–15 days.
• Prophylactic antibiotics are administered during a 24 h
period.
• Skin staples or skin sutures are removed between day 12
and day 15.
Clinical follow-up is planned for days 45, 90, 180, and
360. Telos stress radiographs are obtained at the 1 year visit.
When a young surgeon starts his practice, it can be useful to

check the tunnel position using a 3D CT scan. Very quickly
this surgeon will become very confident and improve his tunnel placement.

The KJ Modification: The KJT or KJG
This surgical intervention combines both an intra-articular
reconstruction using a bone patella tendon bone graft and an
extra-articular augmentation. This may be performed using
the gracilis or occasionally the semitendinosus tendon. We
(PN) introduced this combination in 1996. Alternatively, the
fascia lata may be used, as described in the next section on the
Lemaire extra-articular augmentation and its short graft variant. These techniques improve the rotational stability of the
knee and are indicated in cases where there is a clearly posi-

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

tive pivot shift test. They are contraindicated in cases where
there is posterior lateral laxity, as the pathological external
tibial rotation may be permanently set by the procedure.

 arvest of the Bone Patellar Tendon
H
Bone Graft
We start with the harvest of the bone patellar tendon bone
graft as described above. The anteromedial skin incision,
which is used for the harvest of the bone patella tendon bone
graft, is extended distally for about 2 cm.

Harvest of the Gracilis or Semitendinosus
For the extra-articular anterolateral plasty, either the semitendinosus or the gracilis tendon can be used. We prefer the
semitendinosus. The pes anserinus is identified. The sartorius tendon is most superficial and covers the gracilis and
semitendinosus tendons. Both can be seen and palpated
under the sartorius fascia. The sartorius tendon is incised in
the direction of the fibers proximally and then “hockey-­
sticked” at its insertion on the tibia. On its undersurface, the
gracilis tendon can be identified proximally and the semitendinosus distally (Figs. 5.52 and 5.53). As the three tendons

Fig. 5.52 Dissection of hamstring tendons (proximally: gracilis, distally: semi-tendinosus)

47

have a conjoint tendon insertion on the tibia, they can be
more easily identified 4–5 cm proximally. The superficial
fibers of the medial collateral ligament can be harmed during
the dissection since they cross deep to the pes anserinus conjoint tendon. Once the gracilis tendon is identified, the tendon is isolated and a vessel loop is applied. If the
semitendinosus is used, the vinculae of the semitendinosus
(including one to the gastrocnemius aponeurosis) are dissected carefully. The distal part of the tendon is whipstitched
using a number 5 suture. Once this is done, the insertion of
the tendon on the tibia is cut. Subsequently the tendon is harvested using a closed stripper (Fig. 5.54). The pulling sutures
are passed through the eye of the closed stripper. The tendon
is maintained under tension while the stripper is progressively pushed proximally with the knee in the figure of four
position. Usually an increase in resistance is felt when the
myotendinous junction is reached. The graft is usually at
least 5 mm in diameter and at least 18 cm in length.

Preparation of the Graft
The muscle fibers attached to the proximal end of the tendon
are removed with the use of an osteotome or the back side of
a blade. The proximal end of this tendon is usually wider and
thinner as this is the muscle-tendon transition area. This side
of the tendon is also whipstitched with a no 5 suture. The

Fig. 5.53 Dissection of semitendinosus tendon in this case

48

R Magnussen et al.

tendon is then passed through the tibial bone block that was
perforated with a 4.5 mm drill (Figs. 5.55 and 5.56). This
construction allows for an intraosseous fixation of the extra-­
articular augmentation when the bone block is impacted in
the femoral tunnel.
Upon wound closure, it is advisable to close the extensions of the sartorius tendon and to have an extra drain positioned in this area.

 urgical Approach for the Extra-Articular
S
Augmentation

Fig. 5.54 Hamstring harvesting using a closed stripper

Fig. 5.55 Bone-patellar
tendon-bone graft. A
4.5 mm diameter hole is
drilled in the tibial bone
block

Fig. 5.56 Composite graft
(patellar tendon and
hamstring tendon)

After the patellar tendon and the hamstring tendon harvest,
the surgical approach for the lateral plasty is made. This
approach is done prior to the arthroscopy. The anterolateral incision is 5–7 cm long and starts just proximal to the
lateral epicondyle and ends at the level of Gerdy’s tubercle
(Fig. 5.57). The iliotibial (IT) band is subsequently divided
in the direction of its fibers (the distal part of the incision
is somewhat more oblique) to the level of Gerdy’s tubercle. Care has to be taken not to harm or transect the lateral
collateral ligament since it crosses the undersurface of the
IT band. The lateral collateral ligament is identified by

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

49

Fig. 5.59 Press-fit proximal fixation

Fig. 5.57 Lateral approach and fascia lata incision

Fig. 5.60 The graft is first passed under the lateral collateral ligament

Fig. 5.58 Femoral tunnel just proximal to the insertion of the lateral
collateral ligament

palpating its anterior and posterior borders with the scissors. Just posterior to the lateral collateral ligament, the
lateral head of the gastrocnemius muscle can be palpated
as well as the posterior lateral capsular structures. The
posterolateral structures and the insertion of the lateral
collateral ligament on the femur form a triangle. The
undersurface of the lateral collateral ligament is dissected.
However one should stay extra-articular, and therefore, we
prefer to do this last step at the end of the intra-articular
procedure, in order to avoid leakage of arthroscopic irrigation fluid. The lateral entry point for the femoral tunnel is
chosen just proximal and ­posterior to the insertion of the
lateral collateral ligament on the femur (Figs. 5.58, 5.59,
5.60, 5.61, 5.62, and 5.63). This entry point determines the

Fig. 5.61 The posterior/inferior bundle is passed deep to the fascia lata

50

R Magnussen et al.

Fig. 5.62 Lateral tenodesis; proximal fixation

Fig. 5.64 Bony tunnel under Gerdy’s tubercle. The awl is making the
medial entry point. The lateral entry has been made, after releasing a
small area of tibialis anterior

Fig. 5.63 Lateral tenodesis; proximal and distal fixation

biomechanics of the lateral extra-­articular augmentation.
Therefore, the direction of the femoral tunnel will be
somewhat more horizontal than for the classic
KJ. Reconstruction of the intra-articular anterior cruciate
ligament is as described in the previous chapter.

Fixation of the Extra-Articular Augmentation
Proximal graft fixation is obtained automatically when the
bone block is impacted in the femoral tunnel (Fig. 5.59). The
two free ends of the tendon are now passed under the lateral
collateral ligament. Distal fixation of the tendon is achieved
under Gerdy’s tubercle. A bony tunnel is made at this location using an awl (Fig. 5.64). To achieve this, it can be useful
to release a small amount of the origin of the tibialis anterior
muscle on the lateral border of Gerdy’s tubercle. The proximal, anterior free end of the tendon is passed from anterior to
posterior through the tunnel. The posterior part is passed

Fig. 5.65 Completed lateral tenodesis fixation

from posterior to anterior in the same tunnel. To achieve this
position, it must first be passed deep to the posterior part of
the fascia lata (Figs. 5.60 and 5.61). The two ends of the
tendon are sutured side-to-side, thus creating a solid fixation
(Figs. 5.63 and 5.65). Tensioning of the extra-articular plasty
is performed after fixation of the intra-articular reconstruction, with the knee in 30° of flexion and neutral rotation.

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

51

Approach
We use a lateral skin incision starting distally at the level of
Gerdy’s tubercle and continuing proximally 15 cm in the
direction of the fibers of the IT band (Fig. 5.67).

 arvest and Preparation of the Fascia Lata
H
Graft

Fig. 5.66 Fascia lata suture

The fascia lata graft, 18 cm long and 1 cm in width, is harvested with the 23 blade (Figs. 5.68 and 5.69). Care is taken
not to section or harm the lateral collateral ligament, which
crosses the incision distally. The dorsal border of the graft
corresponds with the anterior border of the intermuscular
septum. The graft remains in continuity distally with Gerdy’s
tubercle.

Closure of the wound is done over a drain. The IT band is
sutured with interrupted No. 5 sutures (Fig. 5.66).

Lemaire Extra-Articular Augmentation
The extra-articular augmentation using the fascia lata was
described by Marcel Lemaire in 1967. This technique was
modified in Lyon by Professor Dejour and has always been
considered very useful in the treatment for rotational instability, specifically in the case of a clearly positive pivot-shift
test. It does not control anterior tibial translation of the
medial compartment.
Although usually used in addition to an anterior cruciate
ligament reconstruction (KJT), this technique can be used on
its own, although very infrequently. The indications include
residual laxity of the lateral compartment after an isolated
reconstruction of the anterior cruciate ligament and chronic
anterior laxity in the older patient (55 years and up). The
absence of the posterior horn of the medial meniscus is a
theoretical contraindication to this procedure since the posterior horn of the medial meniscus serves at the central point
of rotation in this procedure.

Fig. 5.67 Lateral approach according to Lemaire

Positioning of the Patient
This procedure can be performed under general anesthesia or
with regional anesthesia. The patient is placed in the supine
position. A lateral vertical post is located high on the femur.
The distal lateral post allows the knee to be flexed at 30°. The
tourniquet is positioned high on the femur. A contralateral
post can be applied and allows the surgical table to be
inclined to the contralateral side.
Fig. 5.68 ITB harvest; proximal

52

Fig. 5.69 ITB harvest; distal

Fig. 5.70 ITB preparation

The graft is then prepared by removing all fatty tissues.
The proximal end is whipstitched over a minimum distance
of 2 cm using No. 5 resorbable sutures (Fig. 5.70).

Preparation of the Femoral Tunnels
The entry points for the femoral tunnel are marked. On the
femur, the anterior entry point is situated exactly at the
end of the lateral intermuscular septum on the lateral condyle. This point is easily identified by following the septum from proximally to distally, carefully lifting the

R Magnussen et al.

Fig. 5.71 LCL dissection

vastus lateralis muscle and reflecting it anteriorly together
with the suprapatella pouch using a Farabeuf or Hohmann
retractor.
Care must be taken to obtain hemostasis of the perforating and metaphyseal vessels. The knee is now flexed by
hanging the foot over the edge of the operating table. This
maneuver relaxes the posterior margin of the fascia lata.
The posterior tunnel is located exactly at the top of the triangle formed anteriorly by the lateral collateral ligament
and posteriorly by the lateral head of the gastrocnemius
muscle (see Chap. 9, Fig. 9.2). The LCL is identified easily
just below the epicondyle. If difficulty is encountered in
finding the ligament in this location, it can be traced from
the top of the fibula. Nevertheless, the best landmark for
tunnel location is the anterior border of the lateral gastrocnemius muscle.
The connective tissue and fatty tissue covering the LCL
are stripped from both sides of the ligament, providing
clear visualization. The interval between the deep surface
of the proximal two thirds of the LCL and the underlying
synovium is carefully opened with fine dissecting scissors
(Fig. 5.71). The popliteus tendon can be palpated deep to
the LCL.
The two tunnels are now made using a straight awl
(Fig. 5.72). Using “O’Shaunessy” arterial clamps, the size of
the tunnels is progressively increased. A curved suture guide
is passed through the femoral tunnel from anteriorly to
posteriorly to insert a passing suture to guide the graft
­
through the tunnel (Fig. 5.73).

5

Anterior Cruciate Ligament Reconstruction: Surgical Technique

a

53

b

Fig. 5.72 (a) Creation of the anterior femoral tunnel entry with an awl. (b) The instruments identify anterior and posterior femoral tunnel entry
points in this different case

Fig. 5.73 Passing sutures placed through the femoral tunnel

Tibial Tunnel Preparation

Fig. 5.74 Creation of the lateral/posterior entry of the tibial tunnel
with an awl

The Tibial tunnel passes under Gerdy’s tubercle. A 1 cm incision in the tibialis anterior muscle is made at the posterior
border of the tubercle, where the posterior exit of the tibial
tunnel will be located, and an entry made with an awl
(Fig. 5.74).
The anterior end of the tibial tunnel is located anterior to
Gerdy’s tubercle (Fig. 5.75). Again the entry is made with an
awl and enlarged with the “O’Shaunessy” arterial clamp.
The tibial passing suture is inserted from posteriorly to anteriorly using a curved suture passer.

Passage of the Graft
The graft is routed using passing sutures. It is first pulled
underneath the lateral collateral ligament from distal to proximal taking care not to twist it (Fig. 5.76). The graft should
remain extra-synovial when passing underneath the lateral
collateral ligament but superficial to the popliteus tendon. It
is then passed from posterior to anterior through the femoral
tunnel using the previously placed passing suture (Fig. 5.77a,

Fig. 5.75 Creation of the medial/anterior entry of the tibial tunnel

b). The graft is again passed under the lateral collateral ligament, this time from proximal to distal (Fig. 5.78). Finally it
is passed through the tibial tunnel from anterior to posterior
using the tibial tunnel passing suture.

54

Fig. 5.76 The graft is pulled underneath the LCL from distal to
proximal

a

R Magnussen et al.

Fig. 5.78 The graft is pulled underneath the LCL from proximal to
distal

b

Fig. 5.77 (a, b) The graft is passed through the tunnel femoral

The knee is now positioned near extension and in neutral
rotation, in contrast to the initial description by Lemaire
where the foot was placed in external rotation.

Fixation of the Lateral Augmentation
The graft is secured by suturing the ends together on either
side of the tibial tunnel with 2 or 3 solid sutures using
braided resorbable No. 5 suture. Distally, the two ends of
the traction suture are passed through the edges of the IT
band with a Reverdin needle to additionally secure the
graft. This completes the extra-articular tenodesis

(Fig. 5.79). The remaining IT band is closed with interrupted sutures to prevent herniation of the vastus lateralis
muscle. The rest of the incision is closed as normal. A drain
is inserted under the fascia lata.

Postoperative Care
• The patient is braced in 20° of flexion (full extension
according to M. Lemaire) for 48 h.
• Range of motion exercises of the knee are commenced on
the first postoperative day.
• Full weight-bearing is allowed.

5

55

Anterior Cruciate Ligament Reconstruction: Surgical Technique

Fig. 5.79 Completed Lemaire extra-articular tenodesis
Fig. 5.80 Direction of the osteotomy controlled by fluoroscopy

Variant of the Extra-Articular
Augmentation with a Short Graft
P Christel introduced a variant of the Lemaire technique.
A 10 cm strip of fascia lata is harvested, and the augmentation is made using this single bundle. The distal end of
the graft is left attached to the Gerdy’s tubercle and the
proximal end prepared with a whipstitch. It is passed
underneath the lateral collateral ligament, superficial to
the popliteus tendon in the same manner as the original
technique.
A passing pin is placed in the femur in the area between
the lateral collateral ligament insertion and the medial gastrocnemius insertion. A 20 mm tunnel is drilled over the
pin. The bundle is pulled into it and tensioned and fixed
with an interference screw at 20° flexion and neutral
rotation.
In cases combined with ACL reconstruction, the bundle is
introduced though the 10 mm diameter femoral tunnel. The
bundle is then tensioned by grasping the whipstitch
arthroscopically in the notch. The fixation will be ensured in
this case by the introduction and impaction of the femoral
bone block of the ACL graft from outside in.

 nterior Cruciate Ligament
A
Reconstruction with a High Tibial
Osteotomy
The indication to combine a high tibial osteotomy with a
reconstruction of the anterior cruciate ligament is pre-­arthritis
with varus alignment or associated lateral-side ­laxity. This surgical intervention combines two separate surgical procedures

that are detailed in this chapter and Chap. 17 (Valgus High
Tibial Osteotomy). Here we detail the sequence of the surgical
steps. For over 10 years, we have preferred the opening wedge
osteotomy as it allows a precise correction to be obtained.
Nevertheless, one should always be careful not to change the
tibial slope. To facilitate this, the guidance pins need to reach
the lateral tibial cortex proximal to the tibiofibular joint, so
that the osteotomy is also positioned above it (see Chap. 7).
The procedure starts with the harvest of the patellar tendon
graft followed by the preparation of the femoral and tibial tunnels. Through the same anteromedial approach, the opening
wedge osteotomy can be performed prior to the introduction of
the tendon graft (Fig. 5.80). Once the osteotomy is performed
(see Chap. 17), the desired axis correction is verified with an
image intensifier, the metal bar illustrating the mechanical axis
(Fig. 5.4). Once an adequate correction is obtained, corticalcancellous iliac bone grafts with the correct dimension are harvested. These grafts are introduced posterior to the medial
collateral ligament to help avoid an increase of tibial slope. The
osteotomy is fixed using two to three staples or a plate (Figs. 5.5
and 5.81). If a plate is used, one can avoid placing screws in the
tibial tunnel by leaving a 9 mm drill in the tunnel, and making a
visual check in the tunnel with the arthroscope. Placing the plate
posteriorly will not only help prevent increased slope but also
helps avoid this problem. Once the osteotomy is fixed, the bone
patellar tendon bone graft is introduced into the femoral and
tibial tunnels. The bone block in the tibial tunnel is fixed with a
metal wire on a post (Fig. 5.82) or with FiberWire via a cortical
bone bridge as described above. Isolated fixation on a post or
over a bone bridge in combination with an opening wedge osteotomy is insufficient. Therefore we advise augmentation of the
graft fixation with an interference screw.

56
Fig. 5.81 Postoperative
x-rays of a case fixed with
staples: (a) AP view, (b)
lateral view

R Magnussen et al.

a

b

Postoperative Care
• Flexion is limited to 120° for 45–60 days.
• Weight-bearing is only allowed after 60 days.
• Range of motion exercises are allowed immediately in the
postoperative setting.
• A brace in 20° of flexion is applied between the rehabilitation sessions.

Fig. 5.82 Tibial fixation with wire over a post

6

Anterior Cruciate Ligament
Reconstruction with Six-Strand
Hamstring Tendon Graft
S Orduna, N Darwich, and C Butcher

Introduction
The gold standard autograft, the patellar tendon autograft
(BTB), has been used for many years and is still the choice for
many surgeons. Due to the low donor site morbidity, improvements of soft tissue graft fixation techniques, and satisfactory
clinical outcome studies, we use a hamstring tendon graft. In
this chapter, we describe our current surgical technique for performing ACL reconstruction using a six-stranded autogenous,
triple gracilis, triple semitendinosus graft (TGST).

Diagnosis and Imaging
In addition to a detailed history and examination, the diagnosis is confirmed by MRI and KT 2000 arthrometer measurements comparing both knees. We look for associated
posterolateral corner or meniscal injuries and malalignment
on full-length standing radiographs, all of which may lead to
a failure of the ACL reconstruction.

Graft Choice
Graft choice includes hamstring tendons, patellar tendon, or
the quadriceps tendon. Allograft tendons are used for special
conditions.

We are pleased to include this invited chapter on ACL reconstruction
with hamstrings. We do not perform this procedure, but due to the popularity of the procedure, we include this detailed description.
S Orduna (*) · C Butcher
Healthpoint, Abu Dhabi, UAE
e-mail: s.orduna@healthpoint.ae
N Darwich
Burjeel Orthopedics and Knee Sports Medicine Centre,
Abu Dhabi, UAE

Hamstring tendon graft is our choice for the following
patients:
• Patients whose occupation, lifestyle, or religion requires
“knee walking,” crawling, or kneeling
• Patients with a history of a patellofemoral pain or patellar
tendinopathy
• Patients with open growth plates
The only absolute contraindication to the use of homolateral hamstring tendon grafts for ACL reconstruction is previous surgery done using the hamstring tendons.

Surgical Technique
Anesthesia and Positioning
Most of our patients receive regional anesthesia, especially
adductor nerve blocks for postoperative pain management. A
first-generation cephalosporin is administered intravenously.
A thigh-length anti-embolism stocking and a foam rubber
heel pad are applied to the contralateral leg. A padded pneumatic tourniquet is applied high on the thigh of the operative
leg but is rarely used during the operation. The patient is
positioned supine with a thigh support and two foot supports.
The contralateral side padded hip positioner stabilizes the
patient’s pelvis, and the padded thigh post acts as a fulcrum
to allow application of valgus force to the knee, allowing the
medial compartment to be opened for meniscus or other surgery. The lower extremity is positioned so that a full, free
range of motion can be performed during the procedure
(Figs. 6.1, 6.2, 6.3, and 6.4). Full flexion of the knee will be
required for drilling the femoral tunnel through the anteromedial portal, without risk of cartilage damage to the cartilage of the medial femoral condyle.
Surface markings are drawn for patella, patellar tendon,
anterolateral and anteromedial portals, and the hamstring
harvesting approach (Fig. 6.5).

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_6

57

58

S Orduna et al.

Fig. 6.1 Extension position

Fig. 6.4 Full flexion

Fig. 6.2 90° position

Fig. 6.5 Surface markings

Hamstring Tendon Graft Harvest

Fig. 6.3 120° position

A solution of 5 mg of morphine sulfate with 20 mL of
0.25% bupivacaine and 1:100,000 epinephrine is injected
into the suprapatellar pouch for pre-emptive analgesia. The
skin incision and subcutaneous tissues are infiltrated with a
solution of 0.25% bupivacaine with 1:100,000 epinephrine
for hemostasis and pre-emptive analgesia.

Harvest of the hamstring tendons is performed through a longitudinal skin incision centered over the tibial insertion of
the hamstrings tendons (Fig. 6.6). The landmark for the
superior border of the sartorius tendon is approximately one
finger width below the tibial tubercle or three finger widths
below the medial joint line.
The incision is positioned close to the anterior crest of the
tibia, easily allowing extension for the harvest of a patellar tendon graft in the case of premature amputation of the semitendinosus tendon graft. In revision cases where a BTB graft was used
for the primary procedure, we extend the previous patellar tendon incision 2–3 cm distal to the tibial tubercle to allow harvest
of the hamstring tendons and removal of the previous hardware.
Scissors are used for the dissection of the subcutaneous
tissues to reduce the risk of damage to the infrapatellar
branches of the saphenous nerve, avoiding the scalpel for
this part of the procedure.
The sartorius fascia is exposed by blunt dissection. While
an Allis-Adair tissue forceps is used to grab the distal

6

Anterior Cruciate Ligament Reconstruction with Six-Strand Hamstring Tendon Graft

59

Fig. 6.8 Extension of the fascial incision longitudinally with scissors

Fig. 6.6 Vertical incision

Fig. 6.7 Distal tendons grabbed with Allis-Adair tissue forceps and
incision of sartorius fascia

Fig. 6.9 Inverted L-shaped incision through the Sartorius using the
cautery

tendons, a transverse 1 cm incision is made in the fascia proximal to the superior border of the sartorius tendon (Fig. 6.7).
This is extended proximally with scissors, giving a view of
the deep aspect of the pes anserinus (Fig. 6.8).
The conjoined tibial insertion of the two tendons is
detached from the tibia by making an inverted L-shaped incision in the sartorius insertion using the cautery and scalpel
(Fig. 6.9). The sartorius fascia is grasped with an Allis clamp
and lifted away from the tibia, and the underlying medial
collateral ligament is visualized and protected (Fig. 6.10).
The tibial insertion of the two tendons is sharply released
from the crest of the tibia with scalpel.
A right-angled-type clamp is used to separate the two tendons from the undersurface of the sartorius fascial flap, which
is preserved for later closure (Fig. 6.11). The gracilis tendon
is sharply divided and grasped with a wide Allis-Adair tissue

forceps and then freed from the undersurface of the sartorius
fascia with the knife (Fig. 6.12).
Carefully the interconnecting fascial bands that run
between the two tendons are released (Fig. 6.13). Sharp or
scissors dissection along the superior border of the gracilis is
avoided to prevent injury to the saphenous nerve.
A running suture (approximately five throws) is placed in
the free end of the gracilis tendon with a No. 2 nonabsorbable suture (Fig. 6.14). While maintaining strong traction,
remaining fascial bands are released with the index finger.
A closed tendon stripper is used to harvest each tendon
(Figs. 6.15 and 6.16). The gracilis tendon is harvested first,
by flexing the knee to 90° and advancing the tendon stripper
parallel to the tendon by a slow, steady, rotating motion.
The semitendinosus tendon is harvested in a similar fashion; however, there are more extensive fascial connections

60

S Orduna et al.

Fig. 6.10 The sartorius fascia is grasped with an Allis clamp and lifted
away from the tibia

Fig. 6.12 The knife is used to free the tendon

Fig. 6.11 A right-angled-type clamp is used to separate the two
tendons

that extend from the inferior border of the semitendinosus
tendon to the medial head of the gastrocnemius. These fascial
connections must be released with the scissors to prevent premature amputation of the semitendinosus tendon.
Premature amputation of the semitendinosus tendon can
result if the tendon stripper passes away from the tendon’s
normal path. If excessive resistance is encountered in

Fig. 6.13 Gracilis and semitendinosus tendons

6

Anterior Cruciate Ligament Reconstruction with Six-Strand Hamstring Tendon Graft

attempting to advance the tendon stripper, the tension and
force on the tendon and stripper are decreased, and then the
tendon is re-tensioned before pushing the stripper again.
Rotational movements may help to advance it around the

61

tendon. A successful graft harvest typically results in graft
lengths of 20–26 cm for the gracilis and 24–30 cm for the
semitendinosus tendon.

 reparation of the Six Bundled Hamstring
P
Tendon Graft

Fig. 6.14 A running suture (approximately five throws) is placed in the
free end of the gracilis tendon

As soon as the gracilis tendon is harvested, the assistant
starts the graft preparation while the surgeon changes gloves
and completes the semitendinosus tendon harvesting
(Fig. 6.17).
The grafts are placed on a preparation board (Smith and
Nephew Graftmaster III), and residual muscle fibers on the
proximal end of both tendons are removed by blunt dissection with a metal ruler, a large curette, or one arm of a sharp
scissors (Fig. 6.18).
The proximal end of each tendon is tubularized with a
continuous No. 2 nonabsorbable suture, and the sutures are
tensioned (Fig. 6.19).
The tubularized proximal ends of each graft are sutured to
the center of a polyester tape (Fig. 6.20).
The EndoButton device is inserted into its clamp and the
whipstitches from the distal end of the gracilis, and the semitendinosus tendons are tied to the loop (Figs. 6.21, 6.22,
and 6.23).

Fig. 6.15 Closed tendon stripper applied to the gracilis
Fig. 6.17 The tendon laid out on a preparation board

Fig. 6.16 Harvesting the tendon

Fig. 6.18 The residual muscle fibers are removed

62

S Orduna et al.

Fig. 6.19 The proximal end of each tendon is tubularized

Fig. 6.22 Whipstiches from the proximal end of the gracilis and semitendnosus tendons are tied to the EndoButton loop

Fig. 6.20 The distal ends of each graft are sutured to the center of a
polyester tape

Fig. 6.23 The proximal ends of both tendons have now been sutured to
the EndoButton loop

Fig. 6.21 EndoButton device in the Smith and Nephew clamp

The proximal end of each tendon, with the polyester tape,
is passed through the endobutton loop, creating a loop in the
graft (Fig. 6.24). Then by passing one limb of the polyester
tape on either side of the loop in the graft, a triple-strand
graft is created (Figs. 6.25 and 6.26). The tape and the suture
from the tubularized graft are then tied to the loop (Fig. 6.27).
The same process is repeated for the semitendinosus tendon (Figs. 6.28 and 6.29).

Fig. 6.24 Passing the distal end of the tendon through the EndoButton
loop creates a loop in the graft (shown here held in the surgeon’s
index finger)

6

Anterior Cruciate Ligament Reconstruction with Six-Strand Hamstring Tendon Graft

63

Fig. 6.25 Passing one limb of the polyester tape either side of the graft
loop

Fig. 6.28 Polyester tape being passed around the graft loop of semitendinosus tendon (above), and completed gracilis tendon (below)

Fig. 6.26 Triple-bundle gracilis tendon

Fig. 6.29 Triple semitendinosus and gracilis grafts completed

Fig. 6.27 Tying the tape and suture to the tendon loop

Fig. 6.30 The diameter of the graft is measured

The diameter of the TGST graft is measured with the
0.5 mm incremental sizing block or sizing tubes (Fig. 6.30).
The graft is covered with an antibiotic-soaked swab (vancomycin) and pretensioned at 15–20 pounds on the graft preparation board for the remainder of the procedure (Fig. 6.31).

After pretensionning, the six tendon strands are sutured
together for 30 mm proximally and 20 mm distally with No. 2
nonabsorbable suture, in order to obtain better fixation with
the rigid fix pins in femoral tunnel and the interference screw
in the tibial tunnel (Fig. 6.32). After measurement of the fem-

64

S Orduna et al.

Fig. 6.31 Pretensionning of the graft
Fig. 6.33 Dilatation of the anteromedial portal with scissors

Fig. 6.32 The six strands are sutured together for 30 mm proximally
and 20 mm distally

oral tunnel, the graft will be marked at the appropriate levels,
to aid appropriate positioning of the graft. The polyester tapes
facilitate strong traction during tensioning and fixation and
can be used for additional tibial fixation with a staple.

Arthroscopic Portal Placement
We use two, or possibly three, portals (Fig. 6.5). A routine
high anterolateral portal at the level of the inferior pole of the
patella and adjacent to the lateral border of the patellar t­ endon
is used for viewing. This portal gives a frontal view of the
femoral attachment site of the ACL and is helpful in determining the clock orientation and the anatomic placement of
the femoral tunnel. An anteromedial portal at the level of the
inferior pole of the patella adjacent to the medial border of the
patella tendon is used for instrumentation and viewing of the
medial wall of the lateral femoral condyle if necessary.
We may extend the anteromedial portal distally for a few
millimeters for drilling the femoral tunnel, but an accessory
medial portal located inferior to the anteromedial portal at
the level of the medial joint line is occasionally used if access
to the femoral footprint is difficult. The location for the
accessory medial portal is made by an 18-gauge spinal needle. This portal is located as low as possible, just above the
medial joint line avoiding damage to the medial meniscus.
Placement of the portal too medially produces a shorter
femoral tunnel and risks injury to the medial femoral condyle
by the endoscopic drill. Dilatation of the portal with the blunt

Fig. 6.34 Radiofrequency pencil from anteromedial portal removing
remaining ACL from the lateral wall

arthroscope obturator followed by the Metzenbaum scissors
helps ease future passage of instrumentation (Fig. 6.33).
We will change the scope from anterolateral to anteromedial portals to look for previously undetected injuries, for
example, capsulomeniscal ramp lesion injuries, meniscal
root detachment, or osteochondral injuries.

Notch Preparation
After a routine diagnostic arthroscopy and treatment of any
associated pathology, we start with the intercondylar notch
preparation. The torn fibers of the ACL are removed from the
lateral femoral condyle and the tibial attachment site by a
motorized shaver, electrocautery pencil, or radiofrequency
probe (Fig. 6.34). The radiofrequency probe is preferred to
the shaver blade in order to coagulate and to remove the soft
tissue along the lateral wall of the interondylar notch without

6

Anterior Cruciate Ligament Reconstruction with Six-Strand Hamstring Tendon Graft

damage to the bony anatomy. We feel it is not necessary to
remove all the remaining fibers as they may have a biological
role to play in proprioception.
Use of the anteromedial portal technique allows the femoral tunnel to be positioned freely, and more posteriorly (lower
down) on the sidewall of the lateral femoral condyle than in
the past. This results in a more horizontal orientation of the
ACL graft, which may avoid posterior cruciate ligament
impingement, and in most cases eliminates the need for a
notchplasty. However, a selective notchplasty may be
required in the case of congenitally narrowed notches, more
frequent in female or in chronic cases with notch stenosis
due to the development of notch osteophytes.

65

priate sized offset femoral aimer is passed through the medial
portal; e.g. 7 mm offset in order to perform an 8 mm tunnel
whilst preserving a 3 mm posterior tunnel wall. The posterior
blade of the femoral offset aimer is placed in the over-the-top
position, and the knee is slowly flexed to 120° (Fig. 6.36a, b).
A 2.7-mm drill-tipped guide pin is positioned at the site of
the microfracture awl penetration mark and drilled out

Femoral Tunnel
We aim for a femoral tunnel within the anterior medial bundle footprint, posterior to the intercondylar ridge, and proximal/deep to the bifurcate (resident’s) ridge (see Fig. 5.24,
Chap. 5), leaving a posterior wall of 3 mm. A microfracture
awl is passed through the medial portal and used to make the
starting point (Fig. 6.35).
Verification of the correct starting point can be made by
swapping the camera to the anteromedial portal. An appro-

a

Fig. 6.35 A microfracture awl is passed through the accessory medial
portal and used to make the starting point or footprint for the femoral
tunnel

b

Fig. 6.36 (a) Femoral aimer entrance and view on screen in full flexion position knee. (b) Femoral aimer is placed in the over-the-top position

66

S Orduna et al.

Fig. 6.37 The 2.7-mm drill-tipped guidewire is placed at the previously marked entry point

Fig. 6.39 Measurement of femoral tunnel length

Fig. 6.38 4.5-mm EndoButton drilling through the femoral cortex

through the soft tissues of the lateral thigh (Fig. 6.37).
Inadequate knee flexion can result in the guide pins coming
to lie posterior to the intermuscular septum, placing the peroneal nerve at risk.
A 4.5-mm EndoButton drill (Smith and Nephew
Endoscopy) is used to drill a tunnel through the lateral femoral cortex (Fig. 6.38), and the tunnel length is measured with
a depth gauge (Fig. 6.39).
Tunnel drilling then proceeds progressively in 0.5 mm
increments from 7 mm to the final size, which is equal to the
measured proximal graft diameter. The femoral socket depth
must allow for the length of the TGST graft (usually
25–30 mm) plus the extra 6 mm to allow the EndoButton to
clear the lateral femoral cortex and to flip across the aperture.
The articular edge of the femoral tunnel is smoothed with a
rasp. The tunnel is then visualized arthroscopically, and the
debris is removed from the knee with the shaver.
Next, the Rigidfix guide is introduced, and the two
cross pin femoral tunnels are drilled from the medial side
of the knee (Figs. 6.40 and 6.41). Their correct position is

Fig. 6.40 Introduction of Rigidfix femoral guide into the femoral tunnel

Fig. 6.41 Addition of the curved guide, and drilling of the femoral
cross pin tunnels from the medial side of the knee

6

Anterior Cruciate Ligament Reconstruction with Six-Strand Hamstring Tendon Graft

67

Fig. 6.44 The adjustable tibial aimer is set between 50° and 55°
Fig. 6.42 Arthroscopic confirmation of the position of one of the
Rigidfix tunnels

Fig. 6.43 Fluid emerging from the Rigidfix pins confirms that they are
in the tunnel

Fig. 6.45 Intra-articular position of the tibial aimer

verified arthroscopically for each pin tunnel in turn
(Fig. 6.42) and by the presence of fluid from the cannulated pins (Fig. 6.43).
The ends of a No. 5 nonabsorbable passing suture are
inserted into the eyelet of a passing pin which is introduced
into the femoral tunnel and brought out of the lateral thigh.
The loop of the suture is positioned at the entrance of the
femoral tunnel and will be used later in the procedure to pass
the hamstring graft (see Fig. 6.45).

nence (Fig. 6.45). In relation to the PCL, it is situated 7 mm
anteriorly, with around 2 mm between the ACL and PCL.
The guidewire is drilled into the joint, and once the position
is confirmed as satisfactory, the tunnel is progressively reamed
in increments from 5 mm to the final size, which is equal to the
measured distal graft diameter (Figs. 6.46 and 6.47). The intraarticular aperture of the tunnel is smoothed with a rasp, and the
soft tissue around the external aperture is cleared with an electrocautery pencil and a Cobb periosteal elevator.

Tibial Tunnel

Graft Fixation

The adjustable tibial aimer is introduced into the knee via the
anterior medial portal. It is set between 50° and 55° which
produces a sufficiently long tunnel, between 45 and 55 mm
length (Fig. 6.44).
The intended intra-articular position of the guide pin is at
the level of the posterior edge of the anterior horn of the lateral meniscus and between the medial and lateral tibial emi-

There are an increasing number of choices of graft fixation,
all with their advantages and disadvantages. Healing of hamstring grafts to bone is a long process, and on the femur, we
prefer to use a combination of EndoButton CL and Rigidfix
cross pins, as they both provide 360° contact between the
graft and the bone and have been shown to be effective in
resisting slippage under cyclic loading.

68

Fig. 6.46 The guidewire is drilled into the joint

S Orduna et al.

total tunnel length. For instance, if the femoral tunnel
length measures 48 mm, and 30 mm of graft has been chosen to be inserted into the femoral tunnel, the distance
available for the continuous loop is 18 mm. The next shorter
loop available will be chosen, to increase the stiffness of
the femur-EndoButton CL-TGST graft complex, in this
case a 15 mm loop. The pretensioned graft is marked with
a surgical marking pen at the measured femoral tunnel
length (e.g., 48 mm). A full-length No. 2 flipping suture
and a No. 5 passing suture are passed through the end holes
of the EndoButton. A second No. 5 suture can be inserted
into the same hole as the No. 2 flipping suture and passed
alongside the graft and out of the tibial tunnel which will
aid in the removal of the graft if there are any difficulties
later in the procedure.

Graft Passage and Femoral Fixation

Fig. 6.47 A Kocher holds the guidewire during tibial reaming to prevent it from passing into the joint

The lower bone mineral density of the proximal tibia is
the main cause of concern for tibial fixation. The tibial fixation devices must resist shear forces applied parallel to the
axis of the tibial bone tunnel. Intratunnel tibial fixation with
interference screws seems to demonstrate high initial fixation strength and stiffness with minimal slippage under
cyclic loading conditions.
We prefer intratunnel tibial fixation with a bioabsorbable
interference screw, but we do not hesitate to add additional
cortical fixation with a staple if the fixation does not appear
adequate.

 alculation of EndoButton CL Length and Final
C
Graft Preparation
The required EndoButton loop length is decided by subtracting the intended graft length in the tunnel from the

The loop of No. 5 suture previously placed in the femoral
tunnel is retrieved arthroscopically and pulled out of the tibial tunnel. This is used to pass the FiberWire No. 2 flipping
suture and other No. 5 passing suture out of the lateral thigh.
Under arthroscopic visualization, the EndoButton and the
attached hamstring tendon graft are passed across the joint
and into the femoral socket by use of the passing suture
(Fig. 6.48). The previously placed insertion mark guides the
passage of the graft 6 mm deeper, to allow the EndoButton to
pass outside the lateral femoral cortex and flip. Correct
deployment is verified by rocking the EndoButton against
the lateral femoral cortex and subsequent tensioning of the
graft (Figs. 6.49 and 6.50). When this tension is applied, the
previously placed mark at the insertion length will appear at
the aperture of the femoral tunnel.
If any doubts exist about secure deployment of the
EndoButton, fluoroscopy can be used to check the position
of the EndoButton. It should be placed over the cortical
bone.
Once the EndoButton is flipped, the two Rigidfix cross
pins are introduced from the medial side of the femur.
Tension on the graft is maintained during this procedure, at
90° of flexion (Figs. 6.51 and 6.52).

Graft Tensioning
Equal tension is applied to both ends of the six-stranded
hamstring tendon graft. The knee is cycled from 0 to 90° for
a minimum of 30 cycles to allow the EndoButton CL to settle
on the femoral cortex and remove creep from the graft construct. We fix the graft with the knee positioned between 0
and 20° of flexion, ensuring full extension is not limited by
the graft isometry.

6

Anterior Cruciate Ligament Reconstruction with Six-Strand Hamstring Tendon Graft

69

Fig. 6.48 EndoButton and graft passing through tibial and femoral
tunnels

Fig. 6.51 Tension of the graft is confirmed arthroscopically before
cross pin fixation

Fig. 6.49 Confirming that the EndoButton is flipped

Fig. 6.52 The two Rigidfix cross pins are introduced from the medial
side of the femur

Fig. 6.50 Tension is applied to the graft in order to confirm EndoButton
fixation

Fig. 6.53 A bioabsorbable interference screw for tibial fixation

70

S Orduna et al.

No. 0 absorbable suture. The subcutaneous tissue is closed in
layers with fine absorbable sutures.

Postoperative Care

Fig. 6.54 Final assessment of the ACL graft

After isolated ACLR full weight-bearing is permitted, a
range of motion brace and crutches are supplied, and knee
movement is limited to 0–90° for 4 weeks. If there has been
a meniscal repair or osteochondral procedure, non-weight-­
bearing is continued for 4 weeks.
Prophylaxis of thrombosis is by early mobilization, anti-­
embolism stockings, and a low molecular weight heparin
daily for 2 weeks.
The patient is seen at 7–10 days for suture removal and
postoperative radiographs.

Tibial Fixation

Complications

The bioabsorbable interference screw is our choice for tibial
fixation. The direction of the tibial tunnel is identified by
passing a 1.1 mm guidewire.
A bioabsorbable screw 1 mm larger than the tibial tunnel
diameter is inserted until flush with the cortex (Fig. 6.53). If
there was inadequate torque during the insertion of the
tapered screw or if the patient has soft bone, supplemental
tibial fixation with a soft tissue staple is performed.
The stability and range of motion of the knee are checked.
It is important to verify that the patient has full range of
motion before leaving the operating room. The arthroscope
is inserted to the knee, and graft tension and impingement
are assessed (Fig. 6.54).
After confirmation that the patient has a full range of
motion and negative Lachman, anterior drawer and pivot
shift tests, the passing and flipping sutures are cut close to
the skin and pulled out of the lateral thigh.

The risk of complications such as infection, deep venous
thrombosis, and loss of motion are the same as for ACL
reconstructions performed with other graft types. However,
we are unaware of reports of extensor mechanism rupture or
patellar fracture after ACL reconstruction performed with
hamstring tendon grafts. Complications unique to hamstring
tendon grafts include premature amputation of the hamstring
tendons, saphenous nerve injury, bleeding at the hamstring
tendon harvest site, and hamstring muscle strains in the postoperative rehabilitation period.
The risk of premature amputation of the tendons can be
minimized by following the recommendations outlined in
the section on graft harvest. If the gracilis tendon is amputated and the semitendinosus is successfully harvested, it is
possible in most cases either to triple or quadruple the
­semitendinosus tendon, depending on its length. In these
situations, the EndoButton CL can still be used for femoral
fixation; however, because of the shorter length of the graft
construct, alternative tibial fixation is obtained by tying the
polyester tape around a fixation post or an extra-small nonbarbed staple. If necessary, this tibial fixation can still be
augmented with a 25–30 mm bioabsorbable screw. If the
semitendinosus tendon is amputated, it will be necessary to
use an alternative autograft (such as the patellar tendon or
quadriceps tendon), or allograft tissue (if available and preoperative consent has been obtained).

Closure
A closed suction drain is inserted under the sartorius fascia
up into the hamstring harvest site to help prevent postoperative hematoma formation and medial ecchymosis. This is
removed after 24 h. The sartorius fascia that was preserved
during the graft harvest is repaired back to the tibia with a

7

Revision Anterior Cruciate Ligament
Reconstruction
R Magnussen, G Demey, P Neyret,
and C Butcher

Introduction
Revision anterior cruciate ligament (ACL) reconstruction is
increasingly common. These procedures must be carefully planned
and are often fraught with technical difficulties. The surgeon must
address the following clinical questions prior to surgery:
––
––
––
––

How are the previous tunnels positioned?
Which graft should be used?
Is a one-stage or two-stage reconstruction required?
What graft fixation will be utilized?

To answer the questions, the cause(s) of failure of the
prior ACL reconstruction should be identified. The answers
to these critical questions will guide whether revision surgery is indicated, and if so, what technique is best. The indication will depend not only on these technical issues but on
the expected functional results. There may be disparity
between the expectations of the surgeon, and those of the
patient, which may encourage conservative treatment.

Causes of Failure
It is important to obtain a detailed clinical history, including
the initial injury mechanism and information about the prior
reconstruction such as graft type, surgical technique, and
intraoperative findings including meniscal and articular carR Magnussen
Centre Albert Trillat, Lyon, France
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France

tilage status. One should also determine the postoperative
rehabilitation protocol, the time to return to sport, and any
subsequent surgical procedures such as the resection of the
Cyclops lesion or subsequent meniscal tear.

Technical Error
Technical error is the most common cause of recurrent instability following ACL reconstruction.

Incorrect Tunnel Position
This error is by far the most common. The positions of tibial and
femoral tunnels can be evaluated on plain radiographs (see analysis of the causes of failure) and computed tomography (CT)
scans. Our experience has been that three-­
dimensional CT
reconstructions are quite useful in evaluating tunnel position.
Errors in femoral tunnel position are more common than
those involving the tibia, but both can be present. On the
femur, tunnels are often too anterior (Fig. 7.1), leading to
impingement in the notch and a loss of extension. Placement
of the femoral tunnel too far posterior can lead to graft laxity
in flexion or excessive tension in extension. Vertical positioning in the notch is also common, leading to poorer control of tibial rotation.
In the tibia, the tunnel may be too far posterior, leading to
a vertical graft that poorly controls anterior translation
(Fig. 7.2), or too far anterior, leading to impingement of the
graft in the notch with extension. Similarly, lateral tunnel
placement can lead to impingement of the graft on the medial
border of the lateral femoral condyle, possibly leading to
abrasion and graft rupture.

P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_7

71

72

Fig. 7.1 Lateral plain radiograph demonstrating anterior femoral tunnel malposition

Poor Fixation
Graft fixation may be inadequate or insufficient. This problem can occur when there is inadequate contact between an
interference screw and the bone block of a patellar tendon
graft. Advancing the screw too far or not far enough can lead
to this problem, as can divergence of the screw and graft.
Additionally, poor bone quality may diminish the fixation
strength of an appropriately placed screw. This problem
commonly occurs in the cancellous bone of the tibia, which
is why we prefer to utilize double fixation in the tibia.
 oor Graft Quality
P
Use of a graft that is too small or of poor quality can lead to
early reconstruction failure. The quality of allograft tissue is
variable and highly dependent on the sterilization process.

R Magnussen et al.

Fig. 7.2 Lateral stress radiograph demonstrating anterior tibial translation that is poorly controlled by a vertical graft with a posterior tibial
tunnel

Associated Lesions
Lesions of the posterolateral or posteromedial corner (including ramp lesions, meniscal root tears) that are not addressed
at the time of ACL reconstruction frequently lead to persistent postoperative instability and failure. Of specific interest
is whether a medial meniscectomy was required. Excision of
significant portions of the medial meniscus can lead to
increased stress on the graft and contribute to failure.
Excessive posterior tibial slope (α > 13°) can also contribute
to increased graft stress and failure (Fig. 7.3). A narrow
notch is also sometimes encountered, which may lead to
graft impingement and attrition.

7

Revision Anterior Cruciate Ligament Reconstruction

73

Clinical Examination
In addition to signs of anterior laxity, one should identify the
scars from previous surgeries, palpate the patella and tibia to
identify any bone loss if a prior patellar tendon graft was taken,
search for any additional instability (especially posterolateral or
posteromedial corner), and evaluate for excessive recurvatum,
varus or valgus alignment, or diffuse ligamentous laxity.

Review of the Initial Operative Report
Critical information includes the type of graft and fixation
that were used as well as any information regarding intraoperative complications or challenges.

α
X

X

Radiographic Examination
One should obtain:

Fig. 7.3 Drawing of a lateral view of a proximal tibia demonstrating
measurement of the posterior tibial slope based on the angle (α) between
the long axis of the tibia and medial joint line

Re-traumatic Rupture
True traumatic graft ruptures are rare causes of failure and
are ultimately a diagnosis of exclusion. To affirm this
cause, we prefer to have a documented examination objectively describing control of the anterior laxity by the previous reconstruction and a significant traumatic re-injury.
Re-injury is rarely caused by trivial injuries and is often
accompanied by a “pop” and hemarthrosis. Patients often
attribute failure of their prior reconstruction to a trivial
trauma, and it is critical to rule out other etiologies of failure so the mistakes of previous surgeries are not repeated
at revision.

Biologic Failure
Biologic failure has been described as failure of the ACL graft
tissue to revascularize and undergo ligamentization. The ligamentization process is significantly slower and often incomplete in allografts, making them more susceptible to this mode
of failure. This etiology is again a diagnosis of exclusion.

Analysis of Failure Causes
The preoperative evaluation should be complete.

–– Comparative anteroposterior and lateral radiographs of
the knee at 30° of flexion in a single-leg stance.
These views allow for the evaluation of tunnel position in
both the coronal and sagittal planes. Additionally, they
allow for the assessment of tunnel enlargement.
–– Objective radiographic measurements of anterior laxity
(anterior tibial translation with differential Telos).
These views allow quantitative evaluation of anterior laxity and comparison to the contralateral side. One can prefer Telos stress device assessment.
–– Bilateral standing anteroposterior radiographs in 45° of
flexion (Schuss view).
This view is the most sensitive for the detection of tibiofemoral osteoarthritis.
–– Axial patellar view in 30° of flexion.
This view detects patellofemoral arthritis and patellar
maltracking.
–– Long cassette views of the lower extremities if malalignment is suspected clinically.
This view determines the mechanical axis and can influence decision-making regarding the need for associated
osteotomy.

CT with 3D Reconstructions
This study is essential. It not only gives precise information on
the positioning of the tunnels but also quantifies bone defects
that are often poorly assessed on plain radiographs. Analysis of
the axial cuts is particularly useful in evaluating the relationship
of the femoral tunnel with the notch (in terms of orientation,

74

R Magnussen et al.

Fig. 7.4 Axial CT image demonstrating ideal positioning of ACL femoral tunnel in this plane. Note the tunnel is centered on the cut in which
the notch resembles a Roman arch

positioning, and filling). Ideal positioning is indicated when the
femoral tunnel is visible on the cut in which the notch forms the
shape of a Roman arch (Fig. 7.4). The axial cuts also allow evaluation of graft fixation, the position of the interference screw,
and its resorption. A complete description of the femoral tunnel
position requires information from both the axial and sagittal
cuts, making its analysis difficult.
On the tibial side, the axial cuts allow analysis of the position of the tibial tunnel. Sagittal plane analysis details tunnel
orientation and graft fixation. Again, evaluation of the exact
point of entry of the tunnel into the joint is complex and
requires the use of data from several cuts.
The diameters of the tunnels and any other bony defects
can be easily assessed with CT images. Bone loss near the
entry site of the tunnel into the joint is especially important
(Fig. 7.5). In the case of a bony defect greater than 15 mm,
there is a risk of a windshield-wiper effect and subsequent
graft loosening. Bone graft fillers may be utilized in these
cases (see “Surgical Technique” section). In case of revision
following failed double-bundle surgery, this type of tunnel
enlargement is frequently seen.
The 3D reconstruction allows one to integrate information from numerous planes into a single image and is the
single most important aid in understanding tunnel position.
While quantitative analysis is difficult, the general graft position and tunnel orientation can be qualitatively assessed
(Fig. 7.6). These views provide an accurate preview of the
view of the notch that will be encountered intraoperatively
and are indispensable planning tools.

Fig. 7.5 A sagittal CT image of a knee demonstrating enlargement of
the ACL tibial tunnel

Fig. 7.6 A 3D CT reconstruction of part of a distal right femur viewed
from distally. One can visualize the prior femoral tunnel and note its
improper vertical position in the femoral notch

7

Revision Anterior Cruciate Ligament Reconstruction

MRI
Allows evaluation of the appearance of the prior graft and determination of the status of the menisci and articular cartilage. It is
quite useful in diagnosis of graft failure and other predictors of
outcome but is less useful than CT for preoperative planning.

Surgical Technique

75

the notch (Fig. 7.7). A limited notchplasty with a small
osteotome may be helpful in a pre-osteoarthritic knee to aid
in visualization. It is imperative to fully visualize the posterior part of the lateral wall of the notch in order to ensure
appropriate femoral tunnel placement (Fig. 7.8).
Fat pad resection is generally minimized, but some resection is required to allow accurate tibial tunnel placement
(Fig. 7.9).

General
Our preferred technique for revision ACL is nearly identical to
that for primary ACL reconstruction (see Chap. 5). The primary
difference occurs when repeat harvest of the ipsilateral patellar
tendon is not possible, requiring contralateral harvest. Other
considerations include the impact of previous tunnels and bone
loss on placement of the tunnels for the revision surgery.
An image intensifier can be quite useful in case of removal
of retained hardware or if an associated osteotomy is
performed.
Physical examination is repeated under anesthesia to
assess the degree of anterior laxity and detect any associated
instability.

Choice of Graft
Patellar tendon autograft is our preferred graft for revision
Fig. 7.7 An arthroscopic view of the notch in a right knee demonACL reconstruction in order to attain bone-to-bone fixation. In strates a vertical ACL graft with the tibial tunnel placed too far
addition, bone blocks can help fill an expanded tunnel if tun- posterior
nels from the prior surgery are to be reused and allow adjustment of the position of the graft by rotation of the bone block.
We consider re-harvest of an ipsilateral patellar tendon to
be possible 18 months after prior harvest. When re-­
harvesting, the scar must be enlarged frequently.
When ipsilateral quadriceps tendon harvesting or
ipsilateral re-harvest of patellar tendon are impossible
due to significant bone loss of either the patella or tibia,
we propose a contralateral harvesting of the quadriceps
tendon. It provides a broad, thick tendon with a bone
block. This decision must be made preoperatively in
order to inform the patient and to drape appropriately.
Reconstruction using hamstrings may be possible in
selected circumstances.

Joint Exploration
Anterolateral and anteromedial portals are made. A systematic assessment of the joint should be performed evaluating
all articular cartilage surfaces and the menisci. The shaver is
used to clear fat and scar tissues and achieve a clear view of

Fig. 7.8 An arthroscopic view of the lateral wall of the notch in a right
knee with clear visualization of the posterior portion of the lateral femoral condyle. The prior femoral tunnel is clearly visible in a too anterior/higher and shallow position (circle)

76

Fig. 7.9 Arthroscopic view of the tibial surface following resection of
fat and scar. Note the too posterior tibial attachment point of the prior
graft (circle)

R Magnussen et al.

Fig. 7.10 Arthroscopic view demonstrating over-drilling of the guide
pin with a 9 mm drill. Note the use of a curette to prevent guide pin
advancement during drilling

Tunnel Placement
There are two scenarios: the tunnels from the initial reconstruction are correctly positioned or there is an error in the
position of one or both.

Previous Tunnels Are Correctly Positioned
Removal of Hardware
When the original position of the tunnels is correct, the previous hardware is often an obstacle to tunnel preparation and
placement of the new graft. If the original hardware was
metal and intraosseous (such as a metal interference screw),
it must be removed. It is therefore important to have the
appropriate screwdriver available. This information should
be gleaned from the original operative report.
In the tibia, we find it useful to remove all hardware,
including any cortical fixation (staples or screws) in addition
to intraosseous hardware. On the femoral side especially,
fluoroscopy can be useful in identifying and extracting hardware that has become overgrown by bone.
Drilling of the Tunnels
If the original tunnels are well placed, they are frequently
reusable. It is often sufficient to drill a second time through
the same tunnel at the desired diameter, especially in the
tibia.
The tibial tunnel is created as in the prior ACL reconstruction by first placing a guide wire through the old tunnel, then
over-drilling with a drill, the diameter of which is equal to
the desired tunnel size (usually 9 mm) (Fig. 7.10). Care must

Fig. 7.11 View with the arthroscope through the tibial tunnel demonstrating its clean appearance after drilling and removal of scar tissue

then be taken to clean the tunnel with a curette and/or shaver
to remove any residual material (absorbable fixation, etc.)
still in the tunnel (Fig. 7.11).
If tunnel expansion is demonstrated in the preoperative
workup, this finding may affect graft choice. The size of the
bone block can be enlarged to a point to deal with this problem. Tunnel enlargement near the joint surface may potentially affect the graft position, which must be carefully
monitored. However, as the enlargement usually affects the
cancellous bone, achieving a solid primary fixation is a
difficulty.

7

Revision Anterior Cruciate Ligament Reconstruction

Backup fixation is routinely used on the anterior tibia.
The cortical fixation ensures appropriate graft tension and
minimizes stress on the primary fixation. The cortical fixation is performed with a FiberWire loop passed through the
bone block graft and tied over a bone bridge on the tibial
tuberosity. Primary fixation is generally achieved with an
interference screw 9 or 11 mm in diameter. If significant tunnel expansion has occurred, this fixation may not be sufficient. In this case, a useful trick is to place a second
interference screw to augment the first. This second screw
will both aid in fixation and help fill the area of osteolysis.
The femoral tunnel is generally performed using an
“outside-­in” technique with a standard drill guide and guide
pin. Often the prior femoral tunnel was made by the “allinside” technique and cannot be easily recreated using our
preferred “outside-in” technique. In this case, the tunnel is
drilled in the standard outside-in manner and may intersect
the previous tunnel near the notch. As fixation is achieved
on the femoral cortex using a press-fit technique as with a
primary reconstruction, this intersection has no effect on
fixation.

Incorrect Prior Tunnel Position
Hardware
As above, hardware must be removed if it will interfere with
the creation of the new tunnels. However, it may be difficult
to remove the hardware in the femur. In the case of poor
positioning of the femoral tunnel, it may be possible to leave
the old hardware in place. Unnecessary removal of hardware
may weaken the bone or lead to enlargement of bony defects
and should be avoided.

a

77

Tunnels
Malposition of the Tibial Tunnel

It is easy to drill a new tunnel in anatomic position if the
initial tibial tunnel is very poorly positioned. Prior hardware
can be ignored, and the new tunnel can then be drilled in the
usual manner.
In contrast, if the previous tunnel was only slightly offset
from the ideal position, particularly if the previous tunnel was
too posterior, independent tunnel entry into the joint cannot be
achieved. The resulting tunnel is then very large, complicating
both accurate graft positioning and fixation. One can compromise tibial tunnel position a bit without affecting outcome, but
this solution has its limits. In case of excessive enlargement of
an already malpositioned tibial tunnel, consideration should
be given to a two-stage reconstruction (see below).
In practice, it seems much easier to correct a tibial tunnel that is placed far too anterior and much harder to deal
with a tunnel placed too far posterior. Too far lateral tibial
tunnels can also be observed, although this malpositioning
is generally small and correctable by placing the interference screw on the lateral side of the new graft and/or rotating the bone block.
Again, if the original hardware proves impossible to
remove, or removal of the initial hardware will lead to a very
large tunnel opening, it may be best to leave it in place.
Malposition of the Femoral Tunnel

If the femoral tunnel is poorly positioned (commonly noted
to be vertical in the notch), it is quite easy to drill a new
tunnel in the correct position on the lateral wall using the
“outside-­in” technique (Fig. 7.12a, b). This type of tunnel

b

Fig. 7.12 Arthroscopic view of a new femoral tunnel being created. (a) The curette holds the guide pin. The prior femoral tunnel can be visualized
higher/more anteriorly in the notch. (b) The guide pin is over-drilled in a lower/more posterior and deeper position in the notch

78

R Magnussen et al.

Fig. 7.13 Axial CT cut of the distal femur of a left knee. The previous
vertical femoral socket is seen as well as the new femoral tunnel drilled
with the “outside-in” technique

placement generally completely avoids any intersection
with the old tunnel (Fig. 7.13). Because the femoral bone
block is fixed in the lateral cortex and the lateral part of the
condyle, fixation will be solid even if the aperture is
enlarged.
When tunnels are only slightly malpositioned, it is easier
to correct a femoral tunnel placed too far posterior and harder
to correct a femoral tunnel placed too far anterior. The drilling of a second femoral tunnel in these cases may lead to
increased risk for femoral fracture. A two-stage reconstruction may be indicated (see below). We rarely consider an
over the top positioning of the graft, but this option can be
very useful.

Fixation
When the prior tunnels are used, the fixation of a new graft
can also be performed in the usual way by interference
screws and a cortical backup on the tibia.
In the case of a bone defect or poor bone quality, we recommend the use of two screws in the same tunnel associated
with cortical backup. This can be achieved with a FiberWire®
loop through bone tunnels on the anterior tibia or use of a
wire through the graft bone block and around a screw with a
washer (Fig. 7.14).

Fig. 7.14 Intraoperative view of double tibial fixation. This is achieved
with a wire through the bone block around a screw/washer and an interference screw

Two-Stage Reconstruction
When there is a significant bone loss that may compromise
fixation and positioning of the new graft, a two-stage reconstruction is indicated. The first stage includes removal of the
prior graft and hardware followed by bone grafting of the
tunnels. The iliac crest should be prepped into the operative
field. The previous graft is then completely excised using a
shaver and/or basket. The tunnels are cleaned, fibrosis
excised, and the previous hardware removed. Fluoroscopy
may be useful for locating intraosseous hardware. The
cleaned tunnels can be grafted with cancellous bone from the
anterior iliac crest. ACL reconstruction is then performed
3–6 months later.
This procedure is rarely performed in our practice and
should be considered in extreme cases including severe tunnel enlargement or failed double-bundle reconstruction in
which the two tunnels have eroded into one large defect
(Fig. 7.15a–g). The exception is the case of slightly posterior
tibial tunnel or slightly anterior femoral tunnel. In these
cases, attempts to correct the tunnel position will likely lead
to an enlarged entry site into the joint and placement of the
graft in the same position as in the prior reconstruction. In

7

Revision Anterior Cruciate Ligament Reconstruction

79

a

b

c

d

e

f

Fig. 7.15 (a) Sagittal CT image demonstrating tunnel enlargement and
coalescence following a double-bundle reconstruction. A two-stage
reconstruction is indicated. (b) Iliac crest bone harvest. (c) Debridement

of the femoral tunnel. (d) and (e) Femoral tunnel grafting with cancellous bone. (f) Cancellous graft seen in the tibial tunnel. (g) Postoperative
radiographs showing bone grafts in situ

80

R Magnussen et al.

these cases, we recommend a two-stage reconstruction even
in the absence of tunnel enlargement.

g

Combined Procedures
In specific clinical situations, revision ACL reconstruction
can be combined with additional procedures to improve the
odds of successful outcome or address associated pathology.
Note that these combined procedures can also be discussed
in primary evolved anterior chronic laxity, or where there is
a medial compartment femorotibial pre-osteoarthritis.

 CL Reconstruction and Valgus-Producing
A
High Tibial Osteotomy

Fig. 7.15 (continued)

a

The addition of a valgus-producing high tibial osteotomy is
indicated in the presence of early medial tibiofemoral
­arthritis or in cases with significant genu varum, especially
associated with a lesion of the posterolateral corner ligament
complex. In case of significant isolated genu varum (tibial in
most cases), the osteotomy is designed to protect the graft as
increased stress in the medial compartment likely contributed to the failure of the initial graft (Fig. 7.16a, b). We consider genu varum to be significant when the hip-knee-ankle
angle exceeds 8° of varus. In cases of an associated injury to
the posterolateral corner, the osteotomy will serve to protect
b

Fig. 7.16 Anteroposterior (a) and lateral plain (b) postoperative radiographs following revision ACL reconstruction and associated opening-­
wedge high tibial osteotomy

7

Revision Anterior Cruciate Ligament Reconstruction

a

81

b

Fig. 7.17 (a) The direction of the osteotomy is proximal to the superior tibio-fibular joint. (b) This posteriorly located joint acts as a hinge if the
osteotomy is too distal and will induce an anterior opening effect

both the ACL reconstruction and repair of the posterolateral
corner injury.
Two specific points are important when HTO is combined with ACL reconstruction. First, the final alignment
must take into account the desire of the patient to return to
sports. A hypercorrection may limit sports activities,
depending on the type of sports. In this situation, a moderate
hypercorrection (between 0 and 3°) is performed even if the
longevity of the osteotomy is reduced. Second, the technique must be adapted in order not to increase the tibial
slope. It is essential that the direction of the osteotomy is
proximal to the posterolateral superior tibio-fibular joint
(Fig. 7.17a). This posteriorly located joint acts as a hinge,
and if the osteotomy is too distal, it will induce an anterior
opening effect (Fig. 7.17b).
A detailed description of the performance of the osteotomy is found in Chap. 17 and will not be repeated here.

 CL Reconstruction and Anterior Tibial Closing
A
Osteotomy
This procedure is rarely indicated; however, it must be considered in patients with a failed ACL reconstruction combined with a tibial slope greater than 14° (Fig. 7.18).
Osteotomy is performed to reduce anterior tibial translation
induced by excessive posterior tibial slope. This technique
does not require elevation of the anterior tibial tuberosity,
and thus its position remains unchanged.
The surgical approach is identical to the valgus-producing high tibial osteotomy. The patellar tendon graft is

Fig. 7.18 Lateral stress radiograph of a knee demonstrating a failed
ACL reconstruction (increased anterior tibial translation noted) associated with a tibial slope of 14°

82

a

R Magnussen et al.

b

Fig. 7.19 (a) Drawing of a sagittal section through the proximal tibia
demonstrating the path of the anterior closing osteotomy. Anteriorly,
the cut is just proximal to the anterior tibial tuberosity. The posterior
hinge is centered at the junction of the PCL facet and posterior tibial

cortex just distal to the PCL insertion. (b) Intraoperative fluoroscopic
image of a proximal tibia with excessive slope and guide pin tip at the
appropriate position

harvested, and the femoral and tibial tunnels are created
prior to performance of the osteotomy. The anterior closing-wedge osteotomy is performed by preserving a posterior hinge centered at the tibial PCL insertion
(Fig. 7.19a, b). An anteroposterior guide pin is placed on
both sides of the patellar tendon just proximal to its insertion, about 3.5 cm below the joint line (Fig. 7.20). The
anterior portion of the superficial MCL medially and the
proximal portion of the tibialis anterior origin laterally
will need to be elevated to provide complete visualization.
The pins are placed with a cranial trajectory and should
meet the posterior tibial cortex near its junction with the
PCL facet. The positioning of the pins is controlled by
fluoroscopy (Fig. 7.19b).
The osteotomy is performed with an oscillating saw
below the pins on both sides of the patellar tendon, and a
second osteotomy is then made proximal to it. Depending
on the position of the anterior tibial tuberosity, a small
vertical coronal osteotomy may be necessary posterior to

it to allow a sufficient angle of cut. When planning the
degree of correction, the calculation must take into
account the measured bone abnormality, but also the clinical abnormality. A patient with no significant recurvatum
will tolerate a larger correction. If the correction of slope
required to control the anterior tibial translation produces
excessive recurvatum, then a re-­tension of the posterior
capsule is mandatory. We generally aim to reduce the tibial slope by about 5–10°. As about 1 mm of anterior closure allows for a correction of about 2°, the second
osteotomy should begin between 2 and 5 mm proximal to
the first and converge posteriorly. The bone wedge is
removed retaining the posterior hinge. The posterior cortex is fenestrated with a 3.2 mm drill bit in order to aid the
closure (Fig. 7.21) of the osteotomy. A new radiograph is
obtained and if correction is appropriate, the osteotomy is
secured by a staple on both sides of the patellar tendon
(Fig. 7.22). The tibial tunnel is then over-drilled using the
same diameter drill used to create the tunnel. The graft is

7

Revision Anterior Cruciate Ligament Reconstruction

83

Fig. 7.22 Intraoperative photograph demonstrating staple fixation on
either side of the patellar tendon

Fig. 7.20 Intraoperative photo of right knee showing the two guide
pins in situ (blue stars). The pins are placed just above the tibial tuberosity. Note the elevation of the superficial MCL medially and tibialis
anterior muscle laterally to provide complete visualization

Fig. 7.23 Intraoperative photograph demonstrating double tibial ACL
graft fixation

passed. Double fixation with wire around a screw as well
as an absorbable interference screw is preferred (Figs. 7.23
and 7.24). An additional locked plate (e.g. Tomofix)
placed medially can be useful to obtain better stability of
the osteotomy.
Fig. 7.21 Intraoperative photograph demonstrating fenestration of the
posterior tibial cortex with a 3.2 mm drill

84

R Magnussen et al.

Fig. 7.24 Postoperative plain lateral radiograph demonstrating tibial
post graft fixation, and staple fixation of the osteotomy. Note the correction of the excess posterior tibial slope as well as the anterior tibial
translation

Fig. 7.25 Drawing depicting reefing of the posterior medial capsule
through drill holes in the femoral condyle

Reefing of Posteromedial Soft Tissues

Lateral Extra-Articular Tenodesis

Reefing of the posteromedial soft tissues is usually sufficient
to control any hyperextension secondary to the anterior deflection osteotomy. Rarely posterolateral reefing is also required.
Reefing is performed by placing a retention suture in the
superficial medial collateral ligament and oblique popliteal
ligament and advancing the semimembranosus, or reefing
the posterior capsule to bone (Fig. 7.25). This procedure is
useful for control of anterior tibial translation in single-leg
stance and control of recurvatum. Rehabilitation will include
bracing to maintain 5° flexion for 45 days.

A lateral extra-articular tenodesis is performed to protect the
new ACL graft and allow better control of the pivot-shift. We
prefer to utilize a 10 mm width bundle of the iliotibial band.
The specifics of the technique are described in Chap. 5 and
will not be detailed here (Fig. 7.26).
The lateral tenodesis is justified for several reasons in
revision cases. The presence of prior tunnels potentially
alters tunnel position and may compromise control of the
laxity. The addition of an extra-articular tenodesis may better
control the laxity and associated pivot-shift. Additionally,

7

Revision Anterior Cruciate Ligament Reconstruction

85

Meniscal ramp lesions are recognized using trans-notch
arthroscopic visualization and use of the posteromedial
portal. The ramp lesion with its characteristic slip of the
posterior medial capsule must be sutured once reduced.
Also, the meniscal root tear (particularly of the lateral
meniscus associated with ACL rupture) needs to be fixed.
The technique is more complex and necessitates specific
instrumentation (see Chap. 4).

Conclusion

Fig. 7.26 Diagram demonstrating the completed lateral extra-articular
tenodesis

failure of the prior graft demonstrates that this patient is
prone to repeat instability, and everything should be done to
potentially increase stability. Neither the surgeon nor the
patient wants to face another failure.

Associated Meniscal Lesions
Meniscal ramp lesion and meniscal root tears need to be
addressed in the context of knee revision ACL reconstruction. A meniscal allograft may also be discussed in case of
previous total meniscectomy.

Successful revision ACL reconstruction requires a detailed
analysis of the reason for the failure of the prior reconstruction. The etiology of prior failure, prior graft choice,
prior surgical technique, and previous tunnel placement
influence the revision surgical technique. A two-stage surgery is discussed but is rare in our practice. The technique
of “outside-­in” drilling combined with cortical fixation
solves most technical challenges. Associated injuries and
anatomical factors must be taken into account and be
treated by either bony procedures (valgus-producing or
anterior closing tibial osteotomies) or soft tissue procedures (lateral extra-articular tenodesis or repair of associated ligamentous injuries). The results of revision ACL
surgery are inferior to those of primary reconstruction, and
return to sports uncertain, and it is mandatory to discuss
the outcomes with the patient, particularly if the surgeon
foresees technical difficulties.

8

Reconstruction of the Posterior
Cruciate Ligament
E Servien, G Demey, R Magnussen, P Neyret,
and C Butcher

Introduction

a

This chapter describes our technique for arthroscopic single-­
bundle reconstruction of the posterior cruciate ligament
(PCL) with quadriceps tendon autograft. This technique can
be adjusted to achieve a double-bundle reconstruction.

Indications
In our practice, the majority of isolated PCL ruptures are treated
conservatively. Our indications for surgical treatment are:
• Acute PCL ruptures associated with significant associated
peripheral laxity (posterior drawer differential >10 mm):
multi-ligament knee injuries (Fig. 8.1a, b)
• Persistent functional instability in chronic ruptures

b

E Servien
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Fig. 8.1 (a, b) Bartlett posterior stress view. (a) Setup. (b) X-ray
showing posterior drawer of 22 mm

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_8

87

88

E Servien et al.

Surgical Technique
Patient Positioning and Initial Setup
The patient is positioned on the operating table in the supine
position. A horizontal post is positioned distally on the table to
hold the knee in a 90° flexed position. A lateral support holds the
knee in this position, the thigh resting on the support with slight
external rotation of the hip. Fluoroscopy is used routinely to
control the correct positioning of the tibial tunnel. The image
intensifier is positioned prior to the establishment of the sterile
field, and the arch is positioned over the table to allow lateral
images to be obtained when the knee is placed in 90° of flexion
(Fig. 8.2). The image intensifier is moved in this position on its
base up to the level of the patient’s head, so that it does not interfere during the rest of the procedure.
Prior to prepping the surgical limb, the grade of the posterior drawer test is performed (Figs. 8.3a, b and 8.4a, b).
a

Fig. 8.2 Preoperative positioning of the image intensifier

b

Fig. 8.3 (a, b) Examination under anesthesia

a

Fig. 8.4 (a, b) Posterior tibial translation under fluoroscopy

b

8

Reconstruction of the Posterior Cruciate Ligament

Harvesting of the Quadriceps Tendon
The quadriceps tendon graft is harvested through an anteromedial skin incision beginning at the superior pole of the
patella and extending 6–8 cm proximally. Following exposure of the tendon, it is sometimes necessary to elevate some
of the distal fibers of the rectus femoris muscle to achieve a
sufficient graft length. The incision of the tendon is in line
with its fibers. The width of the graft must be 10 mm for a
length of 8 cm. In order not to breach the capsule, which
causes leakage of arthroscopy fluid during the procedure, we
try to only take the two most superficial layers of the quadriceps tendon, leaving vastus intermedius. The dimensions of
the patellar bone block are 10 mm wide by 20 mm long. This
is outlined in the periosteum using a 23 scalpel blade. Two
holes are then drilled into the bone block using a 2.7 mm
drill. These holes are used to pass the metal wire that is used
for traction when positioning the graft.
The anterior cortex is then cut with an oscillating saw
along the periosteal incision. A 10 mm Lambotte osteotome
is used to elevate the bone plug to a thickness of 8 mm. Once
the graft has been harvested, it is taken to a side table to be
prepared for implantation by the assistant surgeon. The edges
of the quadriceps tendon are closed with a No. 2 braided
absorbable suture. Some more recent instrumentation has
been developed by C. Fink (Karl Storz Minimally Invasive
Quadriceps Tendon Harvesting System, see Chap. 33, Fig.
33.22a). This allows the graft to be harvested through a
smaller transverse incision just proximal to the patella, but it
needs some experience.

89

tibial fixation. The free wire ends should be sufficiently long
(20 cm) for this fixation. Allograft can also be used if necessary. Its preparation is identical (Fig. 8.6).

Arthroscopy
An anterolateral portal is used for the arthroscope and an
anteromedial portal for the instruments. We now routinely
used in our practice an accessory posteromedial portal
that is useful to fully visualize and clean the tissue off the
posterior aspect of the tibia at the outlet of the tibial tunnel (Fig. 8.7). Coblation is particularly helpful for the
preparation of the posterior aperture of the tibial tunnel.
We believe that notch clearance and debridement of the
PCL, however, should be minimized in order to enhance
the biological integration of the graft. We also try to preserve the meniscofemoral ligaments (Fig. 8.8a, b). A thorough diagnostic arthroscopy is performed to assess the
anterior cruciate ligament and look for chondral and
meniscal pathology.

Fig. 8.6 Prepared quadriceps tendon allograft

Graft Preparation
After stripping any remaining muscle from the tendon graft,
the end of the graft is tubularized with a whipstitch for a
length of 5 cm with nonabsorbable suture, typically
FiberWire®. These will facilitate passage of the graft. The
bone plug and graft are then trimmed with a rongeur, so that
it can pass easily through a 10 mm sizing tube (Fig. 8.5).
Through the two drill holes in the patellar bone block, a
0.5 mm diameter metal wire is introduced in a figure of “8,”
which will allow distal control of the graft and double-distal

Fig. 8.5 Prepared quadriceps tendon autograft

Fig. 8.7 Debridement of the tibial footprint via the posterior medial
portal

90

a

E Servien et al.

b

Fig. 8.8 (a, b) Arthroscopic view of the notch following shaving of the synovium. This highlights the residual fibers of the PCL that are preserved
in order to optimize biological integration of the graft

Tibial Tunnel Preparation
The preparation of the tibial tunnel is performed with the
knee flexed to 90°. This helps to protect the popliteal neurovascular structures. We use a specific tibial drill guide,
Phusis® (Fig. 8.9a, b). If this is not available, an alternative is the Smith & Nephew (ACUFEX) instrumentation.
The latter guide is not fully stabilized to the tibia however,
and ­visualization of the back of the knee through a posteriomedial portal as well as flouroscopy becomes more
important (Fig. 8.9c). The arm of either guide is inserted
into the knee via the anteromedial portal and through the
notch and positioned in the PCL fossa on the posterior
tibia (Fig. 8.10). The tip of the guide is positioned under
arthroscopy, and a double check is made with fluoroscopy
to control the ideal location for the tibial tunnel: The recommended landmark for placement of this guide is
approximately 1.5 cm distal to the articular edge of the
posterior plateau, which corresponds to the junction of the
middle and distal 1/3 of the posterior tibial facet
(Fig. 8.11). The bullet portion of the Phusis drill guide is

placed on the anterior medial aspect of the proximal tibia.
A vertical incision is made approximately 3 cm medial to
the tibial tuberosity, and the guide is applied to the bone.
The guide is then secured to the tibia with two short pins
(Fig. 8.12). The guidewire is drilled under fluoroscopic
control to prevent injury to the popliteal vessels (Fig. 8.13).
In the sagittal plane, the guidewire forms an angle of 55°
(first setting of the guide) with the tibial diaphysis. The
bullet is then removed, keeping the guide on the posterior
aspect of the tibia in place to protect the vessels. The tibial
tunnel is made using cannulated reamers over the guidewire, under fluoroscopic control. The tunnel diameter is
gradually increased from 6 to 9 mm and then to the definitive 11 mm (Fig. 8.14a–c). The guidewire is then removed.
The arthroscope is introduced into the tibial tunnel, and
the shaver is used to debride the remnants of the ligament
(Fig. 8.15a, b). This stage is critical in order to allow easy
passage of the graft. In this case, the arthroscope is placed
via the anterolateral portal, and the shaver can be used via
the posteromedial portal.

8

Reconstruction of the Posterior Cruciate Ligament

a

91

b

c

Fig. 8.9 (a, b) Specific PCL tibial drill guide, set at an angle of 55°. (c) Positioning of Acufex tibial PCL guide (posteromedial portal view). The
gradations will help position the wire 15 mm distal to the articular surface

Fig. 8.10 The tibial guide is inserted through the anterior medial
portal
Fig. 8.11 Fluoroscopic control of the position of the tibial guide

92

E Servien et al.

Fig. 8.12 Arthroscopic control and pinning of the guide to the tibia
Fig. 8.13 Placement of the guidewire under fluoroscopic control

a

b

c

Fig. 8.14 (a–c) Progressive reaming of the tibial tunnel (6 mm diameter drill bit, then 9 mm and 11 mm)

8

Reconstruction of the Posterior Cruciate Ligament

a

93

b

Fig. 8.15 (a, b) The arthroscope is inserted into the tibial tunnel to visualize debris and remnants of the PCL at the posterior exit of the tibial
tunnel

Drilling the Femoral Tunnel
The single-bundle technique that we describe aims to
reconstruct the anterolateral bundle of the PCL. We use an
outside-­in femoral tunnel guide. The arm of the guide is
introduced through the anteromedial portal. The tip of the
guide is placed such that the guidewire will exit through
the center of the femoral insertion of the anterolateral
bundle of the PCL. This goal is achieved with the knee at
90° of flexion. In this position, the intra-articular position
of the tunnel opening in the axial plane is at 1 o’clock in
the right knee and 11 o’clock in the left knee. The anterior
border of the tunnel lies between the condylar wall and
roof of the notch.
A 2 cm incision is made over the anteromedial aspect
of the medial femoral condyle. The distal edge of the vastus medialis is identified and retracted upward to avoid
injury to the muscle belly (Fig. 8.16). The bullet of the
outside-in femoral guide is advanced to bone. A guidewire
is then passed through the condyle under arthroscopic
control. The bullet and guide are removed, and the end of
the guidewire is held in a curette (Fig. 8.17a, b). The femoral tunnel is drilled using a cannulated reamer. Like the
tibial tunnel, an initial tunnel of 6 mm in diameter is

Fig. 8.16 Incision for the femoral tunnel. The vastus medialis is elevated to allow placement of tunnel without injury to the muscle

drilled and then enlarged using a reamer of 10 mm in
diameter (Fig. 8.18). The intra-­articular opening of the
tunnel is then debrided by successively introducing the
arthroscopic shaver through the anteromedial portal and
the femoral tunnel.

94

a

E Servien et al.

b

Fig. 8.17 (a, b) Insertion of the guidewire. Drilling the femoral tunnel is facilitated by placement of the special curette to maintain the guidewire
in an optimum position

Fig. 8.19 Preparation of metal traction wire

Fig. 8.18 Femoral tunnel

Graft Passage
The graft can be inserted with traction using a 0.5 mm
metal wire or FiberWire. In the case of the metal wire, it is
bent at its end to form a loop (Fig. 8.19). This is then
passed up through the tibial tunnel inside a 6 mm cannulated reamer. Under fluoroscopy, the tip of the metal wire
is then passed through a slot at the end of the tibial guide.
The wire locks into the slot due to the loop at its end
(Fig. 8.20). The tibial guide and the wire are then removed
through the anteromedial portal (Fig. 8.21). The wire
should be secured at both ends with a Kocher. If a FiberWire
is used, a doubled suture loop is passed up the tibial tun-

nel, retrieved with a curved grasper, and pulled out of the
anteromedial portal. This is then used to pass another doubled suture in a retrograde fashion, passing the loop out of
the tibial tunnel distally.
The graft is now passed from distal to proximal, through
the tibial tunnel using the wire, or FiberWire suture loop
(Fig. 8.22). It is inserted so that the bone remains in the
tibial tunnel, using counter tension on the distal wires. The
progression of the graft through the tibial tunnel is monitored using the image intensifier. A probe inserted through
the posterior medial portal acts like a pulley to modify the
orientation of the traction. It is considered sufficiently
advanced when the end of the bone block is flush with the
intra-­articular end of the tibial tunnel (Fig. 8.23). The other
end of the graft is delivered into the notch, and the passing
sutures (FiberWire) are easily retrieved using a Kelly clamp
through the femoral tunnel (Fig. 8.24).

8

Reconstruction of the Posterior Cruciate Ligament

95

Fig. 8.22 Passage of the graft from distal to proximal

Fig. 8.20 The wire is aimed at the hole in the tibial guide using a drill bit

Fig. 8.23 Positioning the graft bone block near the posterior aperture
of the tunnel, confirmed by fluoroscopy

Fig. 8.21 The guide is withdrawn bringing the wire out of the anterior
medial portal

Next the bone block of the graft is fixed in the tibia. A
guidewire is inserted into the tibial tunnel and positioned in
front of the bone block. A 9 × 25 mm absorbable or metallic
interference screw is inserted over the guidewire and screwed
into position. When the screw is flush with the joint line, the
tightening is stopped. The advancement is easily assessed
with the image intensifier by knowing that the screw protrudes 5 mm beyond the tip of the screwdriver (Fig. 8.25).
The arthroscope can be inserted into the tibial tunnel to

Fig. 8.24 The FiberWire is retrieved through the femoral tunnel

96

E Servien et al.

Fig. 8.27 The wires are wound around a cortical tibial screw, providing double fixation

Fig. 8.25 Correct position of the interference screw is inferred by the
position of the screwdriver

Fig. 8.28 Manual reduction of the posterior drawer, and placement of
the femoral interference screw

Fig. 8.26 Appearance of the screw and bone block in the tibial tunnel
using the arthroscope

check the position of the screw in relation to the bone block
(Fig. 8.26).
The tibial fixation is supplemented with a 4.5 mm cortical
screw on the anterior cortex of the tibia. The two strands of
wire are wound around the screw, and the screw is tightened
(Fig. 8.27). The tibia is correctly reduced in 90° of flexion,
and the graft is tensioned (Fig. 8.28). The femoral fixation is
with a 9 × 25 mm absorbable interference screw in an outside-­in manner. The tension of the graft and the absence of
residual posterior tibial translation are evaluated clinically
and arthroscopically (Fig. 8.29).

Fig. 8.29 Arthroscopic control

8

Reconstruction of the Posterior Cruciate Ligament

Postoperative Care
Following deflation of the tourniquet, hemostasis is
achieved. A suction drain is placed intra-articularly, and the
wounds are closed. The knee is locked in extension using an
extension splint. This should include a wedge or a pillow
under the calf to prevent posterior tibial translation due to
gravity. The surgeon verifies the presence of distal pulses
and normal capillary refill of the limb prior to awakening the

97

patient. An AP and true lateral radiograph of the knee is
taken. DVT thomboprophylaxis is continued for a period of
15 days, and antibiotics are given for 24 h. Staples are
removed from the wounds on the 15th postoperative day.
Postoperative follow-­up is scheduled 45, 90, and 180 days
and 1 year postoperatively. Rehabilitation is designed to
prevent posterior tibial translation and can be done in the
prone position.

9

Posterolateral Corner and Lateral
Collateral Ligament Reconstruction
E Servien, R Magnussen, P Neyret,
and C Butcher

Introduction
This chapter describes the technique we are using for the
reconstruction of the posterolateral corner and the lateral collateral ligament of the knee. Posterolateral corner injuries are
complex injuries that remain under-diagnosed. The resultant
laxity from this injury can be classified as:
• Horizontal plane laxity (posterolateral corner, PLC)
• Frontal plane laxity (lateral collateral ligament, LCL)
• Combined laxity in both planes (PLC and LCL).
Careful clinical examination allows the surgeon to define
the extent of the laxity and make the correct diagnosis
(Table 9.1).
AP and lateral radiographs of the knee both supine and
standing should be performed, including long leg films to
assess lower extremity alignment. Monopodal stance x-ray
may give more functional information than bipodal views.
Comparative stress radiographs should also be considered to
assess passive varus/valgus laxity, anterior and posterior tibial translation, and recurvatum (Fig. 9.1). MRI is routinely
ordered to assess the cruciate ligaments, menisci, and articular cartilage as well as the LCL and PLC.

E Servien
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Table 9.1 Clinical findings

Posterolateral
corner
(horizontal
plane)
Lateral
collateral
ligament
(frontal plane)
Combined

Lateral
Varus hypermobility
laxity test
0
+

Posterior
drawer in
external
rotation
+

Recurvatum
test
(Hughston)
+

+

0

0

0

+

+

+

+

Patient Positioning and Initial Setup
The patient is positioned on the operating table in the supine
position. A horizontal post is positioned distally on the table to
hold the knee in a 90° flexed position. A lateral support holds
the knee in this position, the thigh resting on the support with
slight external rotation of the hip. A thorough examination is
carried out under anesthesia in order to confirm the injury.

Technique
A 6–8 inch curvilinear incision is made along the lateral
aspect of the thigh starting from the posterior aspect of the
lateral femoral condyle extending midway between
Gerdy’s tubercle and the fibular head to 1 cm below the
level of the fibular neck. The fascia lata is incised in the
middle in line with its fibers to its insertion on Gerdy’s
tubercle. Depending on the level of injury, the incision
along the fascia lata may be placed more anteriorly or posteriorly. The first step in the procedure is to identify various anatomical elements of the posterolateral corner of the
knee (popliteal tendon, LCL, lateral epicondyle, fibular
head) (Fig. 9.2). If the popliteal tendon and/or lateral col-

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_9

99

100

E Servien et al.

Fig. 9.2 Posterolateral corner anatomy. Drawing reprinted with permission from LaPrade RF, Ly TV, Wentorf FA, Engebresten L: The posterolateral attachments of the Knee. Am J of Sports Med; 2003, 31,
854-860

Fig. 9.1 Varus stress x-rays (Telos® stress device)

lateral ligament are still present, they should be clearly
identified using a loop (Figs. 9.3 and 9.4). The choice of
graft can either be autograft or an allograft. The harvest
and preparation of these will not be described in this chapter. The principal of the reconstruction remains the same
irrespective of the graft choice.

Lateral Collateral Ligament
Fibular Tunnel
It is essential that the common peroneal nerve is identified
prior to making a tunnel in the fibular head. It is identified
proximally just posterior and inferior to the biceps femoris

Fig. 9.3 Identification of the LCL and popliteal tendon – the latter has
been tagged with blue cotton tape

9

Posterolateral Corner and Lateral Collateral Ligament Reconstruction

101

Fig. 9.4 The LCL has been tagged (red sloop)

Fig. 9.7 Drilling the fibular tunnel from anterior lateral to posterior
medial

Fig. 9.5 Identification of the common peroneal nerve (in this case in
blue tape)

Fig. 9.8 Placement of the guide pin in the lateral epicondyle

and gently explored from proximal to distal. Once identified,
the surgeon must be aware of its position at all times; hence,
it is tagged with a vessel loop (Figs. 9.5 and 9.6). A tunnel in
the fibular head is made using 3.2 mm drill bit, and it is then
enlarged to 4.5 mm. The tunnel is angled from lateral to
medial as it is drilled from the front of the fibular head toward
the back (Fig. 9.7). A drill guide is used throughout to help
protect the nerve and provide support to the fibular head
while drilling.

Femoral Tunnel

Fig. 9.6 All relevant structures have been identified

To reconstruct the LCL, a femoral tunnel is made in the
center of the lateral epicondylar eminence. A guidewire
with an eye is placed directly perpendicular to the lateral
epicondyle (Fig. 9.8). A 7 mm cannulated drill is then used

102

E Servien et al.

Fig. 9.9 Preparation of the tunnel with a cannulated 9 mm drill bit

Fig. 9.11 Position of the femoral tunnels for the LCL and popliteal
tendon grafts
Fig. 9.10 Y-graft

to create a tunnel 25 mm in depth. If reconstruction of both
the LCL and the popliteal tendon is required, then a second
7 mm femoral tunnel will be required. In the case of isolated LCL reconstruction, a larger femoral tunnel of
8–9 mm can be made (Fig. 9.9). The graft is fixed in the
femur with an absorbable interference screw of the same
diameter as the tunnel.

Popliteus Tendon
Graft
If an isolated popliteus tendon reconstruction is required, a
single strand graft with a bone block is used for femoral fixation. To achieve sufficient length (10 cm), an autograft
(quadriceps tendon) or allograft (Achilles tendon) is preferred. If reconstruction of the PLC requires reconstruction
of the popliteal fibular ligament (PFL), a bifurcated graft is

used (Fig. 9.10), and one strand is designed to reconstruct the
popliteal tendon and the other the PFL. The role of the PFL
defined by Gilles Bousquet is to act as a pulley, changing the
orientation and the tension of the popliteus tendon. In these
cases, the reconstruction is almost always combined with an
ACL and/or PCL reconstruction.

Femoral Tunnel
The 7 mm femoral tunnel for the popliteus tendon reconstruction is located at the anatomic insertion site of the popliteus tendon, approximately 11 mm distal, and anterior to
the femoral tunnel for the LCL (Fig. 9.11). The preparation
of the tunnel is identical to that of the femoral tunnel for the
LCL. The graft is fixed with an absorbable interference
screw of 7 mm. In case of a multi ligament reconstruction
(LCL, popliteal and cruciate ligaments), it is sometimes
necessary to perform a lateral reconstruction with a single
tunnel for both the LCL and the popliteus tendon.

9

Posterolateral Corner and Lateral Collateral Ligament Reconstruction

103

Distal Tunnel
PFL (Fibular Tunnel)
The short arm of the bifurcated graft is used to reconstruct
the PFL. It is passed with the other strand under the fascia
lata and then through the tunnel in the head of the fibula from
posterior to anterior (Figs. 9.12 and 9.13). It is fixed with the
LCL with an interference screw.
 opliteus Tendon (Tibial Tunnel)
P
Blunt dissection is carried out to expose the popliteus muscle
belly and posterolateral joint capsule on the posterior aspect
of the tibial plateau (Fig. 9.14). The tibial tunnel is made
using the tibial guide for the ACL. The tunnel is horizontal.
The guidewire is passed from just below Gerdy’s tubercle to
the posterior tibia approximately 1 cm below the joint line. It
is prudent to protect the soft tissues at the back of the tibial
plateau when passing the guidewire. A 6 mm tunnel is drilled
over the wire.
The long strand of the graft is passed from back to front
and is secured with absorbable interference screw placed
from anterior. The grafts are fixed at 30° flexion for the LCL
and 90° for the popliteus tendon, with the foot in neutral
rotation (Figs. 9.12 and 9.13).
Fig. 9.12 Diagram of the popliteal tendon and PFL grafts, with second
tunnel for the LCL (lateral view)

a

b

Fig. 9.14 Exposure of the posterior aspect of the tibial plateau

Fig. 9.13 (a, b) Diagram of the LCL, popliteus tendon, and PFL
grafts. (a) Side view. (b) Posterior view

10

Dislocations and Bicruciate Lesions
S Lustig, R Magnussen, P Neyret,
and C Butcher

Introduction
Knee dislocations usually involve both anterior and posterior
cruciate ligament injury (except in some rare anterior or posterior dislocations) and injury of the lateral and/or medial
knee structures. Dislocations imply a high risk of neurovascular lesions, and an angiogram is often indicated. This
chapter does not deal with knee dislocations where one of
the two cruciate ligaments is not torn.

Classification and Diagnosis
It is not always possible to be precise about the mechanism
of these injuries, nevertheless there are definite patterns of
injury, and it is therefore possible to anticipate ligamentous
lesions and associated complications. A classification helps
to identify these patterns; ours was developed with
F. Rongieras in 1998 and inspired by the successive works of
the Lyon School of Knee Surgery. In addition to the tears of
the cruciates, it takes into account the elementary lesions of
the peripheral structures, which may be either

The limb injuries result from either prolonged application
of force, with patterns of escalating severity, or high-velocity
injury from motorcycle accidents, avalanche, etc. (Fig. 10.1).
Simple bicruciate injuries, or ‘Pentades’ are a result of
this sustained progressive force application. Three typical
types exist, caused by valgus, varus, or hyperextension
(Fig. 10.2). There is opening (gaping) from tearing of the
structures on convex side, as well as bicruciate injury. These
injuries may originate with a “triad.” The medial pentade,
for example, is most often the result of an unhappy Don
O’Donoghue’s triad (Fig. 10.3). The lateral pentade follows
the same principle, but is associated with common peroneal
lesions (Fig. 10.4), and the posterior pentade with vascular
injury (Fig. 10.5).
If the force continues and is not spent, a dislocation
results. In addition to tearing and gaping on the convex side,
there is periosteal stripping on the concave side and associ-

MECHANISMS

• Tearing (associated with rotation in coronal, sagittal, or
axial planes) or
• Periosteal stripping (associated with translation).
S Lustig
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

Low energy,
forced, prolonged
I
N
J
U
R
Y

Triad

Pentade

High energy
moto, etc

Dislocation

Fig. 10.1 Injuries result from either prolonged application of force,
with patterns of escalating severity, or high-velocity injury

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_10

105

106

S Lustig et al.

PENTADES

Medial - Bicruciate

Gaping

Lateral - Bicruciate

Posterior - Bicruciate

Combined lesions
- ligaments
- rare complication

Fig. 10.2 Simple bicruciate injuries or “Pentades” are a result of a
sustained progressive force application. Three typical types exist

a

b

ated translation (hence dislocation). There are five types
(Fig. 10.6); two are the natural progression of the medial and
lateral pentades (Fig. 10.7). Another involves tearing of all
the peripheral structures, with only one corner intact, around
which there is rotational dislocation (Fig. 10.8). In the last
two, there is stripping of the peripheral structures on two
sides, allowing translational dislocation (Fig. 10.9). With the
anterior dislocation, there may be injury to the extensor
mechanism, and with the posterior variant, vascular injury.
In the case of knee dislocation, reduction under general
anesthesia with fluoroscopic control is an emergency.
Radiographs are repeated after immobilization of the
lower limb in a posterior knee–ankle splint to confirm that
the joint remains reduced. Completely torn structures
impart no stability, but stripped structures remain longitudinally functional in tension, providing an intact hinge
c

Fig. 10.3 (a–c) A medial pentade comprises tearing of the medial structures, both cruciates and medial compartment injury. Typically there are
no complications
Fig. 10.4 (a, b) The lateral pentade
involves tearing of the lateral
peripheral structures and often
includes a common peroneal lesion.
The MRI shows lateral soft tissue
injury and evidence of medial
compression with bone bruising

a

b

10 Dislocations and Bicruciate Lesions

107
BICRUCIATE LESIONS
Simple (Pentades)

Combined (Dislocations)

Medial
Lateral

Medial (lateral dislocation)
Lateral (medial dislocation)
Rotatory

Pure (Dislocations)

Anterior
Posterior

Posterior

Fig. 10.6 Classification of bicruciate lesions

Fig. 10.5 Traction injury to the tibial vascular structures is common
with the posterior pentade. Anterior tibial impaction is an important
sign of this injury

a

b

c

Fig. 10.7 (a–c) Lateral and medial dislocation. In these injuries, there is dislocation of the tibia toward the opposite side of the gaping. In this
case, there is medial gaping and lateral dislocation. When there is lateral gaping, the tibia will dislocate medially

a

b

c

Fig. 10.8 (a–c) Rotational dislocation. Tearing of almost all peripheral structures, apart from at one stable point

108

S Lustig et al.

c

b

Fig. 10.9 (a) Pure dislocation from stripping on both sides. (b, c). Anterior dislocation produces traction injury to the posterior vascular
bundle. (d) Posterior dislocation causes extensor mechanism lesions. The dislocated patella is visible

Fig. 10.10 Dynamic x-rays and MRI give
complimentary information

Clinical exam + dynamic X-rays
Laxity
Mechanisms

Operative indication

MRI
Ligament injury and localisation
Meniscus lesions
Osteochondral lesions

once the joint is reduced. The surgeon can exploit this
residual stability in the acute post injury period with
three-point external support, as with a periosteal hinge in
fracture surgery.
The identification of the mechanism and type of lesions of
the peripheral structures is by clinical examination of the
joint, and reference to the radiology, including stress x-rays
and MRI (Fig. 10.10).
• Stress x-rays provide functional information about the
character of the lesions, and thus the indications for
surgery.
• MRI provides the localization of anatomical defects and
informs the strategy of the surgery (Fig. 10.11).
The stress x-rays may be performed with the patient
awake, under anesthesia at the time of the reduction, or at the

Operative strategy

start of a surgical intervention, and should specifically
include
• Valgus and varus laxity
• Anterior and posterior translation
• Medial and lateral translation.
Without performing the last of these stress tests, the stripping capsular-periosteal detachments typical of dislocations
may be missed (Fig. 10.12).
In addition to localizing the ligamentous lesions, MRI is
useful to identify and localize meniscal and cartilaginous
injuries (Fig. 10.13).
A definitive treatment plan can finally be established, taking into account the number, type, and location of lesions, as
well as the limb alignment and other patient factors such as
age and activity level.

10 Dislocations and Bicruciate Lesions
Fig. 10.11 Stress x-ray gives
a functional diagnosis and
indication for surgery—in this
case MCL injury (a). MRI
provides the site of injury to
help plan the surgery—
confirming in this case injury
at its proximal origin (star) (b)

109

a

b

a

b

d

e

Fig. 10.12 Stress x-rays identify a lateral bicruciate lesion and exclude a
dislocation. Varus stress view reveals laxity of the lateral structures (a).
Although anterior and posterior stress lateral x-rays reveal bicruciate

c

insufficiency (b and c), the translation stress views reveal no periosteal
stripping, specifically on the medial side (d, e). These tests should be performed carefully to prevent further injury to the common peroneal nerve

110

S Lustig et al.

Cruciate Ligament Reconstruction
The first step is PCL reconstruction, by arthrotomy or
arthroscopy. We prefer arthroscopic reconstruction.
PCL reconstruction in multi-ligament injuries is the same as
in cases of isolated PCL injury. One must be vigilant for the risk
of extra-articular fluid leakage because of associated capsular
injury, and careful control of the pump pressure is critical.
Joint irrigation is the first step of the procedure to clean out
the hemarthrosis. Intercondylar notch cleaning is minimal to
preserve residual PCL fibers that could heal in the presence of
the PCL graft. PCL graft fixation must be done with careful
attention to the position of the tibia relative to the femur in the
sagittal plane. The tibia should be positioned in its reduced
position (approximately 1 cm anterior to the femur) or even
with a slight anterior tibial translation of 1–2 mm. ACL reconstruction technique is the same as the one used for isolated
ACL reconstruction. If slight posterior laxity persists after PCL
reconstruction, care must be taken not to induce a posterior
tibial translation during ACL graft tensioning and fixation.

Reconstruction Sequence
Fig. 10.13 MRI also reveals pathology of the chondral surfaces and
menisci (circle)

Operative Timing
If possible, surgery is done 5–15 days after the initial trauma,
when the soft tissues are less swollen. Indications for urgent
intervention include vascular injuries, unreducible dislocations, incarceration of the MCL, and extensor mechanism
rupture.
The PCL and collateral ligaments are reconstructed in
the first operative procedure. The ACL is reconstructed in a
delayed manner in order to prevent stiffness and to decrease
operative time as multi-ligament surgeries are long and
complex.
However, simultaneous ACL and PCL reconstruction is
sometimes indicated, particularly when a lateral laxity is
associated, as late instability follows if the ACL is not
reconstructed.

Setup
Patient positioning is the same as is utilized for isolated PCL
reconstruction as is described in detail in prior chapters.
Fluoroscopy is utilized for the PCL.

In case of multi-ligament injury, we recommend fixing first
the PCL graft, then the posterolateral corner, and finally the
posteromedial corner. The ACL may be reconstructed in a
second operation, after recovery from the first surgery
(6 months–1 year later). However, when both the PCL and
ACL are reconstructed in the same surgery, the fixation
sequence is PCL graft first, then ACL graft, and finally lateral knee structures.

Lateral Knee Structure Reconstruction
Lateral collateral ligament (LCL) injuries are rarely isolated.
Acute repair or reconstruction of LCL is often done in association with PCL and/or ACL reconstruction.
Clinical examination is very important for the diagnosis. The LCL is palpated when the patient is positioned in
the “figure four” position and comparison with the noninjured knee is important. The presence of an LCL tear is
confirmed by the presence of lateral opening with varus
stress testing.
The exact location of LCL tear is sometimes difficult to
localize. Standard radiographs, CT, and especially MRI can
help distinguish between a mid-substance tear and a ligamentous avulsion injury with or without bony avulsion from
the femur or fibular head.

10 Dislocations and Bicruciate Lesions

111

Surgical Exposure

Femoral Avulsion

A lateral incision 6–8 cm in length extends from the posterior part of the lateral femoral condyle to the fibular neck. It
is similar to the hockey stick incision described in the chapter on posterolateral corner reconstruction.
The iliotibial band is divided parallel to its fibers, until
its insertion on Gerdy’s tubercle. The next step is to localize anatomical structures including the popliteus tendon,
the LCL, and the biceps femoris tendon. The common
peroneal nerve is dissected from proximal to distal and
protected.
According to the location of the LCL tear, iliotibial band
incision can be moved more or less posterior: more posterior
in case of LCL avulsion from the fibular head and more anterior in case of femoral avulsion.

In case of LCL and popliteus avulsion of the femur, osseous
fixation is needed. Many techniques are possible including
screws, anchors, and wires. If a bony fragment has been
avulsed, it can be fixed with a 3.5 mm screw and washer.
In case of LCL avulsion without bony fragment, transfemoral fixation with a femoral socket can be done or the LCL
can be reattached to the femoral epicondyle with anchors or
with transfemoral sutures. To do so, the femoral footprint is
roughened up to create a trough for healing, the proximal part
of the ligament is sutured with No. 2 FiberWire®, and two
parallel transosseous tunnels directed toward the medial femoral condyle are created. The two ends of the suture previously placed on the LCL are passed through the femur and
tied on the medial femoral cortex (Fig. 10.14).

Fig. 10.14 Repair of a
femoral avulsion

112

Fibular Head Avulsion

S Lustig et al.

–– Fixation with a 3.5 mm screw is easier but sometimes not
enough if the biceps tendon is inserted on the bone frag-

ment, as it will apply high traction forces on this
fragment.
–– A transosseous wire can be used as a suture or a cerclage
and is more effective to resist traction forces. A
0.8–1 mm diameter wire is used. It is passed through the
bone fragment in a U shape. A tunnel is drilled in the
fibular head after protecting the common peroneal nerve.
The wire is crossed in an 8-shape and passed through the
fibular head (Fig. 10.17). At 20–30° of knee flexion, the
wire is tightened until the fragment is reduced and the
LCL is re-­tensioned. The wire is then cut and the free
ends are bent. Cancellous autograft harvested from
Gerdy’s tubercle can be used to enhance consolidation.

Fig. 10.15 Fibular head avulsion fracture

Fig. 10.17 Fixation of a fibular head avulsion fracture

Fibular head avulsion corresponds to avulsion of the LCL
alone or with fabello-fibular ligament, fabello-popliteus ligament, and arcuate popliteus ligament (Figs. 10.15 and 10.16).
If the bony fragment is large, the biceps tendon may also be
avulsed.
Depending on the bone fragment size and shape, it can be
fixed back to the fibular head with a single screw, a wire, or
both.

a

b

Fig. 10.16 (a, b) Operative views showing reduction of a fibular head fracture with biceps tendon and lateral collateral ligament avulsions

10 Dislocations and Bicruciate Lesions

Mid-Substance Ligament Rupture

113

eral epicondyle. The graft is passed in the transtibial tunnel,
then through the lateral gastrocnemius, and then into the
Different suture techniques are possible (U-stitch, frame-­ femoral tunnel. W. Müller has described a reconstruction
stitch). However, transverse rupture is rare. Usually, ruptures with a graft of iliotibial band measuring 10–15 cm detached
are Z-shaped. The distal half of the ligament looks like a from Gerdy’s tubercle, passed in a transtibial tunnel from
round cord surrounded by a sheath, as compared to the anterior to posterior, and fixed on the femoral insertion of
­proximal part which is in the capsular tissue. It is thus diffi- popliteus.
cult to identify the proximal tissue and perform an end-toend suture of the torn ligament. If it is possible, we prefer
non-­resorbable, braided suture (e.g., Ethibond®, Mersuture®, Postoperative Care
TiCron®, FiberWire®) for such repairs.
We usually prefer reconstruction of the lateral structures During the first 45 postoperative days, the patient wears a
because repaired tissue is often insufficient. A 6 × 1 cm band of posterior knee splint in extension and remains non-weight-­
biceps tendon can be harvested and fixed to the lateral femoral bearing. Knee mobilization is begun on day 1, limited to 60°
epicondyle in a bone socket or with an anchor. Another option is until day 21, and then to 95° until day 45. Any valgus/varus
to use an ipsilateral or contralateral gracilis tendon autograft.
stress is avoided for 90 days, with the use of a range of
motion brace. Weight-bearing is then progressive from day
45, and flexion is no longer restricted. Sports are forbidden
for 6 months. Until 9 months postoperative, ligaments may
Posterolateral Reconstructions
be thick and sometimes painful.
This procedure includes reconstruction of the essential anatomic structures of the posterolateral corner: the popliteus
tendon and popliteo-fibular ligament. In case of bony avul- Medial Knee Structure Reconstruction
sion, with or without a bone fragment, they should be fixed
acutely, as for LCL injuries. In case of mid-substance liga- Type of Injury
ment tears, the remnants of torn ligaments can be sutured,
but as with LCL injuries, augmentation with a graft is Medial knee structure injuries may be a mid-substance
mandatory.
ligament tear or a bony avulsion of the ligament
G. Bousquet described a reconstruction of the PLC called insertion.
the “petit poplité,” which prevents tibial external rotation and
Mid-substance tears are usually classified into three
posterior subluxation of the lateral tibial plateau.
grades. The surgeon plans his surgery according to the medial
The first step is reconstruction of the popliteo-fibular liga- knee laxity found during clinical examination and measured
ment which is done with a 0.5 × 7 cm band of the biceps on stress radiographs and according to associated cruciate
femoris tendon left attached on the fibular head. This band is ligament injuries. When medial knee laxity is present in full
passed under the iliotibial band, around the popliteus tendon, extension, other ligamentous injuries may be associated:
from anterior to posterior and from medial to lateral (in order
to pull the popliteus tendon laterally and distally), and is then • Posteromedial corner injury (posterior-oblique ligament
fixed to the lateral capsule and to the fibular head with the
(POL) as described by Hughston)
foot in neutral rotation.
• PCL injury, which may be difficult to diagnose clinically
The second step is to tighten the recurrent popliteus ten- • Deep MCL avulsion on the femoral side, which may not
don with two or three resorbable sutures, which tightens the
heal and can lead to chronic laxity if not diagnosed.
posterior condylar recess to the anterior border of the lateral
retroligamentar arthrotomy (and eventually to the LCL).
As described by W. Muller, deep MCL tears are often not
The third and last step is to tighten the fabello-popliteus at the same level as superficial MCL tears. MCL lesions are
ligament with a bundle of iliotibial band measuring 1 × 5 cm often considered to be a minor injury by the surgeon, but not
left attached on its two extremities. This band is fixed to the by the patient who is quite restricted in his physical
fabella or to its fibrous nucleus with single stitches of resorb- activities.
able sutures.
Reconstruction with iliotibial band has also been
described. Jaeger reported a reconstruction with an iliotibial Arthroscopy
band graft measuring 15–20 cm long, harvested from Gerdy’s
tubercle. Two tunnels are drilled: a transtibial anteroposte- Arthroscopy may be useful to confirm the diagnosis. It
rior tunnel of 6 mm diameter from Gerdy’s tubercle and a helps the surgeon to determine if the MCL tear is above or
transcondylar anteroposterior tunnel below the femoral lat- below the meniscus. In case of increased space above the

114

meniscus, the MCL is torn on the femoral side, and in case
of increased space below the meniscus, the MCL is torn on
the tibial side.

Mid-Substance Rupture
The MCL can heal with a knee brace allowing flexion and
extension but restricting valgus loading of the knee. Surgical
reconstruction is rarely needed except in occasional cases of
complete rupture. If needed, an end-to-end MCL repair is
done. When associated with an ACL reconstruction, an
anteromedial surgical approach is used. The skin incision is
longitudinal, 2–3 cm medial to the anterior tibial tubercle
(ATT). MCL end-to-end sutures favor healing. For superficial MCL tears, we prefer to whip-stitch the entire ligament
to tighten it before suturing and fixing it. Anchors can also be
used. An augmentation with the gracilis tendon as described
by Helfet can also be done.

Repair of Bony Avulsions
In case of femoral avulsion, an oblique skin incision over the
medial epicondyle allows good visualization of the lesion.
In case of a large bony avulsion (e.g., the whole medial
epicondyle), fixation can be achieved with a 3.5 mm screw and
washer, transcondylar FiberWire sutures, or with a staple.
In case of tibial avulsion, the skin incision is vertical
and medial. The sartorius is incised in a reverse L shape,
and the hamstring tendons are elevated to expose the
avulsed superficial MCL. Anchors can be used to reinsert
it on the tibia.

MCL Reconstruction
If a direct suture is not sufficient or not possible, a reconstruction can be done. The patient is supine with a mid-thigh
tourniquet in 90° of knee flexion. A medial incision extends
from the level of the patella to 3 cm below the ATT. Dissection
is done to expose the medial epicondyle.
• Reconstruction with hamstring tendons. The sartorius fascia is retracted and the gracilis tendon is exposed, dis-

S Lustig et al.

sected but left attached to the tibia. An open stripper is
used to detach it proximally. A vertical tunnel is created
deep to the femoral medial epicondyle by drilling two
4.5 mm holes, 10–15 mm apart. O’Shaw claws of increasing diameter are used to join these two holes and create
the tunnel. The gracilis tendon previously prepared and
whip-stitched with an absorbable suture is passed through
the femoral tunnel and sutured back on its tibial insertion
with a similar suture. Note that when we can choose
another graft option, we prefer not to harvest the ipsilateral hamstrings, in order not to weaken the medial structures and control of tibial rotation.
• In case of chronic laxity, it is difficult to determine clinically if medial laxity is due to an isolated MCL tear, isolated injury to the posteromedial corner, or both. MRI can
help localize the tear, but can be inconclusive in such
chronic cases. Usually it is advisable to repair both. The
superficial MCL can be whip-stitched over its entire
length and sutured in a tightened position. For the posteromedial corner, one or two anchors are inserted on the
posterior side of the medial epicondyle via a medial retro-­
ligamentar arthrotomy. The surgeon looks for the “lunule
sign” described by H. Dejour above the meniscus. It corresponds to an avulsion of the posteromedial capsular
recess. It is then anchored on the posterior and proximal
side of the condyle, taking care not to overtighten it,
which could restrict extension. The POL itself is stitched
and anchored back on the posterior side of the medial epicondyle. Finally, in case of major and chronic laxity, an
MCL and POL reconstruction with quadriceps tendon
autograft as described by Engebretsen can be done.

Postoperative Care
During the first 45 postoperative days, the patient wears a
posterior knee splint in extension and remains non-weight-­
bearing. Knee mobilization is begun on day 1, limited to 60°
until day 21, and then to 95° until day 45. Any valgus/varus
stress is avoided for 45 days. Weight-bearing is then progressive, and flexion is no longer restricted. Sports are forbidden
for 4 months. Until 9 months postoperative, ligaments may
be thick and sometimes painful.

Synovectomies of the Knee

11

P Archbold, A Pinaroli, and P Neyret

Indications
In this chapter, the technical principles and in particular the
surgical approaches required to perform a synovectomy are
discussed (septic arthritis, tumor pathology, and synovectomy after TKA are excluded). A synovectomy may be indicated in:
• Pigmented villonodular synovitis
• Inflammatory diseases
• Rare pathologies: chondromatosis, osteochondromatosis,
haemangiosclerosis, desmoid tumors
• More specific synovitis
It must be stressed that a synovectomy is particularly indicated in the young patient with no cartilage loss. For this
reason, it is rare for a patient with advanced arthritis secondary to an inflammatory arthropathy to require a synovectomy.
A “total” synovectomy in the strictest sense is an overstatement since the configuration of the knee joint is very complex and renders a “total” synovectomy virtually impossible.
It would be more appropriate to speak in terms of a “reduction” synovectomy.
The lesions that have not been eradicated during the
total synovectomy are better treated with adjuvant chemotherapy or radioactive isotopes. The surgeon must therefore
balance the advantages of a total synovectomy with respect
to the surgical “goal.” In order to make decisions, some
objective parameters have to be taken into account. To do
so, the surgeon has to know each of the different surgical
approaches to the knee. He can then select one or more
approaches depending on the particular situation. This het-

P Archbold · A Pinaroli
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

erogeneity of treatment options makes analysis of the
results even more difficult; however, they also make this
type of surgery more interesting.

Preoperative Planning
A diagnostic MRI (with gadolinium injection) is useful
in locating the lesions. These images help the surgeon’s
decision-­making process in planning how to address each of
the individual lesions. Perhaps most importantly, they help
evaluate whether there is extra-articular extension of the disease in the soft tissues posteriorly. The size and the location of
the lesions will define the surgical technique chosen (arthroscopy or open surgery) and the surgical approaches needed.
Plain radiographies of the knee including anteroposterior
(AP) view, single-leg weight-bearing view (AP and lateral),
schuss (flexed PA) view, and a skyline view of the patella (at
45° of flexion) are mandatory. They help to assess any joint
space narrowing and identify any bone lesions.
Other imaging techniques have little importance in the
detection of lesions except for maybe the arthro-CT, which
has the advantage of showing the articular cartilage in great
detail. If extension in the proximity of vascular structures is
suspected, an MRI angiogram can be of great help. In those
specific situations, the availability of a vascular surgeon at
the time of the intervention is required.

Surgical Techniques
Arthroscopic Synovectomy
Limited Synovectomy
This intervention is indicated in focal PVNS. For the arthroscopy, the patient is positioned as for general arthroscopy (see
Chap. 2).

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_11

115

116

P Archbold et al.

A tourniquet is positioned high on the thigh and inflated
to 300 Hg. Its use is helpful for visualization during the
arthroscopy since resection of synovitis is typically associated with significant intra-articular bleeding. Use of intravenous tranexamic acid can be considered.
The arthroscopic portals are chosen depending on the
location of the lesions. The frequently used portals are
described in detail in the chapters on arthroscopy and
meniscectomy.
For lesions in the suprapatellar pouch or in the notch, the
anterolateral and anteromedial portals are frequently sufficient. In order to access lesions situated in the femoral gutters,
a superolateral or superomedial portal is necessary. For lesions
situated more posteriorly in the knee, that is, behind the PCL
or behind the femoral condyles, a posterolateral or posteromedial portal is necessary as described by Philippe Beaufils.
A posteromedial portal is made with a knee in 90° of flexion. The entry point of this portal is made at the edge of the
posteromedial border of the condyle, 1 cm above the joint line.
This portal allows visualization of the posterior aspect of the
medial condyle and the medial portion of the distal PCL. The
round-tipped obturator is then reintroduced into the sleeve,
gently perforating the synovial tent covering the PCL while
staying in contact with the posterior condyles. The arthroscope
is reintroduced into the sleeve, and the posterior part of the
lateral condyle can be visualized. Transillumination of the
posterolateral region is of major importance: the portal has to
be ventral to the biceps tendon to eliminate the risk for nerve
damage. The posterolateral entry portal can be made with an
11 blade in 90° of flexion following spinal needle localization
(Fig. 11.1). The posterolateral portal allows the use of a shaver
(Fig. 11.2). Shaving of the lesions in this region can thus be
done under arthroscopic control (Fig. 11.3). Radiofrequency
ablation (coblation) is a useful option to limit bleeding.

The instruments used in this technique include a 30°
arthroscope, a biopsy grasper, and a 5.5 shaver blade. A 70°
arthroscope can be of help in the case of a posterior synovectomy. The use of an arthroscopic pump (pressure set at
40 mm Hg) helps in visualization due to the possibility of
excessive intra-articular bleeding.
When localized PVNS is suspected, we recommend performance of several synovial biopsies in order to confirm
that the PVNS is not diffuse.

Fig. 11.1 Posterolateral entry portal with an 11 blade

Fig. 11.3 Shaving of the posterior compartment synovium

Total Synovectomy
This surgical approach can be used for diffuse forms of
PVNS and for nonspecific synovitis without extra-articular

Fig. 11.2 Shaver used through the posteromedial portal (posterolateral
portal view)

11

Synovectomies of the Knee

involvement. The volume of the lesions must be fairly
limited.
The setup and instrumentation are identical to that
required for a “limited” synovectomy. Four portals are used:
anterolateral, anteromedial, superolateral, and superomedial,
which allow for a total synovectomy in the anterior compartment of the knee using a shaver (Fig. 11.4).
In PVNS, the technique introduced by P. Beaufils is of
particular interest when the lesions are situated just posteriorly to the PCL without extra-articular involvement into the
popliteal fossa (Fig. 11.5).

117

Open Synovectomy (with Arthrotomy)
Limited Synovectomy
This technique can be used for focal PVNS. The arthrotomy
allows for a complete “en bloc” resection of the lesions.
Many of these lesions could also be addressed arthroscopically. Nevertheless, some areas in the knee are more difficult
to access arthroscopically, and in some cases, the surgeon
could be uncertain about the completeness of the excision
(because of the limited accessibility by instruments but also
because of the difficultly in interpreting the arthroscopic
images). Furthermore, the arthroscopy does not allow an “en
bloc” resection of the lesions. Yet, an arthroscopic approach
has the advantage to allow a complete exploration of the
joint cavity and to perform multiple biopsies. These considerations frequently justify a combination of both techniques.
The surgical approach largely depends on the location of the
lesion(s) (Figs. 11.6 and 11.7). Each common approach to
the knee can thus be used, always taking into account the
existence of previous skin incisions.
Total Synovectomy
The patient is placed in a supine position with a vertical lateral and a horizontal distal post. A tourniquet is inflated high

Fig. 11.4 Synovectomy of the anterior compartment

Fig. 11.5 Synovectomy of the posterior compartment using P. Beaufils
technique

Fig. 11.6 Cutaneous incision after open limited synovectomy

118

P Archbold et al.

Fig. 11.7 Resection of localized pigmented villonodular synovitis

on the thigh. This technique is used for diffuse PVNS and
nonspecific synovitis without involvement of the popliteal
fossa. We typically use an anteromedial and an anterolateral
skin incision more proximal (Fig. 11.8) with enough distance
to prevent skin necrosis. This method avoids one single large
midline incision that would require significant subcutaneous
dissection, resulting in an increased risk for skin necrosis.
Arthrotomies posterior to the medial and lateral collateral
ligaments allow for intra-articular access posterior to the
PCL. Posteromedial lesions situated between the medial head
of the gastrocnemius muscle (MGM) and the semimembranosus are accessible by the medial approach. The posteromedial and posterolateral approaches not only give intra-articular
access but also allow extra-articular lesions to be addressed.
The Anteromedial Approach
An anteromedial vertical skin incision is made with a knee in
90° of flexion starting 1 cm proximal to the patella and extending distally to a point just medial to the tibial tuberosity.
The length of incision is about 8–10 cm. An extended
resection of the suprapatellar pouch can thus be done using a
scalpel with the knee in extension and the extensor apparatus
retracted with a Farabeuf retractor. (For a complete resection
of this area, it is advised to combine this approach with an
anterolateral approach.) A Volkman retractor is used to
retract the anteromedial capsule. A synovectomy of the
medial gutter can also be performed with the help of a large
grasper (Fig. 11.9). At first sight, this technique could seem
rather imprecise and less elegant, but in our hands, it is considered very efficient and reproducible. Synovial tissue is
easily caught without any resistance between the teeth of the
instruments while ligamentous and capsular tissues are much
tougher. The scalpel has to be carefully manipulated in the
presence of the cruciate ligaments and collateral ligaments,
but allows easy identification of the different planes of dissection and allows one to quickly proceed with an “en bloc”

Fig. 11.8 Cutaneous incision after open total synovectomy

Fig. 11.9 Synovectomy of the medial gutter performed with a large
grasper

11

Synovectomies of the Knee

119

Fig. 11.10 Medial parapatellar arthrotomy and intercondylar notch view

Fig. 11.11 Posteromedial anatomy

resection. The synovectomy underneath and above the
meniscus can be performed using the same approach and the
same instruments but care has to be taken not to damage the
cartilage. The intercondylar notch is easily accessible using
this approach. A specific patellar retractor can reflect the
patellar tendon and the extensor mechanism laterally
(Fig. 11.10). The same instruments can be used to perform a
synovectomy of the Hoffa fat pad and the cruciate region
(arthroscopic grasper).

Approach with Subcutaneous Undermining (Preferred
Technique)

The Posteromedial Approach
Direct

The skin incision is vertical with the knee in 90° of flexion
and centered on the posterior border of the medial condyle
(Fig. 11.11). The anteromedial arthrotomy can be used to
palpate the posterior border of the medial collateral ligament, and thus, the exact location of the posteromedial skin
incision can be determined. The skin incision extends from
the superior and posterior borders of the medial condyle to
the tibial insertion of the semimembranosus approximately
8–10 mm distally to the joint line. The anatomic advantages
of this skin incision are disputable. Moreover, lesions of the
saphenous nerve and its branches are frequently noted.

This approach requires an extended skin incision 3 cm proximal to the superior edge of the patella to 2 cm distal to the
regular medial approach. This skin and the subcutaneous fat
are elevated and retracted using a Farabeuf retractor. The
knee is now placed in a 90° flexed position in the figure-of-­
four position. The posterior border of the medial collateral
ligament serves as a landmark as does the posterior edge of
the tibial plateau and the medial femoral condyle. The vertical arthrotomy is performed just posteriorly to the medial
collateral ligament and stops just superior to the medial
meniscus. One has to take care not to cut the posterior corner of the medial meniscus. For an extended exposure, it is
sometimes necessary to release the posterior capsule for a
couple of millimeters from the femoral condyle. By doing
this, it is possible to obtain a good view in the posteromedial
compartment. At the end of the procedure, the released capsule can be reattached using an anchor or suture. The synovectomy in the posteromedial compartment and posterior to
the posterior cruciate ligament is performed with the use of
a large grasper. The knee is always kept in a flexed position
and in varus in order to maximally relax the posterior capsule (Fig. 11.12).

120

Fig. 11.12 The synovectomy in the posteromedial compartment and
posterior to the posterior cruciate ligament is performed with the use of
a large grasper

Anterolateral Approach
With the knee in a 90° flexed position, a second anterolateral
skin incision is performed starting from the mid part of the
patella and extending 10 cm proximally in the direction of the
anterior border of the iliotibial band (ITB) (Fig. 11.13). In
order to avoid skin necrosis, the anterolateral and anteromedial skin incision should be at least four finger widths apart. It
is also advisable to place the anterolateral (superolateral) skin
incision proximally and the anteromedial (inferomedial) skin
incision more distally to avoid skin necrosis. If a posterolateral
arthrotomy is planned and the skin is to be undermined, the
anterolateral skin incision should be made somewhat more laterally, that is, on the middle part of the ITB. An extended lateral approach necessitates longer skin incisions notably at the
height of Gerdy’s tubercle distally that could compromise the
cutaneous vascularization in the prepatellar and infrapatellar
regions. The lateral parapatellar arthrotomy starts distally at
the inferior pole of the patella and goes up 3 cm in a vertical
direction always staying lateral to the midline. The lateral condyle, the superior border of the lateral meniscus, and the lateral portion of the infrapatellar fat pad can thus be visualized.
With the knee in full extension, an incision in the quadriceps tendon allows for visualization of the suprapatellar
pouch. The complete extensor mechanism can now be
lifted with a Farabeuf retractor (Fig. 11.14) or a specific
patellar retractor. A synovectomy can now be performed
in the same way as with a medial parapatellar arthrotomy.
In order to have access below the meniscus, the skin incision has to be extended distally (always keeping attention
to the minimum required distance between both skin incisions), and then a small horizontal arthrotomy is performed underneath the meniscus. This synovectomy on
the undersurface of the meniscal body and on its posterior

P Archbold et al.

Fig. 11.13 Anterolateral approach and ITB exposure

Fig. 11.14 Suprapatellar pouch exposure through a lateral parapatellar
approach

part are more difficult with this open approach than with
an arthroscopy.
Posterolateral Approach
Direct

In combination with an anterolateral approach, a direct vertical arthrotomy just posterior to the lateral collateral ligament
and superior to the lateral meniscus can be performed. The
exact location of the lateral collateral ligament can be found
with the knee in the “figure-of-four” position. The skin incision is now performed with the knee in 90 degrees of flexion
and neutral rotation. Additional transillumination with the
arthroscope can provide some additional help. A skin inci-

11

Synovectomies of the Knee

121

Fig. 11.16 Dissection of the lateral collateral ligament

Fig. 11.15 Posterolateral anatomy

sion is made from the posterosuperior border of the lateral
condyle vertically down to tibial plateau always staying just
posterior to the lateral collateral ligament (Fig. 11.15).
Approach with Subcutaneous Undermining (Preferred
Technique)

With the ITB as landmark, the posterolateral compartment
can be accessed above, through, or under the ITB.
Access to the Posterolateral Capsule

The posterolateral knee capsule can be accessed through the
ITB (incising it in the direction of the fibers) or just anterior
to it. The lateral collateral ligament and the anterior border of
the lateral head of the gastrocnemius muscle can be palpated
with the Metzenbaum scissors (Fig. 11.16). The knee should
always be flexed. Thus, a vertical arthrotomy can be performed just posterior to the lateral collateral ligament
(Fig. 11.17). A small artery is frequently present, and care
should be taken to achieve hemostasis. Attention should be
paid not to cut the posterior corner of the lateral meniscus or
the popliteus tendon. The latter blocks access to the tibial
plateau, especially to its posterior part and to the posterior
recess. The posterior border of the lateral tibial plateau and
femoral condyle can now be palpated. The synovectomy of

Fig. 11.17 Vertical arthrotomy posterior to the lateral collateral
ligament

the posterolateral suprameniscal recess and the posterior cruciate ligament can be performed using a large grasper
(Fig. 11.18). As for the medial side, the approach can be
extended by dissecting the posterior capsule from the femoral condyle, this provides access to the posterior part of the
lateral tibial plateau. The exposure of this zone is very difficult through the previously described approach. It can be
necessary to go posterior to the distal ITB. This allows visualization of the popliteal hiatus and the posterior border of
the lateral tibial plateau. If increased exposure is required,
incision of the lateral meniscus in the red-red zone is
necessary.
Posterior Approach
This approach is needed in diffuse PVNS with lesions posterior to the PCL or extra-articular lesions in the popliteal
fossa. In some rare cases, one can encounter localized PVNS

122

P Archbold et al.

Fig. 11.18 Posterolateral synovectomy using a large grasper

or a benign tumor located posterior to the PCL that must also
be addressed via this approach. The first surgical step is
always the anterior synovectomy, either arthroscopically or
using arthrotomies. After skin closure, the patient is placed
in the prone position. The surgical field is left in place. The
surgical team turns the lower limbs of the patient while the
anesthesia team turns the thorax and head of the patient.
Rotation of the patient is performed clockwise for a right
knee and counter-clockwise for a left knee when viewed
from the patient’s feet.
A new stocking and extremity sheet are applied, and the
flexion crease of the knee is marked by placing a skin marker
in the flexion crease and flexing the knee. The marker is then
slowly retracted while exactly marking the flexion crease. A
sterile adhesive incision drape is then applied.
The posterior approach is performed according to
Trickey. A lateral vertical skin incision of approximately
5 cm is performed just medially to the biceps tendon. It is
continued horizontally in the flexion crease from lateral to
medial to the insertion of the medial gastrocnemius muscle.
The incision is then extended distally in a vertical direction,
for about 7 cm. An angled skin incision is to be avoided. The
skin incision is performed with a knee in extension
(Fig. 11.19). With the support under the foot and the knee in
slight flexion, the subsequent surgical steps are performed.
First, one identifies the small saphenous vein, which is never
easy. The head of the medial gastrocnemius muscle is
retracted medially and its fascia is incised vertically. This
exposes the hamstring tendons superficially and the deeper
semimembranosus tendon. These tendons guide the surgeon
to the posterior area of the tibia and thus to the “safe zone”
avoiding potential damage to the neurovascular structures in
cases where the posterior border of the posterior cruciate
ligament needs to be exposed. The posterior tibia is covered
with the popliteus muscle. The neurovascular elements can

Fig. 11.19 Posterior skin incision (prone position right knee)

be retracted carefully by placing a Homan retractor in contact with the tibia. Finally a capsulotomy is performed. The
arthrotomy is vertical and extending toward the posteromedial border of the lateral condyle. In specific cases, it can be
necessary to perform a partial section of the head of the
medial gastrocnemius muscle in its tendinous part for
approximately 15 mm. This allows an improved view of the
posterior capsule. The posterior synovectomy in the zone
behind the posterior cruciate ligament can now be performed using the large grasper.
Some extensive lesions in the popliteal fossa are situated
more superficially. The popliteal neurovascular structures
should be identified and carefully retracted using a Farabeuf
retractor (Fig. 11.20). The posterior extra-articular lesions
are carefully dissected (Fig. 11.21). When in close contact
with the vascular structures, the assistance of a vascular surgeon is sometimes advisable, especially in revision operations with adhesions that require exploration both medial and
lateral to the neurovascular structures. We routinely ask for
their assistance when the lesions extend laterally to neurovascular bundle.

11

Synovectomies of the Knee

Fig. 11.20 Dissection of tibial nerve and popliteal blood vessels

123

Fig. 11.21 Resection of a large posterior lesion

Combined Synovectomies
The indications for this surgery is typically a revision operation for diffuse PVNS with limited lesions in the anterior
compartment but extensive posterior lesions or lesions not
accessible by arthroscopy. The first surgical step is an
arthroscopy of the anterior compartment as described previously. It allows for arthroscopic evaluation and multiple
biopsies. After closure of the arthroscopic skin portals, the
patient is turned in a prone position as described previously,
and a Trickey approach to the posterior compartment is
performed.

Total Knee Arthroplasty
Placement of a total knee arthroplasty offers the opportunity
to perform an extensive synovectomy.
This surgical intervention is performed using a single
classic approach to the knee. The bony cuts allow a more
extended exposure to the different compartments. Only in
cases of extensive extra-articular lesions, an additional
approach may be necessary. A primary total knee arthroplasty can be justified in the elderly with articular destruc-

tion but is less frequently performed in PVNS of the knee
compared to that of the hip. Nevertheless the exposure during the TKA does not allow removal of extra-articular
pathology.

Postoperative Care
Because of the risk of extensive intra-articular bleeding after
a synovectomy and hence increased risk of skin necrosis,
anticoagulants are avoided in our practice. To prevent VTE,
weight-bearing, mobilization of the ankle and foot, and early
ROM exercises are encouraged. We routinely request ultrasound in order to confirm the absence of DVT in the early
postoperative period. Postoperative stiffness is a well-known
complication in this type of surgery. In order to limit the risk
after a total synovectomy, the position of the knee is frequently changed from extension to flexion using a specifically designed flexion brace. The knee is held in this flexed
position during 1 or 2 h every 6 h.
Continuous passive motion is allowed after 4 or 5 days,
when the risk of bleeding has decreased. Time should always
be taken to look for any signs of skin problems.

124

P Archbold et al.

Specific Cases
Pigmented Villonodular Synovitis
The aim in the treatment of this type of pathology should be
a single procedure involving a total synovectomy with multiple biopsies (Fig. 11.22). Limited PVNS does not need specific postoperative surveillance. In diffuse PVNS, we prefer
a total synovectomy with an arthrotomy. In case of surgical
failure, a chemical synovectomy can be performed 4–6 weeks
after the surgical intervention. Recurrence should be looked
for during follow-up (clinical examination and MR imaging
every year during the first 3 years following surgery and following this if there is any clinical suspicion of recurrence).
One must remember that findings at pathology (clean margins) do not predict recurrence. Recurrence rates decrease
with the passage of time, but the surgeon must carefully follow these patients for many years.

Primary (Osteo) Chondromatosis
Arthroscopic treatment of this pathology is in most cases
effective. In contrast to our aggressive treatment for PVNS,
we believe that the symptomatic, limited treatment of primary osteochondromatosis is appropriate. Of course, one
should take into account the extent of the pathology in every
patient. The typical grains of rice can be attached to the synovial tissue (early stage of primary osteochondromatosis)
necessitating a debridement of the synovial tissue using the
shaver. In a later stage, joint lavage and a limited synovec-

Fig. 11.22 Resected pigmented villonodular synovitis

tomy in those regions of synovial hypertrophy can suffice.
This type of surgery can be repeated if needed, and in many
cases, the extent of the patients’ symptoms diminish over
time. Malignant transformation is rarely described in the literature. It is very important to identify and debride the
regions underneath the lateral and medial meniscus. In order
to remove the grains of rice that are located in the popliteal
hiatus, manual pressure in the posterolateral region of the
knee and the popliteal fossa is applied while repeatedly flexing and extending the knee. This forces the grains of rice into
the knee joint from which they can be removed easily.

Surgical Management of Chondral
and Osteochondral Lesions

12

P Archbold, T Aït si selmi, C Bussière, P Neyret,
and C Butcher

Basic Principles
Different surgical techniques exist for the treatment of chondral
and osteochondral lesions. A distinction has to be made between
those techniques that debride or microfracture the subchondral
bone in an attempt to stimulate a healing response, and those
that transplant cartilage. Cartilage transplantation encompasses
the transplantation of osteochondral grafts or a chondrocyte cell
suspension. Techniques may also be thought of as palliative
(e.g., chondroplasty), reparative (e.g., osteochondral grafting),
or restorative (e.g., autologous chondrocyte implantation).
Microfracture and abrasion will result in the formation of a
fibro-cartilaginous repair tissue with biochemical and biomechanical characteristics inferior to those of articular cartilage.
This fibrocartilage is characterized by an extracellular matrix
which mainly consists of type I collagen rather than type II collagen, and an absence of differentiated chondrocytes. The aim
of chondrocyte cell transplantation is to reproduce a hyalinelike cartilage with differentiated chondrocytes and an extracellular matrix rich in type II collagen and proteoglycans. The
future of chondral restoration is likely to involve techniques
using mesenchymal stem cells and gene therapy.

Diagnosis and Preoperative Planning
Multiplanar imaging is mandatory to visualize the lesion,
localize it, measure its depth (grade III or IV lesion according to ICRS specification), differentiate between chondral
and osteochondral lesions, and to evaluate its size. These
P Archbold · T Aït si selmi · C Bussière
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

variables will define which type of graft as well as the surgical approach is necessary. An arthroCT scan was the imaging
modality of choice, but a 3 T (and now 7 T) MRI gives excellent assessment. The addition of intra-articular gadolinium
(arthroMRI) may give further morphological information.
These examinations not only allow evaluation of the articular
lesions, but help confirm that sufficient meniscal tissue
remains. Confirmation of the integrity of the ligamentous
structures is essential by clinical examination and aided by
MRI. Preoperative planning also includes plain radiographs.
A “schuss” view allows visualization of possible kissing
lesions, which will preclude cartilage restoration techniques.
A long leg film evaluates the axial alignment and will indicate the necessity for associated osteotomy. For posteriorly
located lesions, a lateral radiograph in knee hyper-flexion
will indicate if the lesion is accessible.

Surgical Techniques and Indications
The majority of the techniques described in this chapter are
performed on a regular basis in our department. Techniques
such as chondrocyte transplantation are only performed in a
clinical research setting.
Indications for the different techniques continue to
evolve. Location, size, and depth are primary indicators.
Patellar lesions respond less well to osteochondral grafting, and chondrocyte implantation methods may be more
applicable. Smaller defects of the femoral condyles around
2 cm2 may do well with osteochondral grafting. Larger
lesions around 2–4 cm2 may be best treated with chondrocyte implantation, and even larger ones by osteochondral
allografts. Thus fortunately the majority of lesions, which
are small, can be treated in most units around the world
with simple equipment, whereas the less common larger
lesions demand treatment with procedures that require
more logistical support.
Other factors to be included in the algorithm for treatment
include age, weight, and activity level, as well as alignment,

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_12

125

126

state of the menisci, and stability. For the majority of patients,
a technique providing durable and complete surface restoration would be ideal. This is not always feasible or appropriate however, and in higher level professional sports players,
the need for rapid return to play may counterintuitively discourage the use of the ideal restorative techniques.

Perforations of the Subchondral Bone
Numerous methods exist to perforate the subchondral bone.
They are probably indicated for small lesions. However, it is
important to know that the success of cartilage transplantation
techniques for cases where the marrow stimulation techniques
that have failed may be inferior to their success in virgin cases.
• Pridie drilling: the perforations of the subchondral bone
are performed using a drill. After debridement of the
lesion, multiple perforations with an interval of 2–3 mm
are made using a 2.0 drill. The depth of the perforations is
approximately 15 mm. After the surgery, the tourniquet is
deflated to verify bleeding from the perforations.
• Microfracture according to Steadman: these perforations
are performed using a microfracture awl.
• Abrasions: this technique performs an abrasion of the subchondral bone using a high speed burr, or coblation with a
frequency and irrigation that is adapted to cartilage.

Osteochondral Grafting and Mosaicplasty

P Archbold et al.

Surgical Technique
The patient is placed in a supine position. A vertical lateral
and a horizontal distal post are positioned and a tourniquet
is used. The surgical approach depends on the location of
the lesion. Most frequently, an anteromedial parapatellar
arthrotomy is used. In the case of a lateral lesion, a lateral
parapatellar approach is used. This approach can be associated with an osteotomy of the tibial tubercle to gain
access to the posterolateral compartment. The arthrotomy,
either medial or lateral, is performed in a subvastus fashion. The knee joint is systematically explored for associated lesions.
In the first step, the bottom and edges of the lesion are
debrided. This step allows evaluation of the dimensions and
the depth of the lesions. The number of plugs needed and
their diameter can then be chosen. The osteochondral grafts
should cover at least 70% of the lesion.
In the second step, the lesion area is prepared using a specific calibrated drill. The direction of the drill hole is perpendicular to the articular surface. The depth of the hole should
be 15 mm in case of a chondral lesion and 25 mm with an
osteochondral lesion (OCD) (Fig. 12.1).
Using a tubular harvester, the first osteochondral plug is
harvested in the donor area. The primary donor area is the
medial trochlea followed by the lateral trochlea or the
intercondylar notch area. Again the direction of the harvester should be perpendicular to the articular surface
(Fig. 12.2). The harvester is calibrated so the correct depth

Following the first founding symposium of the International
Cartilage Repair Society in Fribourg in 1997, we started this
technique, with the support of R. Jakob.

Principles
Mosaicplasty encompasses the transplantation of osteochondral plugs for the treatment of chondral and osteochondral
lesions. This technique was originally described by Matsusue
(1981) and was popularized by L. Hangody during the 1990s.
In the English literature, Vladimir Bobic from the UK is one
of the leading authors. The osteochondral autografts are harvested from the medial or lateral border of the trochlea or in
the intercondylar notch. Small osteochondral grafts transplanted into the lesions and arranged in a mosaic-like fashion. Although initially described for the treatment of femoral
condyle lesions, this technique has been extended to other
joints. Donor areas now also include the contralateral knee
and the proximal tibio-fibular joint (J. Espregueira-Mendes).
Allograft is an alternative source.
Fig. 12.1 Mosaicplasty—specific calibrated drill. The direction of the
drill hole is perpendicular to the articular surface

12

Surgical Management of Chondral and Osteochondral Lesions

127

Fig. 12.2 Mosaicplasty—first osteochondral plug harvested using a
tubular harvester in the donor area

Fig. 12.4 Mosaicplasty—graded adjustable plunger

Fig. 12.3 Mosaicplasty—osteochondral plug sizing (length 15 mm,
width 4.5 mm)

Fig. 12.5 Mosaicplasty—insertion of the osteochondral plug in the
acceptor tunnel

of osteochondral plug can be acquired. The plug is measured to confirm the longitudinal dimension (Fig. 12.3).
Harvesting the osteochondral plug is done by rocking the
harvester or by rotating it. This depends on the type of instrumentation. The graft is extracted from the harvester by gently taping on the osseous end of the graft or using an adapted
pusher through the harvester.

Dilatation of the acceptor drill hole is done with a calibrated dilator. The insertion of the osteochondral plug in
the acceptor tunnel is performed using a graded adjustable
plunger (Fig. 12.4). This allows exact control of progression and final depth of the osteochondral plug in the acceptor tunnel without the application of undue force
(Fig. 12.5).

128

P Archbold et al.

Fig. 12.6 Mosaicplasty—final aspect. The levels of the osteochondral
plugs are in line with the adjacent cartilage. Note the trochlear donor
site

Fig. 12.7 Osteochondral allograft. Note the perfect match (appropriate
size and curvature)

The level of the osteochondral plug should be in line
with the adjacent cartilage. Too prominent a graft or too
deep a graft should be avoided (Fig. 12.6). The commercial kits available today make the technique more
systematic.

Autologous Chondrocyte Transplantation

Postoperative Guidelines

Chondrocytes Cultures

The patient should wear a brace. Weight bearing is prohibited for 45 days and thromboprophylaxis is considered for
this period. Continuous passive motion is allowed on the first
postoperative day. The patella should be mobilized. Open
and close kinetic chain exercises are prescribed. Return to
sports is allowed after 6 months.

Autologous chondrocytes are proliferated in vitro. First, a
cartilage biopsy of approximately 200 mg is harvested
arthroscopically from the medial trochlea or the intercondylar notch. The cells are isolated by enzymatic digestion of
the matrix and are subsequently cultured as a monolayer in
order to obtain the desired cell quantity (approximately 10
million cells).

Osteochondral Allograft
When addressing very large lesions (over 6 cm2), harvesting
sufficient autologous cartilage for a classic mosaicplasty is
impossible. Specific large diameter instruments are available
in order to use the same technique with allograft material, but
results are questionable. For very large lesions, a monoblock
allograft with custom dimensions is optimal. In these cases,
the donor condyle should be of the same size and curvature as
the native condyle to obtain a perfect match (Fig. 12.7).

The evaluation and preparation of lesions and the surgical
approach has been described previously. Autologous chondrocytes transplantation can address lesions up to 5 mm in
thickness.

Implantation
 echnique According to Brittberg and Peterson
T
(ACI: Autologous Chondrocyte Implantation)
The cells are transplanted as a cell suspension. To contain
the cells within the defect, a periosteal or collagen membrane is needed to cover the defect. This membrane is
sutured to the defect edges as has been described by
Brittberg (Fig. 12.8).

12

Surgical Management of Chondral and Osteochondral Lesions

129

Fig. 12.10 Cartipatch®—14 mm diameter grafts

Fig. 12.8 Autologous chondrocyte implantation (ACI, Brittberg, and
Peterson)

Fig. 12.9 Three-dimensional collagen matrix (Geistlich)

Cell transplantation with the use of a three-dimensional matrix improves chondrocytes re-differentiation
and thus ensures the correct production of different extracellular matrix proteins. Different types of matrices are
available on the market: such as a sponge-type matrix
(Fig. 12.9) or gel (alginate and agarose gel such as

Cartipatch) (Fig. 12.10). The following is the description
of the Cartipatch technique. Although in our department
we now use a different matrix, the technique is very
similar.

Cartipatch Technique
The Cartipatch technique is very similar to the mosaic
plasty. Preoperative planning and preparation however is
mandatory. A biopsy harvest arthroscopy should be performed as well as preoperative imaging to evaluate the
lesion. According to the size of the lesions, a number of
Cartipatch grafts can be prepared. The Cartipatch graft is
available in three different diameters: 10, 14, and 18 mm.
For lesions close to the intercondylar notch, primary stability of the Cartipatch graft is obtained as long as the graft
is contained along at least two thirds of its circumference.
Specific instrumentations including calibrated drill bits are
available to prepare the recipient area (Fig. 12.11). Trial
components allow evaluation of the position and the height
of the defect with respect to the normal cartilage
(Fig. 12.12). The graft is subsequently introduced into the
prepared defect using a needle (Fig. 12.13). The needle
will guide the positioning, evacuate the air, and temporarily fix the graft in the defect in case of multiple grafts
(Fig. 12.14). At the end of the intervention, the tourniquet
is deflated to observe possible expulsion of the graft. The
knee is then cycled to test primary stability of the grafts
within the defect. The knee is immobilized for 48 h, thromboprophylaxis is prescribed, and weight bearing is not
allowed for 45 days. Continuous passive motion of the
knee between 0 and 90° is prescribed for 1 month. Sports
are allowed after 1 year.

130

P Archbold et al.

Fig. 12.11 Cartipatch®—calibrated drill bit

Fig. 12.13 Cartipatch®—insertion of the graft using an IM needle

Fig. 12.12 Cartipatch®—trial component

Associated Procedures
Fig. 12.14 Cartipatch®—single 18 mm diameter graft

Specific associated conditions can negatively affect the outcome of a cartilage lesion. Therefore, these conditions have
to be corrected previously or concomitantly. These associated lesions include ligamentous laxity (e.g., anterior cruciate ligament rupture) and meniscus lesions. Malalignment of
the lower limb exceeding 5° can be addressed by an osteotomy and should be discussed within the context of the chondral lesions or other associated lesions. Meniscal allograft
transplantation can be discussed if the patient is deficient in
this regard.

Conservative Surgical Techniques
Fixation
A traumatic osteochondral lesion or osteochondritis dissecans (OCD) lesion can be treated with fixation to preserve
the patient’s original cartilage. Several types of fixation

12

Surgical Management of Chondral and Osteochondral Lesions

Fig. 12.15 Fixation—trans-tendinous portal, according to Gillquist

131

Fig. 12.16 Fixation—fluoroscopic control of the location and direction of the screw

device are available including sutures, pins, and screws
(absorbable or not). The common characteristic is that
these devices can be inserted completely and that they do
not harm the opposing cartilage. Herbert, or other types of
headless screws, are very useful for this. This technique can
be performed as open surgery or under arthroscopic control. During the arthroscopy, a needle inserted perpendicular to the defect surface will illustrate the trajectory and
will indicate the correct position of the portal. Very frequently, a trans-tendinous portal (according to Gillquist) is
used (Fig. 12.15). Image intensified control can help to
obtain the perfect direction and placement of the screw
(Fig. 12.16).
OCD lesions in a young child have a good chance of healing spontaneously as long as the cartilage is intact. If the
articular cartilage is breached or the fragment has become
unstable, the bony lesion has to be debrided in order to stimulate bony healing. Sometimes, it will be necessary to use
autologous bone graft to fill the underlying bony defect.
Since the bony defect and bone fragment may not be the
same size, care must be taken to avoid articular incongruity
(Fig. 12.17).

Drilling
In a number of conditions, more specifically in the osteochondritis dissecans in the child with no breach of the
articular cartilage, the bony lesions can be perforated
­
arthroscopically from within the knee joint with a 2 mm drill,
or alternatively in a retrograde fashion extra-articularly.

Fig. 12.17 Fixation—debridement and autologous bone graft are
sometimes necessary to fill the underlying bony defect

132

In the case of an unstable lesion, additional fixation can
be provided by a mosaic plug or screws.

Postoperative Guidelines
Weight bearing is not allowed for 45 days. Rehabilitation
starts on day 1.

P Archbold et al.

Pearls
Kissing lesions are a contraindication for cartilage transplantation (best identified on the Schuss view).
Conservative treatment is the treatment of choice for
OCD lesions in the young child.
If an osteotomy is considered for pain, the therapeutic
value of cartilage surgery should be questioned.

Iliotibial Band Syndrome

13

P Archbold, G Mezzadri, P Neyret,
and C Butcher

Introduction
Iliotibial band syndrome (ITBS), also known as iliotibial
band friction syndrome (ITBFS) or runner’s knee, is a common condition among athletes. It is the leading cause of lateral knee pain in long distance runners. We have frequently
observed this pathology among cyclists as well.
The pain is due to friction between the iliotibial band
(ITB) and the lateral epicondyle, enhanced by tensioning of
the ITB in single-leg stance. The origin of this syndrome is
multifactorial and aggravating factors are consistently found:
sports overuse, poor equipment (shoes and soil), and training
errors such as lack of stretching before exercise.

Treatment Strategy and Indications
Conservative treatment is usually effective and includes
sports rest/activity modification, topical NSAIDs, icing, and
physical therapy (including deep transverse massage by the
physiotherapist). Treatment continues with stretching of the
tensor fascia lata and strengthening of the gluteal muscles.
Formal therapy should be complimented by self-­rehabilitation
(learning self-stretching). Injections are sometimes used for
more refractory cases.
However, a minority of patients remain resistant to these
treatments. Surgical treatment is then discussed with these
highly motivated athletes who continue to experience lateral
P Archbold · G Mezzadri
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

knee pain which prevents their practice of sport. Surgical
treatment is generally offered after 6 months of a well-­
conducted conservative treatment.
The procedure consists of a release of posterior fibers of
the iliotibial tract facing the lateral epicondyle. The results
are generally good and the morbidity is low, with a fast return
to sport frequently observed.

Preoperative Clinical Evaluation
History and physical examination should be systematic and
detailed.
In the history, patients report lateral knee pain that occurs
under physical exercise (running) that usually forces the patient
to stop. There are no episodes of locking, instability, or effusion.
On examination, pain is reproduced with the Noble test.
The patient is placed in the supine position with the knee flexed
at 90°. Pressure is applied to the lateral epicondyle, 2–3 cm
proximal to the lateral joint line, as the knee is progressively
extended. The pain typically occurs at 30° of flexion (Fig. 13.1).
Sometimes palpable crepitus is also present.
Knee range of motion is normal without joint effusion or
ligamentous laxity. The diagnosis is mainly clinical and
requires elimination of all other potential causes of pain in
the lateral compartment of the knee.

Imaging Before Surgery
All patients should have plain radiographs of both knees and
MRI of the painful knee.
The radiographic assessment includes the anterior view,
lateral view, schuss view, and axial view of the patella at 30°
flexion. All images are generally normal in these patients.
The clinical diagnosis of ITBS is sometimes confirmed by
MRI. An area of hyper-intense signal between the ITB and

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_13

133

134

Fig. 13.1 Left knee pain reported by the patient to be adjacent to the
lateral epicondyle, and proximal to the joint line

P Archbold et al.

Fig. 13.3 Dissection of the iliotibial band

Fig. 13.4 Posterior fibers of the iliotibial band are sectioned transversely for 10 mm. The anterior fibers are left intact
Fig. 13.2 MRI, axial section left knee. The arrow shows the high signal interposed between the iliotibial band and the lateral femoral
condyle

the lateral condyle reflects an inflammatory thickening in
this area (Fig. 13.2). One can also find moderate edema of
the lateral condyle. MRI can also rule out other lesions, particularly involving the lateral meniscus.

Surgical Technique
The procedure is performed under either general or regional
anesthesia. The patient is placed in supine position with a
vertical post just lateral to the proximal thigh. The knee is
maintained at 90° of flexion with a horizontal post. A tourniquet is routinely applied.

The procedure is performed in two stages:
The first stage consists of knee arthroscopy with routine
anteromedial portal and anterolateral portals. The knee
arthroscopy is performed to eliminate and treat any alternative intra-articular source of pain. In particular, the lateral
compartment must be confirmed as normal on arthroscopic
evaluation.
The second stage consists of an open approach with the
knee in flexion. A short skin incision 15–20 mm in length is
created parallel to and at the posterior margin of the ITB, at
the level of the lateral epicondyle. The incision is carried
down through skin and subcutaneous tissue until the posterior border of the ITB is visualized (Figs. 13.3 and 13.4). A
transverse incision 10 mm in length is made in the posterior
fibers of the ITB with an 11 blade. The anterior fibers of the
iliotibial band are left intact. The incision of the posterior
fibers leads to the formation of a “V” shaped opening pos-

13

Iliotibial Band Syndrome

teriorly, which faces the lateral epicondyle. The synovium
is not opened and synovectomy is not performed.

Postoperative Care
This operation is performed as an outpatient surgery. Full
weight bearing is allowed immediately postoperatively
without immobilization. The skin sutures or staples are

135

removed between day 12 and day 15. The patient should
initiate physical therapy during the recovery period.
Anticoagulation is generally not indicated. The patient is
generally out of work for less than a week. The resumption
of sports activity takes place during the second month
postoperatively.

Part II
Surgery for Degenerative Conditions

14

Surgical Indications in the Treatment
of Osteoarthritis
P Archbold, JL Paillot, P Neyret, and C Butcher

Introduction
When conservative management of knee arthritis fails, one
of the following surgical procedures may be indicated: osteotomy, unicompartmental arthroplasty (UKA), or total knee
arthroplasty (TKA). Arthroscopy and lavage as well as
arthrodesis will not be described here. The procedure indicated is dependent on the clinical history from the patient, as
well as his or her functional complaints, motivations, clinical
examination, and the radiological findings.
An overview of the anatomic and clinical parameters is
given. The weight of each factor can vary depending on circumstances, and thus there is no true algorithm.
Anatomic factors
Stage of osteoarthritis
Analysis of the deformity
and its reducibility
Ligamentous status (frontal
and sagittal laxity)
Range of motion

Clinical factors
Weight
Age, level of activity, function
Medical conditions (diabetes,
rheumatoid arthritis, use of
anticoagulants)
Surgical history (including sepsis)

The procedure chosen by the surgeon is also influenced
by geographical factors (an osteotomy is more frequently
performed in continental Europe than in the UK or USA),
cultural factors (osteotomy more frequently in Asian and
Muslim countries, arthroplasty more frequently in English
speaking countries), educational factors (UKA is not recognized and taught as a treatment option in certain countries),
and economical factors. Prostheses are more frequently

implanted far from, and osteotomies performed close to the
equator. Today fast recovery, short hospital stay, and a wish
to return to work may also influence the decision. These
influences may originate from a number of sources including
the patient, insurers, lawyers, government, or employers.

Patient Expectations
A patient’s satisfaction following surgery is the result of the
difference between his expectations (expected functional
result) and the obtained functional result (Fig. 14.1).
This equation is therefore dependent on informing the
patient in detail of the risks, benefits, and expected outcome of the surgical procedure that is to be performed.
Importantly, this information must be adapted to the
patient’s level of understanding. Unrealistic patient expectations can be a common reason for dissatisfaction following surgery.

10
Patient’s
expectations

Patient’s
satisfaction

Functional
results

Preoperative
condition

P Archbold · JL Paillot
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

0

Fig. 14.1 A patient’s satisfaction following surgery is the result of the
difference between his expectations (expected functional result) and the
obtained functional result

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_14

139

140

P Archbold et al.

 he Concept of the Functional Envelope
T
Applied to Osteoarthritis
Fig. 14.2 shows the concept of the functional envelope,
described by Scott Dye. The X-axis represents the frequency
of the applied forces/load while the Y-axis represents the
magnitude of the applied forces/load. The area under the
curve defines the functional envelope of the knee. The upper
limit, thus defines the threshold above which a clinical reaction may be observed (discomfort, pain, swelling, stress fracture). The definition of the functional envelope remains a
theoretical concept with a large variation between individuals and over time. It thus remains difficult to determine the
individual upper and lower threshold.
Nevertheless, the profile of the functional envelope can be
modified by medication, surgery, and rehabilitation. Each
type of intervention will modify the functional envelope in a
specific way. Total knee arthroplasty will change the shape
of the curve differently to an osteotomy.
It has to be remembered that:
1. The patient has the possibility to modify his activity (or
his body weight) to re-enter the functional envelope.
2. The aim of surgery is to enlarge this envelope, either increasing the potential frequency of load, the magnitude, or both.
If the area of the envelope might be reduced by the intervention in one or other way, it has to be clearly explained to the
patient. If the patient applies excessive forces, above the
threshold, the risk for failure is increased. This concept of a
functional envelope, and the scheme, are very useful to
explain the situation and therapeutic options to the patient.

Expected Functional Outcomes
The following paragraphs are a simplification of the current
common opinion and the literature. This is of course schematic and disputable but understandable by the majority of
patients.

After osteotomy
1.
2.
3.
4.
5.
6.

Pain: pain free (95%), forgotten knee (80%).
Stability (90%).
Unlimited walking distance.
Normal stair climbing and descent.
No limp, no use of crutches, no swelling.
All sports (impact and contact) are possible but are not
recommended.
7. Full extension, flexion to 145°.
8. Slow recovery: weight bearing is not allowed until
2 months post-surgery, one to 2 days hospitalization,
return to home, functional autonomy and driving
(75 days), slow adaptation to the modified biomechanics
and degree of valgus (4–6 months).
9. Revision total knee arthroplasty is easy (see chapter on
TKA after osteotomy).
Survival rate: 70% to 10 years.
Infection rate: less than 0.5%.

After unicompartmental knee arthroplasty

Fig. 14.2 Concept of the functional envelope
(described by Scott Dye) applied to osteoarthritis.
Situations: Circle, jump from 3 m height; Square,
playing basketball; Star, sitting in chair; Diamond:
walking 10 km

Load

1. Pain: pain free or mild occasional pain (92%), forgotten
knee (70%).
2. Stability (98%).

Zone of
structural failure
Zone of
supraphysiologic
overload

Zone of
homeostasis

Normal envelope

Osteoarthritis

Prosthesis

Frequence

14

Surgical Indications in the Treatment of Osteoarthritis

141

3.
4.
5.
6.
7.

Walking distance of at least 10 km.
Normal stair climbing and descending.
No limp or use of crutches.
No swelling.
Walking on uneven terrain, hiking, skiing, tennis are
possible.
8. Full extension, flexion of up to 145°.
9. Recovery: immediate weight bearing, 1–2 days hospitalization, return to home or rehab center 2 weeks, functional autonomy and driving of a car possible 30 days
postoperatively. Outpatient surgery can be considered in
most of cases.
10. Strict surveillance during follow-up (demanding intervention for the surgeon), revision to TKA possible.
Survival curve: 90% at 10 years after medial UKA. 95%
after lateral UKA.
Infection rate: 0.5% on the 10 years postoperative
period.

After a total knee arthroplasty
1. Pain: pain free or mild and occasional pain (95%), forgotten knee (50%),
2. Stability (98%).
3. Walking distance of at least 5 km.
4. Normal stair climbing.
5. No limping or use of crutches.
6. Swelling of the knee is possible.
7. Hunting, golf, doubles tennis, gardening are expected.
8. Full extension, flexion up to 120°.
9. Slow postoperative recovery for the patient: immediate
weight bearing, 2 to 4 days hospitalization (in some specific
circumstances an out-patient surgery is proposed), rehabilitation center (3–4 weeks), activities of daily life, and driving
of the car possible 30 to 45 days postoperatively.
10. Necessity for long-term follow-up, revision TKA
possible.
Survival curve: 90% at 15 years.
Infection rate: 1.5% in the
postoperative.

10

years

period

Indications
The indication is often a compromise and it should be a
choice made by both the patient and the surgeon. For teaching purposes, we would like to remind you that it is not
always possible to have ideal indications. Sometimes, one
or more criteria will make the indications limited or
disputable.

Osteotomy
• Ideal indications.
–– Clinical exam:
Pain localized to the tibiofemoral joint line.
Normal range of motion.
Normal ligamentous status.
Non reducible deformity (Fig. 14.3a, b).
No inflammatory arthritis.
Less than 70 years old.
No obesity.
–– Radiological findings: (Fig. 14.4a–c).
Partial or complete joint space narrowing in one
compartment.
No contralateral tibiofemoral joint space narrowing or
patellofemoral joint space narrowing.
Extra-articular deformity more than 5°.
• Disputable indications:
Patellofemoral arthritis.
A “cupula”—tibial bone loss in severe osteoarthritis.
Flexion <100° or fixed flexion deformity.
Intra-articular deformity.
Age > 70 years.
Obese women.
This is the ideal indication in case of true osteoarthritis
where a hypercorrection is mandatory. This hypercorrection
is adapted to the wear of the severity of the osteoarthritis
(between 3 and 6°). The situation is different in case of osteotomy combined with meniscal, cartilage, or ligament injury
where the patient wants to return to sports. In this situation,
a normo-alignment or a moderate hypercorrection (between
0 and 3°) is performed even if the longevity of the osteotomy
is reduced (see Chap. 7 Revision ACL reconstruction).

Unicompartmental Prosthesis
• Ideal indications.
–– Clinical examination: (Fig. 14.5a–c).
Pain at the tibiofemoral joint line.
Normal range of motion.
Normal ligament status.
Reducible deformity.
Above 60 years old.
Weight limited to 80 kg.
No inflammatory arthritis.
–– Radiological findings: (Fig. 14.6a–c).
Unicompartmental partial or complete joint space
narrowing.
No contralateral tibiofemoral or patellofemoral joint
space narrowing.
No ligamentous laxity.

142

P Archbold et al.

Fig. 14.3 (a, b) Non-­
reducible deformity

a

a

b

b

c

Fig. 14.4 Full weight bearing X-rays. (a) AP view. (b) Schuss view at 45° of flexion. (c) lateral view (30° of flexion)

Reducible deformity without hyper-correction.
No frontal laxity.
Extra-articular deformity <5°.
• Disputable indications:
Asymptomatic patellofemoral arthritis.
Flexion <100°.

Extra-articular bony deformity between 5 and 8°.
Surgical history including: malunion, HTO, UKA.
Age < 60 years old.
• Contraindications:
Inflammatory arthritis.
Chronic anterior laxity or ligament insufficiency.

14

Surgical Indications in the Treatment of Osteoarthritis

a

143

b

c

Fig. 14.5 Clinical examination. (a) Mild deformity. (b) Reducible deformity. (c) No flexion stiffness

a

b

c

Fig. 14.6 X-ray findings. (a) AP view. (b) Schuss view. (c) Lateral view

Total Knee Arthroplasty
• Indication
Pain localized to the arthritic knee
Any deformity, laxity, or range of movement

the presence of contraindications for a unicompartmental
arthroplasty or an osteotomy. In our opinion, weight is not
a contraindication, and has no influence on wear (Fig. 14.7).
Early mobilization and improved preoperative management have minimized the effects of excessive weight.

The basic indication for surgery is reduced quality of
life due to the degenerative knee pathology. The decision
to proceed to a TKA is the most commonly selected surgical option in the treatment of osteoarthritis, as there are
fewer factors that predict a poor outcome. A “monoculture” surgeon is tempted to propose a TKA for the majority
of his/her patients. Others will only proceed to a TKA in

• Disputable indications.
Early osteoarthritis, where the joint space is still preserved on plain X-ray. Attempting non-operative treatment methods first will be mandatory.
Young age: although there is more logic to perform TKA in
advancing age, this is unavoidable in certain young
patients where other treatment methods are not suitable.

144

P Archbold et al.

Fig. 14.7 Obesity is not a contraindication to TKA

Fig. 14.8 Epiphyseal axis defined by Levigne

Radiological Evaluation

This View Is of Interest:
For osteotomies: it will define the origin of the deformity
(at the level of the femur or tibia) and will thus indicate the
level to perform the osteotomy, the importance of the overall
deformity and the amount of correction that will have to be
performed.
Unicompartmental knee prosthesis: will define the
deformity and will illustrate reducibility (full leg stress
X-rays).
Total knee arthroplasty: will determine the overall deformity, and possible bony defect. It will allow planning of the
femoral and tibial cuts, and therefore predict the need for soft
tissue release.
Stress radiographs in varus and valgus will illustrate intra-­
articular laxity and reducibility of the deformity.

The radiologic evaluation is the same for all three types of
intervention (osteotomies, unicompartmental knee, and total
knee replacement). It includes:

 t the Time of the Consultation (Minimum
A
Work-Up)
–– Single leg AP view: type of arthritis, location, presence of
osteophytes, cysts, foreign bodies, obliquity of the joint line.
–– Single leg lateral view at 30° of flexion: presence of a
cupule, patella height, tibial slope, anterior tibial translation, malunion with flexion deformity. This view is the
most important view for anti-recurvatum osteotomies.
–– Skyline view of the patella in 30° of flexion: to examine
the patellofemoral joint.
–– Bilateral leg stance at 45° of flexion view (schuss view).
This view is excellent to evaluate tibiofemoral joint space
narrowing that is frequently underestimated on the AP view.

Prior to an Intervention
Preoperative planning is essential. It includes:
Bilateral full leg view: allows measuring of different
angles and axes.
–– The mechanical femoral axis is represented by a line connecting the center of the femoral head and the middle of
the tibial spines.
–– The mechanical tibial axis connects the middle of the
tibial spines and the middle of the ankle joint.
–– The mechanical lower limb axis represents the overall
deformity of the lower limb.

Of interest
• Measurement of the constitutional varus.
–– Epiphyseal axis defined by Levigne: line connecting
the middle of the tibial joint line and the middle of the
line connecting ends of the tibial physeal scar. This
axis forms a constant angle of 90°±2° to the lateral
tibial plateau (Fig. 14.8). The constitutional deformity
of the tibia is defined as the angle between the epiphyseal axis and the tibial mechanical axis (Fig. 14.9).
–– Sometimes it is difficult to determine the middle of
the tibial joint line and to perform the measurement.
Therefore, we prefer to determine the level of the
original tibial plateau by the line tangent to the normal contralateral tibial plateau. Subsequently, the
mechanical tibial axis is drawn. The angle between
both axes is the angle alpha. The constitutional varus
is defined by the complementary angle 90-alpha
(Fig. 14.10).
• Measurement of the hip knee femoral angles; this will be
discussed in Chap. 25, Steps and Strategies.

14

Surgical Indications in the Treatment of Osteoarthritis

145

Fig. 14.10 The constitutional varus is defined by the complementary
angle 90-alpha

Fig. 14.9 The constitutional deformity of the tibia is defined as the
angle between the epiphyseal axis and the tibial mechanical axis

Fig. 14.11 MRI coronal
images in the same patient
showing (a). Posterior horn
root tear. (b) Meniscal
extrusion and subchondral
edema

a

 dditional Radiologic Investigations:
A
For anti-recurvatum osteotomies: two long profile hyperextension views of the lower limb. The femoral recurvatum is
the angle defined by the line tangent to the anterior cortex
and the line perpendicular to the Blumensaat line. The tibial
recurvatum is defined by the tibial slope. For both, see
Fig. 14.2a–c, Chap. 20).
CT imaging: this will determine the presence of rotational problems. Certain patients with a frontal valgus or
varus deformity develop a unilateral arthritis at the side of
the convexity of the malunion. This lateralization of the
degenerative process can be explained by the associated
rotational problem. An internal femoral rotational defor-

b

mity will cause lateral tibiofemoral arthritis, while an
external rotational deformity will cause medial tibiofemoral arthritis.
We do not use routinely low dose X-rays imaging developed by Charpak and Dubousset (EOS system) but it does
allow a precise measurement of the deformities in the three
planes.
MRI may be useful in cases of early osteoarthritis to show
evidence of AVN or overload from varus deformity, and
incompetence of the meniscus secondary to root tear and
extrusion (Fig. 14.11). It will thus play a part when the indications appear disputable and guide various surgical interventions including root repair, osteotomy, UKA, or TKA.

Osteotomy: General Concepts
and Indications

15

P Archbold, JL Paillot, P Neyret,
and C Butcher

Introduction
Before the introduction of the total knee arthroplasty into
clinical practice, an osteotomy was the treatment of choice
for osteoarthritis. Today, an osteotomy is considered technically difficult for the surgeon and demanding for the patient.
Nevertheless, in our daily practice osteotomies are an important treatment option for arthritis of the knee because they
allow a return to a high level of activities including sports.
An osteotomy delays the need for a total knee prosthesis in
young active patients. Obviously, the following variables
have to be taken into account: the type of arthritis, clinical
and radiological criteria, and patient expectations. In this
chapter, we will not discuss the criteria that make us chose an
osteotomy over a total knee prosthesis for degenerative knee
pathology, but rather which type of osteotomy is indicated in
different clinical situations.

The Goal of the Osteotomy
Once we have decided what type of osteotomy to perform,
we need to decide the exact goal of the procedure beforehand. Much work has gone into improving the accuracy of
osteotomy, including better preoperative imaging and measurement, and patient-specific guides and computer navigation. But consistently hitting a suboptimal target will not
achieve the clinical success that is being sought (Figs. 15.1,
P Archbold · JL Paillot
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

15.2, and 15.3). Overcorrection is poorly tolerated and risks
degeneration in another compartment, and undercorrection
may lead to early failure of the procedure. Defining the ideal
anatomical goal of the correction in each individual, however, is a complex and evolving subject. Broad guidelines
have been established; for instance, the correction to 3–6°
valgus after osteotomy in medial femorotibial osteoarthritis.
However, logic tells us that these figures need fine tuning
to the individual’s circumstance. In addition to recommendations defined by research into lower limb biomechanics,
there are many clinical parameters which may suggest a
particular target for each patient. These include anatomic
factors such as the weight, torsional profile of the lower
limb, and knee laxity, as well as general factors such as the
age, and level and type of activity the patient aspires to. In
the past, experience has been the backbone of the art form
called osteotomy. Gradually, we are learning the science
behind the art.

Type of Arthritis
Medial Osteoarthritis
Certain factors support the use of a tibial osteotomy:
–– the origin of the varus in medial gonarthrosis is usually on
the tibial side and is usually in the proximal metaphyseal
region.
–– the clinical outcome of an osteotomy in medial osteoarthritis is reported to be good, reliable, and durable with a
survivorship of approximately 70% at 10 years.
–– an osteotomy restores the morphology with a horizontal
joint line.
–– technically, the objective of this procedure is to obtain
an overcorrection between 3 and 6° of valgus, as measured on the mechanical tibiofemoral angle between
183° and 186°.

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_15

147

148

Fig. 15.1 Consistent and accurate achievement, but incorrect target.
Poor clinical results

P Archbold et al.

Fig. 15.3 Correct target, consistent and accurate achievement. Good
clinical results

Opening Wedge Tibial Osteotomy
Advantages:
–– very precise correction
–– fewer problems with the peroneal nerve
Disadvantages:
–– need for bone graft with large corrections, consolidation
is more difficult (8–10 weeks)
–– tensioning of the extensor system and to a less degree the
medial collateral ligament and medial tendinous
structures
We prefer an opening wedge high tibial osteotomy in the
young patient with preosteoarthritis or limited osteoarthritis.
Fig. 15.2 Correct target, but inconsistent and inaccurate achievement.
Variable clinical results

Closing Wedge Tibial Osteotomy
Advantages:
–– easy consolidation (7–8 weeks)
–– natural tendency to decrease posterior tibial slope

15

Osteotomy: General Concepts and Indications

149

Disadvantages:
–– peroneal nerve at risk
–– more variability in the obtained correction

 pening Wedge Femoral Osteotomy
O
Since the origin of the valgus knee is often situated in
the distal femur, an osteotomy of the distal femur seems
logical. Nevertheless, we have to understand that a correction by osteotomy is only obtained in the frontal plane,
in extension (P. Chambat). The anatomy and alignment is
not changed in flexion and thus a valgus knee will persist
in flexion after a distal femoral osteotomy. Therefore, the
indication for a distal femoral osteotomy is a valgus knee
in extension (Figs. 15.4, 15.5, and 15.6). If the knee is
well aligned in extension but a joint space narrowing is
observed on the tunnel view, the options for treatment are
a medial high tibial closing wedge osteotomy or a unicompartmental prosthesis. For the moment, we believe
that the classification of the valgus knee according to the
origin of the deformity is not yet well understood and
that deformities at the level of the diaphysis are not yet

We prefer a closing wedge high tibial osteotomy in the
somewhat older patient with advanced osteoarthritis. In case
of evolving osteoarthritis secondary to chronic anterior laxity, this is the technique of choice.

Lateral Osteoarthritis
–– this type of OA is of mixed origin both in the femur (hypoplasia of the lateral femoral condyle) and in the tibia.
–– the clinical outcome is less reproducible.
–– we aim for a correction between 0 and 2 degrees of
varus.

Fig. 15.4 A correction by femoral osteotomy is only obtained in the frontal plane, in extension (P. Chambat)

150

P Archbold et al.

included. A distal femoral osteotomy requires rigid fixation is associated with more blood loss and has a high risk
for arthrofibrosis.
We generally perform a distal femoral osteotomy in
younger patients with a valgus of distal femoral origin. The
patients should be well motivated.

 edial Closing Wedge Tibial Osteotomy
M
This type of osteotomy on the contrary will have an effect
both in extension and in flexion. It is indicated and justified in those valgus knees of a mixed origin. However, it is
accompanied with a risk of significant obliquity in the joint
line. This obliquity, if superior to 10°, can generate excessive
stress on the patellofemoral joint, especially on the medial
side. We propose a medial closing wedge high tibial osteotomy for the patient around 60 with a high level of activities including sport, with a valgus knee of mixed origin or of
tibial origin which is less than 8°.

Clinical and Radiological Criteria
Fig. 15.5 After femoral osteotomy, in extension the valgus has been
corrected

 ge
A
In a young patient with limited or early medial gonarthritis,
we prefer an opening wedge high tibial osteotomy.
Weight
Morbid obesity has a negative influence because of both loss
of correction in the osteotomy and of difficulties during the
non-weightbearing period.
 rthritis Secondary to ACL Rupture
A
Because the wear pattern is located more posteriorly on the
tibial plateau (due to the ACL rupture) decreasing the tibial
slope will limit the anterior tibial translation. Therefore, a
closing wedge high tibial osteotomy is more appropriate.
Origin of the Deformity
–– if extra-articular (constitutional or malunion), the osteotomy is considered “corrective” since it will correct the
bony deformity.
–– if intra-articular (wear), the osteotomy is considered “palliative” because the wear deformation is compensated by
creating a bony deformity.

Fig. 15.6 After femoral osteotomy, the valgus remains in flexion

15

Osteotomy: General Concepts and Indications

Expectations of the Patient
The preoperative level of activity and the expected postoperative level of activity of the patient will influence the indications for an osteotomy. We are more likely to treat an older
patient with a high level of activity, including sports, with an
osteotomy.

Advice to Give to the Patients Before Surgery
–– adapt your home (carpets, animals, stairways) to decrease
the risk of a fall.
–– physiotherapy should be initiated preoperatively to learn
how to walk with crutches.
–– advise weight loss preoperatively (this is possible in the
young patient but it remains difficult in the older
patient).
–– advise to quit smoking since this has a negative effect on
the achievement of union and on wound healing.

I n Conclusion Our Main Indications Are
as Follows
Medial Osteoarthritis
Opening wedge high tibial osteotomy:
–– young patient.
–– early OA: stage 1 and 2.
–– specific case: combination ACL reconstruction and
osteotomy.
–– in the exceptional case of a constitutional varus knee
without OA (constitutional varus superior to 8°, if bilateral or with more than 4 finger widths of space between
the condyles). In this rare cases, the aim is to leave some
residual varus (2–3°).
Closing wedge high tibial osteotomy:
–– older patient but active
–– stage 3 and 4

151

–– patella infera
–– chronic anterior laxity with posterior wear on the tibial
plateau
Femoral osteotomy and double osteotomy are exceptional: these techniques are indicated in secondary arthritis
due to malunion, Vit D deficiency, etc.

Lateral Arthritis
Tibial osteotomy:
–– to correct abnormalities of mixed origin (femoral and
tibial) only if the obliquity of the joint line will not be
superior to 10° after your osteotomy and in a valgus knee
inferior to 8°.
–– we prefer a medial closing wedge osteotomy.
–– lateral opening wedge high tibial osteotomy with a revision osteotomy of the fibula is only indicated secondary to
an excessive lateral closing wedge high tibial osteotomy
with an overcorrection.
Femoral osteotomy:
–– valgus knee of femoral origin.
–– valgus with a fixed flexion deformity or a hyper extension of
more than 20°: this pathology can be addressed more appropriately with a femoral osteotomy than with a tibial osteotomy. However, the morbidity of the femoral o­ steotomy is
more significant and has to be integrated into the indications
flowchart in order to prevent complications.
In case of a large deformity:
–– double osteotomy combining a lateral distal femoral
opening wedge osteotomy and a medial closing wedge
high tibial osteotomy can be considered.

16

Varus Distal Femoral Osteotomy:
Lateral Opening
P Verdonk, R Magnussen, P Neyret,
and C Butcher

Introduction

a

b

In this chapter, we present the surgical steps to perform an
opening lateral wedge distal femoral osteotomy for valgus
deformity; fixed either with a 95 angled blade plate, or a
locked plate. The overall aim of this osteotomy is to correct
the mechanical axis of the lower limb to a normal varus
(0–3° of varus). In general, it is better to slightly overcorrect
than to under correct. During preoperative planning, one can
determine the desired angle of correction and the opening
that will be needed to obtain this correction.

Radiological Workup
See chapter on surgical indications for osteoarthritis.
The radiographs serve not only to determine the proper
indications but also to measure the correction needed
(Figs. 16.1, 16.2 and 16.3). A torsional deformity is likely if
the valgus is due to femoral fracture.

Surgical Technique: 95° Plate
With the knee in 90° of flexion, a lateral skin incision starts
15 cm proximal to the joint line and ends at a level of Gerdy’s
tubercle (Fig. 16.4). The fascia lata is incised slightly anteriorly
in the direction of its fibres and the lateral vastus muscle is elevated. The perforating arteries of the vastus lateralis are care-

P Verdonk · R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire 69300, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

Fig. 16.1 Schuss X-rays (a) have better sensitivity for osteoarthritis
diagnosis than full extension X-ray (b), particularly for lateral femorotibial OA

fully coagulated or ligated. Subsequently, the vastus lateralis is
elevated from the intermuscular septum and the lateral border of
the femoral diaphysis, using a periosteal elevator. The patella
tendon is identified and a limited lateral arthrotomy is performed: this is to visualise the orientation of the trochlea and the
condyles. Two guide pins are inserted into the joint: one at the
femoro-tibial joint line, another in the patello- femoral joint
(Fig. 16.5). The guide pins act as a guide to help orient the surgeon to accurately place the blade plate. This step reduces the
radiation due to imaging. Next, the osteotomy site is prepared.
The osteotomy is horizontal, just proximal to the lateral part of
the trochlea. An additional anterior coronal osteotomy may be
added to increase stability. With the knee in extension, the suprapatellar pouch is elevated and with the knee at 90° of flexion soft
tissues on the posterior side of the metaphyseal region are elevated. With the oscillating saw, a landmark is made on the lat-

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_16

153

154

P Verdonk et al.

eral side of the femur perpendicular to the horizontal osteotomy.
This mark will serve as a guide to determine the rotation following osteotomy (Fig. 16.5).

Introduction of the Blade

Fig. 16.2 Long leg films (full weight bearing) are required to measure
the femoro-tibial mechanical axis, the femoral mechanical axis, and the
tibial mechanical axis and to diagnose a lower limb discrepancy

Fig. 16.3 In case of
rotational deformity, a
CT-Scan is required to
measure the femoral neck
anteversion/retroversion
according to the posterior
condyles line

The blade should be introduced into the epiphyseal region,
30 mm proximal to the joint line. The blade plate is 5.6 mm in
thickness, 16 mm in width, and the distance between the screw
holes is 16 mm. The guide for the blade plate should be introduced anteriorly and proximally to the femoral insertion of the
lateral collateral ligament. The angle of insertion depends on the
level of the deformity. If the deformity is situated at the diaphyseal level, the blade should be introduced obliquely to the joint
line (Fig. 16.6). To obtain a varization of 10°, the angle should
be set at 75° (85°–10°; complementary angle to the anatomical
medial distal femoral angle (95°)—angle of correction). If the
deformity is situated at the metaphyseal level, the blade should
be introduced parallel to the joint line (Fig. 16.7). This is the
most common situation. When introducing the blade parallel to
the joint line, an automatic correction to the normal anatomical
femoral valgus of 5° is automatically obtained by introducing a
95° angle blade plate. In other words, if the femur is normal, no
correction would be obtained when the blade plate is introduced
parallel to the joint line. If we are confronted with a combined
deformity or mixed with a metaphyseal component (lateral condyle hypoplasia or diaphyseal malunion), the angle of introduction should be even smaller and the blade plate should be
introduced at a smaller angle. This preoperative planning is
essential to evaluate the correction needed.

16 Varus Distal Femoral Osteotomy: Lateral Opening

Fig. 16.4 The skin incision is from 15 cm proximally to the joint line
to the Gerdy’s tubercle

155

Before

After

Fig. 16.6 Diaphyseal deformity: the blade should be introduced
obliquely to the joint line. The correction angle will be equal to the
femoral deformity angle

Fig. 16.5 Preoperative view showing the lateral cortical of the femur
(left knee). A proximal arthrotomy is necessary. Two wires (left arrows)
are inserted into the femoro-tibial joint and patello-femoral joint. A
rotational landmark is superficially done on the femoral cortex using
the saw (right arrow)

Intra-operative Control
The position of the blade can be checked using the image
intensifier. The angle of correction can now be measured on
a printout by drawing a line tangent to the medial and lateral
condyle and another line tangent to the blade.

The Osteotomy
The femoral osteotomy is performed with an oscillating saw.
The medial cortex should not be cut (the saw should “knock
on the door” of the medial cortex as said by Henri Dejour).
The blade plate is introduced, and the medial cortex is weakened using a 3.2 mm drill bit. Two or more osteotomes are
then introduced into the osteotomy. It is however the impaction of the blade plate that will progressively open up the

Before

After

Fig. 16.7 Metaphyseal deformity: the blade should be introduced parallel to the joint line. An automatic correction to the normal anatomical
femoral valgus of 5° is automatically obtained

osteotomy once in contact with the diaphysis. Temporarily, a
screw is placed in the distal part of the oval screw hole
(Fig. 16.8a). The blade plate is now impacted. The screw is
then in the proximal zone of the hole (Fig. 16.8b).
Subsequently, a screw is introduced in another screw hole
while the former is taken out (Fig. 16.8c). The impaction of
the blade plate is continued and the osteotomy will progressively open up until the blade plate is in full contact with the
lateral side of the femoral diaphysis (Fig. 16.9).
Progressive impaction allows opening of the osteotomy.
Provisional fixation with one screw helps to control the correction and gives additional stability. By playing with the

156

P Verdonk et al.

a

b

c

d

impaction and the positioning of the screws, one can either
augment or decrease the amount of opening. If the blade plate
is impacted with the screw left in place, the correction will be
halted. To the contrary, if an additional screw is again placed
in the distal part of the screw hole and the former screw is
taken out, the correction can be augmented (Fig. 16.8c). Final
fixation of the blade plate is achieved by four cortical screws
of 4.5 mm diameter (Fig. 16.10). Cortical and cancellous iliac
crest bone grafts are used to fill the osteotomy site. The soft
tissues and skin are closed over a drain, which is introduced
underneath the fascia lata.

Surgical Technique: Locked Plate
Fig. 16.8 A screw is temporarily placed in the distal part of the oval
screw hole (a). The blade plate is then impacted. The screw is in the
proximal zone of the hole (b). Subsequently, a screw is introduced in
another screw hole while the former is taken out. First situation: the
blade impaction enhances the corrective effect if the first screw is
removed (c). Second situation: the blade is impacted without removal of
the first screw stopping the corrective effect (d)

a

b

Fig. 16.9 The impaction of the blade plate is continued and the osteotomy will progressively open up until the blade plate is in full contact
with the lateral side of the femoral diaphysis

The technique is different in several aspects:
• It is possible to make a smaller approach in some cases when
using a locked plate, making a shorter incision, and only
elevating the very distal part of the vastus lateralis. The plate
can be slid under the vastus lateralis, and screw insertion performed through the skin and muscle. We prefer routinely a
loner approach in order to well position the plate along the
shaft. An arthrotomy is optional and may be directed by any
desire to treat intra-articular pathology (like a lateral partial
facetectomy). In this case, a formal but limited lateral parapatellar approach may be more appropriate.
• The osteotomy must be performed and opened before
the application of the plate. Consequently, great care
must be taken to maintain stability while opening the
osteotomy. To aid in this regard, the osteotomy may be
more oblique, aiming towards the flare of the medial
femoral condyle where the bone is less brittle
(Fig. 16.11). Subsequent opening of the osteotomy is
slow and controlled with one or two laminar spreaders,
avoiding complete fracture of the medial hinge, or a sagittal plane deformity. Once satisfactory alignment has
been confirmed with a metal rod or cord, the plate can be
applied and fixed. If using bone graft, additional stability can be achieved at this point by the addition of
wedges, applied anterior or posteriorly to counter any
tendency to anterior or posterior tilt.
• The particular plate must be analysed for screw position,
and the osteotomy started on the lateral cortex at an appropriate place to maximise the number of screws in the distal
fragment. In practice, starting just proximal to the lateral
superior geniculate vessels will be sufficient. The varying
anatomy in this region means that the contoured plate may
not fit the bone perfectly, but the temptation to prioritise fit
over screw position must be resisted. The most distal
screws may be needed to be short to avoid entering the
intercondylar notch. Our concern is the amount of correction. In fact, once the plate is perfectly adapted to the lower

16 Varus Distal Femoral Osteotomy: Lateral Opening

157

Fig. 16.10 (a, b) Postoperative X-rays showing a
95° angle blade plate with 4
proximal cortical screws. The
final femoro-tibial angle is
good

a

b

Fig. 16.11 (a, b) Postoperative X-rays showing a
locked plate. The osteotomy
is oblique, aiming towards the
medial flare of the medial
femoral condyle

a

b

158

extremity the angle of the distal shaft and the line passing
through the 2 condyles is 95° and the anatomical angle
number 1 is 90°. This plate is perfect for treatment of fracture, sequellae of supracondylar fracture or epiphysiodesis
without osteoarthritis. But it doesn’t allow for a proximal
deformity of the mid diaphysis malunion or a correction of
articular deformity. To do so one needs to bend the plate or
to use a customised plate. The current plates available (e.g.
Tomofix, Synthes) are very strong, and altering the shape is
problematic.
• Future directions
A solution could be a customised plate. The plate will be
designed in order to fit the bone after correction. We are considering this option with both locking screws and normal
screws in oval holes in order to combine the advantages of
the blade plate with a controlled progressive correction and
the rigidity of the fixation with locking screws.

P Verdonk et al.

Post-operative Guidelines
Continuous passive motion is initiated immediately post-­
operatively. The flexion should be limited to 120° for the first
15 days post-operatively. Non-weight bearing is continued
for 2 months and an extension brace is applied. Complications
are observed somewhat more frequently than after a tibial
osteotomy. Specifically, blood loss can be significant and
stiffness of the knee and delayed union are more frequent.
Complications can be minimised by careful surgical technique and adherence to a specific post-operative rehabilitation protocol and a rigid fixation.

Valgus High Tibial Osteotomy: Lateral
Closing and Medial Opening

17

R Debarge, F Trouillet, G Demey, R Magnussen, P Neyret,
and C Butcher

Introduction

3. Development of patellofemoral arthritis.

In patients who have osteoarthritis of the medial compartment of the knee in association with genu varum, a high tibial osteotomy (HTO) remains an important surgical option.
The clinical outcome at 10 years continues to be favorable in
more than 70% of the patients if the frontal angular malalignment has been corrected to 3–6 degrees of valgus.
The main reasons for failure are:

Two surgical techniques are available. The medial opening wedge HTO may require the use of a tri-cortical bone
graft from the iliac crest (or bone substitute) for large corrections and the lateral closing wedge HTO requires an osteotomy of the fibula neck. The clinical outcome is more
predictable in patients who are not obese. Therefore, we generally provide information on a weight loss program preoperatively. In the young, sports-minded patient, the osteotomy
still remains the option of choice above an arthroplasty, particularly if the deformity is from extra-articular bone deformity. We prefer until today to use a plate with locking screws
(e.g., Tomofix) in medial opening wedge osteotomies. This
technique has the benefit of not requiring a bone graft
although with large corrections it may be advisable.
We are working on the development of a customized plate
with a new and optimal design. This will allow individualized correction in all planes and the fixation will ensure better control of the rotation, despite using a smaller plate.
Radiological evaluation: See the chapter on surgical indications and osteoarthritis. In the case of evolved osteoarthritis, the
amount of opening or closing of the osteotomy needed to obtain
a valgus correction of 3–6° is calculated with respect to the
width of the tibia at the level of the osteotomy and the angular
correction needed (Fig. 17.3). In early osteoarthritis, and particularly when the patient wants to continue to practice sports,
the target is less and must be adapted (between 0 and 3°).

1. Initial under correction with the presence of a residual
varus deformity. Even if a temporary improvement is
often observed, after 3–5 years a reoperation is often
required.
2. Overcorrection with progressive lateral arthritis. In cases of
overcorrection, the adaptation of the patient to the new
alignment is very difficult and takes more than 1 year.
During this period, the patient complains not only of ankle
pain, but also feels very uncomfortable with the deformity.
Most often the patient doesn’t accept this overcorrection
and a reoperation must be proposed before this, in the
2-year follow-up period. Any distal valgus deformity, for
instance hindfoot valgus from posterior tibial tendon disorder, may exacerbate this problem. This needs to be sought
preoperatively; it may not be obvious on alignment X-rays
(Figs. 17.1 and 17.2).
R Debarge · F Trouillet · R Magnussen
Centre Albert Trillat, Lyon, France
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France

Lateral Closing Wedge HTO

P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

Set-Up

C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

The patient is placed in the supine position. A tourniquet is
generally used. The patient is draped using an extremity
sheet (Fig. 17.4). The image intensifier is positioned

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_17

159

160

R Debarge et al.

Fig. 17.2 The patient needed early conversion to TKA

Fig. 17.1 Valgus hindfoot deformity. This must be taken into account
when calculating the overall limb alignment

temporarily to ensure sufficient access to the entire limb,
including the hip. A slightly oblique, almost horizontal,
anterolateral skin incision is used. It starts 1 cm above the
anterior tibial tuberosity and proceeds laterally to 1 cm below
the fibular head (Fig. 17.5). The fascia of the proximal portion of the origin of the tibialis anterior is released as a
Z-plasty. Subsequently, the tibialis anterior muscle and the
long toe extensor muscle are released from the tibial metaphysis using a large periosteal elevator (Fig. 17.6).

Osteotomy of the Neck of the Fibula
The neck of the fibula is identified and exposed. A periosteal
elevator is slid around the neck always staying in contact with
the bone. This maneuver protects the peroneal nerve (Fig. 17.7).
Four holes are now drilled in the neck using 3.2 mm drill.
With the use of the osteotome, the four holes are interconnected
and the segment is removed using a large grasper. The two dis-

tal holes are joined first (Fig. 17.8); if the proximal holes are cut
first, the distal cut becomes difficult due to the increased mobility of the fibula shaft. Free mobility of the shaft confirms completion of the osteotomy. Care must be taken that the peroneal
nerve is not in contact with the osteotomy site.

Tibial Osteotomy
Specific instruments help perform the osteotomy and achieve
its fixation in a reproducible way. The osteotomy is performed proximal to the tibial tubercle in an oblique direction
in both coronal and sagittal planes.
Identification of both starting point and direction of the
osteotomy with imaging is not necessary if the following
rules are respected.
–– Laterally, the osteotomy should start distal to the proximal tibiofibular joint and should cross the tibia proximal
to the tibial tubercle. In this direction, there is no danger
to the tibial plateau (Fig. 17.9).
–– The patellar tendon should be protected during the procedure.
–– Always use imaging to control the amount of alignment
correction that is to be obtained during the operation.

17 Valgus High Tibial Osteotomy: Lateral Closing and Medial Opening

161

Fig. 17.5 Oblique skin incision

Fig. 17.3 Femorotibial mechanical angle of 170°: a correction of 13°
(10° + 3°) is planned for the osteotomy

Fig. 17.6 The tibialis anterior muscle and the long toe extensor muscle
are released from the tibial metaphysis

We currently use the Lepine HTO plate for the fixation
(Fig. 17.10). This blade plate/screw system has been specifically designed to minimize subcutaneous irritation. The
improvement of the fixation is due to the locking screws. We
first use a normal 4.5 mm bi-cortical screw, producing good
compression. The second screw (6.5 mm locking screw) is
placed, followed by exchange of the 4.5 mm screw for another
locking screw. The fixation achieved is very rigid and a variety
of plate and screw sizes accommodate different tibial widths.
(a) Introduction of the guide pin parallel to the joint line.

Fig. 17.4 Patient set-up

A small guide pin is introduced at the level of the joint
line and an alignment guide is placed over it (Fig. 17.11).
This guide will position the second guide pin parallel to the
joint line and 1 cm distal to it.

162

R Debarge et al.

Fig. 17.7 Protection of the peroneal nerve

Fig. 17.9 The correct direction of the osteotomy

Fig. 17.8 The two distal fibula holes are connected first with an
osteotome

Fig. 17.10 HTO blade (Lepine®)

17 Valgus High Tibial Osteotomy: Lateral Closing and Medial Opening

163

Fig. 17.13 Blade chisel introduction over the guide pin

Fig. 17.11 Introduction of the guide pin parallel to the joint line

Fig. 17.14 Preparation of the socket and screw holes for the blade

(c) Socket and screw hole preparation for the blade.
Fig. 17.12 Blade chisel introduction over the guide pin

The box preparation guide with drill guides already
assembled is introduced over the guide pin and impacted
(Figs. 17.14 and 17.15). Four drill holes are made with 6 mm
diameter.

(b) Blade reamer introduction over the second guide pin
(Figs. 17.12 and 17.13).

(d) Introduction of the HTO blade.

The length of the blade should be 1 cm shorter than the
total width of the tibia.

The blade, with screw guides already assembled, is introduced and impacted into the socket (Fig. 17.16).

164

R Debarge et al.

Fig. 17.15 Preparation of the socket and screw holes for the blade

Fig. 17.17 The posterior surface of the tibia is protected by a large
radiolucent protector, and anteriorly the patellar tendon is retracted

Fig. 17.16 Introduction of the HTO blade. The screw guides must be
in place before seating the blade

Fig. 17.18 Freehand distal cut of the osteotomy

(e) Distal cut of the closing wedge osteotomy.
Many surgeons use a guide pin for the distal cut of the osteotomy, but we do not feel this is necessary. The posterior surface
of the tibia is protected by a large radiolucent protector, and
anteriorly the patellar tendon is retracted (Fig. 17.17). An oscillating saw is used to perform the distal cut freehand (Fig. 17.18).
(f) Proximal cut:
The appropriate angled cutting guide (6–8–10°) is introduced in the distal cut of the osteotomy, and the proximal
cut is now performed using this angle (Fig. 17.19). The cutting guide should be introduced and fully impacted onto the
medial cortex (Fig. 17.20). An oscillating saw is used to
make the cut and the bone wedge is removed (Fig. 17.21).

Fig. 17.19 Proximal cutting guide (6-8-10) being introduced into the
distal cut

17 Valgus High Tibial Osteotomy: Lateral Closing and Medial Opening

165

Fig. 17.20 The cutting guide should be fully impacted onto the medial
cortex

Fig. 17.22 The medial cortex is weakened with a 3.2 mm drill

Fig. 17.21 Bone wedge removal

Fig. 17.23 Closing the osteotomy using the reduction clamp and
inserting a temporary compression screw

(g) Closing the wedge and image intensifier control of the
obtained mechanical axis:

(h) Fixation of the osteotomy.

The medial cortex is breached with a 3.2 mm drill
(Fig. 17.22). Distal to the osteotomy a temporary unicortical
screw is positioned. This screw will be used as the fixation
point for the reduction clamp. The wedge is closed with the
reduction clamp (Fig. 17.23). Using a long metal rod positioned on the center of the femoral head and in a middle of
the ankle joint, the mechanical axis of the limb is evaluated.
The axis should pass just lateral to the lateral tibial spine
(Figs. 17.24 and 17.25).

One bi-cortical 4.5 mm screw, and then the first locking
screw are introduced through the blade into the distal tibia.
The 4.5 mm screw is then replaced by the second 6.5 mm
locking screw to complete the fixation (Figs. 17.26 and
17.27). The muscle insertions are closed over a drain. The
skin is closed with interrupted sutures.
Two specific complications can be observed after a closing wedge HTO. We must carefully check the neurovascular
status at the end of surgery and also to pay attention to
uncontrolled pain postoperatively.
–– Common peroneal nerve lesion
–– Compartment syndrome

166

R Debarge et al.

Fig. 17.26 Completed fixation

Fig. 17.24 Preoperative fluoroscopic control using a long metal rod

Fig. 17.25 Note the axis hypercorrection

Fig. 17.27 Postoperative X-ray

17 Valgus High Tibial Osteotomy: Lateral Closing and Medial Opening

167

Medial Opening Wedge HTO

Tibial Osteotomy

Set-Up

The joint line and tibial tuberosity are marked with a pen and
a 10 cm anteromedial vertical skin incision is used for exposure of the proximal tibia (Fig. 17.28). The pes anserinus
tendons are retracted, or partially disinserted at the proximal
aspect. The superficial medial collateral ligament is incised
at the level of the osteotomy (Fig. 17.29). The posterior surface of the tibia is exposed using a large periosteal elevator.
During the osteotomy, this periosteal elevator is left in place.
Anteriorly, the patellar tendon is retracted using a Farabeuf
retractor.

The osteotomy is performed proximal to the tibial tubercle
and through the superficial medial collateral ligament,
which has previously been incised. The plane of the osteotomy is almost horizontal in the sagittal plane (different
from the closing wedge medial high tibial osteotomy which
is more oblique). The option exists to make a biplanar osteotomy, creating a vertical cut in line with the anterior tibial
cortex. The tibial tuberosity then may be kept with the
proximal or distal fragments (Figs. 17.30 and 17.31). This
allows a more distal osteotomy, and placement of a fourth
screw in the epiphysis when using the Tomofix.
First, two 2.5 mm Kirschner guide pins are introduced
from the medial side (Fig. 17.32). The first one is close to
the anterior cortex, and the second close to the posterior cortex. This will allow enough room to pre-position the Tomofix
plate between them, and check its position relative to the
joint line fluoroscopically. Laterally, the tips of these guide
pins should be just proximal to the head of the fibula, particularly if one does not want to increase the tibial slope. An
image intensifier is used to correctly position the guide pins,
adjusting as necessary (Fig. 17.33). Using an oscillating

Fig. 17.28 Skin incision

Fig. 17.29 Incision of the superficial medial collateral ligament

The patient is placed in the supine position. A tourniquet is
applied. An extremity sheet is used for the knee and a small
square field is applied over the ipsilateral iliac crest. A small
bump is positioned underneath the ipsilateral buttocks to
obtain a better exposure of the iliac crest.

Skin Incision

168

R Debarge et al.

Fig. 17.30 Biplanar osteotomy. The tibial tuberosity kept with the distal fragment

Fig. 17.32 Laterally, the tips of these guide pins should be just superior to the head of the fibula

breaking the lateral hinge. Using the specifically designed
osteotomes with a distance mark and a raised edge at the
blunt end limits the risk of an excessive introduction of the
osteotomes. If insufficient opening of the osteotomy is
obtained, the remaining anterior and posterior bony cortex
should be carefully broken using an additional osteotome.
Care is taken to open the osteotomy more posteriorly than
anteriorly to control the tibial slope.

Fig. 17.31 Biplanar osteotomy. The tibial tuberosity kept with the
proximal fragment

saw, the tibial cut is now performed underneath these guide
pins, but always staying in contact with them (Fig. 17.34).
The center of the tibia is cut first followed by the anterior
and posterior cortices. The cuts are completed using an
osteotome, especially on the anterior cortex where the
patella tendon can be damaged (Fig. 17.35). Subsequently, a
Lambotte osteotome (thickness 2 mm, corresponding with
approximately 2° of angular correction) is introduced into
the osteotomy. A second osteotome is then introduced below
the first. To gently open the osteotomy, several more osteotomes are introduced between the first two (Fig. 17.36). The
first osteotome should be impacted against the lateral cortex
and the second distal to the first, nearly as far. The third
osteotome is then introduced, and if necessary a fourth and
fifth. Each is impacted less deep than the previous to prevent

Two primary complications can be encountered during this
type of osteotomy:
–– Fracture of the lateral hinge—Frequently observed in significant corrections. This results in an undercorrection of
the deformity. In this situation, the lateral displacement of
the tibia can be reduced by a temporary distal 4.5 mm
compressive screw through the plate just distal to the
osteotomy.
–– Fracture of the lateral tibial plateau—This complication
can occur if the lateral hinge has been insufficiently weakened, if one forcefully tries to open the osteotomy with a
valgus maneuver, or if the osteotomes are not placed
deeply enough. Usually plate and screw fixation suffice to
overcome this complication.
The obtained angle of correction is systematically evaluated using a long metal rod centered on the hip and ankle
(Fig. 17.37). The angular correction is evaluated at the level
of the joint line (Fig. 17.38). If necessary, an additional
osteotome is introduced or removed.

17 Valgus High Tibial Osteotomy: Lateral Closing and Medial Opening

169

Fig. 17.33 An image intensifier is used to correctly position the guide pins. Internal rotation of the leg allows accurate assessment of the pin tips
in relation to the proximal tibiofibular joint

Osteosynthesis
In order to avoid loss of correction in the postoperative
period, the fixation should be strong and stable. We currently
use a locking plate (Tomofix, Synthes®) (Fig. 17.39). Other
types of fixation are also possible (Staples, Surfix Plate,
Chambat Plate). The anatomically pre-shaped Tomofix plate
is inserted into the subcutaneous plane and centered on the
anteromedial tibia. Proximal fixation is achieved first with
three locking screws, which provide wide support for the
subcortical tibial plateau. At this stage, a lag screw can be
placed in the screw hole just distal to the osteotomy site; this

approximates the plate towards the tibia and induces compression at the lateral hinge. For definitive fixation of the
plate, the distal locking screws can now be placed. Finally,
the lag screw can be replaced with a locking screw and an
X-ray taken to check screw length and overall position. In
cases of large correction (over 10°), the osteotomy site is
filled with tri-cortical bone graft harvested from the ipsilateral anterior iliac crest (Fig. 17.40). These grafts are
impacted, taking care not to overcorrect. Bone substitutes are
also available and can be used instead of the bone graft. The
superficial medial collateral ligament is now approximated
over the staples.

170

R Debarge et al.

Fig. 17.34 The tibial cut has been performed underneath the guide
pins, always staying in contact with them

Fig. 17.36 To open the osteotomy, several more osteotomes are introduced between the first two

Fig. 17.35 The tibial cut is completed using an osteotome

Fig. 17.37 Intra-operative evaluation of correction using a long metal
rod

17 Valgus High Tibial Osteotomy: Lateral Closing and Medial Opening

171

Fig. 17.38 The femorotibial mechanical axis is lateral to the lateral
tibial spine. The osteotomy is fairly horizontal in this case

Fig. 17.40 Osteotomy site filled with bone graft in case of correction
over 10°

Postoperative Guidelines
The postoperative guidelines are identical for the closing
wedge as well as the opening wedge HTO.
- With the Tomofix plate patients can be mobilized with
partial weight bearing 15–20 kg on the operated leg the day
following the operation. The knee is mobilized with active
and passive range of movement exercises and the patient is
discharged home when ambulating safely on crutches.
Progressive weight bearing is allowed from 4 weeks.

Fig. 17.39 Fixation of the osteotomy with a locking plate

––
––
––
––

Walking protected by two crutches.
Thromboprophylaxis for 1 month.
Skin sutures are removed around day 12.
Bracing in extension whilst mobilizing for 2 months.

172

–– Flexion is limited to 120° for the first 15 days. Progressive
flexion then follows.
–– Driving a car is not permitted for 8–10 weeks.
–– Physical work is not allowed for 3–4 months.
–– Sports with impacts or contact are allowed 2 months after
bony union has been achieved.
Radiographs are taken 6–8 weeks after the intervention.
If bony healing is observed, full weight bearing can begin.
If delayed union is suspected, progression of weight bearing is delayed and the patient is invited to come back 1
month later.

R Debarge et al.

Future Improvement
Future work may include inclusion of the degree of tibial
rotation in the preoperative plan and postoperative evaluation. Computer assisted surgery can be used to obtain and
evaluate the desired post-osteotomy mechanical axis and is
currently under investigation.
We hypothesize that a customized implant, with personalized specific instrumentation using pre-operative CT scan or
MRI, will allow exact correction of the deformity according
to the preoperative plan, reaching the desired “target.” It
would change drastically the technique.

Varus High Tibial Osteotomy:
Medial Closing

18

R Debarge, P Archbold, P Neyret,
and C Butcher

Introduction

Surgical Technique

The varus high tibial osteotomy is indicated in the young
active patient with lateral arthritis of the knee and a moderately valgus knee with a valgus tibia. This surgical procedure
results in a durable and satisfying clinical outcome up to
8–12 years if the lower limb has been corrected to neutral
alignment. This procedure addresses both the valgus in
extension as well as in flexion. It may result in obliquity of
the joint line however. This surgery should be used as an
alternative to a knee prosthesis (TKA or UKA). The surgical
technique consists of a closing wedge osteotomy on the
medial side of the tibia. Exceptionally, a lateral opening
wedge osteotomy is an alternative option to correct a deformity resulting from an excessive lateral closing wedge high
tibial osteotomy, or in selected pediatric disorders.

Patient Set-Up

Radiological Workup
See chapter surgical indications in arthritis of the knee.
The amount of correction needed to obtain a mechanical
femoro-tibial angle of approximately 180° is calculated with
respect to the width of the metaphyseal area of the tibia
(Fig. 18.1).

R Debarge · P Archbold
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

The patient is placed in the supine position, and a tourniquet
is used. The lower limb is covered with an extremity sheet
(Fig. 18.2). The image intensifier is positioned temporarily to
ensure sufficient access to the entire limb, including the hip.

Incision
An anteromedial, slightly oblique, almost horizontal skin
incision starts 1 cm proximal to the tibial tubercle and continues medially over a distance of 8 cm (Fig. 18.3).
The hamstring tendons are identified and retracted. The
superficial medial collateral ligament (MCL) is incised horizontally at the level of the osteotomy (Fig. 18.4). The proximal fibers of the MCL are elevated proximal and distal to the
incision over distance of a few millimeter, uncovering the
area of the wedge that will be resected.
A periosteal elevator is introduced posterior to the metaphyseal area of the tibia, always staying in contact with the bone
to the lateral side of the posterior tibia. The periosteal elevator
may be replaced by a specific radiolucent retractor, which is
smooth, flexible, and curved (Fig. 18.5). This will protect the
posterior structures optimally during the osteotomy. A
Farabeuf retractor is introduced underneath the patellar tendon to retract and protect it during the osteotomy.

The Tibial Osteotomy
The tibial osteotomy is performed just proximal to the level
of the tibial tubercle, slightly oblique and sloped proximally

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_18

173

174

R Debarge et al.

Fig. 18.3 Skin incision

Fig. 18.1 Femoro-tibial mechanical angle of 186°: a correction of 6°
is planned

Fig. 18.2 Patient set-up

Fig. 18.4 The superficial medial collateral ligament is incised
horizontally

18 Varus High Tibial Osteotomy: Medial Closing

175

Fig. 18.5 Specific
retractor—smooth, curved,
and radiolucent

Fig. 18.7 Pins are placed at the tip of the fibula laterally

Fig. 18.6 Intraoperative fluoroscopic control

from medial to lateral. Two 2.5 mm Kirschner wires will
serve as guide pins for the proximal cut of the osteotomy.
The pins are introduced medially and will emerge laterally
just proximal to the tibiofibular joint. Their correct position
is verified using an image intensifier. (Figs. 18.6 and 18.7).
The proximal cut of the osteotomy is done with an oscillating saw under the two guide pins (Fig. 18.8). First the mid
part of the tibia is cut, then the anterior and posterior cortex. The lateral cortex is left intact and will serve as a hinge
during the procedure. As Henri Dejour used to say, you
should just “knock at the door.” Subsequently, the distal cut
is ­performed. In the sagittal plane, it should be parallel to
the proximal cut, and in the frontal plane, it should converge at the lateral hinge. The distance between both cuts at
the level of the medial cortex has been defined during the
surgical planning. The wedge is removed using a large
grasper. The lateral hinge is now gently perforated with a
3.2 drill to weaken it (Fig. 18.9). Subsequently, the osteotomy will progressively close by introducing an osteotome
into the osteotomy and gently further weakening the lateral

176

R Debarge et al.

Fig. 18.8 Osteotomy with an oscillating saw under the two guide pins

Fig. 18.10 Intraoperative evaluation of the correction using a long
metal rod
Fig. 18.9 Weakening the lateral hinge with 3.2 mm drill holes

hinge. An intraoperative evaluation of the correction is
mandatory. A long metal rod is placed from directly over
the middle of the femoral head to the middle of the ankle
joint (Figs. 18.10 and 18.11). At the level of the knee, this
rod should be in the center of the knee following correction
(Fig. 18.12).
An overcorrection should be avoided. Therefore, the
height of the resected wedge should not be excessive. A fre-

quent error of overcorrection is the fact that the surgeon did
not consider the thickness of the saw blade when making the
resection. The osteotomy is fixed using two to three Blount
or Orthomed staples on the medial side (Figs. 18.13 and
18.14). Use of other fixation devices such as a locked plate
(Tomofix, customized, etc.) are of course possible but are
more prominent. The pes anserinus is closed over the staples.
A drain is positioned in proximity to the osteotomy and the
skin is closed using interrupted sutures.

18 Varus High Tibial Osteotomy: Medial Closing

177

Fig 18.11 Intraoperative evaluation of the correction using a long
metal rod

Fig. 18.13 Postoperative X-ray AP

Postoperative Guidelines
The patient should receive information on the postoperative
guidelines prior to the surgery.
These postoperative guidelines are identical to those for
an opening wedge osteotomy, but must be adjusted to the
type of fixation and resulting stability.

Complications

Fig. 18.12 The rod should be in the center of the knee following correction. In this case, the axis is varus

–– Errors of correction: Overcorrection is more frequent than
under correction.
–– Non-union and fixation failures are rare.
–– Delayed union can be observed in case of an imperfect fit
between the osteotomy cuts.
–– The osteosynthesis material can cause pain or discomfort.
Removal is in many cases sufficient for pain relief.
–– The clinical outcome of a medial closing wedge high tibial osteotomy can decline after approximately 7–20 years.
In most cases, a total knee arthroplasty can be performed
without any major difficulties.

178

R Debarge et al.

Future Improvements
–– Improvement in the calculation of the desired correction—the “target.”
–– Improvements in the reproducibility of the desired correction: computer-assisted surgery and navigation could
result in a more precise evaluation of the mechanical
femoro-­tibial axis.
–– Improvement in the fixation of the osteotomy allowing
earlier weight bearing.
–– Applications of specific growth factors or other biologic
agents to improve early consolidation.

Fig. 18.14 Postoperative X-ray lateral

Varus High Tibial Osteotomy:
Lateral Opening

19

R Debarge, P Archbold, P Neyret,
and C Butcher

Introduction
Indications for a lateral opening wedge tibial osteotomy
include: cases where an overcorrection has been made following a valgus-producing, lateral closing wedge high tibial
osteotomy (Fig. 19.1a, b) or in valgus malunions following a
tibial plateau fracture (Fig. 19.2).
It aims to compensate for a valgus malunion by raising
the joint line laterally. Its advantage over a closing medial
osteotomy is that it does not lower the joint line. It requires
an autologous bone graft harvested from the ipsilateral anterior iliac crest and an osteotomy of the fibula neck. This technique is commonly practiced in pediatric surgery but
continues to have a bad reputation when used in adults due to
the risk of injury to the common peroneal nerve.
The thickness of the opening wedge to obtain a normocorrection is calculated based on the width of the tibial epiphysis
and the desired angle of correction. In the cases of an iatrogenic
overcorrection, one should aim to leave a discrete valgus.

Surgical Technique
Set Up
The patient is placed in the supine position. A tourniquet is
applied. An extremity sheet is used for the knee and a small
square field is applied on the ipsilateral iliac crest. A small
bump is positioned underneath the ipsilateral buttock to obtain
R Debarge · P Archbold
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

a better exposure of the iliac crest. An image intensifier is
required. This technique has similarities to that used for a lateral closing valgus tibial osteotomy as described previously.

Incision
A slightly oblique, almost horizontal, anterolateral skin incision is used. It starts 1 cm above the anterior tibial tuberosity
(ATT) and proceeds laterally to 1 cm below the fibular head (if
the procedure is required due the result of trauma or a previous
osteotomy, we use the scar that is already present) (Fig. 19.3).
The fascia of the tibialis anterior is released from the tibia.
Subsequently, the tibialis anterior muscle and the long toe
extensor muscle are elevated from the tibial metaphysis using
a large periosteal elevator. In order to mobilize and protect the
common peroneal nerve from tension, these muscles must be
released more distally than in a lateral closing wedge osteotomy. In cases of revision surgery, more care must be taken as
the tissues are often scarred, and the common peroneal nerve
is at risk. A neurolysis of the nerve is performed in all cases by
careful proximal to distal dissection. This is performed irrespective of the location of fibular osteotomy.

Osteotomy of the Neck of the Fibula
The neck of the fibula is identified and exposed. A periosteal
elevator is placed around the neck, always staying in contact with
the bone. This maneuver protects the peroneal nerve from direct
injury (Fig. 19.4). Two holes are now drilled in the neck using a
3.2 mm drill (Fig. 19.5). With the use of an osteotome the 2 holes
are interconnected (Fig. 19.6). The fibular shaft should be mobile.
Care must to be taken that the peroneal nerve is not in contact
with the osteotomy. In the absence of a malunion and if local
conditions (dissection) permit, we prefer to perform the osteotomy at the neck of the fibula; otherwise, it is possible to perform
the osteotomy in the distal third of the tibial shaft.

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_19

179

180

R Debarge et al.

a

b

Fig. 19.1 (a, b): X-rays showing high overcorrection after closing wedge HTO

Osteotomy of the Tibia
A large retractor is positioned behind the posterior cortex of
tibia to protect the neurovascular structures, and the patellar
tendon is anteriorly retracted by a Farabeuf (Figs. 19.7 and
19.8). The direction of the osteotomy in the frontal plane is
marked with two 2.5 mm threaded K-wires under fluoroscopic
control. The wires are passed parallel to each other in a slightly

proximal trajectory ending about 1 cm below the medial tibial
plateau. The osteotomy is performed proximal to the tibial
tubercle in an almost horizontal direction in the sagittal plane
(Fig. 19.9). If necessary a coronal plane cut behind the ATT
can be made to allow a more horizontal osteotomy (Fig. 19.10).
Care needs to be taken in the placement of the pins, as the
anatomy of the proximal tibia has been changed by the previous osteotomy or trauma, and landmarks may be altered.

19

Varus High Tibial Osteotomy: Lateral Opening

181

Fig. 19.4 Fibular neck exposure and protection of the peroneal nerve
(bottom left)

Fig. 19.2 X-rays showing sequalae of lateral tibial plateau fracture

Fig. 19.5 Two holes are drilled in the fibular neck using a 3.2 mm drill

Fig. 19.3 Skin incision must take into consideration previous scar

182

R Debarge et al.

Fig. 19.6 Fibular neck osteotomy with the use of an osteotome

Fig. 19.8 A Farabeuf retractor retracts the patellar tendon (top right)

Fig. 19.7 A retractor is positioned behind the posterior cortex of the
tibia to prevent vascular lesions. The peroneal nerve is also protected by
the retractor (arrow)

Using an oscillating saw, the osteotomy is now performed
underneath these guide pins, but always staying in contact
with them. The center of the tibia is cut first followed by the
anterior and posterior cortices. The cuts are completed using
an osteotome, especially on the anterior cortex where the
patella tendon can be damaged. It is necessary to have an
intact medial hinge for this type of osteotomy. We mobilize
this hinge by weakening it with a number of 3.2 mm drill
holes (Fig. 19.11). Subsequently, a Lambotte osteotome
(thickness 2 mm, corresponding with approximately 2° of
angular correction) is introduced into the osteotomy. A second osteotome is now introduced under the first. Depending
on the desired correction, more osteotomes are then introduced between the first two (Fig. 19.12a, b). It is important

Fig. 19.9 Tibial cut under the guidewires using the oscillating saw

19

Varus High Tibial Osteotomy: Lateral Opening

183

not to drive the third osteotome too far in order to avoid fracturing the medial cortex.
The natural tendency is to open the osteotomy site more
anteriorly than posteriorly, with potentially deleterious
effects on tibial slope. One must be careful to avoid this
mistake. Placing the osteotomes more posteriorly can minimize this risk. The degree of correction is monitored by
checking the amount of opening at the fibular osteotomy
site. The correction is maintained with Méary retractor
(Fig. 19.13).
Two complications can be encountered during this type of
osteotomy:

Fig. 19.10 A coronal plane cut behind the tibial tubercle can be done
to allow more horizontal osteotomy

–– Fracture of the medial hinge. Frequently observed in
larger corrections, this complication results in under-­
correction of the deformity. If this problem occurs, it
should be fixed by placing a staple over the medial hinge
via a direct medial approach.
–– Fracture of the medial tibial plateau. This problem is
observed if the medial hinge has been insufficiently
weakened, if one forcefully tries to open the osteotomy
with a varus maneuver, or if the osteotomes are not
placed sufficiently deep. Performing the osteotomy distal to the previously placed pins helps prevent this.
Usually medial sided fixation suffices to overcome this
complication.
The obtained angle of correction is systematically evaluated using a long metal rod centered on the femoral head and
center of the ankle mortise (Fig. 19.14a, b). The angular correction is evaluated at the level of the knee joint line. If necessary, an additional osteotome is introduced or removed to
dial in the appropriate correction.

Fig. 19.11 The hinge is weakened with 3.2 mm drill holes

a

Fig. 19.12 (a, b): The wedge is progressively opened using osteotomes

b

184

Fig. 19.13 The correction is maintained with a Méary retractor

a

b

R Debarge et al.

Fig. 19.15 Fixation using staples. The osteotomy site is filled with
bone graft

Osteosynthesis
In order to avoid loss of correction in the postoperative
period, the fixation should be strong and stable. We use two
symmetrical Orthomed staples. One is placed between
Gerdy’s tubercle and the tibial tuberosity, the other placed
more posteriorly, both between the epiphysis and diaphysis
(Figs. 19.15 and 19.16). The staples converge toward the
center of the shaft. The osteotomy site is filled with tricortical bone graft harvested from the ipsilateral anterior iliac
crest. These grafts are impacted, taking care not to overcorrect. This type of fixation has our preference in this very special indication. A locking plate can be used for fixation;
however, it can be difficult to place due to the proximity of
the osteotomy to the joint line, or the abnormal shape of the
proximal tibia (Fig. 19.17a, b). The wound is closed by
repairing the tibialis anterior and extensor digitorum muscle
bellies over the anterolateral staples. A small diameter
“Redon” suction drain, trimmed to produce approximately
six holes, is placed and the wound is close in layers.

Postoperative Guidelines
––
––
––
––

Fig. 19.14 (a, b): Intraoperative fluoroscopic control using a long
metal rod centered on the femoral head and center of the ankle mortise.
Visible are the posterior retractor, the osteotomes, and the lucency from
removal of the previous fixation device

No weight bearing for 2 months.
Walking protected by two crutches.
Thromboprophylaxis for 1 month.
Bracing in extension for 2 months, but early cautious
mobilization of the knee.
–– Flexion is limited to 120° for the first 15 days. After that
date flexion can be progressively advanced.
–– The drain is removed on postoperative day one, or the
same day of outpatient surgery.
–– Skin sutures are removed around day 12.

19

Varus High Tibial Osteotomy: Lateral Opening

185

a

b

Fig. 19.16 (a, b): Postoperative X-rays of the osteotomy stabilized with two staples

a

b

c

Fig. 19.17 (a–c): Pre-operative (a), and postoperative (b, c) X-rays of a post-traumatic valgus knee treated by osteotomy, and stabilized with a
locked plate. The plate appears well adapted to the bone, but this is not always the case

186

–– Driving a car is not allowed for 10 weeks.
–– Physical work is not allowed for 3–4 months.
–– Sports are allowed 6 months after bony union has been
achieved.
WARNING!
Two specific complications can be observed after a lateral
opening wedge high tibial osteotomy (HTO)
–– Peroneal nerve lesion—particularly in large corrections
or in revision surgery
–– Compartment syndrome

R Debarge et al.

The patient is reviewed 2 months after the intervention.
If radiological bony healing is observed, weight bearing can
be started. If delayed union is suspected, partial weight
bearing is allowed and the patient is invited to come back in
1 month.

Flexion High Tibial Osteotomy:
Anterior Opening

20

P Archbold, P Verdonk, E Servien, P Neyret,
and C Butcher

Introduction
The surgical management of a genu recurvatum (hyperextension of the knee) is rare. It should be considered for symptomatic patients who suffer from significant asymmetrical
genu recurvatum (more than 20°). Distinction should be
made between this pathology and an idiopathic symmetric
genu recurvatum, or a secondary genu recurvatum due to a
bony or ligamentous lesion.
It is of major importance to evaluate the recurvatum clinically as well as radiographically. The evaluation should be
compared to the contralateral side (Fig. 20.1).

Radiological Workup
(See Chap. 14, surgical indications in the treatment of osteoarthritis). The aim of this workup is to quantify the overall
recurvatum in both knees and to calculate the amount of
recurvatum residing in the femur or/and tibia (Fig. 20.2).
Comparison with the contralateral knee will be useful.

Indications
–– Deformity secondary to poliomyelitis: the recurvatum
should not be corrected completely so as not to lose the
stabilizing effect on the lower limb. This effect on stabil-

ity is important in these patients who frequently lack a
functional quadriceps muscle.
–– Chronic posterior laxity: reducing the posterior tibial
translation.
–– Bony recurvatum at the level of the tibia (negative tibial
slope) due to a fracture malunion or secondary to growth
plate arrest of the anterior aspect of the proximal tibial
physis (Fig. 20.3).

Technique
This technique has been described by Henri Dejour and
F. Lecuire (Fig. 20.4). The anterior opening wedge osteotomy is performed at the level of the tibial tubercle with a
posterior hinge. The hinge is situated at the level of the
insertion of the fibers of the posterior cruciate ligament and
the attachment of posterior knee joint capsule on the tibia.
The anteromedial skin incision is made in line with the
medial border of the patellar tendon (Fig. 20.5). An osteotomy of the tibial tubercle is performed. The bone block
should be 6–8 cm long and should reach into the metaphyseal bone (Fig. 20.6 and Chap. 3, Episodic Patellar
Dislocation). Guide pins are introduced anteriorly approximately 4–5 cm below the joint line directed posteriorly and
aimed at the level of the insertion of the posterior cruciate

P Archbold · P Verdonk
Centre Albert Trillat, Lyon, France
E Servien
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Fig. 20.1 Asymmetric genu recurvatum

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_20

187

188

P Archbold et al.

a

b

c

Fig. 20.2 Radiological measurements of femoral and tibial recurvatum. (a) Global recurvatum (angle between anatomical femoral and tibial
axes). (b) Tibial slope. (c) Epiphyseal femoral angle (between the femoral anterior tangent line and the line perpendicular to Blumensaat’s line)

ligament fibers and proximal to the insertion of the posterior
capsule (Fig. 20.7). It is imperative to control the progress of
the pins with fluoroscopy to prevent inadvertent damage to
the neurovascular structures. Medially, a large periosteal
elevator is inserted underneath the fibers of the superficial
medial collateral ligament. Laterally, the tibialis anterior
muscle insertion is partially released from bone.
The osteotomy is completed with an oscillating saw
underneath the guide pins and always staying in contact
with the guide pins (Fig. 20.8). The osteotomy should be
situated proximal to the tibiofibular joint. Opening of the
osteotomy is achieved by the sequential introduction of several ­osteotomes (the same as surgical technique for medial
and lateral opening wedge high tibial osteotomies, Chaps.
17 and 19).
Generally, each osteotome or 1 mm of opening achieves a
correction of about 2°. The final correction should take into
account not only the bony genu recurvatum measured radiologically, but also the clinical genu recurvatum. A bony
recurvatum of 20° but with only a clinical recurvatum of 10°
should not be corrected by 20°. This degree of correction
could result in a clinical flexion deformity that is poorly tolerated by the patient.
Fig. 20.3 Negative (medial) tibial slope of 7°

20 Flexion High Tibial Osteotomy: Anterior Opening

189

Fig. 20.4 Surgical technique for anterior opening wedge osteotomy

Fig. 20.5 Skin incision

Fig. 20.6 Tibial tubercle osteotomy

Remark An anterior tibial osteotomy frequently increases
varus of the tibia. Therefore, the osteotomes should be inserted
from the medial side during opening to minimize this effect.

The posterior cortex is weakened with a 3.2 mm drill (as in
medial and lateral closing wedge HTO technique). The final
correction should be controlled clinically to avoid a hypercor-

190

P Archbold et al.

Fig. 20.8 Tibial osteotomy distal to guide pins to avoid epiphyseal
fracture
Fig. 20.7 Intra-operative fluoroscopic control of guide pins
positioning

a

b

c

Fig. 20.9 (a–c): Fixation by two Blount staples; intra-operative view (a), and post-operative X-rays (b, c)

rection, and resulting flexion. The osteotomy is fixed by two
Blount staples on either side of the tibial tubercle (Fig. 20.9a–
c). Nowadays, we do not hesitate to place medially a locked
plate (e.g., Tomofix) in order to obtain better stability (particularly in case of osteoporosis that is frequent in polio) and
to start early rehabilitation (Fig. 20.10a–c).
Cortical and cancellous iliac crest bone grafts (or bone
substitutes) are needed to fill the osteotomy (Figs. 20.10 and

20.11). The tibial tubercle osteotomy is fixed using two anterior to posterior 4.5 mm AO screws (Fig. 20.12). Both may
be placed distal to the osteotomy, or one above and one
below. The patella height should not be modified. In others
words, the tibial tuberosity bone block is proximalized with
respect to the distal tibial fragment, by the same amount as
the opening wedge osteotomy, in order to avoid a patella
infera.

20 Flexion High Tibial Osteotomy: Anterior Opening

a

b

191

c

Fig. 20.10 (a–c): Fixation by medial locked plate and lateral staple; intra-operative view (a), and post-operative X-rays (b, c) showing positive
slope of 6°

Fig. 20.11 Bone grafts to fill the osteotomy

Fig. 20.12 Tibial tubercle fixation

192

Post-operative Guidelines
–– Non-weight bearing with crutches for 2 months unless
locked plate fixation is used, in which case partial weight
bearing is allowed.
–– Progressive mobilization of the knee, limited to 90° for
60 days (to ensure consolidation of the tibial tubercle
osteotomy).

P Archbold et al.

–– Bracing at 10° of flexion.
–– Post-operative radiographs should include a lateral radiograph of the knee to measure the obtained correction.

21

Double Osteotomy
S Lustig, MF AlSaati, R Magnussen, P Neyret,
and C Butcher

Introduction
A double osteotomy is indicated in the following situations:
–– A situation in which an isolated osteotomy (of the femur
or of the tibia) to correct a major angular deformity (>10°)
in the frontal plane, either valgus (Fig. 21.1) or varus
(Figs. 21.2 and 21.3a), would result in an oblique joint
line (Fig. 21.3b). This obliquity would create shear forces
across the knee joint that can lead to early failure of the
intervention. A distal femoral osteotomy combined with a
proximal tibial osteotomy is able to correct the axis of the
lower limb while maintaining an acceptable obliquity of
the joint line (Fig. 21.3c).
–– A situation in which an attempted single site correction
with an opening wedge osteotomy results in too much
opening, compromising the stability of the osteotomy.
–– A situation in which correction with a single closing wedge
osteotomy would be too large and result in poor coaptation
of the proximal and distal bone segments, which can cause
problems for future total knee arthroplasty.
–– The treatment of osteoarthritis secondary to malunion of
the femur. In these cases, the aim of the procedure is to
address the frontal or torsional malunion of the femur by
a femoral osteotomy and to address the arthritis with a
tibial osteotomy (Fig. 21.4). It is of major importance to
know that femoral malunions situated close to the knee

S Lustig
Service d orthopedie de l Hopital de la Croix Rousse, Lyon 69004,
France
MF AlSaati · R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

joint are more significant than those situated at a greater
distance from the joint. A femoral osteotomy can only
correct a deformity in extension and not in flexion.
Certain difficulties and complications are inherent to a
double osteotomy.
1. The risk for a delayed union or malunion are increased
compared to an isolated osteotomy.
2. Calculation of the correction remains difficult and complicated. In the case of a femoral malunion, one can perform both interventions separately starting with the
femoral derotation osteotomy and then the tibial osteotomy at a later stage. During the two-week period between
the procedures, one can assess the first correction by long
films and CT scan, and plan the second with more accuracy. If a computer assisted navigation is available, both
correction in the frontal and horizontal plan can be combined during the same intervention.
Nevertheless, indications for a double osteotomy remain
rare and in this chapter we will not discuss proximal or
diaphyseal femoral osteotomies that are indicated in isolated
torsional problems.

The Principles
Varus Knee
In a varus knee with a mechanical axis less than 165°, the
combination of a lateral closing wedge distal femur osteotomy with a lateral closing wedge high tibial osteotomy or
medial opening wedge high tibial osteotomy is indicated.
The advantage of an opening wedge high tibial osteotomy is
preservation of the length of the lower limb. The skin incision is placed laterally on the femur and crosses the midline
at the level of the tibial tubercle to continue medially on the

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_21

193

194

S Lustig et al.

Fig. 21.1 Major angular deformity in the frontal plan in valgus (right
knee)

Fig. 21.3 (a): Case of major
angular deformity in varus. (b):
An isolated tibial osteotomy to
correct a major angular
deformity would create an
oblique joint line. (c): A normal
or acceptable joint line obliquity
after correction of a major
angular deformity becomes
possible using a distal femoral
osteotomy associated with the
proximal tibial osteotomy

a

Fig. 21.2 Major angular deformity in the frontal plan in varus (both
knees)

b

c

21

Double Osteotomy

a

195

b

c

Fig. 21.4 (a): Medial femoro-tibial osteoarthritis caused by rotational malunion after femoral fracture (external rotation of the right limb). (b):
External rotational deformity measured by CT scan (right limb appears on the right of this image) (c): Clinical deformity

tibia. Alternatively, an isolated lateral femoral incision can
be combined with an isolated medial tibial incision. In cases
of a closing wedge high tibial osteotomy, (Fig. 21.5), a laterally based long skin incision is typically used.

otomy with a closing wedge medial high tibial osteotomy is
indicated (Fig. 21.6). This combination results in an acceptable orientation of the joint line while lowering the risk of
injury to the peroneal nerve.

Valgus Knee

Malunion with Torsional Problem

In a valgus knee with a mechanical axis greater than 190°, a
combination of an opening wedge lateral distal femoral oste-

In cases of osteoarthritis secondary to a femoral malunion in
combination with a torsional problem greater than 15° and a

196

Fig. 21.5 Postoperative X-rays (see case Fig. 21.2)

frontal deviation greater than 10°, we advise the combination
of a derotation osteotomy on the femur and a tibial osteotomy to address the frontal plane deformity (Fig. 21.7).

Surgical Technique
On the Femur
The approach has been described in detail in the chapter on
femoral osteotomy for varization.
1. Lateral Opening Wedge Osteotomy for Valgus Knee
(See Chap. 16).
2. Lateral Closing Wedge Osteotomy for the Varus Knee
The area for the osteotomy is prepared. Two additional
Kirschner guide pins are introduced in the femur as guide
pins for the future osteotomy. One pin is introduced paral-

S Lustig et al.

Fig. 21.6 Postoperative long leg films after correction of a major valgus deformity

lel to the joint line approximately 50 mm proximal to the
joint line. The second pin is introduced proximally to the
first on the lateral cortex but converging with the first
medially. This represents the angle and the wedge that
will be resected. The quadriceps muscle is retracted at a
level proximal to the trochlea with the knee in extension;
the posterior side of the knee is cleared. A superficial
­longitudinal mark on the lateral cortex of the femur with
the oscillating saw can serve as a landmark to determine
the rotation (Fig. 21.8). The blade plate has to be introduced in the epiphyseal area approximately 30 mm proximal to the joint line. The blade is 5.6 mm thick, 16 mm in
width, and the distance between the holes is 16 mm. Its
entry point is anterior and proximal to the lateral collateral ligament. The entry angle has been determined by
pre-­operative planning and a specific reamer is used. For
a calculated valgus correction of 8°, the guide instrument
is set at 93° (85° + 8°; this is the complementary angle to

21

Double Osteotomy

197

Fig. 21.7 Pre- and
postoperative X-rays after
correction of an external
rotational malunion
associated with a medial
compartment osteoarthritis
(see case Fig. 21.4)

the desired anatomical angle of 95°, plus the angle of correction). The blade is subsequently introduced into the
femur. The correct angulations are again checked using
the image intensifier.
3. Derotation Osteotomy in the Case of Femoral Malrotation
If the location of the malunion is in the proximal femur,
it is logical to perform an inter-trochanteric osteotomy
instead, thereby realigning the muscles more anatomically. If performed distally, the area of the osteotomy is

prepared in the same manner as above. Two superficial
saw marks are made on the lateral cortex indicating the
desired angle of the derotation (Fig. 21.8). By doing
this, an isolated derotation osteotomy can be performed
as well as a derotation osteotomy in combination with
an opening wedge or a closing wedge femoral osteotomy. The derotation osteotomy should not interfere
with the patella tracking or create a step on the anterior
cortex.

198

S Lustig et al.

On the Tibia
For these surgical techniques, please see Chaps. 17 and 18.
The bone graft obtained in case of a closing wedge femoral
osteotomy is used to fill the opening wedge tibial osteotomy.

Postoperative Guidelines
The postoperative guidelines are identical as for a high tibial
osteotomy, but with attention to preventing the stiffness that
is possible after femoral osteotomy.
Fig. 21.8 Two rotational landmarks are superficially done on the femoral cortex using the saw

Patellar Femoral Arthritis and the
Lateral Partial Patellectomy

22

S Lustig, LN Favarro Francisco, R Magnussen, P Neyret,
and C Butcher

Introduction

Radiographic Workup

Isolated patellofemoral arthritis (PFA) is a relatively rare
condition. Above the age of 55, the incidence is estimated at
8% in females and 2% in males. The condition is bilateral in
over 70% of the patients and in 80% of cases trochlear dysplasia can be identified as the major etiological factor.

The radiographic workup includes the following plain radiographs: anteroposterior weight bearing (including 45 degrees
flexed “schuss” view), lateral weight bearing, and patellar
axial skyline views. Primary etiological factors for lateral
patellofemoral arthritis include trochlear dysplasia, patella
alta, and trauma. One should always exclude inflammatory
joint disease, tibiofemoral arthritis, and the sequelae of complex regional pain syndrome. The differential diagnosis
should include chondrocalcinosis. In our opinion, meniscal
calcifications, which are frequently observed on plain radiographs, are not indicative of chondrocalcinosis but indicate
small calcifications secondary to a prior hemarthrosis.
Chondrocalcinosis is characterized by the typical recess
above the trochlea and the typical radiographic sign of the
patella femoral joint shaped like a “saw.” Patellofemoral
degenerative lesions can be classified according to Iwano, in
four stages (Fig. 22.1a–d):

Clinical Evaluation
Anterior knee pain is typically observed in patients with lateral patellofemoral arthritis. Ascending and descending
stairs generally increases the pain. These patients are often
unable to rise from a chair or squat without using their hands
without significant pain. The ability of Muslims to pray sitting on the heels is limited. The nature of the pain is never
excessive and generally does not interfere with activities of
daily life. Walking on flat ground is usually not limited,
which helps to differentiate between patellofemoral arthritis
and tibiofemoral arthritis. Swelling of the knee is intermittently present. Manual pressure and manipulation of the lateral or medial facets usually evokes this specific pain. Range
of motion of the knee is normal or near normal. Signs of
patellar instability are generally absent.

S Lustig
Service d orthopedie de l Hopital de la Croix Rousse, Lyon 69004,
France
LN Favarro Francisco · R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Stage 1: Presence of osteophytes, joint space width narrowing, remodeling of the subchondral bone.
Stage 2: Joint space width narrowing less than 3 mm.
Stage 3: Joint space width narrowing more than 3 mm.
Stage 4: Absence of joint space. The patella can have the
shape of a beret (Fig. 22.2).
CT imaging can help determine possible indicators for
patellofemoral instability. An MRI is useful to evaluate the
tibiofemoral compartment.

Treatment
Options
Treatment options for lateral patellofemoral arthritis are
numerous. Conservative therapy is generally prescribed for
the early stages of patellofemoral arthritis (modifications of

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_22

199

200

S Lustig et al.

a

b

c

d

Fig. 22.1 Iwano classification of isolated patellofemoral osteoarthritis. (a): Stage 1, Slight; (b): Stage 2, Joint space width narrowing less than
3 mm; (c): Stage 3, Joint space width narrowing more than 3 mm; (d): Stage 4, Absence of joint space

Non-arthroplasty Surgery
 rthroscopic Debridement, Chondroplasty,
A
and Lavage
The efficacy of these procedures is disputed. Clinical
improvement is not reproducible and often of short duration.
Therefore, we do not recommend or perform them.

Lateral Patellar Release

Fig. 22.2 Béret Basque

ADL’s, NSAID’s, physiotherapy, and injections of corticosteroids or hyaluronic acid). Nevertheless, surgery can be
indicated if the symptoms are severe or if they fail to respond
to conservative therapy. Surgical treatment is determined by
several factors: the age of the patient, the patient’s profession, the patient’s function in activities of daily living, and
the clinical symptoms.

Isolated lateral patellar release is controversial and therefore
not generally indicated for anterior knee pain. We do consider
this procedure in cases of lateral facet overload characterized
by pain over the lateral patellar facet and increased patellar tilt
(indicative of a pathologically tight lateral retinaculum) in
patients without significant patellofemoral arthritis on radiographs. Some authors have described Z-lengthening of the lateral retinaculum in such cases with good results.

Tibial Tubercle Osteotomy
The principles of this technique are to decrease patellofemoral
contract pressure and transfer the weight bearing area from the
area of wear to another region with intact articular cartilage. It is
therefore contra-indicated in cases of complex regional pains
syndromes and generalized diffuse patellofemoral arthritis.

22

Patellar Femoral Arthritis and the Lateral Partial Patellectomy

(a) Anteriorization of the Tibial tubercle.
Introduced and popularized by Maquet, the aim of this
procedure is to reduce the contact pressure of the patellofemoral articulation. This technique is fraught with complications and therefore it is not performed as a routine
procedure. If we desire anteriorization of the tibial tubercle,
we generally perform an anteromedialization of the tubercle
as described by Fulkerson.
(b) Medialization of the tibial tubercle.
For lateral patellofemoral arthritis secondary to episodic dislocation of the patella, this technique is usually performed. A
medialization of about 5 mm is usually desired. This technique
can be combined with a partial lateral patellar facetectomy. We
have abandoned the performance of a VMO advancement in
such cases due to complications during rehabilitation. This technique is routinely combined with a lateral patellar release if the
retinaculum is noted to be pathologically tight.
(c) Distalization of the tibial tubercle.
This technique is indicated in the cases of a patella alta.
Although this technique seems logical in terms of displacement of the contact zone, the postoperative period is long
and characterized by a persistent swelling of the knee.
Commonly, pain reduction is incomplete.

 ateral Vertical Partial Patellectomy (Lateral
L
Facetectomy)
See complete description below.

Total Patellectomy
A total patellectomy is characterized by a subsequent
weakness of the extensor apparatus. It also results in a
large scar. Therefore, this technique should be used only
in cases of severe post-traumatic arthritis. It should be
noted however that resection of up to 25% of the total
width of the patella does not affect patellofemoral
biomechanics.

Arthroplasty
Patellofemoral Arthroplasty
The success rate of patellofemoral arthroplasty varies from
44% to 90%. The clinical outcome is superior to an isolated
resurfacing of the patella. Reasons for failure include pro-

201

gressive tibiofemoral arthritis, implant malpositioning, and
malalignment of the extensor apparatus (failure to correct an
increased TT-TG). Due to the inconsistency in the results, we
do not perform this procedure routinely.

Total Knee Arthroplasty
In the older patients, the total knee arthroplasty remains the
treatment of choice for lateral patellofemoral arthritis. Pain
reduction and improvement of function are excellent.
Patients’ satisfaction is especially high in patients with significant preoperative functional restrictions and limited
expectations.

 artial Lateral Patellectomy or Lateral
P
Facetectomy
This procedure is technically easy to perform; however, the
indications are limited. Ideally, the patients should be
between 40 and 65 years of age and only limited during certain physical activities (ascending and descending stairs).
Walking distance and flexion should be within normal limits.
Conservative therapy should have been tried for at least
6 months. The patients should not be obese and should be
normally aligned in the frontal plane. Palpation of the lateral
border of the patella should evoke pain.
Plain radiographs, including the schuss view, should confirm absence of pathology in the tibiofemoral joint. The
patellar skyline view should show an osteophyte or a typical
“beret” aspect of the patella.
CT imaging allows evaluation of the TT-TG distance. It
should be within normal limits. If an excessive TT-TG is measured, a medialization of the tibial tubercle may be indicated.

Surgical Technique
The patient is placed in the supine position with a tourniquet
applied. An arthroscopy can be performed initially to evaluate the tibiofemoral joint as well as to remove potential loose
bodies in the knee joint. In our experience, arthroscopy is
rarely required (Fig. 22.3).
A lateral parapatellar skin incision is created, centered for
the patella and measuring 5–6 cm in length. The lateral retinaculum is released (Fig. 22.4). The release should not
involve the distal fibers of the vastus lateralis. The knee is
now placed in extension and the articular surfaces of the
patella and trochlea are inspected (Fig. 22.5).
The pre-patella soft tissues are carefully released from the
anterior border of the patella over a distance of 1 cm using a
15 blade (Fig. 22.6). Between 1 and 1.5 cm of the lateral
patella facet, including its osteophyte are then resected using
an oscillating saw. The articular surface is protected by

202

S Lustig et al.

Fig. 22.3 Arthroscopic view: patellofemoral osteoarthritis with apparent subchondral bone

Fig. 22.5 Lateral retinaculum release and inspection of the patella and
trochlea

retractors or by placing a sponge/swab into the patellofemoral joint (Figs. 22.7 and 22.8). The resected area generally
goes from the lateral insertion of the patella tendon to the
lateral insertion of the vastus lateralis and must be ­sufficiently
large. Very frequently, the inexperienced surgeon will be
somewhat disappointed about the obtained resection on the
postoperative radiograph (Fig. 22.9).
The capsule is not closed in the middle and distal portions
of the patella. Hemostasis should be carefully performed
(bone wax may be applied to the resection area). An intra-­
articular drain may be left for 24 hrs.

Postoperative Guidelines

Fig. 22.4 Lateral parapatellar skin incision

Ice application should be limited to 3–4 days post op.
Thromboprophylaxis is continued for 10 days. As for all surgery on the patellofemoral joint, the brace should be in 20°
of flexion at rest and in extension for walking. Isometric
quadriceps contractions and continuous passive motion
should start of the first postoperative day. Weight bearing is
allowed immediately. Walking with crutches is allowed for
3–5 days after which quadriceps training is started.

22

Patellar Femoral Arthritis and the Lateral Partial Patellectomy

Fig. 22.6 Release of the pre-patella soft tissues from the anterior border of the patella

203

Fig. 22.8 Final view after resection

Rehabilitation should be progressive, slow, and should not
provoke pain. Return to normal activities of daily living is
allowed after 1 or 2 months. Downhill walking is not allowed
for 2–3 months, and squatting for 6 months.

Complications
Complications include hematoma and pain because of an
insufficiently large resection.
No specific complications were noted in our recently published series. Functional outcome is encouraging even if the
radiological results are mediocre. With a mean follow-up of
8 years, no further surgery after facetectomy had to be performed in our series.

Fig. 22.7 Resection of the lateral patella facet (1–1.5 cm)

204

a

Fig. 22.9 Preoperative (a) and postoperative (b) X-rays (skyline views)

S Lustig et al.

b

Unicompartmental Knee Arthroplasty

23

S Lustig, A Daher, R Magnussen, P Neyret,
and C Butcher

Introduction
A unicompartmental prosthesis is indicated in unicompartmental arthritis. Patient selection and surgical technique are key factors for a successful outcome (see Chap.
14). Although the indications have expanded in many centers, we would advise caution when selecting patients.
Due to the potential activity possible after unicompartmental knee arthroplasty (UKA), it is tempting to increasingly perform this procedure in younger patients, but the
survival rates need to be considered carefully. Although
we prefer not to perform bilateral total knee arthroplasty
(TKA), bilateral simultaneous UKA is a reasonable option
due to the lower blood loss, and the lower incidence of
systemic complication. Although we continue to perform
UKA with regional anesthesia and a drain, we will consider the use of local intra-articular injection and no drain
to reduce length of stay.
In this chapter, we will detail the surgical technique for a
medial UKA (Fig. 23.1). The surgical technique for a lateral
UKA is comparable; therefore, we will only cover some
specific points regarding a lateral UKA. Our technique
describes implantation of a fixed bearing UKA. Irrespective
of the instrumentation or prosthesis utilized, the general
principle is to produce a level and inclination of implant
interface which corresponds with the pre-disease native

joint line. This will approximate the natural kinematics of
the knee, and thus will require relative normality of the ligamentous and soft tissue structures. It also infers that the
post-operative coronal alignment will mirror that of the predisease, but this must be within acceptable limits to ensure
longevity of fixation.

S Lustig
Service d orthopedie de l Hopital de la Croix Rousse, Lyon 69004,
France
A Daher · R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Fig. 23.1 U-Kneetec® prosthesis

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_23

205

206
Fig. 23.2 AP (a) and lateral (b)
30° knee flexion X-rays

S Lustig et al.

a

b

Radiological workup: see chapter on surgical indications
for knee arthritis (Figs. 23.2a, b, 23.3, and 23.4).
The stress radiograph is essential in the radiological
workup. It will indicate whether the deformation is reducible
but not overcorrectable (Figs. 23.4 and 23.5). Reducibility
will suggest deformity due to articular wear; a good indication for the resurfacing function of UKA. Overcorrection
may either indicate laxity, or wear of the contralateral compartment. Absence of reducibility may indicate constitutional varus. Caution will then be required to avoid persisting
significant varus which may lead to post-operative overload
(Figs. 23.6, 23.7, and 23.8).
For UKA, the mechanical femorotibial axis (mFTA)
of the lower limb should be within certain limits. The
authors propose that they should not exceed 9° of varus
or 14° of valgus. Outside these limits, TKA is generally
preferred.

Surgical Technique for a Medial UKA
Set Up

Fig. 23.3 Schuss views (45° knee flexion) are useful to detect mild
narrowing of the medial compartment

–– See chapter on TKA
–– Tourniquet—provides excellent vision, as sizing and
position of the components are critical.
–– A vertical lateral support is placed at the level of the tourniquet and a distal horizontal support is placed to keep the
knee flexed at 90°. A second horizontal support is useful
to maintain hyperflexion in the absence of a second
assistant.

23

Unicompartmental Knee Arthroplasty

207

Fig. 23.4 Stress radiograph
in varus and valgus. In this
case, the deformity is
reducible and not
overcorrectable
(no medial laxity)

Fig. 23.5 Anterior stress radiograph. Tibial translation is a sign of
osteoarthritis associated with ACL deficiency. It is a cause of failure
(Previous Chap. 22, Fig. 22.5)

Fig. 23.6 Medial compartment osteoarthritis and constitutional varus

208
Fig. 23.7 (a, b) Postoperative X-rays showing
degree of varus alignment

S Lustig et al.

a

Fig. 23.8 Bone scan showing evidence of medial overload

Approach
A paramedian medial skin incision of 8–10 cm begins the
superior pole of the patella and ends at the medial border of
the tibial tubercle (Fig. 23.9). The vastus medialis and the
medial border of the patellar tendon are identified. A medial

b

“midvastus” arthrotomy of the knee is performed (Fig. 23.10).
The anterior horn of the medial meniscus is incised and the
anterior medial tibial plateau is exposed in a limited fashion.
The midvastus approach can go 15 mm into the vastus medialis using the Metzenbaum scissors as proposed by Engh.
This allows adequate exposure of the femoral condyle. The
appropriate retractors are positioned. The articular cartilage
and the status of the anterior cruciate ligament are
examined.
The anteromedial joint capsule is released from the tibial
metaphysis in a triangular fashion. The Trillat periosteal elevator is inserted between the medial border of the medial
tibial plateau and the joint capsule. We do not perform any
ligamentous release. Using an arthroscopic shaver, a burr or
a rasp, the articular cartilage on the distal femur is removed
up to the level of the subchondral bone. This subchondral
bone will serve as a reference for correct positioning of the
UKA.

The Tibia
Orientation and level of the tibial cut are crucial. It determines the good positioning of the unicompartmental

23

Unicompartmental Knee Arthroplasty

209

Fig. 23.9 Skin incision

Fig. 23.11 Positioning of the tibial alignment guide

prosthesis. The tibial alignment guide is positioned
(Fig. 23.11). First coronal and subsequently sagittal plane
alignment will determine the tibial cut. In a tibia with no
extra-articular deformation, the distal end of the cutting
guide is centered over the midpoint of the mechanical axis in
the coronal plane. In the case of metaphyseal tibial bowing,
the tibial cut should be performed perpendicularly to the
proximal tibial epiphysis axis (and not perpendicular to the
mechanical axis).
Fig. 23.10 Medial midvastus arthrotomy

210

S Lustig et al.

In the sagittal plane, a pin is placed into the medial joint
space resting on the anterior and posterior margin of the
tibial plateau. In the medial compartment, this pin will quite
reliably guide the functional slope as well as the bony slope.
The effect of an intact meniscus is to reduce the anatomical
bony slope only slightly. The meniscus is relatively fixed on
the concave plateau, and consequently the slope is constant
throughout the range of motion (the situation in the lateral
compartment is not so simple; the meniscus is mobile on a
convex plateau, resulting in an increasing slope with increasing flexion.
The extra-medullary cutting guide is placed on this pin
to reproduce the tibial slope. Once the cutting guide is correctly aligned in both the coronal and sagittal planes, the
tibial resection height is determined. In extension, the level
of the exposed subchondral bone of the distal femoral condyle is considered the joint line reference. A controlled valgus stress is now applied to reduce the deformity
(Fig. 23.12). Generally, we aim to correct only the wear
component taking care not to overcorrect the axial alignment. Holding the knee in this correct position, the guide
pin is now positioned in contact with the distal femur subchondral bone. The tibial cutting guide is now lowered
13 mm below this reference. The technique of the tibial cut
corresponds to the total thickness of the tibial component in
extension (9 mm) + the distal femoral component
(3 mm) + 1 mm for laxity. The extra millimeter is added to
allow some “physiological” laxity. The cutting guide is
then securely fixed by three to four guide pins through the
appropriate holes. The thickness of tibial cut must be carefully assessed at this point. Thinking it can be compensated
for by a thicker tibial insert is an error. An excessive tibial
cut results in a smaller tibial surface and the quality of the

cancellous bone is poor, possibly risking tibial component
loosening. An excessive tibial cut can also weaken the deep
medial collateral ligament or the posterior part of the
menisco-tibial ligament and explains some cases of iatrogenic laxity.
The guide pins allow a perfect horizontal tibial cut while
guiding the oscillating saw blade (Fig. 23.13). Next, the
vertical tibial cut is performed just lateral to the medial
tibial plateau in the axial direction of the medial axis of the
notch towards the center of the femoral head. The posterior
part of the vertical cut can be completed using an osteotome. The medial tibial plateau can now be removed carefully with a large grasper. The posterior meniscal horn can
be accessed easily with the knee flexed to 90° and stressed
in a valgus position. The cutting guide is removed and the
appropriate medio-lateral sized tibial trial component size
is selected. Overhang of the trial component is not accepted.
The trial component should be easily introducible in flexion and stable during flexion/extension. Primary assessment of the stability of the trial tibial component is
mandatory at this point (Fig. 23.14a, b).

Fig. 23.12 Controlled valgus stress applied to reduce the deformity

Fig. 23.13 Horizontal tibial cut on guide pins

23

Unicompartmental Knee Arthroplasty

211

Femoral Resurfacing
With the tibial trial component in place and the knee now
in extension, a mark is made on the femoral condyle opposite the anterior limit of the tibial component. A special
femoral guide (the crocodile guide) is vertically introduced
and should lie flat on the tibial component with the knee
still in full extension (Fig. 23.15a, b). This step is very
important because the orientation determines the correct
positioning of the femoral cutting guide and the femoral
component in the coronal and horizontal planes. Two guide
pins are inserted through the two holes of the crocodile
guide. With the crocodile guide still in place, the knee is
now flexed to 90°. The posterior border of the crocodile
guide should be parallel to the tibial plateau. A small
adjustment can be made at this point or during the next
step. The crocodile guide is subsequently removed while
the guide pins are left in place. These guide pins will
accommodate the femoral drilling guide. In 90 degrees of
flexion, the femoral drilling guide is placed onto the two
guide pins (Fig. 23.16). The correct position of this guide
is perpendicular to the tibial cut. Although the design of

a

the prosthesis accepts some freedom of alignment, this
should not exceed 6°. In case of excessive malalignment of
the drilling guide, the guide should be realigned. If necessary, realignment is performed by pinning the guide
through the most central hole on the medial-lateral axis of
the medial condyle including the lateral osteophyte. Next,
the guide is removed and the femoral drilling guide is correctly aligned perpendicularly to the tibial cut with the
knee in 90 degrees of flexion. A third pin can ensure the
correct alignment. All remaining holes are now pre-drilled.
The guide is removed and the pre-drilled holes are now
connected with each other by an oscillating saw to create
the femoral recess. The femoral recess is subsequently
enlarged and impacted. The appropriate femoral cutting
block is chosen according to the size and the curvature of
the femoral condyle (Fig. 23.17). The femoral cutting
block size is assessed with a classic posterior femoral condyle reference system. Anteriorly, it should be in between
the femoral condylar mark and the two femoral guide pinholes. The curvature of the block should match that of the
femoral condyle. The cutting block is fixed in place with
pins. Finally, correct femoral alignment is double-­checked.

b

Fig. 23.14 (a, b): Primary assessment of the stability of the trial tibial component

212

a

S Lustig et al.

b

Fig. 23.15 (a, b): Femoral guide positioning in extension

Fig. 23.16 Femoral drilling guide positioning in flexion. If necessary,
the guide can be moved medially or laterally

Fig. 23.17 Femoral cutting block positioning

23

Unicompartmental Knee Arthroplasty

The posterior cut and the posterior chamfer are then performed. No distal femoral cut is performed.
Key Points
1. The cutting block size is determined by the shape and
­curvature of the femoral condyle.
2. It should be at a level of or cover the two previously performed two holes on the anterior border of the distal
femur.
3. The cutting block should be in contact with the posterior
condyle.

Fig. 23.18 Inadequate slope may cause tightness in flexion

213

4. Rotation should not be modified in order to obtain a better
cover of the condyle. Rotation of the femoral component
is solely determined by the tibial cut.
The femoral and tibial trials are introduced and primary stability is verified first in flexion and then in extension. If the tibial component has a tendency to advance in
flexion, one should first suspect the remnant of a posterior
meniscus horn pushing the trial plateau anteriorly, or an
insufficient slope of the tibial plateau causing “booking”
(Fig. 23.18). A slight laxity should be accepted to

214

S Lustig et al.

guarantee under correction. If ligament tightness exists,
the correct position of the femoral guide should be
checked. It should be in contact with the subchondral
bone. If the tightness is still present with an appropriate
slope, one should proceed with an additional tibial cut of
1–2 mm. This cut can be done easily free hand using the
oscillating saw: the tracts of the guide pins, which correspond with approx 1–2 mm in thickness, should just be
sawed away. Under no circumstances should one perform
a ligamentous release to address the tightness. Again
overhang of the tibial trial component is not accepted.

Implantation
If alignment and laxity are satisfactory, the component can
be cemented. A small bony recess is made underneath the
tibial spine to help with cement fixation. Generally, we first
cement the femoral component followed by the tibial component (Figs. 23.19, 23.20 and 23.21). The joint surfaces
are cleaned from debris and the knee irrigated. The knee
joint capsule and skin are closed. The tourniquet is not
released for the closure. Local injections are used for postoperative pain and bleeding. A drain is left
intra-articularly.

Fig. 23.19 Implantation of the cemented components

Post-operative Guidelines
–– Post op radiographs (Fig. 23.22).
–– Weight bearing is allowed on day 1, crutches are used for
1–3 weeks.
–– Removal of the drain when production of less than 50 cm3
(maximum 24 h).
–– Hospital stay for 1–2 days depending on the comorbidities and social circumstances. Nowadays, it seems logical
to consider out-patient surgery when performing a unicompartmental prosthesis in selected patients.
–– Flexion from 0 to 120° until day 45. Unlimited
afterwards.
–– Thromboprophylaxis for 15 days.

 pecific Points for the Lateral
S
Unicompartmental Arthroplasty

Fig. 23.20 Implantation of the cemented components

In general, the surgical technique for a lateral unicompartmental knee arthroplasty is not more difficult than for the
medial one using modern instruments. In fact, the lateral
compartment is more tolerant than the medial one due to

the extrinsic moment arm, which pushes the knee into
varus. Therefore, the indications for the lateral unicompartmental knee arthroplasty can be pushed somewhat further

23

Unicompartmental Knee Arthroplasty

215

Fig. 23.21 Implantation of the cemented components

Fig. 23.22 Post-operative X-rays

going up to 12–15 degrees of valgus alignment in the frontal plane.
Technically speaking, the lateral femoral resurfacing
implant is usually better adapted than the medial as it can
correct the pre-operative hypoplasia of the lateral condyle,
and this is accommodated by the greater laxity of the lateral
compartment.
The shape of the tibial plateau after the tibial cut is not the
same in medial and lateral compartment. A half-moon shape
is observed in lateral compartment and a more oblong shape
in the medial compartment. In order to better cover the tibial
plateau and to prevent overhang the design of the tibial component must be carefully chosen.
The surgical technique for a lateral unicompartmental
knee arthroplasty is very comparable to the medial unicompartmental arthroplasty except for some specific points:

the lateral compartment of the knee joint. The joint capsule
is opened using a lateral arthrotomy. The iliotibial band is
NOT released from its distal attachment on Gerdy’s tubercle
(Fig. 23.23).

Approach
A lateral longitudinal parapatellar incision of approximately
8 cm is made with the knee in 90° of flexion to gain access to

Tibial Cut
A common error in the lateral compartment is the overcorrection of the deformity by excessive varus stress, resulting
in a tibial cut of insufficient thickness. Therefore, we generally opt to incompletely reduce the valgus deformity
(Fig. 23.24). This will result in a thicker tibial cut. The chosen slope is approximately the same as in medial UKA and
limited to prevent excessive loads in the ACL.

Femoral Resurfacing
Generally, we place the femoral cutting guide on the most
lateral part of the femoral condyle and, if present, on the
lateral femoral osteophyte. This will eliminate a potential

216

S Lustig et al.

Fig. 23.24 Incomplete reduction of the deformity in valgus

Fig. 23.23 The joint capsule is released but the iliotibial band stays
intact

conflict between the femoral component and the tibial
spines, most likely in extension (Fig. 23.25). As mentioned earlier, a general error is overcorrection in the coronal plane. This is due to the physiological laxity, which
is present in the lateral compartment and should be
preserved.

Associated Procedures
We do not hesitate to perform a lateral facetectomy of the
patella in the approach (Fig. 23.26).
Comments:
Until 1996, an osteotomy of the anterior tibial tuberosity
was routinely combined with a lateral unicompartmental
knee arthroplasty. Due to the arrival of new and minimally
invasive instruments, the osteotomy is no longer necessary to
position the lateral unicompartmental knee prosthesis in a
correct and reproducible manner. Therefore, this technique
has been abandoned since 1996.

Fig. 23.25 The femoral component is placed laterally to avoid
impingement on the lateral tibial spine

23

Unicompartmental Knee Arthroplasty

217

Complications
General complications are less frequent than for total knee
arthroplasty: fat embolism, DVT are rarely observed. Septic
arthritis remains an exception.
Axial malalignment: Overcorrection is the most frequently observed error in our experience.
One should always take care of:
–– An excessive tibial cut on the medial side for medial UKA
–– Overcorrection in lateral UKA

Fig. 23.26 Associate lateral facetectomy of the patella

Unicompartmental Knee Arthroplasty
(UKA) After UKA to the Other
Compartment

24

Maad AlSaati, S Lustig, R Magnussen, P Neyret,
and C Butcher

Introduction
The concept of unicompartmental knee arthroplasty (UKA)
has experienced an increase in interest in the last decade. The
two main causes of failure are aseptic loosening and the
occurrence of degenerative lesions in the opposite tibiofemoral compartment.
The UKA is known to be less invasive than total knee
arthroplasty (TKA) and has the added advantage of preserving both cruciate ligaments, which results in kinematics more
similar to those of a normal knee. A UKA also has decreased
morbidity and overall costs than a TKA. It allows a shorter
hospital stay and faster return to normal function. Therefore,
it may be tempting to preserve these benefits in case of progressive degenerative change of the opposite compartment
after UKA. This goal can be achieved in selected cases by
performing a second UKA (Bi-Compartmental UKA) rather
than revision to a TKA.

Patient Selection
The selection criteria for a bilateral UKA are summarized in
Table 24.1. The typical patient is one who initially was happy
with his first intervention, but subsequently developed
degenerative change in the opposite compartment.

M AlSaati · R Magnussen
Centre Albert Trillat, Lyon, France
S Lustig
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Table 24.1 Indications and contra-indications for UKA of a second
tibiofemoral compartment
Indications
 No wear or loosening of the original UKA with correct implant
positioning
 “Overcorrection” close to 5° of the first implant
 Good initial result of the first UKA (initial painless period followed
by appearance of secondary symptoms)
 Localized pain in the opposite tibiofemoral compartment
 Osteoarthritis stage C or D (IKDC) or osteonecrosis of the femoral
condyle
 Reducibility of the deformity in the frontal plane
 Healthy cruciate ligaments
 Normal or near normal range of motion
Contraindications
 Absolute
 Inflammatory arthritis
 History of infection
 Ligament damage
 Major bone loss
 Extension deficit of greater than 10°
 Relative
 Patellofemoral compartment osteoarthritis
 Weight greater than 80 kg

Radiological Evaluation
Standard radiological assessment (anterior view, lateral view,
“Schuss view,” stress views, long leg standing view bilaterally)
can specify the stage of osteoarthritis (Fig. 24.1a, b), the reducibility of the deformity (Fig. 24.2a, b), and the mechanical axis.
On a technical level, when performing bi-­compartmental UKA,
the objective is not to obtain a 180° axis, but compensate for the
intra-articular wear in the non-­prosthetic compartment.
The axial view of the patella is also needed to assess the
patellofemoral joint (Fig. 24.3). In case of lateral facet osteoarthritis, a vertical patellectomy can be performed in associated with the UKA (see Chap. 23).
It is important to evaluate both the original prosthesis and
the opposite compartment. In the prosthetic compartment, no
signs of loosening or wear must be found. The prosthesis

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_24

219

220

a

M AlSaati et al.

b

Fig. 24.1 (a, b): “Schuss” and standing AP views. Lateral femorotibial osteoarthritis after a medial unicompartmental arthroplasty

must be positioned well. There should not have been any
degenerative changes in the opposite compartment at the
time of the initial UKA—they should be appeared progressively over time.

reduction software may compliment these modalities in
future.

Scintigraphy

Infection must be excluded using a complete blood count
(CBC) and assessment of inflammatory markers.
A rheumatological workup should be performed if there
is concern for an inflammatory process.

Although not essential, this test can rule out subclinical
implant loosening. It also verifies the absence of overload of
the opposite tibiofemoral compartment.

CT Scan
A CT scan or CT arthrogram provides information on overhang or possible loosening of the initial UKA and can also
help to identify the possible causes of complaints that are
located in the opposite compartment (osteonecrosis, meniscal tear, cartilage damage, foreign body). MRI with metal

Laboratory Tests

Arthroscopy
Arthroscopic evaluation of the joint can influence the
decision between a second UKA and conversion to
TKA. It is important to evaluate both the status of the
polyethylene in the first UKA and the condition of the
other two compartments. It can be performed at the beginning of the intervention.

24

Unicompartmental Knee Arthroplasty (UKA) After UKA to the Other Compartment

a

221

b

Fig. 24.2 (a, b): Stress view. Medial femorotibial osteoarthritis after lateral unicompartmental arthroplasty. Good reduction of the varus deformity, and absence of laxity in the lateral prosthetic compartment

Surgical Technique
Setup is the same as for all knee arthroplasty with knee is at
90° of flexion and a tourniquet at the upper part of the thigh.
We use a cemented UKA with a cobalt-chrome resurfacing
femoral implant and an all-polyethylene tibial component
(HLS Uni Evolution, Tornier ®). The approach is determined
according to the old incision (that is used during the first
UKA).

Medial UKA After a Lateral UKA

Fig. 24.3 Evaluation of the patellofemoral joint

There are two options for the approach of the second
UKA. The first is to create another separate incision (not less
than 8 cm away from the first) between 8 and 10 cm long,
extending from the upper pole of the patella to the medial
border of the ATT (Fig. 24.4). The second option is to reuse
the first incision (in cases where it is sufficiently medial) by

222

Fig. 24.4 Performing a separate incision away from the existing one

extending it 5–6 cm proximally and 2 cm distally, mobilize a
full thickness flap, and perform a medial arthrotomy
(Fig. 24.5). This second option is our preference.

M AlSaati et al.

Fig. 24.5 The other option is our preferred. Reusing the same incision
with proximal and distal extension, a full thickness flap, and a medial
arthrotomy

radiograph, which predicts the degree of valgus opening that
will occur in the OR.
Two or three pins are inserted through the guide to secure
it. The cut is made on the pins with the knee is in flexion. The
sagittal cut is oriented towards the center of the hip. It passes
Tibial Side
to the medial border of the ACL insertion and along the lateral surface of the medial femoral condyle. The posterior
To position the cutting template, a pin is inserted through the aspect of the cut is completed with an osteotome (10 mm in
tibial cutting guide and into the tibiofemoral joint. This pin width) while respecting the PCL. The medial tibial plateau is
gives the orientation of the tibial slope and establishes the grasped with an alligator clamp and all the menisco-capsular
level of the joint line. The tibial cutting guide is then secured attachments are detached. Before removal of the pins, verify
by a second pin parallel to the first in the sagittal plane, and the location of the tibial cut. It should be perfectly on the pins
approximately 5 mm below the joint line. The extra-­ and it must be flat. The underside of the resected medial tibmedullary stem is then aligned to the mechanical axis, or in ial plateau is used to size the tibial trail. We then introduce
the case of tibial deformity, to the proximal tibia.
the tibial trial component to ensure its primary fit and
The knee is placed in extension and the surgeon makes a stability.
controlled valgus maneuver to compensate for intra-articular
wear without overcorrection. The tibial cutting guide is then
lowered 13 mm (3 mm for the thickness of the prosthetic Femoral Side
condyle, 9 mm corresponding to the thickness of the polyethylene plate, and 1 mm for security to avoid overcorrec- The positioning of the femoral and tibial components is intition). The less valgus applied, the more we cut the tibia. This mately linked. Indeed, the height of the tibial cut (but not its
is the advantage of preoperative assessment with the stress orientation) is determined relative to the joint line while the

24

Unicompartmental Knee Arthroplasty (UKA) After UKA to the Other Compartment

223

Fig. 24.7 Application of the femoral drill guide
Fig. 24.6 Placement of the crocodile after performing the tibial cut
(the tibial trial implant has been left in place)

orientation of the femoral cut is dependent on the position of
the tibial trial component.
The tibial trial component is left in place. A special instrument, which we commonly call a crocodile (positioning template) is placed between the abraded femoral condyle and the
tibial trial (Fig. 24.6). While the knee is extended, we make
sure it stays on the tibial trial and does not translate medially.
This instrument has two functions: first to determine the
anterior limit of the femoral implant, second to eliminate any
rotation of the femoral implant relative to the tibial component during flexion-extension. Two pin markers are introduced into the anterior part of the “crocodile.”
The knee is flexed to 120°, and the crocodile and the tibial
trial are removed. On the two pins, the drill guide is positioned
to prepare the femoral slot. Its location and orientation relative
to the femoral condyle are checked to ensure proper component positioning (Figs. 24.7, 24.8 and 24.9).
After determining the appropriate position of the femoral
component, the next step is to determine its size. The size is
not selected according to the size of the tibial component, but
by using anatomical criteria. The rotation should not be modified to look for better coverage of the condyle. This orientation must strictly follow that given by the slot made in the
femur. The appropriately sized block is then fixed by pins to
the condyle and the posterior cut and drill holes are made.

Trials and Implant Fixation
The femoral trial is introduced first followed by the tibial
trial. We check the primary stability of the implants and look

for the presence of a discrete medial laxity when applying
forced valgus. The lack of mild medial articular laxity when
applying valgus suggests that overcorrection has occurred. In
this situation, we should not perform any ligament release
but, after checking that the femoral component was well-­
seated on the subchondral bone, make minimal cuts on the
tibial plateau to increase the laxity. Both final components
are then cemented into position.

Lateral UKA After a Medial UKA
The technique for performance of a lateral UKA is the
same as for a medial UKA, except for some small details.
The setup is similar to that of the medial UKA. As with a
medial UKA, the preferred option is to extend the previous
incision proximally and distally sufficiently to perform a
lateral arthrotomy. If choosing a second incision, we again
respect a distance of at least 8 cm between the two. In this
case, the second incision starts adjacent to the proximal
patella, descending along the lateral border of the patella
and patellar tendon to the proximal tibial tuberosity
(medial to the Gerdy’s tubercle). The arthrotomy is performed from the distal edge of the vastus lateralis tendon
(it is possible to extend it up to 1 cm by staying in the vastus lateralis) to the tibia. The anterolateral capsule is
released without detaching the iliotibial tract from Gerdy’s
tubercle.
An important detail when preparing the femur relates to
the axial rotation. Contrary to what is commonly done, we
do not tilt the femoral component to cover the lateral ­condyle,
but rather place the component perpendicular to the tibial cut

224

a

M AlSaati et al.

b

Fig. 24.8 (a, b): Postoperative X-rays

as guided by the instruments, positioning it as necessary on
the lateral condylar osteophyte (technical principle from
Philippe Cartier). It should also be lateralized as much as
possible to avoid conflict with the tibial spines with extension. The rest of the intervention is essentially the same as
described above.

Postoperative Care
These are the same as after a primary UKA:
• Full weight bearing from day one.
• Range of motion will be increased gradually but it will be
limited to 120° for 45 days.
• Thromboprophylaxis for 15 days.

24

Unicompartmental Knee Arthroplasty (UKA) After UKA to the Other Compartment

a

Fig. 24.9 Skin appearance (a) and long leg view postoperatively (b)

225

b

Total Knee Arthroplasty: Steps
and Strategies

25

C Butcher and P Neyret

Introduction
Although total knee arthroplasty (TKA) is a routine procedure performed in more than a million cases per year worldwide, there are many different approaches to prosthetic
design and the method of implantation. Our preferred operative methods for posterior stabilised TKA in varus and valgus knees are described in Chaps. 26 and 27. All described
methods help to create a reliable and efficient workflow and
standardise the procedure for the surgeon, department, hospital, or region. This reproducibility no doubt has raised
minimum standards in many respects.
However, two important issues mean that a simply formulaic approach is undesirable. Firstly, there is significant variability between patients in both original and pathological
anatomy, and therefore the details of one operative protocol
may not be appropriate to either the individual patient or the
ethnicity. The second one relates to the complexity of the procedure. There are multiple implant variables (geometry, sizing and level of constraint, etc.) and surgical variables (order
and type of bone cuts and soft tissue releases, instrumentation
system, component positioning, etc.), which are interrelated.
The operation consists of many individual steps which are
sequential and interdependent; each one affects the others.
Total knee arthroplasty is like a complex algorithm, but in
conventional TKA the procedure is linear; as the procedure
progresses, each parameter is set and is often irreversible. The
surgeon must therefore constantly reappraise the progress,
anticipate the following steps and their effects, and be ready
to deviate from the standard protocol because as each is completed, the options become more limited.

C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

“Like the chess player, the surgeon always has to remain one
or more steps in advance.”

The problem posed by the multivariate factors in the insertion of the different knee replacement systems may be illustrated by the possible paths passing through the points on a
three-dimensional shape (Fig. 25.1). The path may start at
any point, and then proceed along different routes until all
points have been visited once. Once the first choices are
made, there are a diminishing number of routes available; this
emphasises the linear nature of the decision-making process.
So the surgeon may for instance start at the tibial cut, or either
femoral cut, and then proceed in different ways, with further
dependent or independent cuts, and making soft tissue
releases at different points. How he or she starts will, in turn,
affect each one of the other factors. There will be trade-­offs in
any one chosen sequence, and the best one will minimise the
impact on the final result. Many compromises must be made,
and the ability to prioritise on the go is essential.
The possible remaining paths may also be predicted at any
point with an algorithm, and the subsequent choice of route
decided in advance. In this way, computer navigation and interactive robotics allow several steps to be carried out virtually,
allowing some prediction of their effect before committing to
some of the bone cuts or soft tissue releases. This is a powerful
tool and although superior clinical outcomes have yet to be
proven, the insights gained are informing the way the equation is
understood and used in conventional TKA. Whether in navigated
or conventional TKA, the ability to weigh up the options at any
given point in the procedure and alter the usual surgical protocol
depends on a solid understanding of both the functional anatomy
and the logic behind each suggested step. This chapter aims to
outline the thinking behind our preferred techniques, and by simplifying the algorithm of TKA to help surgeons improve both the
reproducibility and individualisation of the procedure.

P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_25

227

228

a

C Butcher and P Neyret

b

Fig. 25.1 (a, b) ‘Hamiltonian’ cycle on a decahedron; 2D and 3D representation. William Hamilton, an Irish mathematician, invented a
game in 1857, where pegs were placed in holes on a wooden board
showing a 2D representation of a dodecahedron. The challenge was to
find a route that would include all points—a traveller visiting all cities
just once in a twenty city world. In graph theory, of which this is an
example, routes can also be found mathematically with an algorithm

25 Total Knee Arthroplasty: Steps and Strategies

 art I Definitions and Elementary
P
Concepts
The technical goal of TKA is to insert the implants within
the soft tissue envelope in such a way as to provide a well-­
aligned and balanced knee. Alignment and balance are a
function of the level and orientation of the bone cuts and
appropriately managed soft tissue tension, both of which
create the spaces in which to place each component. The aim
of this text is an understanding of the individual spaces, and
the consequences of both the original deformity and the
steps in their creation.
Due to the complexity of both the anatomy and the procedure, it is useful to consider some definitions and concepts, and
these are covered in this first part. In Part II, we describe the
fundamental concept of separate tibial and femoral spaces,
Fig. 25.2 (a, b): (a)
Intra-articular deformity due
to anterior medial wear (right
knee) (b). Dotted line shows
original intact posterior joint
contour

229

which is the basis of our strategy. In Part III, we look at these
spaces in the context of the priorities and principles of the procedure, and in Part IV, how this informs our daily practice.

Deformity
Apart from soft tissue laxity or contracture, deformity due to
bone or cartilage pathology may be of two types:
• Intra-articular
This is usually due to wear of the cartilage or bone (Fig. 25.2),
or less commonly subchondral collapse due to avascular
necrosis or intra-articular fracture (Fig. 25.3). Typically, the
deformity can be corrected passively, and alignment and balance can be achieved without soft tissue management.

a

Fig. 25.3 Intra-articular deformity due to medial condyle fracture with
depression (right knee)

b

230

C Butcher and P Neyret

• Extra-articular
This is due to deformity in the metaphysis or diaphysis,
which may be developmental (Fig. 25.4), due to fracture
(Fig. 25.5) or to metabolic bone disease. It cannot be corrected passively, and surgically correcting the deformity
at the joint level and at the same time achieving balance
will require alteration of the soft tissue envelope from its
original natural state.

a

b

Alignment
• Alignment, balance, and longevity are linked and need to be
considered together. Always a compromise must be made.
• The coronal alignment at or near extension is dictated
mainly by the tibial cut and the distal femoral cut. We aim
for a mechanically aligned limb, with an orthogonal cut
of the tibia, but we accept some varus of the femoral component if it is necessary to achieve rectangular gaps and
soft tissue balance (Fig. 25.6). We do not attempt a kinematic or constitutional tibial alignment, in an attempt to
prevent excessive varus tibial malalignment and possible
premature loosening. Our femoral alignment, however, is
more kinematic and discussed in Parts II and III.
• The coronal alignment in flexion is dictated by the tibial
cut and the posterior femoral cut. This aspect is often
neglected and is also discussed in Part III.
• Sagittal alignment of the tibial cut is orthogonal and referenced from the proximal tibia to prioritise balance. An
orthogonal cut is appropriate for a PCL substituting
design and prevents flexion laxity, anterior tibial subluxation, and excessive forces on the polyethylene. Sagittal
alignment of the distal femoral cut is also dictated by the
distal femoral anatomy, not the line between the femoral
head and knee joint.

Fig. 25.4 (a, b) Proximal ‘constitutional’ varus deformity (a) X-ray
(b) Intra-operative photograph showing the varus inclination of the
joint surface in relation to the intra-medullary rod, which is fully
inserted and shows the anatomical axis

Fig. 25.5 Valgus tibial deformity due to metaphyseal fracture
malunion

Fig. 25.6 An orthogonal tibial cut, but the proximal femoral deformity
has been left uncorrected to improve ligament balance

25 Total Knee Arthroplasty: Steps and Strategies

Cuts
Bone cuts involve the tibia, the femur, and (where appropriate) the patella.
These are defined by:
• Level
This may influence the size of the gap and is therefore
linked to flexion/extension gap equivalence (gap
balancing).
The level of the cut is guided by a reference, which is
usually a point on the least affected part of the joint
(Fig. 25.7). The thickness of bone that is removed on the
reference side will often be replaced by the implant, but
not always. The bone and cartilage resection should be
equal or less than the minimum implant height, and not
more, to avoid unnecessary bone sacrifice. If there is a
remaining defect on the concave side, this will be dealt
with by augmentation, rather than increasing the depth of
resection at the expense of bone on the convex side
(Fig. 25.8).
Fig. 25.7 (a, b): (a) The level of cut is
10 mm referenced from the unaffected
convex side of the joint. Orientation is
90° to the mechanical/anatomical axis of
the tibia (b). Intra-operative photograph
showing a 10 mm stylus applied to the
unworn lateral tibial condyle

231

• Orientation
This influences the limb alignment. It also influences the
shape of the gap (symmetry) and so is linked to ligament
balancing. In many cases a priority, and thus a compromise, will have to be made between alignment and
balance.
‘The orthopaedic surgeon is often mistaken. He must be
cognisant and choose the most acceptable error.’

The cuts may be described as:
• Independent
Each cut is made separately and is referenced from bony
anatomic landmarks or intra-medullary/extra-medullary
guidance from the relevant bone (Fig. 25.9). The technique is often described as ‘measured resection’.
• Dependent
The cut level is referenced using parameters of a previously formed gap (Fig. 25.10). The technique may be
referred to as ‘gap balancing’ or ‘gap resection’.

a

Fig. 25.8 This medial defect is being dealt with by screw and cement augmentation. The presence
of the defect did not alter the thickness of the cut, which was referenced from the unworn convex
side

b

232

C Butcher and P Neyret

Fig. 25.9 The cuts are made
independent of each other,
based on anatomical
landmarks of each bone

Fig. 25.10 The parameters
of one gap are used to make
the second. In this case,
creation of the flexion gap
follows that of the extension
gap

a

b

Tibial Cut
There is only one tibial cut, except in the case of local bone
defects (see Fig. 25.8, Chap. 31). The orientation affects both
coronal alignment and symmetry of the gap throughout flexion. Orientation in the coronal and sagittal planes may be
guided by intra, or extra-medullary references. Axial rotation
is only set by component positioning and tibial preparation,
rather than a cut (Fig. 25.11). The level affects the gap
throughout flexion.

Fig. 25.11 Axial rotation set by trial positioning and keel preparation

25 Total Knee Arthroplasty: Steps and Strategies

233

Femoral Cuts
The orientation of the distal and posterior cuts contributes to the
coronal alignment of the knee in extension and flexion, respectively (Fig. 25.12). The orientation of each also affects the symmetry of the gap in extension and flexion, respectively.
The level of each cut affects the relationship between the
flexion and extension gaps (gap balance).
The anterior cut contributes to the patellofemoral
articulation. The orientation of the posterior and anterior femoral cuts are fixed and linked by the femoral
prosthesis (Fig. 25.13). A compromise may therefore be
necessary to optimise both the flexion and anterior gaps.
Additional cuts are usually required for anterior and
posterior chamfering, and the box for posterior stabilisation if needed.

Fig. 25.12 Coronal alignment in flexion is dictated by the posterior
femoral cut, as well as the tibial cut. Here, the external rotation of the
femoral component contributes 5° varus

Fig. 25.13 The cuts for the femoral prosthesis are fixed and usually
parallel

234

C Butcher and P Neyret

Gaps
The femorotibial gap is a space formed by the tibial cut, the
distal and posterior femoral cuts, and influenced by the
length of the soft tissues. Although often considered as separate flexion and extension gaps, they are clearly continuous
with each other, and this is relevant when considering ‘mid
flexion’ stability (Fig. 25.14).

The patellar femoral or anterior gap is formed by the
patella (cut surface, or articular surface if the patella is not to
be resurfaced), the anterior femoral cut in extension, and the
distal femoral cut in flexion (Fig. 25.15). Again there is really
only one patellofemoral gap, and it is in continuity with the
femorotibial gap in flexion, hence being influenced by the
distal femoral cut.

20 mm

20 mm

Fig. 25.14 Flexion and extension gaps are continuous with each other

x

Fig. 25.15 Patellofemoral gap, in continuity with the femorotibial gap in flexion

x

25 Total Knee Arthroplasty: Steps and Strategies

Implant Space and Prosthetic Interface
The part of the gap that is taken up by the prostheses is
termed the implant space. Ideally, the gap will be slightly
larger than the implant space, so that there is slight laxity to
avoid pain or stiffness from tissue tension, but not too much
laxity, which may cause instability. This observed difference between the gap and the implant space is due to the
soft tissue laxity and the loading conditions, including
intrinsic and extrinsic moments. Loading conditions are
dictated for the most part by the patient’s position and the
alignment. For instance when standing, if the joint is
mechanically aligned and thus more symmetrically loaded

a

235

in compression, the gap and space may be the same. If the
limb is not mechanically aligned, there may be asymmetric
loading and compression on one side but distraction on the
other, causing the gap and space to be different (Fig. 25.16).
In this regard, it is important to consider the coronal alignment in flexion as well as extension, for instance with
crouching or stair climbing.
The two components of the prosthesis meet at the
prosthetic interface. The goal is to place this as near as
possible to the original native joint line, at the level of the
meniscal rim, although this may not always be possible
(Fig. 25.17).

b

Fig. 25.16 (a, b) Different loading conditions produce different
extrinsic moments. The intrinsic adductor moment results from the
varus alignment of the femoral component

Fig. 25.17 Prosthetic interface located at the meniscal rim

236

C Butcher and P Neyret

Tibial and Femoral Segments
The segments are made up of the implant spaces and the
remaining bone distal or proximal to the collateral insertions.
They therefore depend on the chosen resection level, as well as
the component thickness. The surgeon has control over both
tibial and femoral resection levels, but in most primary systems only the thickness of the tibial component. The ­relative
size of each segment determines the position of the prosthetic
interface in relation to the soft tissue attachments. The aim will
be to place the prosthetic interface at the same level as the
native joint line to normalise the tension in the soft tissues
(Fig. 25.18). The relative size of both segments in relation to
the length of the soft tissues determines this tension or laxity.
Increasing each segment means producing a prosthetic interface further away from the respective attachments of the collateral ligaments. As each segment is composed of bone and a
component, increasing the segment would require more bone
(less resection) or a thicker component (Fig. 25.19). This may
possibly require more potential space from soft tissue release.
Reducing the segment would require the opposite; more bone
resection or thinner components. In a primary TKA, the latter
is only possible with the tibial component, and this may not be
possible in many cases where the minimum thickness polyethylene insert has already been used.

Fig. 25.19 (a, b): The tibial
segment can be increased by
resecting less (a), or using a
thicker tibial component (b).
In both illustrated cases, the
soft tissues have not been
altered, and the femoral
segment has been reduced by
bone resection. The overall
distance between the
collaterals is unchanged, but
the prosthetic interface has
been proximalised in relation
to both femoral and tibial
insertions

a

Fig. 25.18 The relative size of each segment determines the position
of the prosthetic interface in relation to the soft tissue attachments.
Here, it is at the original joint line, at the level of the meniscus

b

25 Total Knee Arthroplasty: Steps and Strategies

Soft Tissue Envelope and Ligament Balancing
The soft tissue envelope consists of all the active and passive
stabilisers of the knee although certain structures predominate (Fig. 25.20). It includes the collateral ligaments, the
extensor mechanism, the flexors, the iliotibial band, popliteus, anterolateral ligament, and patellar stabilisers as well as
the various condensations of the capsule such as the posterior oblique ligament. Their practical contribution to the gaps
is usually through contracture, or inadequate length due to a
thinner bone resection on the concave side of an extra-­
articular deformity, and possibly through laxity on the convex side. All of these situations cause an asymmetric gap in
the coronal plane. Ligament balancing, a technique common
to any surgical strategy, refers to the release of soft tissue in
order to create a rectangular, or ‘symmetric’ gap.

Gap Balancing
Whilst ligament balancing refers to a technique aimed at the
formation of a rectangular, symmetric gap, gap balancing is
a surgical strategy aimed at equalising the size of the flexion

237

and extension gaps. After one gap has been formed by a cut
(often the tibial), and possibly a release, the dimensions are
used to guide the level of another ‘dependent cut’, thereby
producing gaps of equal size (Fig. 25.10).

Tensors, Spreaders, and Spacers
The instrumentation required for gap balancing needs to
incorporate a device which tensions the gap in a controlled
manner, in order to measure and transfer the dimensions to
another gap. This will likely be a tensor of some kind. There
are other methods used to assess a gap (Fig. 25.21).
• A tensor allows a dynamic measurement of medial and
lateral implant spaces separately (Figs. 25.22 and
25.23).
• A distractor (or spreader) allows dynamic measurement of the implant space, without separating medial
and lateral (Fig. 25.23).
• A spacer provides a static measurement of the implant
space. However, serial static measurements, plus the
surgeons examination may provide some dynamism
(Fig. 25.24).

Tensor

Distractor

Spacer

Fig. 25.20 The soft tissue envelope consists of all the active and passive stabilisers of the knee

Fig. 25.21 Methods to assess the gaps. These may be static (spacer) or
dynamic (distractor or tensor). A tensor allows independent assessment
of medial and lateral sides of the joint

238

C Butcher and P Neyret

a

b

Fig. 25.22 (a, b): Use of a tensor to assess: (a) The extension gap. (b) The flexion gap
Fig. 25.23 (a, b):
(a) Depictions of a
spreader (left) and a
tensor (right). (b) A
spreader used on the
lateral side. Use of two
of these devices, one
on each side of the
joint, will act as a
tensor

a

a

b

b

c

Fig. 25.24 (a–c): Use of a static spacer to evaluate the tibiofemoral extension gap in a left knee. (a) Neutral. (b) Valgus stress. (c) Varus stress

25 Total Knee Arthroplasty: Steps and Strategies

Computer-Assisted TKA

239

The dependent cut/gap balancing technique is taken a step
further when using navigation; this allows the femoral
cuts and the resulting two spaces and alignment to be simulated before committing to either cut or the soft tissue
release (Fig. 25.25). The progress of the soft tissue balancing can also be assessed objectively in real time (see
Chap. 29). In this way, the soft tissue and bone manage-

ment can be planned independently, in theory allowing the
priorities of balance and alignment to be considered
separately. At this point, the decision-making process has
ceased to be linear; different strategic scenarios can be
rehearsed and the optimum parameters chosen
independently.
The concepts in the next part are designed to help the
surgeon separate some of these parameters during conventional TKA.

Fig. 25.25 Navigation allows the femoral cuts and their effect on
alignment and balance to be planned simultaneously, prior to committing to them. Here, the effect of the chosen level and rotation of the

posterior cut on coronal alignment and the gap symmetry is shown in
the upper picture. The lower picture shows the level and inclination of
the anterior cut in relation to the original anatomy

240

Part II Tibial and Femoral Spaces

C Butcher and P Neyret

a

The concept of flexion and extension gaps introduced by
Insall can be expanded to separate the tibial and the femoral
contributions (Fig. 25.26). This simplifies the analysis and
process of making equal gaps, and inherently highlights the
position of the prosthetic interface.
• Tibial space
The tibial space is defined by the single tibial cut and a
soft tissue release, if performed (Fig. 25.27). The flexion
and extension spaces are continuous with each other, and
the tibial space will be the same size and symmetry in
both positions.
• Femoral space
The posterior and distal femoral cuts define the femoral
space in flexion and in extension, respectively. The aim
of both femoral cuts is to produce a femoral space
which is, like the tibial subspace, constant throughout
flexion and extension (Fig. 25.28). In this technique, as
the tibial space is constant throughout flexion, the gap
balancing is achieved by gap balancing the femoral
space.

b

Fig. 25.26 (a) Flexion and extension gaps. (b) Tibial and femoral
spaces

In this part, the principles will be outlined. Further detail
will follow in Part III.

Fig. 25.27 Tibial space formed by the tibial cut, and possibly a soft
tissue release if necessary

Fig. 25.28 The relationship between the soft tissue attachments and
the distal and posterior cuts (i.e. the femoral spaces) is equalised in
extension and flexion

25 Total Knee Arthroplasty: Steps and Strategies

241

asymmetric tibial space with lateral laxity (Fig. 25.30).
This is termed ‘resection laxity’. A lengthening on the
concave side of the joint will be necessary to achieve a
rectangular space.

Creation of the Tibial Space
Achieving a rectangular tibial space varies depending on the
patient’s pathoanatomy:
1. N
 o Extra-Articular Deformity
Here the deformity is intra-articular from cartilage and
possibly bone wear. A cut orthogonal to the mechanical
axis will be almost parallel to the original joint line
(Fig. 25.29). The tibial space will be rectangular, or nearly
so. No alteration of the soft tissue will be required.
2. E
 xtra-Articular Varus Deformity of the Tibia
The deformity is extra-articular. A cut orthogonal to the
mechanical axis will produce an asymmetric cut, and an

3.	Convex Side Laxity
The soft tissues of the convex side are stretched, due to
chronic varus deformity and weightbearing. They will
need to be tensioned to some degree by filling the gap
with prosthesis. To achieve a rectangular gap, the concave
tissues will also have to be lengthened, more than the
original and natural length, to match the pathologically
lengthened convex side tissues (Fig. 25.31). Attention
will have to be paid to the prosthetic interface, which will
be altered from the level of the natural joint line.

Fig. 25.29 The deformity of the proximal tibia is intra-articular due to wear

242

C Butcher and P Neyret

Fig. 25.30 The deformity is extra-articular. The orthogonal cut creates an asymmetric space, requiring a lengthening on the concave side

Fig. 25.31 The soft tissues of the convex side are stretched and may need to be tensioned. To achieve a rectangular gap, the concave tissues will
also have to be lengthened, more than the original and natural length

25 Total Knee Arthroplasty: Steps and Strategies

Creation of the Femoral Space
To create equal rectangular femoral spaces in extension or flexion, the posterior femoral cut needs to match the distal femoral
cut (or vice versa). If there is asymmetry of medial and lateral
bone resection during the distal cut, this is reproduced in the
posterior cut. In a measured resection strategy, this can be simply achieved by cutting the same amount of bone from the condyles distally and posteriorly (Fig. 25.32). This strategy, with
allowance for wear of the cartilage and bone if necessary,
should produce a joint line that maintains the natural flexion
axis of the knee. Using this technique, the external rotation of
the femoral component is influenced by the asymmetry of the
distal femoral cut, not the asymmetry of the tibial cut. The
asymmetry of the tibial cut will be dealt with separately by
ligament balancing as previously described, and will have
effect in extension and flexion.
As with the tibia, the different patterns of femoral deformity guide the management:

243

rected (Fig. 25.33). The level of the distal cut is referenced from the less worn lateral condyle. If the cut is
­symmetrical, then so is the posterior cut—parallel to
the posterior condylar line (PCA) (Fig. 25.34). If the
distal cut is asymmetric and removes more lateral
femoral condylar bone than medial (Fig. 25.35), the
posterior cut is still made parallel to the PCA, as we
do not wish the posterior cut to produce any internal
rotation of the femoral component.
• Femoral valgus in a varus knee (infrequent—less
than 5% of cases in our experience). If the distal cut
is asymmetric and removes more medial femoral
condylar bone than the lateral, the rotation is set to
do the same from the posterior part of the condyles,
thereby externally rotating the femoral component
(Fig. 25.36). The cartilage wear (2–3 mm) or bone
wear on the medial condyle is considered, so that
the joint lines are reproduced. The posterior cut is
then a measured resection, but it is dependent on the
distal cut.

1. Varus Knees
• Either no femoral deformity or proximal femoral
varus deformity. The deformity, when present, is
from proximal or mid-shaft origin, i.e. proximal to an
intra-­medullary guide, and thus it will not be cor-

Fig. 25.32 The distal femoral bone resection can be used to guide the
posterior resection. The femoral contribution to the flexion and extension gaps will be equal

Fig. 25.33 Proximal varus will not be corrected by the use of an intra-­
medullary guide

244

C Butcher and P Neyret

a

b

c

Fig. 25.34 (a–c): Right knee (a). Distal cut removes equal bone from the medial and lateral condyles. (b, c). The 4:1 cut block is placed to also
remove symmetrical posterior bone from the condyles, i.e. 0° rotation
Fig. 25.35 Left knee. Here more bone will be removed from the lateral femoral
condyle than the medial, even allowing for wear

a

b

c

d

Fig. 25.36 (a–d): Right knee: (a, b) Distal cut removes more bone from the medial femoral condyle than the lateral. (c, d) External rotation
is set to cut the same posteriorly—removing more bone medially than laterally and externally rotating the femoral component

25 Total Knee Arthroplasty: Steps and Strategies

a

245

b

Fig. 25.37 (a, b): Right knee. (a) Distal cut removes significantly more medial than lateral bone due to lateral condylar hypoplasia. (b) The posterior cutting guide is externally rotated

2. Valgus knees
This often results from lateral condylar hypoplasia, but
may originate more proximally in the diaphysis.
• Lateral condylar hypoplasia. The intra-medullary guide is proximal to the deformity and will
correct it. The level of the distal cut is referenced
from the less worn medial condyle. The cut is
asymmetric, removing significantly more bone
from the medial condyle (Fig. 25.37). The rotation is set to do the same from the posterior part
of the condyles, thereby externally rotating the
femoral component.
• Valgus resulting from proximal deformity. The
intra-­medullary guide is located distal to the valgus
deformity and will not correct it (Fig. 25.38). The
posterior cut will be dictated by the asymmetry of
the distal cut as with the other examples. The extraarticular femoral valgus will remain after the procedure; the femoral space in flexion and extension
(i.e. the joint balance) will have been prioritised
over the alignment. If the remaining valgus is calculated to be unacceptable (rare, and usually due to
malunion), then a femoral osteotomy should be
considered first.

Fig. 25.38 The intra-medullary guide is located distal to a mild valgus
deformity and will not correct it

246

Conclusion
The goal is not only to have a tibiofemoral space that is the
same size and symmetrical shape in flexion and extension
(perhaps slightly more space in flexion), but also with a constant prosthetic interface. This optimises the biomechanics
of the collateral ligaments and the patellar femoral articulation. We prefer to separate the parts of the tibiofemoral gap
into tibial and femoral to achieve this.
• The size of the tibial space formed by the cut is equal
in flexion and extension. We make it symmetric with
soft tissue release if necessary.

C Butcher and P Neyret

• The size and shape of the femoral space is equal in
flexion and extension because the posterior cuts match
the distal cuts.
Achieving the goal by considering the femoral and tibial
spaces separately produces a prosthetic interface that is constant throughout flexion and extension.
In Part III, the steps in the creation of these spaces will
now be considered in the context of the priorities of the
procedure.

25 Total Knee Arthroplasty: Steps and Strategies

 art III Priorities and the Sequences
P
of the Steps
Although we strive for the perfect knee in every aspect, it is
helpful to have a basic order of priorities.
1.
2.
3.
4.

Alignment and balance in extension
Alignment and balance in flexion
Balance between flexion and extension
Joint line/Prosthetic interface

This order, however, is different to the sequence of the steps
of the procedure. Many of the steps of the procedure contribute to multiple priorities, either directly, or indirectly, and this
adds to the complexity of the TKA algorithm. First we will
consider the priorities, and then the sequence of the steps.

Fig. 25.39 (a–c): (a) A
strictly orthogonal mechanical
cut in this circumstance
contributes to an unacceptably
asymmetric gap. (b, c) Better
balance is achieved by
referencing from the distal
femoral anatomy at the
expense of alignment

a

247

1. Alignment and Balance in Extension
Overall coronal alignment: A rectangular extension gap
without laxity, orientated to align the tibial component
mechanically remains our first priority. The specific
­techniques for the femoral and the tibial cuts will be detailed,
but first their contribution to the overall coronal alignment
will be addressed:
• On the femoral side, although we aim for a mechanically
aligned limb, we will not correct proximal femoral varus if
in doing so it creates an asymmetric extension gap. In
varus deformity of femoral origin, a strictly orthogonal
distal femoral cut may not only contribute to an
­unacceptable asymmetry of the extension gap, but also the
opposite asymmetry in flexion (Fig. 25.39) (see Priority 2,

b

c

248

C Butcher and P Neyret

Alignment and balance in flexion). We therefore do not
vary our distal femoral cut according to pre-operative femoral mechanical axis measurement in varus knees, but reference the distal femoral anatomy with an intra-­operative
intra-medullary guide. Thus in many cases, we prioritise
the balance of the knee over the overall coronal alignment.
This allows good long-term results, as mild varus of the
femoral component itself has not been shown to lead to
loosening. The degree of overall varus accepted (perhaps
3–8°) will be a function of the quality of the ligament balancing (degree of laxity in the convexity), the level of
activity, and the life expectancy of the patient.

• Despite this kinematic approach to the femoral space, we
do not aim for either a mechanically aligned leg with
varus of the tibia and valgus of the femur (for instance 87°
and 93°, respectively, Fig. 25.40), or an overall kinematic
alignment with varying degrees of varus on the tibial side,
due to concerns over tibial component longevity.
• So in summary, we aim for a mechanically aligned limb,
with an orthogonal cut of the tibia, but we accept some
varus of the femoral component to encourage a balanced
knee (Fig. 25.41).

93°

87ࡈ°

Fig. 25.40 Mechanical alignment with a varus tibial cut and a valgus
femoral cut

Fig. 25.41 An orthogonal tibial cut, but the proximal femoral deformity is left to improve the balance

25 Total Knee Arthroplasty: Steps and Strategies

 istal Femoral Cut
D
Although navigation can estimate the mechanical axis of
the femur, and identify the flexion axis, in conventional
TKA intra-medullary guidance is usually used, partly as
extra-­medullary guidance using fluoroscopy for the femoral head is not reliable. The technique of insertion of the
intra-­medullary guide affects the alignment. We prefer to
enter the distal femur anterior to the PCL insertion, which
is a reproducible and functional landmark in coronal and
sagittal planes (Fig. 25.42). This lies at a point nearer the
exit of the line of the distal intra-medullary canal, termed
Femoral Anatomic Axis II, than that of Femoral Anatomic
Angle I (Fig. 25.43). If one chooses this point, when the
prosthesis is then centred on the condyles, it will be trans-

249

lated laterally with reference to this starting point, inducing slight varus (Fig. 25.44). We have found that using a
7° valgus cut referenced from the intra-medullary rod
then gives the most predictable and satisfactory alignment. If the distal femur is entered in the centre of the
trochlea, the rod will be nearer the Femoral Anatomic
Angle I, and require setting a lower valgus angle on the
instrumentation.
Both of these methods use the distal femur as a reference,
and thus ignore the proximal femur. If there is deformity in
the proximal femur, it will not be corrected and the result is
that the distal femoral cut prioritises the balance of the knee
rather than the alignment. This is in contrast to a technique of
pre-operative imaging, or intra-operative navigation, where
the centre of the femoral head, and thus the femoral mechanical axis will serve as the reference. In this latter technique,
and with proximal varus, there will be an asymmetric cut,

Fig. 25.42 Entry for the intra-medullary reference guide, anterior to
the PCL

Medialised - valgus

Fig. 25.43 Right knee; the Femoral Anatomic Axis II—midline of the
distal femur. Left knee; the Femoral Anatomic Axis I—centre of shaft to
centre of knee

Lateralised - varus

Fig. 25.44 Translation induces small mechanical angle changes

250

C Butcher and P Neyret

removing more bone laterally, creating an asymmetric gap, as
shown above. It will thus prioritise alignment over balance.
We will alter the angle of the cut in two circumstances.
• Lateral femorotibial osteoarthritis. There is a tendency to
post-operative valgus, and we cut at a 5° valgus angle.
• In obese patients where the distribution of fat is largely in
the lower limb. These people may already have a clinical
pseudovalgus of their knees despite having medial femorotibial osteoarthritis, and correcting the deformity either
causes or exaggerates this. They often have a very wide

Fig. 25.45 (a, b): (a) Full
deformity correction in obese
patients may exacerbate or
cause pseudovalgus and
consequent difficulties of gait.
(b) This patient attempts to
reduce the wide based stance
by crossing the knees, but will
be unable to do so during gait

a

based gait to prevent their knees impinging upon each
other during swing phase (Fig. 25.45). This is most
­pronounced when the d­ istance between the centre of the
hips is smaller than the distance between the centre of the
knees when standing, due to the thickness of the medial
soft tissues (Fig. 25.46). The patients compensate with a
Trendelenberg gait and a short stance phase, both of
which are awkward and raise the energy consumption of
ambulation. Correcting the true varus completely will
aggravate this situation and so we reduce the cut angle to
6° valgus, rather than 7° (Fig. 25.47).

b

25 Total Knee Arthroplasty: Steps and Strategies

a

b

Fig. 25.46 (a–c): (a, b) Distance between knees exceeds that between
hips. Full correction will cause a functional deformity of the lower
limbs reminiscent of this famous landmark. A design for stability, but

Fig. 25.47 The left knee has been slightly overcorrected (2°). If the
right is fully corrected, the intermalleolar distance will be excessive

251

c

not mobility. (c) In this case, despite mild varus in both knees, there is
still significant pseudovalgus

252

C Butcher and P Neyret

Tibial Cut
A.	Inclination of Cut: Achieving an Orthogonal Cut
The tibial cut is orthogonal to the mechanical axis in the
coronal plane and sagittal planes. The effect of sagittal slope
on the flexion gap will be discussed later (in ‘Balance
between flexion and extension’). Combined independent
intra-medullary and extra-medullary referencing is possible
with our current instrumentation and is used routinely
(Fig. 25.48). This caters for a number of clinical situations:
(i)

In the case of a straight tibia:
• The intra-medullary (IM) guide is reliable in both
coronal and sagittal planes, and is especially useful
in obese patients (the tibia is often straight, as the
deformity is mainly due to wear in these patients).
• The extra-medullary (EM) guide is satisfactory,
but may be less precise in the obese, in both
planes, due to the difficulty of identifying the
bony landmarks at the ankle (Fig. 25.49).

Fig. 25.48 Right knee. Combined intra- and extra-medullary guidance. The arrow shows the intra-medullary rod, which is fully inserted
to reach the ankle

		 (ii)	In the case of extra-articular tibial deformity (simple
or complex), or a narrow tibia:
• The IM guide may not reach the ankle. In the
case of a tibial deformity, there will be a risk of
under correction (Fig. 25.50). The chosen tibial
aperture may be changed to pass the rod fully,
but further asymmetry of the cut and difficulty
balancing may result (Fig. 25.51) (see Chap. 30,
TKA after valgus high tibial osteotomy).
Subsequent tibial component translation away
from the aperture in an attempt to obtain maximum bone coverage will introduce alignment
change and needs to be considered. For instance,
with a valgus tibial deformity, the aperture would
need to be medialised to pass the rod, leading to
more medial laxity. Subsequent lateral translation of the component produces varus. It is possible to exploit these facts when dealing with
malunions that involve translation as well as
angulation. This may be difficult to plan in the
tibia, as tibial instruments do not generally allow
variable angles of cut, unlike femoral instrumentation (see Chap. 42, Case 10).
• A narrow tibia may preclude the use of IM guidance completely (Fig. 25.52).
• With an EM guide, the proximal and distal reference points are chosen, providing the mechanical
axis of the tibia, and ignoring the anatomic tibial
axis and ­
deformity in between (Fig. 25.53).
Nevertheless, some aspects must be taken into
consideration:
–– With severe deformity, previous HTO or
metaphyseal malunion, templating is advised
as unanticipated impingement of the keel and
the tibial cortex may occur (Fig. 25.54) (see
Chap. 30).
–– Sagittal plane alignment may be difficult to
achieve accurately, especially in obese individuals. Whilst using EM referencing, we
therefore recommend using additional IM
referencing with a short rod in the proximal
tibia, just for the sagittal alignment
(Fig. 25.55). An advantage is that if there is
sagittal extra-articular deformity, the sagittal
cut will prioritise joint balance over sagittal
alignment (Fig. 25.56). Pre-operative clinical
assessment can anticipate if knee and ankle
movement can compensate for the uncorrected deformity without functional loss.
–– Care must be taken when using the anatomy
of the foot as a reference, as this may introduce error due to mobility or deformity (tibial
torsion will also introduce error).

25 Total Knee Arthroplasty: Steps and Strategies
Fig. 25.49 (a, b) Landmarks
at the ankle may be difficult
to find when the patient is
well covered

a

Fig. 25.50 Valgus will remain after a cut perpendicular to this rod. The
distal reference is medial to the mechanical axis

–– The proximal end of the EM guide must be
aligned with the sagittal axis of the tibia
(Fig. 25.57) and parallel to the distal end of the
guide. Any rotation away from this will introduce
a coronal plane change that will affect gap symmetry (Fig. 25.58). This is in contrast to the IM
guide, which will still produce a predictable cut
in both planes if it is rotated in the axial plane.

253

b

Fig. 25.51 A medialised proximal reference point here will produce
further cut asymmetry and difficulty balancing

B. Inclination of Cut: Effect on Tibial Space Symmetry
The influence of deformity on the coronal space symmetry was discussed in Part II. Any asymmetry of the
tibial space from the resection, as well as from contracture or convex side laxity will be dealt with by ligament
balancing (Fig. 25.59). In the varus knee, this space is
made symmetrical and rectangular by a medial collateral

254

C Butcher and P Neyret

Fig. 25.52 A narrow canal may preclude the use of an intra-medullary
guide

a

Fig. 25.53 Extra-medullary guidance has been effective in creating an
orthogonal cut in this mildly deformed tibia

b

Fig. 25.54 Slight keel impingement was necessary to align the component properly in this more severely valgus tibia. Note the lateral
femoral condyle osteotomy for soft tissue balance (See Chap. 30)

Fig. 25.55 Left knee. Combined intra- and extra-medullary guidance
in a varus knee with a varus tibial deformity. The intra-medullary rod is
not fully inserted and is used for sagittal alignment only

25 Total Knee Arthroplasty: Steps and Strategies

255

Fig. 25.56 The sagittal cut is
prioritised for the joint (left),
rather than the alignment
(centre). Bone resection and
balance would be
unacceptable if sagittal
alignment was mechanically
referenced (right)

Fig. 25.57 (a, b): Right
knee. (a) Care is taken to
align the proximal end of the
EM guide, and thus the cut
block, with the sagittal axis of
the tibia. Here, the rotation is
guided by the centres of the
tibial condyles, as well as the
tibial tuberosity. (b) The
proximal part (star) and distal
part (arrow) of the EM guide
are parallel

a

release. As there is only one cut, the symmetrical tibial
space produced by the release will be the same throughout flexion and extension. We believe the influence of
the MCL release will be the same in flexion and extension, with only minor adjustment between the two
possible.
If the asymmetry of the cut is small, a medial release is
not necessary. If it is moderate, either pie crusting or formal MCL release from the tibia can be performed. In the
latter case, we prefer to add a longer stem (total length at

b

least 75 mm) as the blood supply to the medial tibial condyle may be affected, with a risk of late collapse, especially if there is residual varus or obesity (Fig. 25.6).
There is a limit, however to how much deformity can be
compensated by an asymmetric cut and soft tissue balancing without adding constraint to the prosthesis. Larger
extra-­articular deformities (in excess of 10°) nearer the
knee will have a greater effect on the mechanical alignment and may benefit from combined osteotomy and
TKA (see Chap. 42, Cases 2a and 2b).

256

C Butcher and P Neyret

	C. Level of Cut
The level of the tibial cut will affect the height of the tibial
space and segment directly though removal of more, or
less, bone. However, there may also be an indirect effect,
as the level of the cut will also affect the degree to which
the peripheral soft tissues are released from bone. The
increased soft tissue laxity may cause the space produced
by the cut to be larger than the thickness of bone removed,
and this will be relevant to those using pre-operative templates. This principle is still relevant when using navigation, as it is currently impossible to anticipate or precisely
predict this type of resection laxity by virtual surgery
using any current CAS system. The effect may be most
obvious laterally where the anterolateral ligament and lateral capsule insert over a proximal-­distal distance of several millimetres (Fig. 25.60). This attachment is
approximately at the level of a routine tibial cut, but cuts
more than 14 mm in total on the medial side, made for
instance to accommodate augments, or after HTO with
severe deformity hypercorrection, have consequences for
ligament balancing too. These effects may be more pronounced when the patient is short. This is because all
component sizes require the same thickness of bone
resection, and the cut is thus proportionally deeper in a
shorter tibia.

Fig. 25.58 Right knee. External rotation of the proximal end of the
extra-medullary guide (star) in relation to the distal end (arrow) results
in lateral translation and angulation of the cut block. The angle between
the orthogonal intra-medullary guide (solid line) and the extra-­
medullary guide (dashed line) is visible proximally

Fig. 25.60 Coronal MRI showing insertion of the anterolateral capsular structures on the proximal tibia (white arrow). A 10 mm resection
level is depicted (white dotted line). These structures may be more easily damaged in a small knee

Fig. 25.59 The asymmetric tibial space is equalised by medial release

25 Total Knee Arthroplasty: Steps and Strategies

2. Alignment and Balance in Flexion
The ‘flexion gap’ represents the situation through most of the
arc of movement, even though for simplification we consider
it at 90°. Consideration of this gap is challenging in two ways:
• The femoral cut which forms the flexion gap also defines
the patellofemoral gap. Consideration of both become
increasingly relevant as TKA is performed more widely
in cultures which spend a high proportion of their time in
postures of flexion such as squatting, sitting on the floor,
or praying. The surgeon faces a double compromise; he/
she must consider the coronal alignment of the knee in
flexion, ensuring there is no malalignment during stair
climbing and squatting, yet avoid significant asymmetry
of the tibiofemoral gap. At the same time, with one cut
angle and one component, he/she must optimise the
mechanics of the patellar femoral joint.
• The success of the flexion gap balancing is less obvious to the surgeon post-operatively:
–– Detection of both malalignment and laxity in flexion
through clinical assessment is challenging (Fig. 25.61).
–– Radiographic assessment in flexion is difficult,
especially under loaded conditions, and is therefore
not commonly performed (Fig. 25.62).
Intra-operative assessment is therefore key, and this is one
area where the documentation and focus achieved in navigated knees may be an advantage.
The coronal alignment in flexion is dictated by the posterior femoral cut and thus the rotation of the femoral component. Our technique for this has been described in Part II and
relies on equalising the femoral cuts in extension and flexion,
thereby equalising the femoral extension and flexion spaces.
It is recognised that if the natural axial alignment of the fem-

Fig. 25.61 Assessing lateral laxity in flexion. The limb is taken into
full external rotation, and then relaxed until the contact between lateral
femoral condyle and polyethylene is felt at the joint line. Repeated
examination gives an approximation of the maximum lateral gap. A
similar test in internal rotation is made for the medial laxity

257

oral condyles is preserved in TKA by reproducing the PCA
(0° external rotation of the posterior femoral cut), but the
tibial cut is orthogonal and asymmetric, the knee may be in
mild valgus in the flexed position, compared to the pre-­
operative situation. However, this will match the alignment
of the knee in extension, as a mechanically aligned limb will
be in mild valgus compared with the average natural knee.
This is in keeping with a principle of maintaining equal
alignment and tension throughout the range of motion.
Other techniques of deciding the femoral rotation include:
• Using the epicondylar axes: If it is decided that the
­posterior cut needs to be parallel, or at a certain angle to
an epicondylar axis, then it follows that the distal cut
would also have to be the same angle to this axis to
achieve symmetrical femoral extension and flexion
spaces (Fig. 25.63). The relationship of the epicondylar
axes to the femoral mechanical axis, and the distal and
posterior femoral articular surfaces (distal femoral angle
and condylar twist) is variable between patients with significant standard deviation (Fig. 25.64). One can utilise
pre-operative multiplanar imaging, or estimation of the
flexion axis intra-operatively to guide the cuts, but in
practice we simply equalise the resection from the distal
and posterior part of the condyles.
• Using a fixed external rotation: Using a standard external
rotation in relation to the PCA for the posterior cut (such as
3°) ignores the variable distal femoral angle and condylar
twist found across the population and this may represent
an unnecessary compromise for many individuals. The
variation is particularly significant in valgus knees.

Fig. 25.62 Varus stress view in 90° flexion showing 12 mm lateral
opening

258

C Butcher and P Neyret

Fig. 25.63 Posterior and distal cuts at the same angle to the
transepicondylar axis

aTEA

Condylar
twist angle

PCA

Fig. 25.64 Condylar twist angle

• Prioritising the femoral rotation on a significant tibial
deformity, i.e. using the asymmetry of the tibial cut: Using
3° external rotation when an orthogonal tibial cut is in 3°
of valgus in relation to the natural joint line is logical if
the orthogonal distal femoral cut is also 3° varus from the
natural joint line. This situation does occur outside of
textbooks, but certainly not in all knees. Moreover, if the
deformity of the proximal tibia is greater than 3°, say 8°,
it would not be logical to externally rotate to that degree.
In this bone balancing technique, the flexion balance is
prioritised over the flexion alignment. The tibial varus has
been removed by the tibial cut, but subsequently recreated in the posterior femoral cut, in the desire to gain flexion balance. Subsequently, one would have to make a
similar asymmetric distal femoral cut and accept a similar
varus alignment in extension, or alternatively accept different shaped gaps in flexion and extension (with medial
tightness in extension).
• Using gap balancing: If the surgeon chooses gap balancing instrumentation, the femoral rotation then depends on
the asymmetry of the tibial cut and any ligament release.
A careful compromise must be found between balance
and alignment; prioritising the flexion gap first in a varus

knee may mean accepting either varus or medial tightness
in extension. Simulating both cuts and gaps in parallel
with navigation may be the best way to find this compromise without excessive rotation, or malalignment. Using
conventional instruments, the release needs to be considered and performed carefully, and secondary checks utilised to verify the posterior cut angle, for instance
information from pre-operative multiplanar radiology, or
from intra-operative distal cut thicknesses. The natural
laxity of the lateral compartment in flexion, and any persistent medial tightness, may result in excessive external
rotation. The effects of varus alignment in the flexed
weightbearing TKA are not clear, but external rotational
deformity of the distal femur after fracture malunion is
known to increase medial compartmental stress, and the
overall varus in flexion needs to be considered.
The other consequences of excessive or inappropriate external rotation of the femoral component may be
summarised as follows:
–– Notching of the lateral femoral cortex or an increase
in size of the femoral component. An increase in AP
size, and thus ML size, increases the chance of
medial or lateral overhang, and consequent soft
tissue irritation or tightness of the knee. The chance
of size mismatch between the two components also
increases, depending on the prosthetic system used.
–– Abnormal patellar femoral kinematics.
–– Flexion-extension gap shape mismatch in proximal
femoral varus deformity (see Fig. 25.39a); where
the distal cut has removed more bone laterally, but
the externally rotated posterior cut has removed
more bone medially. The extension gap is trapezoidal, wider laterally. The flexion gap is trapezoidal,
wider medially (Fig. 25.65). In this type of deformity, we perform a posterior cut in a neutral position, accepting mild medial laxity in flexion. Only
internal rotation of the posterior cut would equalise
the shape of the tibiofemoral gaps, but at the
expense of the patellar femoral joint.
Our approach to rotation is the natural result of considering the femoral space separately. The cut orientation, and
therefore the space and balance, is the same in extension and
flexion. The advantages are that the technique is individualised, reproducible, and does not require special instrumentation, navigation or pre-operative imaging.
Tibial component rotation is chosen to optimise the patellar tracking as well as the femorotibial congruency, and here
a further compromise may be necessary. Our method is to
mark the natural AP and ML axes on the tibial plateau prior
to the tibial cut using the centre of each condyle, the PCL,

25 Total Knee Arthroplasty: Steps and Strategies

and the tibial tuberosity as references (Fig. 25.66). The axes
can be transferred to the rest of the tibia for use after tibial
resection by marking an anterior point on the anterior tibia
Fig. 25.65 (a–c): Right
knee. Equalising the spaces in
flexion and extension would
require cutting the femur at
the same angle to the
transepicondylar axis distally
and posteriorly, thereby
internally rotating the
prosthesis. (a) Distal cut
removing more lateral than
medial condyle. Trapezoidal
gap wider laterally. (b)
Posterior cut internally
rotated, at the same angle to
the transepicondylar axis as
the distal cut. Trapezoidal gap
wider laterally so flexion and
extension gaps can be
balanced, but unacceptable
for patellar biomechanics. (c)
Posterior cut externally
rotated, removing more bone
medially. Flexion gap wider
medially, extension gap wider
laterally

259

for the sagittal axis, and a drill hole for the centre of each
condyle for the transverse axis, allowing triangulation
(Fig. 25.67).

b

a

Fig. 25.66 AP and ML axes on the tibial plateau prior to the tibial cut,
using the centre of each condyle, the PCL, and the tibial tuberosity as
references

c

Fig. 25.67 The axes can be transferred to the rest of the tibia for use
after tibial resection by marking an anterior point on the anterior tibia
for the sagittal axis, and drilling the centre of each condyle for the transverse axis

260

 . Balance Between Flexion and Extension
3
(Gap Balance)
The balance between flexion and extension is a reflection of
the relative size of each part of the gap. It is recognised that
this traditional analysis ignores the mid flexion balance, but
the better the flexion and extension gaps are balanced, the
less likely there will be instability between them. The priority for us however is the balance in extension, and a small
degree of flexion laxity is acceptable, especially with appropriate orientation of the cuts.
In the absence of pre-operative fixed flexion deformity,
equalising the level of the distal and posterior femoral cuts
will usually produce symmetry, and our method to achieve
this is described in Part II, and below (see ‘Prosthetic interface’). However, there are a number of factors to be
considered:

Pre-Operative Factors
• Pre-operative fixed flexion deformity (Fig. 25.68)
• Large posterior osteophytes (Fig. 25.69)
Fixed flexion deformity is usually and conveniently dealt
with by proximalising the femoral cut, thus reducing the
femoral segment, although the pathology often lies with
contracture of the active and passive posterior soft tissue
structures. However, after removal of posterior osteophytes,
controlled posterior release from femur or tibia is difficult, and
the consequence of over-release (uncontrolled recurvatum
during gait) is difficult to treat without changing to a
constrained prosthesis. This particularly is the case with the
posterior medial tissues including the menisco-tibial ligament
and posterior oblique fibres. They may not heal and require
mechanical stabilisation with a hinged prosthesis rather than
soft tissue management. A compromise of slight residual fixed
flexion deformity in combination with mild flexion laxity will
be more acceptable to the patient than uncontrolled recurvatum
with a perfect flexion gap. There is a trend of improvement in
post-operative fixed flexion deformity over time.

Fig. 25.68 Pre-operative fixed flexion deformity

C Butcher and P Neyret

Operative Factors
• Femoral component condylar geometry.
• Resection or insufficiency of the Posterior Cruciate
Ligament (PCL). The flexion gap increases
preferentially.
• Femoral cut levels.
–– The AP position of the femoral starting hole may
affect both orientation and translation depending on
the instrumentation, and using the PCL as a landmark is reliable. It is appropriate to reference the
sagittal orientation of the femoral component from
the distal femoral anatomy, not a line from the centre of the femoral head to the centre of the knee.
However, an IM rod placed from our recommended
entry hole to the isthmus of the femoral canal may
be in some extension compared to the distal femur,
and we reference 3° flexion from the rod (Fig. 25.70).
–– The level of anterior and posterior femoral cuts is
determined by
–– the chosen reference; anterior or posterior
–– the component size
–– the chosen centre of rotation (see below Part 4,
Prosthetic interface)
• A priority must be made for the level of the anterior or
posterior cut, at the same time accommodating the offthe-shelf prosthetic sizes.
• The level of the posterior cut in relation to the level of
the distal cut affects the gap balancing. It also determines the correct posterior offset to maximise flexion by
preventing posterior impingement (Fig. 25.71).
• The level of the anterior cut will affect the patellar
femoral gap, and in most cases the anterior offset will
ideally remain unchanged to maintain physiological

Fig. 25.69 Posterior osteophytes, usually femoral, limit range of
movement

25 Total Knee Arthroplasty: Steps and Strategies

261

patellar loads. The level of the anterior cut is ideally
adjusted to resect the same thickness of bone from the
anterior condyles and trochlea as will be replaced by
the components although most systems’ anterior femoral reference is simply from the anterior cortex proximal to the trochlea (Fig. 25.72). The trochlear depth
and condylar height will therefore also be dictated by
the geometry of the femoral prosthesis.

• The compromise between anterior and posterior cuts
and component size is further complicated as the
choice of AP component size will dictate the ML
dimensions too, which will in turn affect the tension
of the medial and lateral patellar restraints. To this
end, we recommend choosing a smaller femoral prosthesis, if bone quality allows.
• Tibial cut sagittal orientation.
–– We use a posterior stabilised prosthesis, and a 0° sagittal tibial slope, referenced from the proximal tibia
(Fig. 25.73). This reduces the forces on the posterior
part of the polyethylene, prevents anterior tibial subluxation during weightbearing, and avoids excessive
laxity in flexion in a PCL substituting design.
–– Conversely, a 0° cut in a tibia with an exaggerated
pre-operative slope will tighten the flexion gap in
relation to the extension gap. Using a positive
slope will loosen the flexion gap preferentially, due
to the more posterior position of the femoral condyles on the tibia with increasing flexion and may
be more appropriate to achieve balance in a PCL
conserving TKA.

Fig. 25.70 The distal cut is flexed in relation to an IM rod due to the
femoral bow

Fig. 25.71 The correct posterior offset will affect gap balancing and
prevent posterior impingement in flexion

Fig. 25.72 (a, b): The most
common anterior femoral
referencing is from the
anterior cortex proximal to
the trochlea (dotted line). The
anterior offset of the trochlea
and the femoral condyles
(solid lines) will also be
influenced by the femoral
component geometry

a

b

262

C Butcher and P Neyret

that due to intra-articular wear, keeping the prosthetic
interface at the level of the native joint line is straightforward. The cut is usually symmetric, or almost so. The
thickness of bone that is removed on the convex reference
side will be replaced by the implant. On the tibia, this
reference is usually situated on the convex, longer side.
This will be the centre of rotation of the angular correction and thus there will be no change in level of joint line
on this reference side. The soft tissues on the concave side
are brought out to length, or surgically lengthened with a
release if necessary, in order to restore the height of the
native joint line on the concave side and produce a rectangular space. This may be guided by the level of the meniscal rim.
B. One-Sided Change in Prosthetic Interface
Extra-articular bone deformity (unreducible) (Fig. 25.30).
Where unilateral lengthening has to be performed because
of an asymmetric cut due to deformity, the level of the
prosthetic interface may still be established from the reference side. On the side of the lengthening, there will be
an increase in the size of the segment, and an increased
height of the prosthetic interface in relation to the native
joint line. With increasing extra-articular deformity, there
is increasing change in the level of the interface.
C.	Both Sided Change in Prosthetic Interface

Fig. 25.73 Zero degree tibial slope, referenced from the proximal tibia

4. Prosthetic Interface
General Principles
In the natural knee, the articulating interface is termed the
joint line. In a TKA, the femoral and tibial and patellar segments meet at the prosthetic interface. The position of this is
important in relation to:
• The origins of the collaterals, and other components of
the femorotibial soft tissue envelope
• The position of the patella and the origins of the patellar
femoral and patellar tibial restraints
Several pathological situations may exist, which dictate
the likelihood of a change in the position of the prosthetic
interface, compared to the native joint line:
A. No Change in Prosthetic Interface
Intra-articular deformity, from joint wear (Fig. 25.29).
Where there is no deformity correction needed apart from

Most often in these cases, there is lengthening on both sides
of the joint; the gap produced by the pathology, bone cut,
and soft tissue balancing exceeds the thickness of the bone
resected. The cause may be bony or soft tissue pathology:
• Convex side laxity (Fig. 25.31). This may occur with
chronic and severe deformity (Fig. 25.74). In this case,
both sides of the joint will need to be addressed to produce a rectangular gap; on the concave side, soft tissue
release will need to be performed (becoming longer than
the pre-disease state), whilst the slack on the convex side
will need to be taken up. This will result in a potential
both sided lengthening, with an increase in limb length by
a few millimetres (Fig. 25.75). The space produced will
exceed the thickness of bone resected. The segment may
need to be increased by using a thicker component to tension the soft tissues adequately, and the level of the prosthetic interface will have been altered.
• Iatrogenic lengthening of the ligaments or soft tissue.
This may occur when the anterolateral ligament and
capsule is sectioned during a routine tibial cut, increasingly likely with greater cut depth (Fig. 25.60). If there
is metaphyseal deformity, for instance after an HTO,
the obliquity of the orthogonal cut may increase the
likelihood of convex side soft tissue injury.
• Iatrogenic extra-articular deformity. This is seen in
patients who have previously undergone high tibial

25 Total Knee Arthroplasty: Steps and Strategies

263

osteotomy. The usual reference point will be in a different location with regard to the rest of the plateau
(Fig. 25.76). Less bone will need resecting on the reference side to avoid an unacceptably large gap from the
bone removal, or damage to the peripheral soft tissue
envelope (see above and Chap. 30). There is potential
for change in the level of the prosthetic interface.
• Significant intra-articular deformity from bone wear.
This is more common in lateral tibiofemoral compartment osteoarthritis (Chap. 27) and in rheumatoid
arthritis. There may be no natural point of reference
due to tri-­compartmental disease, and thus the exact
natural joint line may be difficult to define.

Fig. 25.74 With chronic and severe deformity, there may be true convex side laxity
Fig. 25.75 Uncommonly,
laxity on one side of the joint
may require some lengthening
to achieve stability

Where the lengthening is both sided, thicker implants will
need to be used, or bone resection will have to be more sparing. The degree to which these are required will be dictated
by the accepted laxity.
• If less bone is removed from the tibial side, or if a thicker
polyethylene component is used, the tibial segment will be
increased, i.e. there will be a greater distance between the
prosthetic interface and all the soft tissue insertions on the
tibia (MCL, patellar tendon, capsule) and on the fibula
(LCL) (Fig. 25.77). This will clearly have an effect on femorotibial biomechanics. The prosthetic interface will have
been ‘raised’ (proximalised) in relation to the original
native joint line and the patella (Fig. 25.19). If the tibial
segment is larger due to the thickness of the polyethylene
rather than a sparing tibial cut, an additional consideration
will be the resulting forces at the implant bone interface.
These will be a function of the relative distances of the
prosthetic interface (where torque forces are applied) and

264

C Butcher and P Neyret

Fig. 25.78 Torque at the plateau cement bone interface will be
proportional to the length of keel and height of the prosthesis

Fig. 25.76 The lateral joint reference point in this varus knee is
altered by previous surgery. If the cut is made at the usual distance
(e.g. 10 mm) from the lateral plateau, there will be an excessive cut of
the medial tibia. A thinner cut is appropriate, perhaps like that shown
here in red

Fig. 25.79 The analogy of a keelboat

Fig. 25.77 Greater distance between the prosthetic interface and
MCL origin in the left knee

the distal tip of the prosthesis (where the forces may be
maximally resisted) to the implant bone interface (which
can be debonded by repetitive stress) (Fig. 25.78). As the
distance between the prosthetic interface and implant bone

interface becomes larger, so too must the support from the
prosthetic keel. One analogy is a keelboat, where the keel
provides counter-torque to the forces acting on the sail
(Fig. 25.79). If balancing the knee requires a very thick
polyethylene insert, consideration should be given to
achieving joint stability by changing the type of prosthetic
constraint instead, for instance, by using a rotatory hinge
knee. The prosthetic interface and the biomechanics can
then be kept within acceptable limits.
• If less bone is removed from the femoral side, or if the
bearing surface of the femoral component is moved more
distally by the use of augments, the femoral segment will
be increased, i.e. there will be a greater distance between
the prosthetic interface and the soft tissue insertions on
the distal end of the femur (MCL, LCL, capsule), especially in extension (Fig. 25.80). Again there will be effects
on the femorotibial biomechanics; more so than with tib-

25 Total Knee Arthroplasty: Steps and Strategies
Fig. 25.80 (a, b): (a) As the
femoral segment is
lengthened (red arrow), the
tibiofemoral biomechanics
will change, but the
relationship between the
prosthetic interface and the
patella remain constant (blue
arrow). (b) The femoral
segment has been increased
by under-resection and tibial
decreased by over-resection.
The total length of the soft
tissues is unchanged from the
natural state, but the joint line
is distalised

a

265

b

Fig. 25.82 The patellar restraints will be subject to length change and
tension, especially in flexion

Fig. 25.81 The proportional length of the tibial and femoral parts of
the MCL in this patient are shown

ial segment lengthening, due to the shorter distance to the
collateral insertions (Fig. 25.81). With isolated femoral
segment lengthening there will be no change in patellar
height in relation to the femorotibial prosthetic interface
or trochlea; however, the distance from the patellar femoral interface to the origins of the lateral and medial patel-

lar restraints on the femur (MPFL, retinaculum, etc.) will
increase, especially in flexion (Figs. 25.80a and 25.82),
causing increased tension in these tissues.
The limits of lengthening of tibial and femoral segments
relate to the kinematics of the patellar femoral joint and
the collateral ligaments, bearing in mind that the effect of
changes in femoral segment will be proportionally larger.
We suggest a lengthening of tibia and femur in the ratio
2:1. In absolute terms, the limit is probably around 4 mm
for the tibia (e.g. standard 10 mm tibial cut, and increase
of polyethylene of 4 mm, total tibial component 14 mm),
and 2–3 mm for the femur. More than this will suggest the
use of a more constrained prosthesis (rotating hinge) to

266

preserve the level of the prosthetic interface. The limit of
accepted laxity is multifactorial and it is difficult to state
an absolute figures. It relates to the alignment and adductor moments, whether the laxity is medial or lateral, the
flexion/extension balance, activity level, age and weight of
the patient.

Level of Prosthetic Interface in Flexion
This level is a function of the relative sizes of the tibial segment and the femoral segment in flexion. Certain factors will
particularly influence the posterior condylar offset, and thus
the interface in flexion:
• Type of cut. A measured resection (using independent
cuts) is less likely to alter the level than a dependent
cut, which may do so.
• Reference system. A posterior referencing system will
help keep the interface at the same level as the native
joint line, as opposed to an anterior referencing system
which will prioritise the anterior space.
• Prosthetic design and inventory. Increasing number of
sizes and femoral/tibial compatibility will reduce the
need to anteriorise a femoral component to prevent
notching.

a

C Butcher and P Neyret

• Sagittal flexion. Flexion of the distal cut may increase
the femoral segment (Fig. 25.71).
• The centre of rotation. If the centre of the knee is chosen for an externally rotated cut, the thickness of medial
condylar bone removed will be more than the replaced
metal, and the level of the medial interface will have
changed from the native joint line, albeit by a small
amount (Fig. 25.83). The centre can be chosen however
to influence the level change. For instance, in a valgus
knee, there is often pre-existing medial laxity, and the
external rotation can be achieved by ‘building up’ the
lateral condyle, rather than resecting more posterior
medial condyle. The term ‘build up’ will apply to the
use of spacers of some kind to add to the foot of the
rotation guide to remove less bone than is replaced with
metal (Fig. 25.37b), increasing the lateral segment.
Here, the centre of rotation is the posterior articular
surface of the medial condyle, and exactly the same
amount of medial bone will be removed as replaced
with metal (Fig. 25.84). The concave lateral side will
have been lengthened (in flexion) to match the
lengthening in extension that is required to correct
valgus deformity in extension. This technique helps to
prevent medial laxity in flexion, especially when there
is a flexion extension mismatch due to pre-operative
fixed flexion deformity. In this situation, the flexion gap

b

Fig. 25.83 (a, b). Centre of rotation central left knee. External rotation centrally requires resection of more bone posterior medially and anterior
laterally, compared with a neutral rotation

25 Total Knee Arthroplasty: Steps and Strategies

a

267

b

Fig. 25.84 Centre of rotation at posterior medial condyle left knee. (a) The medial joint level is unchanged from pre-operative, but more anterior
lateral resection is required, increasing the chance of anterior notching. (b) Smaller and larger sizes are superimposed, both with the same rotation.
Upsizing reduces the notching, but may cause medial or lateral overhang

will tend to be more lax already, especially on the
medial side. The disadvantage of this technique is the
resulting increase in AP dimension of the femoral
prosthesis (already necessary with external rotation),
which can lead to ML overhang, and component
mismatch. A solution is to use a combination of the two
techniques as a compromise, placing the centre of
rotation on a line between the centre of the knee and the
posterior medial condyle (see Chap. 27, Fig. 25.17).

Sequence of Steps in TKA
The sequence of the steps is often different to the sequence
of priorities. This discrepancy is large due to the currently
used surgical instrumentation. The challenge is to proceed
with the steps but keeping in mind the priorities, thinking a
few steps ahead.
Although various sequences are possible, most often the
tibial cut is made first. This is in part because practically it
can be the first step in all of the general approaches of
TKA. There are several advantages:
• It contributes to the first priority of TKA—alignment
and balance in extension.
• The coronal orientation of the tibial cut is probably
more critical than the femoral in terms of long-term
fixation and can be set at the start of the procedure.
• There is only one cut, and only one tibial space, which
affects the femorotibial gap throughout flexion and extension. This simplifies the subsequent intra-operative
decisions.

• If necessary, the tibial segment can be adjusted:
–– It can be reduced easily later in the procedure by
distalising the level of resection (e.g. if the gap is
tight throughout extension and flexion) (Fig. 25.85).
Proximalising the femoral cut is less simple once
the chamfer/box cuts have been performed.
–– It can be increased with modular polyethylene inserts
or monobloc components of different thickness (e.g. if
the gap is lax throughout extension and flexion).
Adjustment to increase the femoral subspace after the
femoral cut is not possible without augmentation, due
to fixed distal thickness of the component. This means
that if there is laxity in extension, either pre-operative
and missed in the pre-operative assessment, or iatrogenic caused during the tibial cut, it will be difficult to
correct without resorting to a revision prostheses. This
will require augments, stems, and suitable instrumentation, often not available at the time of the primary
surgery. Initially making a conservative distal cut
guards against this, but may require more frequent
re-cutting.
• The extent of the resection laxity due to the tibial cut is
impossible to predict with accuracy. There is therefore
an advantage to perform this step early in the procedure, even when using navigation.
Occasionally in a very stiff knee, and sometimes when there
are posterior tibial osteophytes, it may be difficult to sublux and
cut the tibia first safely (Fig. 25.86). In this situation, one can
start with the femoral cuts. A sizing and rotation guide which
has smaller posterior ‘feet’ facilitates this part of the procedure
(see Chap. 27, Fig. 27.8).

268
Fig. 25.85 (a, b): (a)
Distalising the level of the
tibia is straightforward with
one cut (b). Due to the
number of cuts, complexity of
the shape, and the lower
stability of the instruments on
small areas of bone,
proximalising the femoral cut
with accuracy is less
straightforward, and more
time consuming

C Butcher and P Neyret

a

Fig. 25.86 Posterior tibial osteophyte can prevent anterior subluxation
of the tibia

b

The procedure may progress in a number of ways; the
surgeon chooses a planned sequence of cuts and ligament
releases. At one end of the spectrum, all the cuts are made
initially, in a way that is designed to make balancing
predictable (Fig. 25.87). The releases, also anticipated, are
­performed afterwards. At the other end of the spectrum, with
navigation, one cut is made, and a stage follows where
ligament release and gap simulation with virtual cuts
precedes the definitive bone resection. In between, there are
other possible sequences. The chosen basic sequence is a
reflection of the surgeon’s training, experience, and
equipment available to him.
Our preferred sequence is to start with the tibial cut, for
the reasons stated above. This initiates the tibial space, and if
no ligament balancing is required, completes it. We then create the femoral spaces, starting with a measured resection of
the distal femur. Our alignment in extension is thus set and
prioritised. The balance in extension is prioritised in many
cases by using intra-medullary guidance and ignoring mild
proximal femoral varus if it exists. Also, the medial release
can be performed at any stage as our only dependent cut does
not rely on ligament tension, and so balance in extension can
still be prioritised. Size and shape of the femoral flexion
space is then matched to the extension space by the dependent posterior cut, completing the alignment/balance in flex-

25 Total Knee Arthroplasty: Steps and Strategies
Fig. 25.87 This table depicts
the different ways of
proceeding with TKA and
shows the emphasis on either
cuts or balancing

269

+
Cuts

–

Independent cuts

Dependent cuts
1

2

3
3 bone cuts
(T 2F)

2 bone cuts

2 spaces created

1 space created

Ligament and space balancing

Second femoral
cut

–

ion, and the gap balancing. One aspect that is set early is the
sagittal inclination of the tibial and femoral cuts. These are
not straightforward to revisit. In practice, we find performing
PCL sacrificing TKA with 0° tibial, and 3° femoral cuts lead
to satisfactory gap balancing, but like in any conventional
TKA sequence, there has been an inevitable compromise in
the order of priorities.

1 bone cut
Tibia
Simulation of 2
spaces
Gap and space
<< Balancing >>
2 femoral cuts

+
balancing

Although in most cases the sequence will be a routine for
each surgeon, there are a number of variations of the procedure which can be anticipated pre-operatively, by clinical
and radiological assessment as outlined in the next section.

270

C Butcher and P Neyret

Part IV Everyday Practice
There are various surgical pathways that can be utilised to
perform a TKA. The aim of the clinical and radiological
examination is to identify anatomical and pathological features that predict the need for certain steps, and their consequences. In doing so, some of the algorithm of TKA may be
considered prior to entering the operating room.
The main factors to consider are:
• Soft tissue envelope
• Presence of deformity
• Site of deformity: Tibia vs femur; intra-articular vs
extra-articular
A. Examination
The presence and location of deformity is most accurately
identified by long leg standing radiology, Rosenberg
views, lateral views of the knee, as well as the skyline
view. However, clinical examination gives some important information that is not supplied by routine studies
and may alert the surgeon to the need for special studies
such as stress views, or long lateral views.
1.

2.

Deformity
• Is the deformity reducible? If so, it may be possible to restore the orientation and level of the
natural joint line. If it is not, there may be concave side contracture requiring release, or extraarticular deformity requiring an asymmetric cut
and ligament balancing.
• Is there evidence of sagittal deformity? This
may determine the type and reliability of referencing, and/or affect the flexion/extension gap
balance (Fig. 25.56). Long lateral views may be
required, as the deformity may have been missed
on standard lateral films, and not be visible on
long leg frontal views.
Limited range of motion
• Is there fixed flexion deformity? This may predict difficulty with gap balancing requiring
extra distal femoral resection, attention to posterior osteophytes, and certain releases. This
also informs the interpretation of the frontal
long leg X-rays; flexion deformity and r­ otation
of the limb will produce apparent coronal
deformity.
• Is there limited flexion? This may be from an intra
or extra-articular pathology, but either way predicts difficulty with exposure, and subluxing the
patella or the tibia. The sequence may need to be
altered, with femoral first preparation, and techniques for exposure employed such as a quadriceps snip, or tibial tubercle osteotomy.

3.

Laxity
• On formal collateral testing is there coronal laxity on the convex side, or a lateral thrust during
gait (Fig. 25.74)? This predicts the need for a
significant concave side release, a large gap, and
an altered prosthetic interface in relation to the
natural joint line (Fig. 25.31).
• Is there recurvatum? This suggests a cautious
approach to bone resection to avoid the need to fill
the space with thick polyethylene (Fig. 25.88).
There may be generalised laxity, but the neurology
also needs to be assessed to exclude weakness as a
cause. In both situations, the possibility of needing
increased constraint must be considered. In these
cases, exposure is likely to be straightforward.

B. Radiology
Plain radiology allows a functional assessment from
weightbearing views, which inform about soft tissues as
well as bone and joint deformity. They can be supplemented by stress views to look for reducibility or maximum laxity (see Chap. 26, Figs. 25.3 and 25.5).
Analysis of long leg weightbearing frontal views, and 45°
flexion weightbearing views (Rosenberg) reveal the location
of the deformity. If sagittal deformity is suspected, long lateral views can be obtained. Deformity needs to be defined as:
• intra-articular or extra-articular
• tibial, femoral, or both bones
• metaphyseal or diaphyseal
There are certain ‘families’ of deformity. In the varus
knee, it originates most commonly in the proximal tibia
(Fig. 25.4, and see Chap. 14, Figs. 14.8, 14.9 and 14.10). It
may also arise more proximally in the femoral diaphysis, the
proximal femur, or the rest of the tibia (Figs. 25.33 and
25.89). In valgus deformities, the pattern is often mixed; lat-

Fig. 25.88 Pre-operative recurvatum

25 Total Knee Arthroplasty: Steps and Strategies

Fig. 25.89 Varus in the shaft, as well as the proximal tibia

271

Fig. 25.90 Valgus deformity in the shafts of both femur and tibia, as
well as lateral femoral condylar hypoplasia

eral condylar hypoplasia may coexist with valgus in the
shafts of the tibia and femur (Fig. 25.90).
Measurements are made to localise and define the
deformity:
• mechanical limb alignment (Fig. 25.91)
• mechanical and distal anatomic alignment of the femur
(Fig. 25.92)
• alignment of the tibia (Fig. 25.93)
Care needs to be taken that the alignment X-rays are taken
with the leg in neutral rotation. External rotation will exaggerate the difference between the mechanical and distal anatomic
femoral angle (Fig. 25.94), and a combination of external rotation and flexion will increase the apparent mechanical limb
varus (Fig. 25.95). Planning the distal femoral cut on the basis
of these angles must be done with caution.
Rotational deformity is often neglected as a cause of
osteoarthritis and needs to be sought clinically and, if necessary, radiologically by CT. External rotational deformities of
the femur are most common after fracture and produce
increased adductor moments, leading to medial compartment overload. This could potentially be relevant when planning a TKA in a young patient, with the potential of corrective
osteotomy as an adjunctive procedure.
Fig. 25.91 Hip Knee Ankle angle—HKA

272

C Butcher and P Neyret

Fig. 25.92 Right femur; medial distal anatomic femoral angle. Left
femur; medial (mechanical) femoral angle MFA

Fig. 25.94 External rotation
exaggerates the difference
between the mechanical and
distal anatomic femoral
angles

Fig. 25.93 Medial tibial angle MTA

a

b
ER

IR

25 Total Knee Arthroplasty: Steps and Strategies

273

The intended type of referencing is then considered; Appropriate reference lines are drawn, and thus the inclinamechanical or anatomic referencing of the femur? Intra-­ tion of both femoral and tibial cuts, and balance and post-­
medullary or extra-medullary referencing of the tibia? operative alignment anticipated. For instance:
• In the case of the femur. If there is no deformity in the
femur, mechanical axis referencing from the femoral
head may be appropriate for both alignment and balancing. If there is deformity, and the degree is acceptable, distal anatomic referencing will prioritise the
balance of the knee over the alignment and may be
preferable (Figs. 25.38 and 25.39).
• In the case of the tibia. If there is no deformity, intra-­
medullary referencing is possible, accurate, and an advantage in the obese. If there is deformity, extra-medullary
guidance needs to be considered for coronal alignment;
even if an intra-medullary guide can be passed, it may lead
to error (Figs 25.51 and 25.96). Intra-medullary guidance
may still be advantageous for appropriate sagittal alignment
however (Fig. 25.56). Potential conflict between the tibial
keel and the cortex can also be anticipated at this stage
(Fig. 25.54, and see Chap. 27).
The extent of extra-articular deformity will predict the asymmetry of bone resection in relation to the origin of the collateral ligaments. In turn this will predict gap asymmetry and
the need for ligament balancing:

Fig. 25.95 Long films of the same limb on the same day. With slight
external rotation and flexion, the limb appears in varus (left x-ray). With
slight internal rotation it appears slightly valgus (right x-ray)
Fig. 25.96 (a) Deformity
near the knee creates an
asymmetric cut. This is
exaggerated when the
reference point is lateralised
in an attempt to pass an IM
rod (b). Planning at this stage
is also important to anticipate
keel and cortex impingement

a

• Intra-articular deformity due to wear—cut less asymmetric, and soft tissue release less likely (unless soft
tissue contracture significant) (Fig. 25.29).
• Extra-articular deformity—cut always asymmetric,
and soft tissue release likely (Fig. 25.30).
• Deformity near the knee joint—greater effect on asymmetry of cut (Fig. 25.96).
• Deformity far from knee joint—less effect on asymmetry of cut.

b

274

C Butcher and P Neyret

Examples
1. Medial compartment OA in a knee with no extraarticular deformity.
There is reducible varus, no laxity, and a good range of
motion. The HKA is 176°, MFA 91°, and MTA is 89°
(Fig. 25.97). An orthogonal 10 mm tibial cut referenced from
the lateral side using combined IM and EM referencing,
removes almost the same amount of bone from the medial
and lateral tibial condyles (Fig. 25.98). The 8 mm, 7° valgus,

Fig. 25.97 (a–c): Case 1.
Minimal varus from joint
wear and no bone deformity

a

3° flexion, distal femoral cut removes almost the same bone
from the medial and lateral condyles (Fig. 25.99). The rotation is set to do the same, a 0° posterior cut (Fig. 25.100). No
AP adjustment is necessary to use a size 3 femur, compatible
with a size 2 tibia. Gap symmetry does not require ligament
release, and the gaps are balanced. Post-operative alignment
is mechanical (Fig. 25.101).
• As there is no extra-articular deformity, achieving the
desired alignment for longevity does not require alteration of the patient’s original soft tissue envelope, and
gap symmetry and balancing is automatic.

b

c

25 Total Knee Arthroplasty: Steps and Strategies

a

b

275

a

c

Fig. 25.98 (a–c): Case 1. The tibial resection is almost symmetrical
(the excised tibial plateau is more easily viewed from posterior)

b
a

b

Fig. 25.100 (a, b): Case 1. Equal posterior resection of the condyles;
the result of setting the rotation/sizing guide to 0° external rotation

Fig. 25.99 (a, b): Case 1. Equal distal resection of the femoral condyles

276
Fig. 25.101 Case 1.
Alignment is mechanical in
this case

C Butcher and P Neyret

a

b

c

2. Medial compartment OA in a knee with a varus deformity of tibial origin.
There is unreducible varus, slight lateral laxity, and a good range
of motion. HKA is 170°, MFA 93°, and MTA 82° (Fig. 25.102).
An orthogonal 10 mm tibial cut referenced from the lateral side, using combined IM and EM referencing, removes
more bone from the lateral condyle of the tibia than the
medial (Fig. 25.103). The 8 mm, 7° valgus, 3° flexion, distal
femoral cut removes more bone from the medial condyle
than the lateral condyle (Fig. 25.104). The rotation is set to
do the same, 3° centrally, to equalise the femoral spaces
(Fig. 25.105). An anterior cut adjustment of 1 mm is made to
reduce notching, allowing the use of a size 5 femur, compatible with a size 4 tibia. The resulting extension and flexion
gaps are trapezoidal, larger laterally, but they are balanced.
Release of the MCL from the tibia is performed to obtain
symmetry (Fig. 25.106). Gap balance is satisfactory. A
12 mm polyethylene insert is used to provide appropriate

tension in the mildly stretched lateral envelope. A longer
tibial stem is used to protect the medial tibia, which may be
devitalised by the MCL release. The overall post op alignment is mechanical although with 2° valgus of the tibial
component (Fig. 25.107).
Points of interest:
• The inclination of the posterior femoral cut is not dictated by the tibial deformity or the asymmetrical cut of
the tibia, but dependent on the distal femoral cut.
• The asymmetry of the tibial cut can be converted into
balanced flexion and extension gaps only by soft tissue
release. This rule applies regardless of the actual
amount of asymmetry.
• In this case, there was overcorrection of the tibial
varus. Although an orthogonal cut is the goal, undercorrection would have been preferable to overcorrection, and would have reduced the extent of medial
release necessary.

25 Total Knee Arthroplasty: Steps and Strategies
Fig. 25.102 (a–c): Case 2.
Varus in the joint from wear,
but also of the proximal tibia

Fig. 25.103 (a–c): Case 2.
The tibial cut is asymmetrical
(the excised tibial plateau is
more easily viewed from
posterior)

a

a

c

277

c

b

b

278
Fig. 25.104 (a, b): Case 2.
The distal femoral cut
removes more bone from the
medial condyle than the
lateral

C Butcher and P Neyret

a

b

Fig. 25.105 Case 2. Cutting guide placed to remove more bone from
the posterior medial femoral condyle than the lateral

Fig. 25.106 Case 2. Release of the MCL (arrow) from the tibia is performed to obtain symmetry. The MCL is slightly tensioned with a laminar spreader, and gently separated from the proximal tibia with a curved
instrument, with repeated assessment of gap symmetry to prevent over-­
release. The intact pes anserinus is labelled with the star

25 Total Knee Arthroplasty: Steps and Strategies
Fig. 25.107 (a–c): Case 2. The overall post
op alignment is mechanical although with 2°
valgus of the tibial component

279

a

b

c

3. Medial compartment OA in a knee with a varus deformity of proximal femoral origin.
There is unreducible varus, no lateral laxity, and a good range of
motion of 0/0/120. The HKA is 160° MFA 85° and MTA 83°
(Fig. 25.108). The femoral varus is situated proximally. An
orthogonal 10 mm tibial cut referenced from the lateral side,
using combined IM and EM referencing, removes more bone
from the lateral condyle of the tibia than the medial creating an
asymmetric gap, wider laterally. The 8 mm, 7° valgus, 3° flexion, distal femoral cut is referenced from the distal femur and
removes almost the same bone from the medial and lateral condyles. The rotation is set to do the same, with 0° external rotation, to equalise the femoral spaces. Symmetry of the gaps in
flexion and extension is achieved with a pie crust of the
MCL. There is good balance. The residual limb varus is significant, but the tibial component is almost orthogonal (Fig. 25.109).
Points of interest:
• A distal femoral cut orthogonal to the mechanical axis
would remove significantly more bone laterally than
medially, resulting in a more trapezoidal extension gap,
larger laterally (see Fig. 25.39a). Instead of performing
this, it is decided to accept residual post-operative varus
due to the proximal deformity, and a routine 7° valgus
cut is made, referenced from the distal femur with an
intra-­medullary rod (Fig. 25.39c).

• The distal cut in this case produced equal distal femoral resection. In those cases where there is more
resection from the lateral condyle, it is possible to
produce a rectangular flexion gap that matches the
extension gap (trapezoidal, wider laterally), but this
would require internal rotation of the posterior cut
to remove more bone laterally than medially (see
Fig. 25.65). Any position of femoral external rotation will remove posterior bone medially more than
laterally, producing a flexion gap that is trapezoidal,
wider medially. In these cases, we recommend a
posterior cut parallel to the posterior condylar axis
at an acceptable inclination for patellar femoral biomechanics. Some medial laxity in flexion is then
accepted. A considered compromise has thus been
made between alignment correction, balance in flexion, balance between flexion and extension, and
patellar femoral biomechanics— to avoid an unacceptable extreme of any one (see also Chap. 42,
Figs. 42.44 and 42.45).
• The degree of varus that is acceptable depends on the
patient’s age, activity, BMI and bone quality, and is
controversial. This case is probably at the limit, and
surgical strategy could be discussed. When proximal
femoral varus is considered unacceptably high, the
alignment can be prioritised by performing a femoral
osteotomy and subsequently a routine TKA.

280

C Butcher and P Neyret

Fig. 25.108 (a–c): Case 3.
Varus deformity in the joint,
tibia, and femur. The femoral
varus is proximal

Fig. 25.109 (a–d): Case 3.
Significant varus alignment
from the femoral deformity,
but well-aligned tibial
component. Symmetry and
balance has been prioritised
over overall alignment.
Without access to long films,
the picture is less
controversial (b)

a

a

b

b

c

c

d

25 Total Knee Arthroplasty: Steps and Strategies

4. Lateral compartment OA with lateral condyle
hypoplasia.
Clinically there is unreducible valgus, minimal medial laxity,
and a fixed flexion deformity of 35° with further flexion to
140° (Fig. 25.110). The HKA is 197°, MFA 96°, and MTA
91° (Fig. 25.111). There is stage IV lateral compartment
osteoarthritis with a lateral tibial defect more than 5 mm.
Referencing is intra-medullary in the proximal tibia
only for sagittal alignment and extra-medullary for coronal.
The orthogonal 7 mm tibial cut is referenced from the
medial condyle, taking a symmetric cut. The resulting tibial
defect is less than 5 mm deep and less than 10% of the plateau. The 5° valgus, 3° flexion, distal femoral cut is referenced at a level 10 mm (thickness of the prosthesis +2 mm)
from the medial condyle and removes only 1 mm of lateral
condyle (Fig. 25.112). The posterior femoral cut is externally rotated partly by building up the lateral condyle
(5 mm) and partly by central rotation (2°) (Fig. 25.113).
The resulting flexion and extension gaps are trapezoidal,
wider medially. The extension gap is smaller than the flexion gap, due to the pre-­operative flexion deformity, and the
central femoral rotation. Ligament balancing is carried out
in this case by the lateral approach (releasing ITB) and a
lateral femoral condyle osteotomy, distalising more than

Fig. 25.110 Case 4. Pre-operative lateral X-ray in full extension; fixed
flexion deformity of 35°

281

posteriorising the fragment to help achieve balanced gaps
(Fig. 25.114). Full extension is achieved, but some laxity in
flexion is however accepted. Alignment post-operatively is
mechanical (Fig. 25.115).
Points of interest:
• A kinematic approach has been taken to the femoral
cuts, and the femoral spaces, by building up the deficient lateral condyle in extension and flexion.
• The lateral femoral condyle osteotomy is well suited to
the intra-articular deformity and can be thought of as
creating space for the enlarged femoral segment. It can
preferentially create more space in flexion or extension
to contribute to gap balancing.
• External rotation of the femoral component often
forces a compromise: one can either upsize the femoral component with potential overhang, or if there is
femoral/tibial size mismatch, use a smaller femoral
component with femoral notching or posterior cut
level adjustment. R
­ otating by build-up of the lateral
condyle alone maintains the posterior condylar level
on the medial side, and optimises the gap balance, but
at the expense of a larger femoral size. Using central
rotation is an alternative method to restrict the size of
the implant. In this case, it facilitated the use of a femoral size 2 which was compatible with the tibial size 1,
which in turn was dictated by overhang. Adjusting the
level of the posterior femoral cut is a third option. This
and the central rotation technique reduce the femoral
segment in flexion and exacerbate the gap imbalance
which arises when there is pre-operative flexion
deformity.

282
Fig. 25.111 (a–d): Case 4.
There is stage IV lateral
femorotibial osteoarthritis.
The valgus arises from joint
wear, tibial deformity, and
lateral femoral condyle
hypoplasia

C Butcher and P Neyret

a

b

c

d

Fig. 25.112 Case 4. Lateral femoral condyle hypoplasia is obvious.
The distance between bone and guide is assessed with an osteotome of
known thickness to help reproduce the build-up posteriorly

25 Total Knee Arthroplasty: Steps and Strategies

283

a

b

Fig. 25.113 Case 4. The external rotation has been produced by build­up, but also some central rotation to allow the use of a smaller femoral
component without notching

c

Fig. 25.114 (a–c): Case 4. Sliding lateral femoral condyle osteotomy.
(a) Cut completed, prior to mobilisation. (b) Fragment fixed with a
single screw. (c) Fragment moved more distally than posteriorly to
improve the gap balancing, as well as the gap symmetry

284

a

C Butcher and P Neyret

b

c

d

Fig. 25.115 (a–d): Case 4. (a) Alignment is mechanical. (b) Extra-medullary guidance was necessary. Distalisation of the lateral condylar fragment is apparent. (c) Lateral view (d) A lateral facetectomy has been performed

25 Total Knee Arthroplasty: Steps and Strategies

Summary
The technical goal of TKA is to produce a balanced knee
without malalignment. What constitutes good alignment is
controversial, but for technological reasons we still aim for
an overall mechanical coronal alignment, although our
approach on the femoral side is more kinematic. We have
used this strategy for many years and it has yielded good
results in terms of longevity, patient reported outcomes and
patellar femoral function.
During conventional TKA, the multitude of interrelated steps, and the linear process of the operation, can
make it difficult to achieve a predictable result in every
patient. Success relies on producing specifically sized
and orientated flexion and extension gaps, by a combination of bone cuts and soft tissue releases. A understand-

285

ing of the relationships between these steps, their order
and priority is required whether in conventional or navigated TKA. Not only is this understanding necessary to
use navigation safely, but conventional TKA techniques
will be the mainstay in many regions of the world for the
foreseeable future.
In attempting to simplify the algorithm of TKA, we consider the femoral and tibial contributions to the gaps separately. This simple concept of tibial and femoral spaces and
segments helps comprehend the interchangeable relationship
between bone resection and soft tissue release and provides
a rationale for patient-specific femoral rotation.
As materials and methods evolve, the constraints of unnatural alignment and balance may also be removed, allowing us to
move towards a process of recreating each patient’s anatomy.
The algorithm may become simpler with time.

Total Knee Arthroplasty in Medial
Arthritis: Surgical Technique

26

G Demey, R Magnussen, P Neyret,
and C Butcher

Preoperative Planning
A detailed history, orthopaedic physical and radiological
evaluation is required for preoperative planning. The aim is
to establish which surgical approach is most suitable, to
choose the appropriate prosthetic implant, and importantly to
anticipate any possible intra-operative technical difficulties
that may be encountered.
For the radiological evaluation, see Chapter ‘Surgical indications in Osteoarthritis of the knee’ (Figs. 26.1 and 26.2).
Valgus stress radiographs show whether the varus deformity is reducible (Fig. 26.3). Incomplete reduction of the
joint narrowing is secondary to contracture of the capsular
and ligamentous structures on the medial side of the knee. In
this situation, a surgical release will be necessary. The need
for soft tissue release is also dependent on the asymmetry of
the bony cuts, related to metaphyseal or diaphyseal deformity (Fig. 26.4). This can be anticipated and appreciated
when drawing the resection lines perpendicular to the
mechanical axis. Although we commonly perform varus
stress radiographs as well, lateral ligamentous laxity is more
difficult to interpret. A pseudo lateral thrust is often observed
in these patients. This pseudo thrust is not usually due to true
lateral ligamentous laxity, but rather due to closing down of
the worn medial compartment (Fig. 26.5).
Fig. 26.1 Measurement of mechanical femoro-tibial angle

G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
R Magnussen
Centre Albert Trillat, Lyon, France

Surgical Technique
This chapter describes our technique for a posterior stabilised TKA.

P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

Surgical Approach

C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

Foot supports are placed to position the knee in 90° flexion
and full flexion. We use a sterile disposable tourniquet in

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_26

287

288

G Demey et al.

Fig. 26.2 Measurement of mechanical femoral and tibial angles
Fig. 26.4 Asymmetry of the bony cuts

Fig. 26.3 Valgus stress X-ray shows lack of correctability

most cases (Fig. 26.6). Regional anaesthesia is preferred, and
either peripheral nerve block or local infiltration is used for
postoperative pain control.
The incision is marked, and an adhesive plastic drape is
applied (Fig. 26.7). A paramedian skin incision is made starting 5–6 cm proximal to the patella and ending on the medial
border of the tibial tuberosity. After incising skin and subcutaneous fat, it is important to dissect onto the superficial retinaculum. Undermining is performed between the superficial
and deep retinaculum. Laterally, the undermining should stop
5 mm past the patella. Proximally, the quadriceps tendon and
distally the medial border of the patellar tendon are identified.
At this stage, we suture blue OR towels into the wound to
isolate the skin edges from the rest of the operating field.
Prior to the arthrotomy, cautery marks are made either side of
the intended cut in the region of the vastus medialis insertion
to guide closure at the end of the procedure. The arthrotomy
is performed with a 23 blade and starts approximately 4 cm
proximal to the patella, on the medial side of the quadriceps
tendon. A small cuff of tendinous tissue is left attached to the
muscle to facilitate strong closure. The arthrotomy is continued distally on the medial side of the patellar tendon towards
the medial side of the tibial tuberosity, again leaving a small
cuff of tendinous tissue on the patella. The anterior part of the
medial meniscus is incised while the scalpel blade stays in

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

289

Fig. 26.7 Incision marked and adhesive plastic drape applied

Fig. 26.5 Varus stress X-ray shows varus from medial narrowing
rather than lateral laxity

Fig. 26.6 Set up with knee at 90° flexion. Note the sterile tourniquet
already applied

contact with the anterior border of the tibial plateau.
Subsequently, the medial capsule is released from the anteromedial part of the tibial plateau. This release is triangular
(Fig. 26.8). The deep fibres of the medial collateral ligament
are released using a periosteal elevator on the proximal border of the tibial plateau at the joint line. Subsequently, a total
medial meniscectomy is performed. The knee is now placed
in full extension and the extensor apparatus together with the
patella are dislocated laterally and everted using a Volkmann
retractor. The knee is then placed in flexion with the patella
everted. Care must be taken not to rupture the patellar tendon
or to avulse its insertion at the tibial tuberosity. Proximally,
the synovium is removed to visualise the anterior cortex of
the femur. We resect all of Hoffa’s fat pad and the anterior
horn of the lateral meniscus, the inter-meniscal ligament, and
the footprint of the ACL. The femoral notch is debrided and
all osteophytes are removed.
The tibia is now dislocated anteriorly using a Hohmann
retractor in the condylar notch, taking care to keep the tip
close to bone. The knee is fully flexed, allowing exposure of
the posterior border of the tibia. A second Hohmann retractor
is placed on the lateral side of the tibial plateau to complete
the exposure (Fig. 26.9). Occasionally, placement of the lateral Hohmann first is easier, especially if the patella is low, or

290

G Demey et al.

the extensor mechanism tight. The centre of the tibial condyles and PCL are identified, and the sagittal and coronal
axes of the tibia marked with cautery. It is possible to make
drill holes at the centre of the condyles, parallel to the intramedullary rod, to transfer them to the cut tibia for reference
later in the procedure (Fig. 26.10a, b).
Specific care should again be taken to prevent avulsion of
the extensor mechanism during flexion of the knee between

Fig. 26.8 Release of the anterior medial capsule and periosteum—
triangular

a

Fig. 26.9 Full exposure of tibial plateau. The tibial condylar centres
and axes have been marked

b

Fig. 26.10 (a, b) Drill holes made parallel to the IM guide transfer the marked condylar centres to the cut surface, for reference later in the
procedure

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

291

30° and 100° with an everted patella, and during the anterior
dislocation of the tibia. Anterior dislocation of the tibia can be
difficult in the presence of a patella infera, or in arthritis secondary to chronic anterior cruciate ligament (ACL) instability.
In this situation, the insertion of the patellar tendon on the tibial
tuberosity can be secured using a pin. This pin is inserted and
directed towards the lateral border of the tibia and should be
placed so it does not hinder the next steps of the intervention
(Fig. 26.11). An additional option to reduce the tension on the
extensor mechanism is to dislocate, but not evert the patella.

Tibial Cut
The tibial intramedullary (IM) rod is inserted at the footprint of the ACL. The entry point is opened up using a
curved osteotome (Fig. 26.12). This ensures correct alignment in the sagittal plane. We aim for a tibial cut perpendicular to the long axis of the tibia. However as the
intramedullary guide alone does not always ensure correct
coronal alignment, an additional extramedullary (EM)
aiming device is utilised (Fig. 26.13). This guide is aligned
with the middle of the ankle joint (not the middle of the

Fig. 26.11 A pin is inserted to protect the patellar tendon from avulsion in the case of difficult exposure

Fig. 26.12 Entry point in ACL footprint opened up using a curved
osteotome

Fig. 26.13 Combined intra- and extramedullary guidance

292

G Demey et al.

Fig. 26.15 Reference from the centre of the lateral plateau
Fig. 26.14 The distal end of the extramedullary guide is aligned with
centre of ankle joint

ankle), and the first intermetatarsal space if there is no
foot deformity (Fig. 26.14). The thickness of the tibial cut
for this prosthesis is 10 mm referenced from the centre of
the lateral tibial plateau (Fig. 26.15). This is the nonaffected side in medial arthritic knees. The 0° tibial cutting guide is subsequently fixed with two guide pins
(Fig. 26.16). The IM rod and jig are removed and up to
three further pins are inserted. The surgeon can fine tune
the alignment at this stage (Fig. 26.17). The proximal tibial cut is performed using an oscillating saw, taking care
not to pass too posteriorly either side of the posterior
Hohmann. After the tibial cut, the tibial plateau can be
loosened gently with a broad osteotome and removed with
a grabber (Fig. 26.18). The soft tissues at the periphery,
including the PCL and anterolateral structures, are
released close to bone. A re-cut is sometimes necessary in
the more inaccessible areas such as the lateral and the
posterior cortex of the lateral tibial plateau. An incomplete exposure and removal of bone from the lateral tibial
border and the zone around Gerdy’s tubercle can induce
varus positioning of the tibial component or a medialisation (Fig. 26.19). Care should be taken not to damage the

Fig. 26.16 The 0° tibial cutting guide is subsequently fixed with two
guide pins

popliteus tendon or the patellar tendon. After the tibial
cut, the tibial plateau is sized using different trial sizers
(Fig. 26.20).

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

293

Fig. 26.19 Incomplete exposure and removal of bone from the lateral
tibial border and the zone around Gerdy’s tubercle can induce varus
positioning of the tibial component or medialisation

Fig. 26.17 Fine tuning of the alignment is possible at this stage

Fig. 26.20 Sizing the tibial component

Technical Points
All bone cuts are performed using an oscillating saw. The
surgeon can be protected from splatter of blood and bone
fragments by covering the joint with a transparent plastic
board (Fig. 26.21).

Fig. 26.18 The tibial plateau is loosened gently with a broad osteotome and removed with a grabber

Particular Difficulties
The popliteus tendon is at risk on two occasions. The first is
during the tibial cut and the second is during the posterior
lateral femoral condyle cut (Fig. 26.22).

294

G Demey et al.

Distal Femoral Condyle Cut

Fig. 26.22 Popliteus injury at the time of the tibial cut (arrow)

The knee is now flexed to 90°. The centre of the entry point
for the IM aiming device is situated approximately 1 cm
anterior to the insertion of the posterior cruciate ligament
(PCL), somewhat to the medial side of the trochlear groove
(Fig. 26.23). Sometimes osteophytes must be removed first
to identify the PCL landmark. The entry point is opened up
using a curved osteotome. An entry point which is too posterior will cause excessive femoral component flexion, while
an entry point which is too anterior risks recurvatum and
anterior notching. In order to decrease the risk of fat embolus,
the intramedullary bone marrow is aspirated prior to insertion of the intramedullary rod.
The femoral cutting guide, which incorporates 3° flexion, is applied to the distal femoral condyles and secured
with four pins (Fig. 26.24). In terms of coronal alignment,
preoperative calculation of the HKS angle between the
mechanical and anatomical femoral axis is not reproducible. Except in the case of specific anatomical variations,
we always set the distal femoral cut to 7° of valgus in the
medial arthritic knee (Fig. 26.25) (see ‘Rotation of the femoral component’ later in the chapter). Either an 8 mm or
10 mm cut is made, depending on the preoperative range of
motion. If there is preoperative fixed flexion deformity, the
cut is 2 mm more than the thickness of the component, i.e.
10 mm. During the cut, the tibia is protected with a specific
instrument or a broad osteotome (Fig. 26.26). The patella is
protected with the previously placed Hohmann retractor
and the medial skin with a Hohmann or rake. A clear plastic
sheet can be used again to shield the staff from blood spatter. The resected condylar bone is inspected for thickness,

Fig. 26.23 Femoral entry point anterior to the PCL

Fig. 26.24 The femoral cutting guide, secured with four pins

Fig. 26.21 Transparent plastic board for protection against blood
spatter

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

295

and this will guide the femoral rotation in the next step (see
‘Specific Technical Points’).

Posterior and Anterior Femoral Condyle Cuts
The femoral sizer with chosen rotation is applied to the distal
femur, ensuring the posterior ‘feet’ are in contact with the
posterior condyles (Fig. 26.27a, b). The anterior stylus is
placed in the middle of the anterior femoral cortex and the
femoral size determined and checked with an ‘angel wing’
(Fig. 26.28). If the bone is in between sizes, the smaller is
chosen and the femoral sizer anteriorised such that there will
be no notching. An alternative option is to anteriorise with
the 4 in 1 cutting guide at the next step.
Two pins are placed through the femoral sizer, which
will then set the rotational position of the 4 in 1 cutting
guide. After removal of the femoral sizer the cutting guide
is placed over the pins, with or without adjustment anteriorly (Fig. 26.29). Two further pins are inserted to stabilise
the guide, and the proposed bone resections checked
before proceeding; visual check for the posterior resection,
and with the angel wing for the anterior resection. Anterior
and posterior cuts are made before the chamfer cuts to
maintain maximal stability of the cutting guide
(Fig. 26.30a, c). The popliteus tendon and MCL are protected during the posterior cuts with Hohmann retractors
placed directly on bone.
Fig. 26.25 Distal cut set to 7°

Posterior Stabilised Box Cuts
The box cutting guide is applied to the femur and stabilised
with pins. Medial and lateral femoral osteophytes are
removed, if necessary, to confirm the medial lateral position
of the guide. The box is fashioned with a combination of
specific drills and osteotomes (Fig. 26.31a–e). The guide is
removed, and the remains of the PCL excised (Fig. 26.32).
Curved osteotomes are used to remove posterior osteophytes
identified on preoperative X-rays, facilitated by the use of a
laminar spreader in the intercondylar area (Fig. 26.33).

Trial Reduction and Assessment of Gaps

Fig. 26.26 The tibia, MCL and patellar are protected. The thickness of
bone resected from each condyle is confirmed

At this point, flexion and extension gap assessment is made
with either spacers or tensors (Fig. 26.34a–d). If there is tightness medially, a medial release is performed (Fig. 26.35) (see
‘Release of MCL’ later in the chapter). Another option if the
balance appears reasonable is to insert the trial components to
assess the gaps. The chosen femoral trial is impacted, taking
care to align it longitudinally with the previous cuts. The chosen tibial component with the thinnest polyethylene insert is

296

a

G Demey et al.

b

Fig. 26.27 Femoral sizer. (a) Feet in contact with posterior condyles. (b) Rotation in this case is set to 0°

a

b

Fig. 26.28 (a) Stylus in the middle of the cortex, along with angel wing. (b) Femoral size, in this case 5

inserted in flexion, and the knee slowly extended (Fig. 26.36).
Varus and valgus stress is applied in turn, with and without
the patella reduced. Medial and lateral laxity is assessed in
full extension and at varying degrees of flexion. Rarely is a
larger polyethylene insert required to balance the knee; this
typically is the case when there is preoperative lateral laxity.
Assessment is then made of patellar femoral (PF) tracking
with a no-­touch technique, and of tibial component rotation
with regard to the femoral component and the tibial tuberos-

ity. The centre of the tibial trial will be located behind the
patellar tendon. The tibial component position is then marked
on the anterior tibia in three places (Fig. 26.37).

Patellar Cut
The patella is the critical link to the extensor mechanism.
One must substitute the articular surface without augmenting

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

Fig. 26.29 4:1 cutting guide applied to distal femur. Checking posterior resection before making the cuts. Hohmann levers are close to bone
and protect the MCL and the popliteus tendon

a

c

Fig. 26.30 (a–c): Anterior and posterior cuts before chamfers

297

the total patellar thickness. Overcutting the patella can
weaken it, increasing the risk of a fracture. The knee is
placed in the extended position. The patella is everted to the
lateral side. The proximal and distal soft tissues should be
resected in order to expose the tendon structures. The thickness of the patella is measured. The patellar cutting clamp is
designed using an anterior reference guide (Fig. 26.38a, b).
The aim is to obtain a symmetrical cut, parallel to the anterior cortex of the patella, with a residual thickness of approximately 15 mm. This should always be thicker than 12 mm to
avoid fractures and sometimes goes up to 16 or 17 mm in
larger patellae (Fig. 26.39). The sum of the patellar
­component and the resected patella should never be thicker
than the original patella.
After the patellar cut, symmetry should be checked manually (Fig. 26.40). For sterility reasons, surgical gloves are
changed and direct contact with the patella is avoided by
using a swab while palpating the patella cut. The three patellar component fixation holes are reamed (Fig. 26.41). These

b

298

G Demey et al.

a

b

c

d

e

Fig. 26.31 (a–e): Fashioning the box with specific drills and osteotomes

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

299

The trial patellar component is positioned. If the patella is
larger than the patellar trial component, the lateral bony
overhang is resected free hand using the oscillating saw.
Assessment is made again of PF tracking with a no-­touch
technique and multiple ROM cycles (Fig. 26.42). PF tracking is considered to be the thermometer of the TKR procedure (Bousquet). If every step of the surgical procedure has
been performed correctly, the PF tracking should be perfect
and lateral retinacular release is seldom necessary; however,
if needed it should be done at the end of the procedure with
the tourniquet deflated.

Final Tibial Preparation

Fig. 26.32 Excision of the PCL remnants

The tibia is dislocated anteriorly in the fully flexed position,
and the tibial trial component is positioned. Care must be
taken to prevent impingement between the lateral femoral
condyle and the tibial trial, which results in internal rotational malpositioning. The anterior tibial marks placed previously guide the rotational position, but this is checked in
relation to the tibial tuberosity again, the shape of the tibia,
and previous drill holes if made. Most often, the medial plateau is larger than the lateral, and an appropriately positioned
tibial component will leave an area of uncovered posterior
medial bone (see Fig. 26.20). The correct size of the tibial
component is that size that maximally covers the tibial plateau without overhang.
The tibial trial is pinned in place, and tibial alignment
may be checked at this stage again prior to keel and delta
wing preparation with specific instruments (Fig. 26.43a, b).
Care must be taken with delta wing osteotomy impaction. If
there is medial tibial subchondral sclerosis, initial preparation of the delta wing slot with a saw is advisable to prevent
fracture, and to prevent lateral translation of the trial and
component.

Implantation

Fig. 26.33 Removal of posterior osteophytes with a curved
osteotome

fixation holes should be positioned in order to avoid a horizontal alignment because of the risk of patellar fractures.
Two holes are reamed medially and one laterally.
The patellar component is placed somewhat inferiorly
and medially.

Local anaesthetic infiltration is commenced while the cement
and prostheses are being prepared, starting with the posterior
capsule, and methodically covering all deep tissues
(Fig. 26.44a–e). A thorough pulsed lavage is then carried out
(Fig. 26.45).
The tibial component is inserted and impacted first, and the
excess cement removed. After impaction of the polyethylene
insert, the knee is hyperflexed to facilitate the positioning of
the femoral component. Once the component has cleared the
tibial post, the knee can be positioned in 90° flexion for the

300

a

c

G Demey et al.

b

d

Fig. 26.34 (a–d): Assessment of flexion and extension gaps with (a). Static spacer (b). Spreader (c, d). Tensor (in this case prior to the posterior
femoral cut in a left knee, and in flexion and extension)

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

301

Fig. 26.37 The tibial trial position is marked in three places for
triangulation

a
Fig. 26.35 Medial collateral release by pie crusting with a No.15
blade

b

Fig. 26.36 Tibial trial inserted in flexion

femoral impaction and removal of excess cement (Fig. 26.46).
The knee placed in extension by supporting the heel, maintaining the limb in neutral rotation. The patellar component is
cemented and compressed with the special clamp.
The rest of the local infiltration is completed while the
cement hardens, including all areas of incision in the periosteum, capsule, extensor mechanism, and skin (Fig. 26.47).

Fig. 26.38 (a) Patella measured at 19 mm, and cutting guide applied,
(b) Cutting guide is anterior referencing

302

G Demey et al.

Fig. 26.41 Appropriate position of fixation holes; one laterally and
two medially

Fig. 26.39 Patellar thickness measured as 14 mm

Fig. 26.42 Assessment of patellar tracking

‘The knee bleeds in extension or flexion, fortunately rarely both.’

Fig. 26.40 Manual verification of cut symmetry with a swab

Wound Closure
The tourniquet is released after the cement hardens. To prevent haemarthrosis, tranexamic acid is given intravenously
20 min prior to tourniquet release, and careful haemostasis is
achieved with diathermy. The knee is again thoroughly
washed with pulsed lavage prior to closure.

The knee is taken through a complete range of motion and
laxity checked for the final time.
The extensor mechanism is closed at 90°of flexion with
multiple interrupted non-resorbable sutures, using the previously placed diathermy marks to guide initial placement
(Fig. 26.48). One intra-articular drain may be left in situ. The
subcutaneous layer is approximated with resorbable sutures
to close the potential dead space, and skin closed with staples. A sterile dressing is applied with the knee in some flexion (Fig. 26.49).

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

a

303

b

Fig. 26.43 (a) Final check of tibial alignment. (b) Preparation of the keel and delta wings

Specific Technical Points
Rotation of the Femoral Component
A posterior femoral cut parallel to the posterior condyles
does not induce any rotation of the femoral component. For
us, rotation of the femoral component is only necessary if the
distal femoral cut is asymmetrical. When the distal femoral
cutting guide is only in contact with the distal medial con-

dyle, the distal cut will result in cutting less from the lateral
condyle (Fig. 26.50). This asymmetrical cut can be transferred to the flexion gap by externally rotating the femoral
cutting guide so that less of the lateral posterior condyle is
cut. The exact rotation needed is calculated by measuring the
distance between the lateral distal femoral condyle and the
femoral guide. This measurement may be made using the
thickness of multiple osteotome blades (Fig. 26.51).
Subsequently, the same number of osteotome blades are

304

G Demey et al.

a

b

c

d

e

Fig. 26.44 (a–e) Local anaesthetic infiltration of the posterior capsule and periosteum is methodical and includes all deep tissues

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

305

Fig. 26.45 Pulsed lavage
Fig. 26.47 Local infiltration completed

Fig. 26.46 Impaction of the femoral component

applied to the lateral posterior condyle and thus the femoral
cutting guide is rotated externally. The centre of rotation in
this case is the medial condyle.
An alternative technique is to externally rotate with a central axis. Many instrumentation systems use this principle
(Fig. 26.52). It is important to know that in some of these, the
posterior medial condylar resection will exceed both the
thickness of metal replaced and the amount of bone resected
from the distal medial condyle.
Of course, in a varus aligned knee (mFA < 90°), if more
distal lateral condyle is resected, an asymmetrical distal femoral cut is not translated into internal rotation.

Fig. 26.48 First suture placed at the previous marks

Release of the MCL
According to the initial recommendations of Insall in the
eighties, soft tissue releases should precede the bony cuts. At

306

G Demey et al.

a

Fig. 26.49 The sterile dressing is applied in some flexion

b

Fig. 26.50 The femoral cutting guide is only in contact with the distal
medial condyle, and the distal cut will remove less from the lateral
condyle

Fig. 26.51 (a, b) The asymmetrical distal cut can be transferred to the
flexion gap by externally rotating the femoral cutting guide so that less
of the posterior lateral condyle is removed. The axis of rotation is the
posterior medial condyle

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

307

frequently ending up in an ‘all or nothing’ situation; an excessive release of the MCL with excessive laxity on the medial
side. We use the pie crust technique to release the MCL. To
achieve this, we use an 11 blade to make multiple small perforations in the superficial MCL from the inside out.
Subsequent testing in flexion allows us to progressively test
and release the MCL and thus to obtain ligamentous balance
in flexion and extension (see Fig. 26.35). It is our experience
that this procedure can be performed in cases with up to 6° of
constitutional (extra-articular) deformity. The pie crust technique can lengthen the MCL by 6–8 mm both in flexion and
extension. We do not share the view with Whiteside that a
selective release of either the posterior or anterior fibres of the
MCL to increase the extension and ­flexion gap, respectively,
is routinely possible. In cases of more extensive deformities,
however, caution is required as the pie crust procedure can
result in complete sectioning of the MCL.
• Distal release of the MCL
In case of a varus deformity exceeding 6–8°, a distal MCL
release is performed on the tibial side. The release is performed close to the bone using a periosteal elevator, leaving
the pes anserinus tendons in continuity (Fig. 26.54a, b).

Fixed Flexion Deformity (FFD) Correction
Fig. 26.52 External rotation of 3° with a central axis (left knee)

present however, we feel it is more logical to perform any
releases after the bony cuts. In the case of a constitutional
tibial varus with a proximal metaphyseal varus deformity, a
tibial cut perpendicular to the longitudinal axis will result in
asymmetric bone resection, and subsequent laxity on the lateral side (Fig. 26.53a, b). This lateral laxity can increase
when the anterolateral soft tissues are severed with a thick
bony cut on the lateral tibial plateau (see ‘Anterolateral capsular structures’ below). In these cases, a medial release of
the soft tissue structures is required. Usually, the medial
approach releases the capsule and the deep MCL sufficiently
for adequate balancing. However, if this is not sufficient, as
with a significant constitutional varus deformity, several surgical techniques can be used to achieve an adequate medial
soft tissue release:
• Pie crust of the MCL
Insall proposed to release the superficial MCL on the distal tibial side. This may be considered an extensive release

When large posterior osteophytes are present, one should
remove them. These osteophytes are best observed on the lateral plain radiographs. These osteophytes tent the posterior
capsule and result in a FFD. Frequently, contracture of the
semimembranous muscle and tendon is responsible for the
FFD (Fig. 26.55a). This tendon can be released from the posterior aspect of the medial proximal tibia (Fig. 26.55b). In our
hands, a preoperative FFD is addressed effectively by a thicker
distal femoral cut of 10 mm, instead of the normal 8 mm, thus
enlarging the extension gap. The final implantation of a 10
mm polyethylene spacer (minimum thickness for this prosthesis) resolves the FFD. A posterior capsular release on the
other hand is of little value to resolve FFD in a varus knee.

Lateral Patellar Release
Instability of the patella is very rarely observed in varus
TKA. If detected during trialing, the surgeon should look for
femoral component malrotation. Most commonly, the femoral component has been positioned in excessive internal rotation. If necessary, a lateral patellar release should be

308

a

G Demey et al.

b

Fig. 26.53 (a) Constitutional tibial varus. (b) Asymmetric cut

performed at the end of the intervention when the definitive
components are in place and the tourniquet deflated. A lateral release is performed from within the joint using a 23
blade and the knee in full extension (Fig. 26.56). Sectioning
of the lateral structures starts at the superior border of the
patella and is extended distally. Care is taken to achieve
haemostasis.

Anterolateral Capsular Structures
The anteromedial approach in a varus knee preserves the lateral capsular structures. However, in case of a thick tibial cut,

the anterolateral structures can be severed. Thus resection
laxity results, exacerbating any preoperative lateral laxity. To
better understand this phenomenon, we use the example of
an ACL tear complicated with a Segond fracture (Fig. 26.57).
This type of fracture results from the avulsion during the torsion trauma of the anterolateral capsular structures from the
tibial plateau (Fig. 26.58). A thick bony cut on the lateral
tibial plateau can have the same effect, as can the anterior
dislocation of the tibia during surgical exposure. These structures are readily identifiable during surgery as a rope-like
structure tensioned between the anteriorly dislocated tibial
plateau and the femoral capsule (Fig. 26.59). If the tibial cut
is higher than its insertion, the structures remain intact and

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

309

a

a

b

b

Fig. 26.55 (a, b) Insertion of semimembranous tendon identified and
isolated (arrow) (c). Tendon released from bone (arrow)

Fig. 26.54 (a, b) Distal release of the MCL (arrow)

functional. In a thick or asymmetrical tibial cut, its insertion
is resected (Fig. 26.60).

Difficulties During the Patellar Preparation
The patella is infrequently significantly worn in medial
OA. However, if this is the case the patellar cut can be very
difficult. In case of chondrocalcinosis, the wear of the patella

can be extreme with a saw-like pattern on the Merchant view
(Fig. 26.61). In this situation, we propose to preserve the lateral osteophyte during the positioning of the cutting guide.
This will help to stabilise the guide and to obtain a flat and
symmetrical cut of the patella (Fig. 26.62a, b). In cases of
severe patellar wear, one can consider avoiding patellar
resurfacing altogether.

Postoperative Care
Compressive bandages are applied immediately after the
intervention. A compressive Velpeau bandage is additionally applied to the knee, but is removed in recovery, one
hour after the intervention. The lower limb may be immobilised with a brace in extension. Low molecular weight heparin is used as thromboprophylaxis and is started on the
evening of surgery.

310

G Demey et al.

Fig. 26.58 MRI showing anterolateral capsular structures

Fig. 26.56 Lateral retinacular release from inside the joint

Fig. 26.59 Anterolateral capsular structures (intra-operative view)

Fig. 26.57 Segond fracture

Fig. 26.60 Asymmetrical and thick tibial cut. The anterolateral capsular structures are resected (arrow)

26

Total Knee Arthroplasty in Medial Arthritis: Surgical Technique

311

a

Fig. 26.61 Typical aspect of chondrocalcinosis

b

The foot is regularly checked for adequate perfusion and
neurological status. Although signs are often poorly defined,
both calves are frequently palpated to check for DVT. If in
doubt a duplex ultrasound is performed.
Close attention is paid to the respiratory function of the
patient. Sudden desaturation may be a sign of pulmonary
embolism. A spiral CT scan or VQ scan can rule out this
pathology.
Good quality orthogonal plain films are obtained in the
postoperative period to exclude fracture, which might change
the postoperative management.

Rehabilitation
Rehab is started one day after surgery. A CPM is applied as
well as progressive active training. The aim is to regain passive and active flexion and extension. Flexion is limited to
95° for the first 6 weeks. This protects the sutures, limits
pain, and avoids hematoma formation. Generally, patients

Fig. 26.62 (a, b) Extreme wear of the patella. We propose to preserve
the lateral osteophytes during the positioning of the cutting guide

leave the hospital on day two to three and may transfer to a
rehabilitation centre. A patient visit is scheduled 2 months
after surgery with long leg films.

Total Knee Arthroplasty in Lateral
Arthritis: Specifics and Surgical
Techniques

27

P Archbold, J Pernin, G Demey, P Neyret,
and C Butcher

Surgical Approach
A lateral paramedian skin incision is made. Proximally the
quadriceps tendon is identified and distally the lateral border
of the patellar tendon exposed. The tendon at the superolateral part of the patella is marked with cautery in order to
facilitate the closure at the end of the procedure (Fig. 27.1)
(all figures show a left knee). The arthrotomy is made with a
longitudinal incision of the quadriceps tendon on its lateral
side leaving a small cuff of tendinous tissue attached to the
vastus lateralis muscle allowing later closer. The patella is
dislocated and the arthrotomy is continued distally, lateral to
the patellar tendon onto the anterolateral tibial plateau. When
dissecting at the level of the patellar tendon, we prefer to
bring a portion of the fat pad laterally with the retinaculum
(Fig. 27.2). This maneuver results in additional soft tissue for
use during closure and can be quite useful in cases in which
a significant valgus deformity is corrected with the TKA. The
lateral capsule is released close to the bone on the anterolateral border of tibial plateau. The capsule remains in continuity with the tendinous origin of the tibialis anterior muscle.
The insertion of the ITB is released subperiosteally from
Gerdy’s tubercle with the scalpel. Because of the continuity
of the ITB proximally with the tibialis anterior muscle distally, we prefer this digastric dissection (Fig. 27.3).
The lateral exposure is completed with the resection of
the anterior corner of the lateral meniscus. The Trillat perios-

P Archbold · J Pernin
Centre Albert Trillat, Lyon, France
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

Fig. 27.1 Lateral approach. Diathermy marks at level of proximal
patella (stars)

teal elevator is used to release the capsular structures from
the lateral tibial plateau at the level of the joint line. In specific cases, this lateral release is continued further posteriorly
reaching the posterior border of the lateral tibial plateau
(Fig. 27.4). The popliteal tendon can thus be visualized completely (Fig. 27.5).
Sometimes the visualization of the posterior medial tibia
improves after the tibial cut. Prior to this, care must be taken
to protect the MCL and posterior medial capsule while using
the saw (Fig. 27.3).

Anterior Tibial Tuberosity Osteotomy
Dislocation of the patella is more difficult in the lateral approach
than in the medial approach. When there is excessive tension
on the patellar tendon and insufficient exposure of the tibial
plateau, an osteotomy of the anterior tibial tuberosity (ATT)
can be performed. If we used this artifice relatively frequently

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_27

313

314

P Archbold et al.

a

b

c

d

e

Fig. 27.2 (a–e) The fat pad is maintained on the retinaculum

until 1996, now for more than 20 years this osteotomy has been
rarely performed, but it could be useful in stiff knee, or in case
of patella infera. Since this technique differs from that used
when performing an ATT osteotomy for episodic patella instability, the specifics are detailed in the following paragraphs.
The osteotomy must be of a certain thickness and length
in order to be in the cancellous bone and to create an area of
contact sufficiently large to achieve union of the osteotomy.

However, the osteotomy must not be too thick due to the
risk of fracture of the tibial epiphysis. The transition of the
osteotomy into the anterior cortex distally has to be progressive and smooth (Fig. 27.6). The osteotomy should not be
performed distally with a transverse bone cut because this
could weaken the anterior cortex of the tibia and result in a
fracture. At the end of the intervention, the osteotomy is
fixed with two bicortical 4.5 mm screws. No washer is used.

27 Total Knee Arthroplasty in Lateral Arthritis: Specifics and Surgical Techniques

315

Fig. 27.5 Popliteus tendon (arrow)

Fig. 27.3 The capsule and ITB remain in continuity with the tendinous
origin of the tibialis anterior muscle

Fig. 27.4 Dissection carried out to posterior border of tibial plateau

Fig. 27.6 Tibial tubercle osteotomy. It must be thick and long enough

Another option is to use three 3.5 mm screws. The holes in
the anterior cortex should be made prior to the osteotomy. It
is very important that each screw is 2 mm longer than the

distance to the posterior cortex, as this achieves optimal
­fixation. We do not use the technique using metal wires
(Whiteside) or resorbable wires (Vielpeau).

316

P Archbold et al.

Bone Cuts
Tibial Cut
Once the tibia is dislocated anteriorly, the center of each condyle, and the coronal and sagittal axes are marked with diathermy (Fig. 27.7). In some cases, sufficient anterior dislocation
to expose the posterior medial tibia is difficult. Making the distal
femoral cuts, and possibly the posterior and chamfer cuts, first
may be helpful. In the latter case, having a femoral sizing guide
with small posterior “feet” will facilitate this (Fig. 27.8).
As with the varus knee, we use combined intra- and extramedullary guidance, and this is advantageous in the valgus

Fig. 27.7 The coronal and sagittal axes are marked with diathermy

a

Fig. 27.9 (a, b) Combined intra- and extramedullary guidance

knee due to the valgus diaphyseal deformity that is sometimes encountered, which precludes passage of the rod
(Fig. 27.9). A tibial cutting guide normally designed for the
contralateral knee is well suited to cut the plateau from the
lateral side of the patellar tendon (Fig. 27.10). As in the varus

Fig. 27.8 A femoral sizer with small posterior “feet” facilitates femur
first preparation

b

27 Total Knee Arthroplasty in Lateral Arthritis: Specifics and Surgical Techniques

317

Fig. 27.11 The resected plateau; usually thicker medially

Fig. 27.10 A right cutting guide is well suited to cut the left plateau
from the lateral side of the patellar tendon

knee, the reference for the tibial cut is the contralateral plateau. In a varus knee, the cut is made 10 mm inferior to the
lateral tibial plateau, which is convex (this assumes a 10 mm
prosthetic tibial space). For a valgus knee the reference is the
medial plateau, which is concave. A cut 10 mm inferior to
the medial convexity would be excessive and would lower
the joint line. We therefore always cut 7 mm inferior to the
medial plateau. It is seldom necessary to perform re-cuts.
The resected plateau is often asymmetrical (thicker medially) due not only to the lateral wear, but also the extra-­
articular deformity (Fig. 27.11).

Femoral Cuts
Fig. 27.12 Valgus angle set at 5°

The HKS angle is always set at 5° (Fig. 27.12). The proximal
deformity of the femur, which is different in each patient, is
NOT routinely considered in this technique. We consider our
technique simple and more reproducible than the measurement of an individual HKS angle. In other words, we do not
correct (in the majority of cases) the proximal extra-articular
deformity intra-articularly. Transferring the individual HKS
to the distal cut could result in an asymmetrical distal femoral cut and difficulties in balancing the collateral ligaments.

The theory about the rotation of the femoral component is
completely applicable in the case of a total knee arthroplasty
in a valgus knee (see Chap. 25). Frequently in these cases,
hypoplasia of the lateral condyle is observed. This hypoplasia can be seen very easily once the intramedullary femoral
guide is positioned in 5° of valgus (Fig. 27.13a, b). Frequently
the distal cutting guide is not in contact with the lateral femo-

318

a

P Archbold et al.

b

Fig. 27.13 (a) Hypoplasia of the lateral condyle. (b) There is good contact with the medial condyle

ral condyle because of both wear and hypoplasia of the lateral condyle. The distal cut is thus asymmetric, less being
resected from the lateral condyle (Fig. 27.14). It is in this
situation that the asymmetrical distal cut is transferred to the
posterior femoral cut, thereby externally rotating the femoral
component. The posterior cut of the lateral condyle must be
less than that of the medial condyle. There are three choices
for the external rotation.
Firstly the center of rotation can be the posterior medial
condyle. In this circumstance, although one cuts less of the
posterior lateral condyle, no more posterior bone than usual
will be resected from the medial condyle. This is particularly
useful to prevent aggravation of pre-existing medial laxity.
This is performed by making a lateral buildup for the guide
with some sort of spacer—we use an osteotome (Fig. 27.15).
The thicker the osteotome, the more external rotation is produced. The disadvantage is that the anterior posterior (AP)
size of the component will increase, leading to component
mismatch or medial lateral femoral overhang. If a smaller
component is used to avoid mismatch or overhang, lateral
femoral notching may result.
The alternative is to make the center of rotation the middle of the knee (Fig. 27.16). Then less is cut from the lateral

condyle but more than usual from the posterior medial condyle. It can create more medial laxity but it limits the
increased AP size of the femoral component.
The third option is a combination of the first two; a compromise where neither the AP size of the component nor the
medial laxity is increased too much (Fig. 27.17). This is a
common choice in our practice.

Lateral Releases
 steotomy of the Lateral Condyle According
O
to Burdin
The knee is flexed at 90°. The synovial tissue covering the
lateral condyle is incised and the popliteal tendon and lateral
collateral ligament are identified. Longitudinal and transverse marks are made to allow estimation of the displacement of the lateral fragment after osteotomy. The osteotomy
is performed with a fine oscillating saw blade, parallel to the
long axis of the femur (Fig. 27.18). The lateral fragment is
approximately 1.5 cm in thickness (or in other words approximately one-third of the width of the lateral condyle). The

27 Total Knee Arthroplasty in Lateral Arthritis: Specifics and Surgical Techniques

319

Fig. 27.14 Little bone has been resected from the lateral condyle

Fig. 27.15 Externally rotating the sizer guide by “buildup” with an
osteotome; the center of rotation is the posterior medial condyle

Fig. 27.16 Externally rotating centrally with the guide 3°; center of
rotation in middle of the knee

Fig. 27.17 A compromise of the two methods. Note 3° central rotation
(arrow), plus a “buildup” posterior laterally (circle)

320

osteotomy is completed carefully with the use of an osteotome. The posterolateral structures must be released with a
knife to allow free movement of the bone block distally and
posteriorly (Fig. 27.19a, b).
Ligament balancing in flexion and extension is performed using a spacer. If the flexion gap is tight laterally,
the osteotomy will slide posteriorly (with the knee in flexion). It will slide distally (with a knee in extension) if the
knee is tight in extension (Fig. 27.20a, b). With the use of
the electric cautery one can now mark the optimal position
of the sliding condyle osteotomy for fixation using a cortical screw diameter 4.5 with, or without, a washer

P Archbold et al.

(Figs. 27.21 and 27.22). The fragment can be trimmed as
necessary.
The osteotomy at the level of the condyle has the advantage that it allows a controlled release of the lateral ligamentous structures. The osteotomy can be moved either distally
or posteriorly independent from each other to address tightness in extension or in flexion. Currently we prefer the osteotomy above the soft tissue release of the lateral collateral
ligament and the popliteal tendon (Fig. 27.23). Indeed,
release of the lateral structures associated with the use of a
total knee arthroplasty design where the anterior cruciate
ligament is not substituted could result in “retroligamentary”
anterolateral laxity. This pattern of instability is even more
significant and exaggerated in cases of residual varus.
However, the soft tissue releases are unable to precisely control the lengthening and the resultant lateral laxity. Insall
replaced the soft tissue release of the lateral collateral ligament and the popliteal tendon by a pie crust of the posterolateral soft tissue structures (with risk for the common peroneal
nerve) through a medial arthrotomy.
The post-operative instructions are slightly modified in
the case of a condylar osteotomy. Toe touch weight bearing
is allowed with the use of a splint during 45 days. Range of
motion exercises follows the conventional rehab protocol.

Pie Crust of the ITB

Fig. 27.18 The osteotomy is fashioned with a fine saw. Note the longitudinal and transverse marks made to allow estimation of subsequent
displacement

a

The pie crust technique encompasses multiple incisions in a
staggered fashion in the body of the ITB tendon. We rarely
perform this technique unless an important retraction remained
in extension at the end of the intervention (Fig. 27.24).

b

Fig. 27.19 (a, b) The osteotomy, completed with an osteotome, is mobilized with sharp dissection allowing free movement of the bone block
distally and posteriorly

27 Total Knee Arthroplasty in Lateral Arthritis: Specifics and Surgical Techniques

321

b

a

Fig. 27.20 (a) The condylar fragment has been moved distally in this case (arrow). (b) In this case it has been moved posteriorly (blue arrow) as
well as distally (green arrow). The fragment is being trimmed distally (star)

a

b

Fig. 27.21 (a, b): Post-operative X-rays of case in Fig. 27.20b. Note the slight notching required to prevent tibiofemoral size mismatch from the
externally rotated femoral component

322

a

P Archbold et al.

b

Fig. 27.22 (a, b) Post-operative X-rays of case in Fig. 27.19. Note the tibial tuberosity osteotomy, and the slight notching again to prevent size
mismatch and overhang

Fig. 27.23 Subperiosteal release of the lateral structures (LCL and
popliteus tendon)

27 Total Knee Arthroplasty in Lateral Arthritis: Specifics and Surgical Techniques

323

Surgical Sequence
A valgus deformity of the knee is a dynamic phenomenon,
which is not yet well understood. In contrast to the varus knee,
the preoperative radiographic evaluation (full leg and stress
X-rays) does not allow one to foresee certain difficulties that
can be encountered during the surgical procedure. However,
one can predict if an additional soft tissue release is necessary
once the tibial cut is performed. If after releasing the capsule
at the level of the joint line on the lateral tibial plateau, the
flexion or extension gap remains trapezoidal we routinely perform an osteotomy of the lateral femoral condyle.
At the end of the procedure, contracture of the IT band
can be the cause of a soft tissue imbalance in extension. In
this very rare situation, we perform a complementary pie
crust of the IT band. Although release of the biceps tendon or
osteotomy of the fibular head is discussed in the literature,
we have never needed this technique to obtain correct alignment and ligamentous balance. If significant laxity is present
in the elderly patient, one can use a more constrained prosthetic design. The use of the type of prosthesis, however, has
to remain limited because of the potential complications
associated with the use of more constraining implant. In our
department, this decision has to be made prior to the intervention since we do not have this type of prosthesis permanently available in our hospital.
Fig. 27.24 Pie crust of the ITB: multiple incisions in a staggered
fashion

Computer-Assisted Total Knee
Arthroplasty

28

S Lustig, R Badet, Maad AlSaati, P Neyret,
and C Butcher

Introduction
The theoretical aspects of computer-assisted total knee
arthroplasty (TKA) were discussed in Chap. 25, and we will
consider in this chapter the practical applications of computer-assisted TKA. We have described conventional TKA in
prior chapters and will discuss here a similar technique with
computer assistance.
The use of “navigation” surgery using the computer has
developed over the past 20 years since the first computer-­
assisted TKA in a human was performed under the direction
of D. Saragaglia. The goal of this technique is to have more
accurate and reproducible surgery (while at any time having
the option of switching to a traditional guide system). We
describe the PLEOS navigation system (Fig. 28.1), but the
principles are similar with most navigation systems. Navigation
systems allow us to use one of three surgical strategies.
• Performing cuts independently
• Performing cuts dependently
• Simulating the distal and posterior femoral cuts (after the
tibial cut is made)

S Lustig
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
R Badet · M AlSaati
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Fig. 28.1 PLEOS navigation system

Setup
Patient positioning and setup is the same as for a conventional knee arthroplasty.

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_28

325

326

S Lustig et al.

Surgical Approach
There is no difference from the standard TKA approach. We
perform an anteromedial approach in cases involving a varus
knee and anterolateral approach in cases involving a valgus
knee.
The cruciate ligaments are resected and the tibia dislocated forward with a Hohmann retractor. This maneuver is
required for acquisition of all necessary landmarks. All
osteophytes are removed.

Introduction of the Sensors
One sensor is affixed to the tibia and one to the femur for
detection by the camera. Both are placed percutaneously and
in positions that will not interfere with access to the knee
during the procedure.
The tibial sensor is positioned 10 cm below the skin incision so as not to interfere with the tibial cutting guide. The
two threaded pins are drilled into the medial cortex of the
tibial with spacing corresponding to the width of the sensor,
which is then attached to the two pins. Two percutaneous
femoral pins are similarly placed 10 cm above the skin incision to avoid impingement on the femoral cutting guide.
Other systems may allow placement of the pins within the
wound proximally, and in the metaphyseal bone.
One then checks the positioning of the computer and the
receiving antenna. The tibial and femoral sensors must be
clearly visible to the cameras in both full extension and maximum flexion of the knee.

Fig. 28.2 Acquisition of the center of the hip. We must make small
rotary movements clockwise

Acquisitions
Femur

Fig. 28.3 Acquisition of the anterior femoral points

The center of the femoral head is defined by putting the knee
in extension and performing a slow, repeated circumduction
movement with the entire lower limb. It is important to
ensure that the pelvis is immobilized at this stage (Fig. 28.2).
The remaining points are acquired using a probe with an
attached optical sensor:
• The anterior cortex is identified with three points. It is
important to identify the most anterior part of the cortex
to avoid notching the femur during the anterior femoral
cut (Fig. 28.3).
• Three points mark Whiteside’s line (Fig. 28.4).
• The knee center is identified as a point about 5 mm anterior to the femoral insertion of PCL (Fig. 28.5).
• The most distal part of the medial and lateral condyles is
marked by scratching in the corresponding areas
(Fig. 28.6a, b). They provide the reference level (height)
for the distal femoral cuts.

Fig. 28.4 Acquisition of the bottom line of trochlea (Whiteside line)

28 Computer-Assisted Total Knee Arthroplasty

327

Fig. 28.5 Acquisition of the femoral center, 5 mm above the femoral
insertion of the PCL

Fig. 28.7 Acquisition of the insertion zone of the ACL to the tibia

a

b

Fig. 28.6 Acquisition of the distal part of the femoral condyles (a: medial, b: lateral)

• In this system, the posterior condyles are identified using
a special instrument which is applied against the femur; it
has two appendages that rest at the back of the condyles
and in the bearing zone distally. The probe is inserted in
this instrument and the computer to indicate the flexion or
recurvatum positioning of the tool. It is positioned at 0°
(or 3° depending on the surgeon’s choice).

the corresponding areas. They provide the reference for
the level of cut (Fig. 28.8a, b).
• The center of the PCL insertion at the posterior plateau,
and the medial third of the anterior tibial tuberosity (ATT)
(Fig. 28.9). The straight line joining the two points provides the sagittal axis of the tibial plateau.
• The medial and lateral malleoli of the ankle (Fig. 28.10).

Tibia

Axis and Ligaments

The following points on the tibia are acquired using the
probe tip:

The overall axis of the lower limb is measured by the navigation system, at an angle close to full extension (Fig. 28.11).
The varus and valgus laxity is evaluated in extension. This
measurement is performed by applying stress in varus and
valgus while the navigation system notes the maximum
value in degrees.

• The ACL insertion site, between the tibial spines (Fig. 28.7).
• The deepest area of the medial tibial plateau and most
prominent point of the lateral plateau, by scratching on

328

S Lustig et al.

a

b

Fig. 28.8 Acquisition of the tibial plateau (a: medial, b: lateral)

Fig. 28.9 Acquisition of the insertion zone of the PCL to the tibia

Fig. 28.10 Acquisition of the medial and lateral malleolus

The same measurements of laxity are carried out at 90° of
flexion.
Finally, the maximum flexion is registered.

sagittal planes (Fig. 28.12). The height of the desired tibial cut
is then adjusted (for our current prosthesis usually 10 mm with
respect to the lateral tibial plateau in case of genu varum and
7 mm from the medial tibial plateau in case of genu valgum)
(Fig. 28.13).
Once the parameters are validated, two pins are used to
“save” the position for the tibial cutting guide and the cut is
carried out with a saw. The actual cut is then compared with
the predicted cut by placing the sensor device on the cut surface (Fig. 28.14).

Tibial Cut
The knee is in 90° of flexion with the tibia dislocated anteriorly. A Hohmann retractor is positioned behind the tibia
while another is applied to the lateral side of the lateral tibial
plateau to maintain the patella eversion and provide
visualization.
The tibial guide is fixed by a pin that is placed at the center
of the tibial epiphysis, in the insertion of the ACL. The onscreen target icon will direct the surgeon in placing the guide
correctly to achieve the chosen cut; we use 0° in the coronal and

Balancing
The 10 mm tibial spacer is used to evaluate the knee balance
before making the femoral cuts.

28 Computer-Assisted Total Knee Arthroplasty

329

Fig. 28.11 Initial assessment with measurement of the deformity
(HKA angle), the laxity in extension/flexion and range of motion
Fig. 28.13 Setting up the tibial cutting guide with control of the varus-­
valgus, slope, and height of cut

Fig. 28.12 The onscreen target icon will direct the surgeon in placing
the tibial guide correctly and achieve the chosen cut

The balance in flexion is checked first (Fig. 28.15). The
computer shows the medial and lateral spaces as well as the
planned femoral cut. A minimum space of 18 mm is required
for our current prosthesis (10 mm tibial and 8 mm femoral
spaces). At this point, a medial or lateral release can be performed to balance the space in flexion, and further adjustments made to the intended cuts and femoral component
position. We can thus control not only the rotation, sizing,
and offset, but also the balancing in flexion without the use
of a multifunction cutting guide.

Fig. 28.14 Verification of the tibial cut with the specific sensor positioned on the cut

The extension space is then checked (Fig. 28.16). The
computer shows the medial and lateral spaces as well as the
planned femoral cut. Again a minimum space of 18 mm is
required for our current prosthesis. The position needed to
obtain the desired mechanical axis of the femur and balance
in extension is selected.

The femoral instrumentation is applied and pinned to the
femur, guided by the on-screen target icon. This will provide
the chosen distal femoral cut as defined in the previous step
(appropriate varus/valgus and flexion). The desired depth of
the femoral cut is then set (usually 10 mm from the distal
femur) (Fig. 28.17).

Once all parameters are considered satisfactory, the virtual positioning and sizing are validated.

Femoral Cut

330

S Lustig et al.

Fig. 28.15 Control spaces in flexion

Fig. 28.17 Setting up the femoral cutting guide with control of the
varus-valgus, of the flexion and the cutting height

Fig. 28.16 Control spaces in extension

Fig. 28.18 Control femoral cut with the specific sensor positioned on
the distal cut

Once the distal cut is made, it is compared with the predicted cut by placing the sensor device on the cut surface
(Fig. 28.18).
The next instrument guides the two positioning pins for
the 4-in-1 femoral cutting guide. It is applied to the distal
femoral cut with its two posterior feet positioned in contact
with both posterior condyles. Adjustment of rotation and the
anterior-posterior positioning is made according to the previous chosen values. Once positioning is verified, the two positioning pins are introduced, and the navigation removed. The
4-in-1 cutting guide is applied, and the four remaining femoral cuts are performed.

Trials
The trials are placed and checks are made of the overall axis
of the lower limb, and the laxity in flexion and extension (the
same parameters as at the beginning of the operation).
The patellar cut is not navigated.

Applying the Final Implants
The final implants are cemented in place. We can confirm the
final tibiofemoral axis, and once the cement hardens we can
evaluate the varus/valgus laxity (Fig. 28.19).

28 Computer-Assisted Total Knee Arthroplasty

331

Closure and Post-operative Care
All navigation pins are removed and closure and post-­
operative care proceed as for a non-navigated case.

Fig. 28.19 Final check of the HKA angle, laxity in extension/flexion
and maximal flexion

Robotic Assisted Unicompartmental
Knee Arthroplasty

29

S Lustig, C Batailler, E Servien, and P Neyret

Introduction

Setup

Robotic surgery is a tool that improves the precision of bone
cutting and ligament balancing for unicompartmental knee
prostheses. The goal is not to replace the surgeon but to allow
him to improve his performance. The term ‘augmented surgeon’ seems appropriate.
On the basis of numerous data published in AngloSaxon journals, systems consisting of a constrained
robotic articulated arm working from scanner data
acquired before the intervention are used in the United
States, including the MAKO system (Stryker®). The main
constraint of these systems may be the need for an preoperative CT scan.
Another approach to robotic surgery has resulted from
the natural evolution of the preoperative image, with a
stage of bone morphing instead. We will present the technical aspects of this evolution, which used to be an open
platform, but is now related to only one company (Navio;
Smith and Nephew®). No specific preoperative imaging is
required, as standard radiographs simply allow the indication to be made (Fig. 29.1). Note also the technique for
medial UKA is very similar to lateral when using
robotics.

The patient is supine with a lateral holder and a distal holder
to maintain the knee at 90° (Fig. 29.2). A pneumatic tourniquet can be installed at the root of the thigh according to the
operator’s habits.
The NAVIO PFS console consists of three elements:

S Lustig (*) · C Batailler · E Servien
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
e-mail: sebastien.lustig@chu-lyon.fr
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

–– An infra-red camera (like that of a conventional surgical
navigation system) that must be installed in the 1-m environment of the operating field, facing the operator, so as
to permanently visualize the femoral sensor and the tibial
sensor.
–– A touch screen covered with a sterile protection. It is
located within the reach of the operator, most often at the
level of the contralateral hip.
–– A console controls the robotic milling cutter and milling
irrigation. The handpiece can be held in one hand and is
connected to the console by a cable and by the pipes
bringing water for irrigation.
The incision (for a medial UKA) is medial parapatellar,
typically the upper pole of the patella to 1 cm below the joint
line over a length of about 10 cm.
The first step is the positioning of the femoral and tibial
sensors, most often percutaneously for the tibia, and by a subvastus approach for the femur (or alternatively passing through
the quadriceps). The two sensors must be visible throughout
the procedure and for the extremes of knee movement.

Acquisition of Points of Interest
To verify that the sensors are stable throughout the intervention, a reference point is identified at the tibia and femur. At
any time during the procedure these can be checked with the
probe to verify that the sensors have not moved.

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_29

333

334

a

S Lustig et al.

b

Fig. 29.1 (a, b) Pre-op X-ray showing MFTOA grade III

The hip centre is acquired by repeated circumduction
movements with a maximum error of 0.9 mm.
The medial and lateral malleoli are then acquired at the
ankle directly with the probe.
The axis of knee flexion is acquired by performing a complete flexion-extension movement without stress in varus
valgus.
Some discrete points or surface on the medial femoral
condyle are acquired. The femoral acquisition is done
with a bone morphing phase of the area of interest, always
using the probe. The same sequence is then repeated in
the tibia.
The flexion axis of the knee is again recorded with stress
in reduction on all the articular amplitudes.

Fig. 29.2 Positioning the patient with the femoral and tibial sensors
positioned on the medial side of the limb, the operator positioning laterally to the knee to allow the sensors to be viewed at all times

29 Robotic Assisted Unicompartmental Knee Arthroplasty

Fig. 29.3 Placement planning of the femoral implant: 0° of varus, 0°
of rotation and 15° of flexion

335

Fig. 29.4 Global knee balancing planning according to the positioning
of the femoral and tibial implants

Planning
This is one of the essential steps of this robotic system,
because it allows a real dynamic planning (taking into
account the reducibility of the deformation).
We begin by deciding on the desired position for the
femoral component, in the three planes of the space
(Fig. 29.3). The first step is to choose the size of the desired
implant, but it can be changed at any time. A separate
screen allows to visualize the exact positioning with respect
to the shape of the femoral condyle that has just been
recorded. The touch screen allows to rotate a 3D view of
the femoral condyle with the implant positioned according
to the planning, in order to visualize exactly the final position of the latter. The angular values corresponding to the
positioning of the femoral implant are visible at all times:
varus/valgus, flexion, rotation. The choice of the implant is
to cover the bone surfaces as much as possible, keeping the
height of the joint space and avoiding having a conflict with
the tibial spines.
The same steps are then performed for the tibial part. We
decide first of all on the size of the implant and the thickness
of the polyethylene. The varus/valgus positioning, the tibial
slope and the positioning relative to the tibial spines, as well
as the rotation of the implant, are then chosen. Again the
touch screen can rotate 3D images to accurately visualize the
positioning of the implant in the three planes of space.

Fig. 29.5 Adjustment of the mediolateral positioning of the implants
in order to centre the femoro-tibial contact point

The next step is to visualize the consequences of our
planning in terms of angular correction (preoperative versus post-operative) between 0° and 120° (Fig. 29.4). At this
stage we can modify the positioning of the tibial implant
(varus/valgus, slope, rotation, cutting height) and femoral
implant (varus/valgus, flexion, rotation, cutting height) and
visualize live the consequences on the final angular correction. These parameters take into account not only static

336

S Lustig et al.

Fig. 29.7 Milling of the femoral condyle. The burr retracts automatically when leaving the planned area

Fig. 29.6 Positioning of the spacers to perform the milling of bone
surfaces

acquisitions but also initial dynamic acquisitions, taking
into account the reducibility of the deformation at each
degree of flexion.
The last stage of the planning consists in visualizing the
contact points between the two implants during the arc of
flexion. This makes it possible, if necessary, to lateralize or
medialize one or the other of the implants to better centre the
point of contact (Fig. 29.5).
It is possible to navigate between the different screens of
the planning, and once the desired result is obtained, the final
choice is validated.

Preparation of Bone Surfaces
Once the planning is validated, we begin the preparation of
the bone surfaces (Fig. 29.6). The assembly of the milling
cutter and the irrigation system and then the calibration
phase take a few seconds. A final control step makes it possible to visualize the area that will be milled and verifies that
it corresponds visually to the area where it is desired to position the implant.
It usually starts with the femur which is more easily
accessible, but it is also possible to start with the tibia if
desired. An automatic control system only mills the planned
area. If one leaves this area, the milling tool retracts, making

it impossible to remove the bone in an unwanted area by
­mistake (Fig. 29.7). At any time, the depth of the bone
remaining on the screen is visualized by means of colour
patches, which makes it possible to position the cutter efficiently. It is the surgeon who has complete freedom of movement, but the robotic system simply retracts the cutter when
it is outside the planned area. The knee is gradually mobilized to reach the most posterior zone with the burr. It is
sometimes necessary to prepare the tibia before accessing
the most posterior part of the femoral condyle (Fig. 29.8).
Once the femur has been prepared, the tibia is shifted to
the same visual planning control system before starting to
mill. It begins on the most anterior part of the tibia, and gradually we mill the planned surface. One option is to prepare
only the anterior part of the tibia with the Navio burr. Then a
tibial guide is placed and the posterior part of the tibial plateau is cut with the saw.
A rasp finally makes it possible to flatten the bone sections. We end with the removal of the meniscus, which is at
this stage very easily accessible.
The last step is to mill the keels or pegs of the femoral
implant, which is specific to each type of implant. This preparation is done under visual control of the planning visible
on the screen (Fig. 29.9).

29 Robotic Assisted Unicompartmental Knee Arthroplasty

a

337

b

c

Fig. 29.8 Display on the screen of the countersunk areas and the bone remaining to be removed (a distal femur, b posterior femur, c tibia)

a

b

Fig. 29.9 Milling of the femoral block (a positioning of the milling cutter, b control on the screen)

338

a

S Lustig et al.

b

Fig. 29.10 Positioning of the trial implant (a) and control of the balance over all the articular amplitudes (b)

Testing

Post-operative Guidelines

We can then perform trials manually and on screen.
Simultaneously we can verify both the alignment obtained
and the balancing over the full range of motion (Fig. 29.10).

No special precautions are necessary, the post-operative care
is similar to conventional medial UKA technique (see
Chap. 27), with immediate full weight bearing and knee
mobilization without limitation.

Definitive Implants
Cementation and fixation of final implants is done according
to the operator’s habit. We can once again check the angular
correction and knee balancing with the implants in situ. The
sensors (femoral and tibial) are removed and the closure is
performed without particularity. At the location of the sensors we simply close the skin with some staples.

Results
In our experience with this system, the results and in particular the positioning of the implants have been significantly improved. The radiographic analysis in the three
planes was satisfactory, corresponding to the desired planning (Fig. 29.11). We did not encounter any particular difficulties, with an operating time rapidly falling below 60 min
after the first cases that were necessary to take control of the
system. Our preliminary results seem to confirm the good
results recently reported with this system by our AngloSaxon colleagues, and the indications currently extend to
primary total knee arthroplasty and bicruciate retaining
arthroplasty.

29 Robotic Assisted Unicompartmental Knee Arthroplasty

a

c

Fig. 29.11 Final implant (a) and post-op X-ray (b, c)

339

b

Total Knee Arthroplasty After Valgus
Osteotomy of the Tibia

30

G Demey, H Hobbs, P Neyret,
and C Butcher

Introduction
The need for a total knee arthroplasty (TKA) after a valgus
high tibial osteotomy (HTO) presents several difficulties.
The osteotomy induces a bone deformity that can lead to
unbalanced bone cuts if the resection is based on ligament
balancing techniques. This potential difficulty should be
anticipated and planned for prior to surgery, by carrying out
a thorough clinical and radiologic examination. Specifically,
radiographs should include:
•
•
•
•

AP and lateral single leg stance views
Axial view of the patella at 30° of knee flexion
Stress views (varus and valgus)
Full leg views (weight bearing)

Analysis of the tibial shape categorizes the deformity of
the epiphysis into:
• Translation
• Angulation
This is then quantified by the mechanical tibial angle
(mTA) and the tibial slope in the sagittal plane (Fig. 30.1).
The overall limb alignment is measured with the mechanical
femorotibial angle (mFTA), showing either undercorrection
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
H Hobbs
Centre Albert Trillat, Lyon, France

(mFTA ≤ 180°) or overcorrection (mFTA > 182°). The angular direction of the previous osteotomy is of lesser importance. These figures plus the templating of the position of the
implant allow anticipation of:
• Mismatch of the bone and the implant, with subsequent
stem/cortex impingement
• Estimated cut level and inclination, and the resulting
induced spaces
With this information, the surgeon can plan the procedure. Angulation over 9° suggests a combined TKA/tibial
osteotomy, and significant translation either an offset stem or
custom prosthesis (Figs. 30.2 and 30.3):
• Operative technique:
–– approach; medial or lateral, tibial tuberosity osteotomy
–– guidance; intramedullary, extramedullary, or combined tibial guidance for coronal cut
–– balancing; sliding osteotomy of the lateral condyle of
Burdin; combined TKA/metaphyseal tibial osteotomy
• Type of prosthesis:
–– constrained or unconstrained
–– custom prosthesis; offset keel

Surgical Technique
The usual setup for a TKA is employed.

P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

Incision

C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

The choice of surgical approach is essential because it facilitates the releases and ligament balancing.

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_30

341

342

a

G Demey et al.

c

b

Fig. 30.1 Assessment of overall alignment (a), translation and angulation of the epiphysis (b), tibial slope and patellar height (c)

Fig. 30.2 The analysis of a deformity guides
treatment. A combined TKA and tibial
osteotomy can be considered in cases with
significant translation, and angulation over 9°

Type of Tibial Malunion
Translation

custom/offset stem

Angulation

<9º condylar osteotomy

or
combined tibial osteotomy

A previous vertical incision can be used and extended, if
necessary, since it is close to the midline.
A previous horizontal incision cannot be re-used. The new
incision is created vertically in the midline, if possible crossing
the previous scar at right angles (Fig. 30.4).
The location of the incision alone does not influence the
choice of whether a medial or lateral arthrotomy is used.
• In cases of undercorrection (varus knee), a medial parapatellar arthrotomy is preferred.
• In case of overcorrection (valgus knee), a lateral parapatellar arthrotomy is preferred (Fig. 30.5).

>9º combined tibial osteotomy

In cases with multiple scars where there is a possibility of
skin necrosis, we prefer the most lateral scar. Sometimes the
opinion of the plastic surgeons is sometimes needed.

Removal of equipment
Previous hardware is not routinely removed. When hardware
removal is required, we remove it during the TKA procedure
when possible in order to avoid two interventions and hospitalizations. Nevertheless, a two-stage surgery is preferable

30

Total Knee Arthroplasty After Valgus Osteotomy of the Tibia

a

343

b

Fig. 30.3 (a, b) Pre- and postoperative X-rays for TKA with an osteotomy of the lateral condyle. One may observe that the lateral compartment
has been lengthened by the procedure. It is difficult to solve this kind of pathology without a condylar osteotomy or a corrective tibial osteotomy

when there is a question about infection. Microbiology advice
is required if there is a history suggestive of infection.
The previous scar from the osteotomy may be used independently to remove the hardware but there is concern of
skin necrosis, especially if used in a one-stage procedure.

Exposure

Fig. 30.4 Case with previous horizontal incision. The new incision
cross the previous scar at right angles

If there is a patella infera or significant knee stiffness (flexion ≤ 90°), there is a risk of avulsion of the patellar tendon
while flexing the knee to dislocate the patella. If the tension
on the patellar tendon insertion is excessive, a 2 mm pin may
be placed in the tendon as an artificial constraint once the
patella is dislocated (Deschamps). This strengthens its
attachment and helps avoid avulsion during flexion and subsequent anterior dislocation of the tibia.

344

G Demey et al.

Fig. 30.6 Tibial tubercle osteotomy

Fig. 30.5 Lateral parapatellar arthrotomy in case of overcorrection
(valgus knee)

Osteotomy of the tibial tuberosity is sometimes necessary.
One must attempt to avoid this however, by initially performing an arthrolysis, releasing the condylar gutters to gain exposure. When performed, the tubercle osteotomy has the
advantage of allowing a proximal transfer of the tibial tubercle
when the patella is low. It is necessary to osteotomize a piece
of tibial tubercle long enough (≥6 cm) and deep enough (it
must reach the cancellous metaphyseal bone) and perform the
fixation with two screws of 4.5 mm diameter (or three screws
of 3.5 mm diameter) in order to avoid a non-union (Fig. 30.6).
A lateral retinacular release is not routinely performed. If
necessary, this release is performed from within the joint to
limit undermining in the pre-patella region.
The posterior femoral cut can be made at this time to facilitate the exposure and subsequent dislocation of the tibia. A
guide with small posterior “feet” is an advantage in these circumstances, as the posterior access may be limited causing
difficulty applying a regular guide (See Chap. 27, Fig. 27.8).

Tibial cut
The objective is to obtain a mechanical tibial angle of 90°,
cutting perpendicular to the mechanical axis of the tibia in the
frontal and sagittal planes. We use intra- and extra-medullary
guides as a double check to determine the correct cutting
angle on the tibia (the extra-medullary guide gives the varus-

valgus alignment). Two difficulties are then encountered: restoration of the joint line and compromise between coverage
and conflict between the tibial stem and the tibial cortex.
This must be planned preoperatively when templating to
ensure that there is no conflict between the tibial stem and
tibial metaphysis. Similarly, one must draw the proposed
cuts and evaluate their asymmetry.
The cutting height is difficult to determine. Due to the
tibial shape, it is difficult to use the plateau to determine the
cutting level and space height. The lateral compartment has
been “reduced” by the osteotomy and now also has unusual
cartilage wear. The medial compartment also has osteoarthritis but with bone loss.
• In the case of varus/undercorrection, the tibial cut is perpendicular to the long axis of the tibia and 7 mm thick
referenced from the lateral tibial plateau (for a combined
tibial component thickness of 10 mm) (Fig. 30.7).
Multiple drill holes in the medial tibial plateau may be
necessary, to help with cement fixation.
• In the case of valgus/overcorrection, the tibial cut is perpendicular to the long axis of the tibia and 7 mm thick referenced
from the medial tibial plateau (again for a combined tibial
component thickness of 10 mm) (Fig. 30.8).
The translation created by the epiphyseal osteotomy may
require a compromise of tibial coverage to avoid conflict
between the keel or stem and the tibial cortex. Sometimes an
offset tibial keel or a custom made ­prosthesis is necessary
(Figs. 30.9 and 30.10). Preoperative planning is therefore
important so that these implants may be ordered if required.
A line from the middle of the ankle joint to the center of the

30

Total Knee Arthroplasty After Valgus Osteotomy of the Tibia

345

Fig. 30.8 Overcorrection (valgus knee): tibial cut of 7 mm from the
medial tibial plateau in case of overcorrection
Fig. 30.7 Undercorrection (varus knee): tibial cut of 7 mm from the
lateral tibial plateau

knee shows the ideal alignment, but may need to be altered
to avoid cortical conflict (Fig. 30.11). Translating medially
will still provide an orthogonal prosthesis, but with a more
asymmetric cut and difficulty with gap symmetry. We recommend using a long tibial stem (75 mm from the tibial cut)
in cases of previous opening-wedge osteotomy, especially
when synthetic bone substitute was used.
The concept of tibial slope in total knee arthroplasty is
important because increased tibial slope can lead to anterior
tibial translation/subluxation of the tibia when weight bearing. It is for this reason that we choose a postoperative tibial
slope of 0°. The problems from asymmetric bone cuts in the
coronal plane can also be found in the sagittal plane, and
tibial slope should be carefully evaluated before surgery. We
measure the angle between the proximal tibial anatomic axis
and the medial tibial plateau, which is less variable than
using the anterior tibial cortex; the latter is affected by rotation on the radiographic films (Fig. 30.1).

In the common valgus/overcorrection deformity, soft tissue balancing involves resection of osteophytes and a lateral
release. We perform the lateral release in the following order:
–– Release of the IT band insertion on Gerdy’s tubercle during the lateral parapatellar approach
–– Pie crusting of the IT band if tight in extension (extended
release)
–– Pie crusting of the posterior lateral corner if there is mild
contracture in flexion and extension
–– Osteotomy of the lateral condyle (of Burdin) if there is
severe contracture in flexion and extension (see previous
chapters for the surgical technique) (Fig. 30.12a–c)
–– Rarely, a femoral release of the popliteus and LCL in
extreme cases (Fig. 30.13)
Finally, in cases of severe valgus/overcorrection
(mTA > 100°), a corrective osteotomy is performed prior to
the TKA so that the definitive procedure can be performed
under the best conditions. We try to avoid doing the corrective osteotomy and the TKA in a s­ ingle surgery. We also usually do a sliding osteotomy of the lateral condyle in these
cases.

346

G Demey et al.

Fig. 30.9 Case of significant tibial translation.
A tibial component overhang may result from
conflict between the keel and the tibial cortex.
Options include tibial osteotomy, an offset tibial
keel, or a custom made prosthesis

Fig. 30.10 Custom made
prosthesis with an offset tibial
keel. (a) Preoperative
planning; (b) Intraoperative
view; (c) Postoperative X-ray

a

b

c

30

Total Knee Arthroplasty After Valgus Osteotomy of the Tibia

347

Fig. 30.11 A variety of tibial
positions can be chosen.
These have different effects
on tibial coverage, keel/cortex
impingement, and cut
asymmetry which affects gap
symmetry and balancing

Fig. 30.12 (a–c) Osteotomy
of the lateral condyle. This
provides balancing in
extension and in flexion

a

b

c

348

G Demey et al.

Rest of procedure
The distal femoral cut is most often performed with a 6° valgus angle. Cementation and closure proceed as usual.

Postoperative
Weight bearing and crutches from day one. Flexion is limited
to 95° for 60 days and then without limitation. An extension
splint is used for walking until the quadriceps is able to lock
the knee in extension. If a proximalization of the tibial tubercle has been performed, the extension splint is utilized for
2 months.

Fig. 30.13 Subperiosteal release of the popliteus tendon and LCL
from the femur

Revision Unicompartmental Knee
Arthroplasty

31

G Demey, R Magnussen, P Neyret,
and C Butcher

Introduction

Indications

Unicompartmental knee prosthesis (UKA) has excellent
results for unicompartmental tibiofemoral osteoarthritis.
However, poor results and failures may occur.
UKA revision to total knee arthroplasty (TKA) is common. Some surgeons think that this procedure is as easy as a
primary TKA. We do not agree even if it may be easier than
a TKA revision.
Surgical history of the knee must be known to plan the
revision surgery: type of UKA (bone resection versus resurfacing) and cause of failure (metallosis, loosening, wear,
tibial plateau fracture, etc.).
UKA revision is not limited to revision to TKA. Sometimes,
only one of the two components needs to be changed.
Arthroscopy after UKA may be indicated in cases of chronic
and unexplained pain (most frequently after medial UKA),
but this indication is very rare.

The cause of failure must be known. Caution must be exercised when there is unexplained pain (usually medial UKA)
and one of the following causes sought. Usual causes, and
thus indications for UKA revision, are:
• Aseptic loosening.
• Implant malpositioning.
• Polyethylene wear or fracture (Fig. 31.1).

Revision of a UKA to a TKA
Surgical planning and techniques are described here, but
details of surgical technique to implant TKA are not. We
focus on specifics of a UKA revision.

G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

Fig. 31.1 Metallosis and polyethylene wear

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_31

349

350

• Osteoarthritis of one of the two other compartments
(opposite tibiofemoral or rarely patellofemoral).
• Sepsis is rare (<0.5%), but if present we prefer two-­stage
revision.
Sometimes, causes of failure are multiple.

Pre-operative planning
Apart from understanding the cause of the failure of the
UKA, a primary goal of the planning is to anticipate the type
of prosthesis we shall implant.

G Demey et al.

Firstly, if there is bone loss on the tibia then an augment
and long tibial stem are required, but there is usually no need
to increase the constraint. Most of the primary TKA systems
allow for the long tibial stem and tibial augment and are thus
sufficient for the case. Conversely, if there is bone loss in the
femur, a revision TKA system must be available, as most
primary TKA systems do not allow for a long femoral stem
or femoral augments.
Secondly, if the failure is due to laxity in the convexity of
the deformity, the degree of constraint of the revision TKA
must be considered and anticipated.
The process of pre-operative preparation and planning
involves clinical examination, along with biological and
radiological screening. Infection must be ruled out with history, clinical exam, inflammatory markers, radiographs, and
bone scintigraphy.
Standard radiographs needed are:
AP single-leg stance view
Lateral single leg stance view with 30° of knee flexion
“Schuss” view with 45° of knee flexion (Fig. 31.2)
Stress valgus/varus radiographs
Standing long-axis view to measure both lower limbs’
axes and angles, as in the planning of primary TKA
• Contralateral knee radiographs
•
•
•
•
•

Fig. 31.2 Schuss view showing polyethylene wear

Computed tomography is useful to both diagnose failure
of a UKA (osteolysis, prosthesis oversizing, loosening) and
plan surgery. Measurement of the posterior condylar angle
indicates if the position of the femoral component would
allow the cutting guide to be used without its removal,
thereby providing appropriate rotational alignment of the
new femoral component (Fig. 31.3).

Fig. 31.3 CT scan shows that the position of the femoral component in relation to the surgical trans-epicondylar axis will provide appropriate
rotational positioning of the femoral guide, and therefore the new prosthesis

31

Revision Unicompartmental Knee Arthroplasty

351

Technetium (tc99) scintigraphy and labelled leucocyte
scintigraphy (Leucoscan) may confirm if loosening is septic
or aseptic.

cave side of the limb is more easily be compensated by a
TKA.
• We use posterior stabilized TKA, so sagittal laxity is
rarely a technical problem. In cases of undiagnosed anterior laxity, metallosis occurs when the polyethylene wears
Technical difficulties—planning
down to the metal backing. In this situation, precise evaluation of bone lesions is more difficult and a CT scan
Technical difficulties are mainly due to bone loss and ligashould be carefully evaluated.
ment laxity.
• Implants: these can be left in place initially if they are
• Bone loss is more frequent. It can be evaluated on pre-­
perfectly positioned. Careful analysis of the lateral radiooperative radiographs (Fig. 31.4) and CT scans. However,
graph and the CT scans reveal the axial and sagittal alignthe true extent of bone loss can only be known during the
ment, and the long leg films the coronal. However, the
surgery, after removal of the components.
femoral component may prevent adequate positioning of
• Frontal plane laxity is evaluated on valgus and varus
cutting guide and it is often necessary to remove it prior to
stress radiographs. In case of lateral laxity with failure of
making the bone cuts.
a medial UKA, ligament balancing must be considered
and can be difficult. In our experience, the need for revision to a hinged TKA is rare however. Laxity on the con- Medial UKA Revision

Surgical Approach
Skin incision for a medial UKA revision to a TKA is made
extending the previous scar proximally, and distally if needed.
A medial parapatellar arthrotomy is performed (Fig. 31.5).
Tibial tubercle osteotomy is usually not necessary.
The cause of UKA failure is confirmed, and wear and
fixation of implants are evaluated. Synovial and bone biopsies are taken to look for wear debris or infection.
Tibial Cut
The tibia is easily anteriorly dislocated (Fig. 31.6). The tibial
component and cement are carefully removed to prevent
­further damage to the medial tibia and minimize further bone
loss. The femoral implant can often be left in place at this

Fig. 31.4 Pre-operative X-ray showing a bony defect of the medial
plateau in case of a 9 mm cut (dotted line)

Fig. 31.5 Medial parapatellar approach

352

G Demey et al.

stage. An osteotome is used to prepare the introduction point
of the intra-medullary guide. The landmark is the ACL
footprint.
The tibial cut is done the same way as in a primary
TKA. The bony reference for the tibial cut is the native lateral compartment and the thickness of the resected bone
should be equal to that of the tibial component (Fig. 31.7a,
b). An oscillating saw is used. The goals are to avoid further
bone loss and to reproduce the tibiofemoral joint line.

This standard level cut may be proximal to the residual
medial compartment bone after removal of the tibial component. If this is the case, a further cut of 5 mm or 10 mm can
be carried out parallel to the lateral compartment tibial cut to
accommodate a metal augment (Fig. 31.8).
When the tibial cut is done, the trial tibial implant with
augments is impacted (Fig. 31.9). The CT may guide the
rotational positioning, and the posterior border of the tibia
and the tibial tubercle are intra-operative landmarks. The
tibial keel is prepared. When augments are necessary, a

Fig. 31.6 Tibial exposure

Fig. 31.8 Medial compartment tibial re-cut (4, 8, or 12 mm)

a

b

Fig. 31.7 Intramedullary guide introduction (a) and fixation of the tibial cutting guide (b)

31

Revision Unicompartmental Knee Arthroplasty

353

long keel must also be used. We do not hesitate to use long
and thin keels (10–12 mm diameter) to avoid conflict of the
keel and cortex and facilitate implant positioning
(Fig. 31.10). Some systems provide opportunity for offsetting the keel to help in this regard, and modular sleeves are
also an option. Finally, custom implants can be used, like
the tuliped tibial keel shown in Fig. 31.11a. The tibial
hemi-epiphysis is filled and the prosthesis lies on medial
cortical bone, similar to the femoral stem of a total hip
prosthesis (Fig. 31.11b).

Fig. 31.9 Metal augment filling the medial space

Fig. 31.10 A long tibial keel is used in case of metal augment

Femoral Cuts
The knee is positioned at 90° of flexion. The femoral guide
for the posterior cut must be applied on the distal and posterior condyles, as described in the TKA chapter.
This can be done without removing the femoral component if it is well positioned, as determined before surgery
with radiographs and CT scan. Rotational malposition and
over-/under-sizing of implant must be excluded. In case of
malpositioning of the femoral component, it is removed and
the posterior defect is filled with augments placed on the
femoral cutting guide (Fig. 31.12a, b).
Generally we prefer to remove the femoral component
before placing the distal femoral guide. Distal augments are
not needed as the guide usually rests on the native condyle.
The femoral entry point is prepared with an osteotome
above the medial side of the notch (just anterior to the PCL
origin). The femoral medullary canal is prepared with a
drill bit and is generally placed in 7° valgus. The stem of
the femoral cutting guide is then inserted in the femoral
medullary canal and the femoral cutting guide rests against
the distal femoral condyle(s). The guide must also be
applied on the posterior femoral condyles (or on the lateral
condyle and the posteromedial augment if the femoral
implant had to be removed).
The antero-posterior size of the implant is measured with
the guide on the anterior femoral cortex. The appropriate rotation is assessed and decided according to the pre-­operative
CT scan. Based on this evaluation, the rotation guide may be
applied to the component if it is deemed to be perfectly positioned. If the femoral component of the medial UKA was
oversized, an excessive internal rotation of the femoral TKA
component can occur. However, if the femoral component of
the UKA has been removed, excessive external rotation must
be prevented with a posterior augment/spacer. Thus, control
of the femoral component rotation based on the posterior condyles is more unpredictable than in a primary TKA. In addition to the CT scan evaluation, the epicondylar axis and
Whiteside’s line can be used as further landmarks. If there is
any doubt about the position of the femoral component, we
remove it before setting the rotation.

354

a

Fig. 31.11 (a, b) Tuliped tibial keel lies on medial cortical bone

G Demey et al.

31

Revision Unicompartmental Knee Arthroplasty

355

b

Fig. 31.11 (continued)

a

b

Fig. 31.12 The femoral component is removed (a) and the femoral cutting guide is positioned (b)

If the component is retained whilst applying the initial
cutting guide, the guide must be taken off once appropriately
positioned, to remove the femoral component. It is then
replaced to make the distal and posterior femoral cuts
(Fig. 31.13).
The femoral guide is applied to the native condyle. The
distal femoral cut and the chamfers are done as described
before (Fig. 31.14a, b).

Filling bone defects
If bone loss is moderate and contained (sparing most of the
periphery), filling is done with autograft from the bone cuts
or with cement (Fig. 31.15). A long keel must be used if the
bony support is not strong enough. If bone loss is large or

uncontained, a metallic augment in association with a long
keel should be used (see chapter TKA revision). Augments
can be posterior, ­distal, or both.
Trial implants are then positioned, using a 9 mm insert
(Fig. 31.16a, b, c)

Ligament balancing
In our experience, PCL resection and use of a posterior-­
stabilized TKA make the ligament balancing easier. However,
increased lateral laxity associated with medial UKA failure
can make ligament balancing difficult.
The first release is done during the surgical approach,
releasing the deep MCL. As described before, medial release
can be increased by “pie-crusting” the MCL. If needed, a

356

complete release of the distal superficial MCL can be
­performed. In cases of flexion contracture, release of the
semimembranosus tibial insertion is performed.

G Demey et al.

Final Implants
Any impingement between the tibial augment and the MCL
must be prevented. Different size and thickness of augments
must be available (Fig. 31.17).
The tibial component is cemented first, cementing the
tibial keel as well. If the tibial keel is long (>75 mm), a poly-

ethylene cement restrictor may be inserted into the medullary canal in order to prevent migration of cement into the
distal canal.
The polyethylene is positioned and the knee is hyperflexed to insert the femoral component. It is impacted with
the knee in 90° of flexion. A cement restrictor is also used
if a long femoral keel is used (>75 mm). The knee is
extended to compress the cement and the patellar button is
cemented to complete the procedure (Figs. 31.18 and
31.19).

Fig. 31.13 Posterior femoral cut

Fig. 31.15 Moderate bone loss, mostly contained

a

Fig. 31.14 Distal femoral cut (a) then anterior and chamfers (b)

b

31

Revision Unicompartmental Knee Arthroplasty

a

357

b

c

Fig. 31.16 Positioning of the trial implants. (a) tibial, (b) femoral, and (c) insert

Lateral UKA Revision
The surgical technique is the same except for the surgical
approach and the level of the tibial cut.

Surgical approach
For lateral UKA revision to a TKA, the skin incision is created
by extending the previous scar proximally and distally
(Fig. 31.20). In case of multiple scars, the most lateral one is
reused. We prefer a lateral approach over a medial one because
soft tissue release is easier and automatic, and some complications may be avoided including skin necrosis and patellar
necrosis. The exposure is easier than with a medial approach,
and a tibial tubercle osteotomy is rarely necessary (Fig. 31.21).
Fig. 31.17 Medial augments attached to the medial component

Preparation of the tibia
The reference for the bone cut is the native compartment.
The cut thickness from the native medial plateau should be
less than the thickness of the tibial implant (e.g. 7 mm below
the native medial tibial plateau, assuming a 10 mm tibial
component). This is because the medial plateau is concave,

358

G Demey et al.

Fig. 31.18 Cemented components
Fig. 31.19 Postoperative X-rays.
(a) AP (b) lateral view

Fig. 31.20 Lateral parapatellar approach

a

b

31

Revision Unicompartmental Knee Arthroplasty

359

Fig. 31.22 Cutting guide positioning. The cut is 6 mm below the
medial tibial plateau
Fig. 31.21 Tibial tubercle osteotomy in case of difficult exposure

unlike the convex lateral plateau which is the reference for
medial UKA revision (Fig. 31.22). As with medial UKA
revision, this cut may be above the remaining lateral tibial
bone after removal of the lateral tibial component. A minimal cut of the lateral compartment can be done parallel to
the medial compartment cut. A 5 mm or 10 mm difference
between the lateral and medial tibial plateau will be compensated by the corresponding metallic augment
(Figs. 31.23, 31.24, and 31.25). Minimal bone resection is
preferred.

Preparation of the femur
As the posterior femoral condyle is absent (after removal of
the implant), one must avoid positioning the femoral cutting guide in internal rotation. The guide is placed on the
distal medial femoral condyle and external rotation is
assured by placing an augment/spacer on the posterior
aspect of the lateral femoral condyle. Alternatively, external rotation can be achieved by placing one or more osteotomes between the posterior lateral condyle and the guide
(Fig. 31.26). Posterior cuts may be done first and then distal
cuts and chamfers, depending on the instrumentation
(Figs. 31.27 and 31.28).

Fig. 31.23 Case of 6 mm cut below the medial tibial plateau. The tibial surface is now flat and augments are not required

Ligament balancing with trial implants
Medial laxity associated with lateral UKA failure can lead
to difficult ligament balancing. The first release is done
during the anterolateral surgical approach where the lateral
capsule is released and iliotibial band is released from
Gerdy’s tubercle (but left in continuity with tibialis anterior

360

G Demey et al.

Fig. 31.26 External rotation achieved by placing one or more osteotomes between the posterior lateral condyle and the guide

Fig. 31.24 Femoral guide positioning: The contact is obtained
between the guide and medial condyle

Fig. 31.27 Posterior cuts

Final implants
Cemented implants are positioned and the knee is extended
to compress the cement (Fig. 31.29).
Fig. 31.25 Femoral guide positioning: The contact is obtained
between the guide and medial condyle

muscle fascia). Laxity in flexion and in extension is checked
after positioning the trial implants. If medial laxity in
extension persists, “pie-crusting” of the IT band is done
with an 11 blade. To do so, multiple transverse incisions are
done.

Postoperative care
Postoperative rehabilitation is the same as after primary
TKA (see Chaps. 26 and 27).
In case of tibial tubercle osteotomy, flexion is limited at
95° for 45 days. Two knee braces are worn during the first
45 days: one in extension for walking, and one at 30° of flexion at rest. A radiograph is done at day 45 to ensure adequate
healing of the osteotomy before removing the braces and
increasing knee flexion.

31

Revision Unicompartmental Knee Arthroplasty

361

Fig. 31.30 Degenerative lesions of the contralateral compartment
(arthroscopic view)
Fig. 31.28 Distal femoral cut

 rthroscopy and Unicompartmental Knee
A
Arthroplasty
In case of unexplained pain after UKA, arthroscopy is a useful diagnostic and therapeutic tool. It helps in diagnosis of:
• meniscal lesions of the opposite compartment
• impingement between the femoral implant and anterior
tibial spine or patella
• arthritis of the native patellofemoral or tibiofemoral compartments (Fig. 31.30)
• pain due to neo-meniscus formation
• metallosis
• integrity of the polyethylene
It also allows removal of extruded cement, fibrous scars,
meniscal tissue (residual or neo-meniscus) (Figs. 31.31 and
31.32), or hypertrophic synovitis. Flexion and extension
kinematics are also checked.

Technique

Fig. 31.29 Cemented implants

The technique has been described in the chapter on
Arthroscopy. Portals should be created carefully to prevent
damaging the femoral component. Patellofemoral compartment exploration is done first to look for cartilaginous
lesions, synovial hyperplasia, or impingement between the
femoral component and patella. Care must be taken not to

362

damage the femoral and tibial components with the arthroscope or the instruments.
The second step is exploration of the notch, in 90° of knee
flexion. The ACL is palpated, and if Hoffa’s fat pad is hypertrophied, it can be partially excised. The medial and lateral
tibiofemoral compartments are explored by positioning the
knee in valgus and then in the “figure of four” position.
Wear or metallosis may be present. Metallosis is very difficult to diagnose: indirect signs are synovial hypertrophy
and polyethylene wear. It is very rare to see black synovium
or synovial fluid. Component fixation and excess cement are
checked. Loosening is sometimes obvious. However, it is
difficult to know by palpation what moves: the implant alone
(loosening), or the implant plus whole bone segment to
which it is fixed. Fibrosis at the bone-implant tibial junction
is excised in order to check for any micromotion.
A contralateral meniscal lesion or meniscal proliferation
(neo-meniscus previously described after total meniscectomy) can be excised. Care must be taken not to damage the
polyethylene with the shaver. Postoperative care includes
full weight bearing and early mobilization.

G Demey et al.

Fig. 31.32 Resection of the neo-meniscus (arthroscopic view)

 evision of Unicompartmental Knee
R
Arthroplasty to a Second Unicompartmental
Knee Arthroplasty
Replacement of one of the two components only can be done
in case of obvious malpositioning (Fig. 31.33) or oversized
components creating joint pain (Fig. 31.34). However, the

Fig. 31.31 Neo-meniscus (arthroscopic view)

Fig. 31.33 Malpositioning of the femoral component in varus

31

Revision Unicompartmental Knee Arthroplasty

363

literature suggests a high rate or poor outcomes from such
procedures and the patient should be informed of the risk of
persistent pain.

Technique
The initial surgical approach is reused. Biopsies are done to
rule out infection. Look for wear or metallosis signs. The
malpositioned implant is removed with an osteotome, minimizing bone loss (Fig. 31.35). Correction of malpositioning
may require techniques such as the use of screw augmentation (Figs. 31.36, 31.37, and 31.38). However, a TKA (with
an augment and long keel) must be available in the operating
room and the patient must be informed that revision may
require a TKA.

Fig. 31.35 Removal of the malpositioned implant with minimal bone
loss

Fig. 31.34 Tibial component overhang

364

Fig. 31.36 Use of screw to to correct malpositioning – initial drilling
(technical trick)

Fig. 31.38 Postoperative X-rays (see case Fig. 31.33)

G Demey et al.

Fig. 31.37 Use of screw to correct the malpositioning – insertion of
screw (technical trick)

Revision Total Knee Arthroplasty:
Planning and Technical Considerations

32

S Lustig, R Magnussen, P Neyret,
and C Butcher

Planning
Preoperative planning is essential and must include understanding the cause of failure of the first arthroplasty.

Screening for Infection
The search for possible infection is necessary with clinical history (local swelling, pain, or symptoms in the joint contemporaneous with an infection in a different location), radiographic
evaluation, laboratory tests (white blood cell count ESR, CRP),
and bone scan (possibly supplemented by scintigraphy with
labeled neutrophils and medullary scans). Aspiration is easily
performed (repeated as necessary), and arthroscopic biopsies
may be obtained if there is doubt about the diagnosis.

 valuation of Bone Loss
E
(Radiographic ± Contralateral, CT Scan)
Plain radiographs are not always an accurate assessment of
bone loss. We obtain a CT scan, especially when preoperative metallosis is suspected, in order to better assess the bone

S Lustig
Service d orthopedie de l Hopital de la Croix Rousse, Lyon 69004,
France
R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

loss. This is often more significant than is seen on plain
radiographs.

Assessing the Size of the Implants
Knowing the size of the implants previously used, either
from operative reports or through the use of templates, is
critical. Templates can be used to size the contralateral knee
as well to provide insight into whether the implants are
appropriately sized.

 eed for Specific Hardware in Order to Remove
N
the Implants
It is important to know the type of implant in order to provide
the specific ancillary equipment for removal when needed.
This is particularly true in cases of more constrained prosthesis with long keels, augments, and sleeves. With hinged prostheses knowledge of the mechanism will be required in order
to separate the femoral and tibial implants.

 valuation of the Collateral Ligaments (Stress
E
Radiographs)
This step is essential. AP stress radiographs are performed routinely to assess the state of the collateral ligaments. Exces­sive
laxity (especially due to impairment of medial collateral ligament) requires the use of a constrained prosthesis for the revision. In the event of mechanical failure of a hinged prosthesis,
these views can also show a frontal plane mobility, which is
always abnormal for such implants.

C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_32

365

366

S Lustig et al.

Assessment of the Joint Line

Choice of the Implant and Modularity

An AP full leg length weight-bearing radiograph (showing
both lower limbs in full) is required. One can measure the
level of the contralateral tibiofemoral joint line (ratio between
length of the total lower limb and femoral length, and distance from anatomical reference points of the knee) and
compare the joint line with the operated knee. We can also
use initial radiographs (when available) prior to the first
prosthesis in order to assess joint line modification relative to
the initial joint line. The analysis is, however, unreliable in
cases with flexion contracture.

It is necessary to have a modular implant which includes longer
stems and also tibial and femoral augments or sleeves. Good
stability of the implant in the metaphysis is the key of success
in difficult revision. We used to order customized tuliped prosthesis in this situation but now sleeves are available in several
systems (Fig. 32.1). Often, appropriate templating can anticipate needs. However, removal of the implants can lead to
defects larger than those initially present. We draw attention to
the fact that the bone defects are systematically underestimated,
especially in cases of metallosis. We should keep this idea in
mind and have a low threshold for ordering a CT.

Choice of Constraint
Use of an unconstrained prosthesis (posterior stabilized which
we favor, “deeply dished,” or PCL retaining) is determined by
the quality of the collateral ligaments. This choice will then
depend on several anatomical factors, but must also take into
account the age and motivations of patients. Too much varus/
valgus laxity, or difficulty obtaining flexion extension gap balance requires the use of a prosthesis with more constraint.
Collateral laxity (evaluated before surgery or anticipated due
to the reoperation) is for us the main element that suggests the
use of a rotating hinge prosthesis. Factors that can lead us to
consider a hinge prosthesis as a first choice are:
• major preoperative stiffness, because it may require a significant release of the collateral ligaments
• a significant frontal laxity or recurvatum, including an
association with neurological disease
• cases in which distraction in order to compensate for laxity would induce a significant change in the joint line and
patella infera
• the revision of a hinge prosthesis
In all these situations, there is an insufficiency of the
peripheral soft tissue envelope.
We also have a low threshold for using a rotating hinge for
very elderly patients.
We must anticipate these needs before the surgery, and
each time doubt exists we must be able to use a more constrained prosthesis. In all cases, we recommend that surgeons
less experienced in this surgery have a more constrained
prosthesis in the operative room before starting the revision
surgery.

Technical Principles
In order to make it easier to understand, we will describe the
surgical technique in two parts—the phase of explantation
and the phase of implantation of the new components.
However, although these two sequences follow one another
chronologically, with respect to a “functional” surgery one
cannot conceive of one without thinking of the other.
Nevertheless in case of revision for infection, one dissociates
these two phases more formally.

Explantation
Setup
Setup does not differ from that used for a primary TKA. The
patient is placed supine with a tourniquet at the top of the
thigh. A distal wedge keeps the knee to 90° of flexion during
the procedure.
 urgical Approach and Exposure
S
Prior incisions are marked with a marking pen as is the
planned incision for the revision procedure. The skin incision used for the first prosthesis is generally reused and is
often enlarged on both ends. If there are multiple prior incisions, the most lateral is used.
We will describe first approach with a medial arthrotomy;
however, the technique with a lateral arthrotomy is quite similar.
No subcutaneous undermining should be undertaken before
opening the superficial fascia. When this subfascial plane is
reached, the dissection can be continued without undue risk.
One can dissect and expose the anterior surface of the patella but
this should not continue laterally to the lateral edge of the
patella. The quadriceps tendon and the medial edge of the patellar tendon are identified. A medial arthrotomy is performed
using a 23 blade. It begins with the longitudinal opening of the

32 Revision Total Knee Arthroplasty: Planning and Technical Considerations
Fig. 32.1 Revision TKA
with a prosthetic system
which utilizes sleeves for
metaphyseal support. (a)
Pre-operative x-ray. (b, c)
Anteroposterior and lateral
post-operative x-rays

a

c

367

b

368

quadriceps tendon on its medial border (sometimes difficult to
identify), leaving a thin strip of tendon next to the vastus medialis muscle, which will allow a solid closure.
The arthrotomy continues adjacent to the patella and
along the patellar tendon to its insertion at the superior
medial edge of the anterior tibial tuberosity. The medial capsule is then released directly off of the bone on the anteromedial surface of the tibial plateau. This is a triangular-shaped
release of the medial capsule.
The deep fibers of the MCL are released with a Trillat
elevator to the proximal edge of the tibial implant, along the
joint line.
The knee is then placed in extension and the extensor
mechanism is dislocated laterally by everting or dislocating
the patella with a retractor. Care must be taken to release
adhesions in the infra-patellar region and reestablish the condylar recesses/gutters. Proximally the synovium abutting the
anterior aspect of the femur is widely resected in order to
expose the area proximal to the trochlea of the prosthesis.
The Hoffa fat pad, if it had been preserved during the previous surgery, may also be resected.
One rarely needs to section the lateral retinaculum to
facilitate eversion of the patella. The knee is then flexed
slowly with the patella dislocated with or without eversion
(Fig. 32.2). This part of the procedure can be dangerous, and
care must be taken not to avulse the insertion of the patellar
tendon at this point. We sometimes use the technical trick of

S Lustig et al.

placing a pin in the insertion of patellar tendon on the tibial
tuberosity to prevent this (Fig. 32.3).
If the stiffness is significant (<70° of flexion) or if the
patella is very low, one must sometimes perform an osteotomy of the anterior tibial tuberosity early in the procedure.
We usually prefer to perform a quadriceps snip (oblique section superiorly and laterally). If we are unable to expose the
knee by retraction of the extensor mechanism, an osteotomy
of the tibial tuberosity may be necessary. We perform it with
an oscillating saw (ideally with an osteotome) with the knee
near 90° of flexion. The length is at least 6 cm, and made
from medial to lateral, leaving it pedicled on the muscles and
fascia of the anterolateral compartment. The osteotomy is
completed by releasing the top of the tibial tuberosity behind
the patellar tendon in extension with a Lambotte osteotome.

Synovial Biopsies
Biopsies are always performed to rule out infection. This
screening includes a frozen histologic section to quantify
neutrophils as well as bacterial cultures. More than ten neutrophils per high-powered field is suggestive of infection.
 racking the Level of the Joint Line (Implants
T
in Place)
The level of the prosthesis is recorded on both the femur and
the tibia. One can use specific devices for this step or simply
mark the femur and tibia a set distance from the prior joint
line (generally between 6 and 10 cm) (Fig. 32.4a–c). One can
then quantify any change in joint line location caused by
changing implants. In the rare case that implantation does
not proceed, it is important to document the distance from
the holes to the joint line for use at second stage surgery.
Implant Removal
Removal of the polyethylene is the first step (Fig. 32.5). With
the knee flexed to 90°, the interface between metal and polyeth-

Fig. 32.2 Medial parapatellar arthrotomy and lateral dislocation of the
patella without eversion

Fig. 32.3 Use of a pin to secure the distal insertion of patellar tendon

32 Revision Total Knee Arthroplasty: Planning and Technical Considerations

a

b

369

c

Fig. 32.4 Tracking the level of the joint line: (a) With a specific device. (b, c) With 3.2 mm holes drilled in the tibia and femur

Fig. 32.5 Removal of the polyethylene

ylene is opened with a twist using a Lambotte osteotome, and it
is removed with a Kocher. Sometimes it is necessary to extend
the leg in order to get more space and facilitate the extraction.
We note the degree of wear and location of any wear.
One then focuses on the femoral implant. The bone–implant
interface is detached gradually with a Lambotte osteotome if it
is not loose (Fig. 32.6a). We can use osteotomes of different

widths. Specially designed osteotomes are available to access
difficult parts of the interface, for instance in the femoral notch
(Fig. 32.6b). The component should not be levered with the
osteotomes however, as this technique impacts the bone. A Gigli
saw, preferred by some to free the anterior part of the femoral
component, is not easy to use in our experience and counterintuitively may remove more bone. We prefer to introduce a thin
saw that stays in contact with the femoral implant, slowly working around the femoral pegs in an organized fashion. The femoral component can then be removed from the femur (Fig. 32.7).
The tibia is then exposed with two retractors, one behind
the tibia to dislocate the tibia forward, and a second laterally
on the tibial plateau, retracting the patella. Removal of the
tibial implant is sometimes difficult. One can initiate the cut
using a very thin saw under the anterior part of the tibial
implant before proceeding with osteotomes. The patellar tendon may impede the passage of the saw, and the osteotomes,
under the lateral portion of the tibial implant. Access to these
areas as well as the posterior interface may again be facilitated with special osteotomes (Fig. 32.8).
To reduce possible impaction of the tibia with osteotomes,
it is possible to insert a second osteotome, between the metal
and the first blade, then repeat the same exercise, alternating
from the medial to the lateral side (Fig. 32.9). If an all-­
polyethylene tibial component is present, using the saw for
removal is often possible. Many components allow a proprietary extractor to be inserted centrally to complete the
removal of the tibial component (Fig. 32.10). Others will
benefit from a generic extractor (Fig. 32.11).
Removal of the cement also requires time and patience.
Removal of the tibial cement should be performed with the

370

S Lustig et al.

a

Fig. 32.7 Removal of the femoral component

b

Fig. 32.8 Access to the posterior interface is possible with a specially
designed reverse osteotome (Shukla Medical NJ)

Fig. 32.6 (a) The bone–implant interface is detached with a Lambotte
osteotome. (b) Specifically shaped osteotomes are available to access
different aspects of the interface, including the femoral notch (Shukla
Medical NJ)

Lambotte osteotome. One can crack the cement into pieces
that can be removed more easily. The cement plug may also
be drilled and then removed with forceps. If a cemented
prosthesis with long stem is present, it is sometimes necessary to puncture the cement cap distally and then use flexible
reamers of increasing size to gradually remove the cement. A
cortical window is used as a last resort. When using an osteo-

tome on the cement, care must be taken that pieces of cement
do not fall back into the wound after contacting non-sterile
surfaces such as the theatre lights and masks.
When removing cementless implants, one must be cautious, methodical, and patient, because there is a risk of
removing a significant amount of bone attached to the
implants during the extraction of the prosthesis. Working
with the osteotome, one seeks to free the bone surface from
the implants as much as possible before attempting their
removal.
In all cases, it is worth taking the time to observe the
implants and note any particular wear of polyethylene.
These observations can provide insights into reasons for
implant failure and should be included in the operative
report.

32 Revision Total Knee Arthroplasty: Planning and Technical Considerations

371

Fig. 32.9 Removal of the tibial component using stacked osteotomes

Patellar Button
The decision whether to remove or leave the patellar button
is sometimes difficult. Removal is not always necessary as it
can weaken the patella or lead to an excessive thinning (less
than 12 mm) thereof. However, we must take into account
the presence of a wear or loosening, either of which may
impose the need to change the patellar component. Other
factors can influence the decision, including the shape of the
prior patellar implant (domed, “policeman hat”) and its congruence with the future trochlea.
“Better” is sometimes the enemy of “good.”

One should leave the original button in place when the
situation is acceptable and infection absent.

Fig. 32.10 Proprietary extractor for the tibial component

When deciding on removal, one should know that it is difficult to remove metal-backed implants. We seek to use thin,
narrow osteotomes, but fracture of the patella is always a
risk. We can try to create a plane of detachment by sliding a
thin saw blade behind the metal component and complete the
separation using a thin Lambotte osteotome. Removal is
much simpler when the patellar button is all-polyethylene. A
saw is used to remove the implant. The remaining pegs are
removed with a drill. A small drill (2.7 mm) is placed in the
middle of each peg and drilled. The plug comes out sometimes or is just weakened, in which cases it is necessary to
use a very small curette in the channel formed by the drill bit
to remove the remaining polyethylene of the peg.

372

S Lustig et al.

Fig. 32.11 Generic extractor for tibial components (Shukla Medical
NJ)

Once the components are removed, three tissue biopsies
are taken from under the tibial prosthesis, and three from
under the femoral prosthesis for histology and culture, and if
necessary, from under the patella button.

 leaning (Cement, Granuloma)
C
All remaining cement must be carefully removed at this time
along with all soft tissue adherent to areas of the femur to
which the new implant will be applied. One must take special care to clean the posterior aspect of the femoral condyles. Maximally flexing the knee and pulling the femur
forward facilitates this cleaning. A good assessment of bone
loss can only be performed after this cleaning is completed.
Principles of Reconstruction
Once implants are removed, the key concerns are addressing
bony defects and ensuring correct alignment of the new implants.
Two approaches exist:

Fig. 32.12 Tibial cutting guide fixation. The extramedullary guide has
been used for coronal alignment and subsequently removed

satisfactory tibial plateau is carried out first using a specific tibial guide (Fig. 32.12). An intramedullary guide
sets the slope at 0° and an extramedullary guide sets the
• the use of intramedullary rods (cones, sleeves and aug- mechanical axis at 90° in the frontal plane. We fix the cutments) to establish the mechanical axes and fix the cutting ting guide with pins after making pre-drilled holes. A
minimal resection (1 mm may be sufficient) provides a
guides. The components are then placed on these rods.
• reconstruction of the joint surfaces first (with intra- or regular bone surface (Fig. 32.13). In case of significant
extra-medullary guides) and the use then of intra-­ medial or lateral segmental defect, the tibial cutting guide
allows for lateral or medial cuts of 5 or 10 mm from the
medullary rods to improve fixation.
initial cut. After removal of the cutting guide, the size of
the plateau is evaluated using trial implants of increasing
Reconstruction of the Tibia
size. Metal augment trials of 5 or 10 mm are placed on the
trial tray as necessary. The thickness of the wedges and
We first reconstruct the tibial side (concept of the tibial polyethylene are estimated, based on the previously made
platform introduced by B. Mandhuit). A cut to ensure a tibial drill hole.

32 Revision Total Knee Arthroplasty: Planning and Technical Considerations

373

• the medial-lateral size of the femur
• the anterior-posterior dimension of the femur
The femoral mechanical axis is measured by placing an
intramedullary guide set with 7° of valgus. The position of
the trial femoral component is then adjusted to ensure that it
contacts the medial and lateral aspects of the 7° intramedullary guide (it is also therefore necessary that this femoral
trial has an empty intercondylar notch that allows the passage of the intramedullary rod) (Fig. 32.16).
If support is not obtained on the two femoral condyles, a
distal wedge is added (with or without appropriate resection)
until there is both good support for the trial and good contact
between the trial and femoral cutting guide. This assessment
of femoral valgus does not commit to a particular level of the
femoral component in relation to the joint line.

Flexion Gap Management
Fig. 32.13 Minimal resection to obtain a regular bone surface

The second step is to balance the flexion gap. It is not done
through ligament release, but by filling gaps with the implant.
This is what T. Ait Si Selmi called “bone balancing.” To do

Fig. 32.14 Femoral trial positioning

Reconstruction of the Femur
A femoral component trial (specific to the revision of prosthesis) is then placed (Fig. 32.14). Its size is determined by
integrating:
• the size of the previous implant
• the templates placed on radiographs of the contralateral
knee or index knee before the first prosthetic implantation
• the size of the tibial component (to avoid a mismatch of
femur and tibia) (Fig. 32.15)

Fig. 32.15 Tibial trial positioning

374

S Lustig et al.

a

b

Fig. 32.16 The position of the trial femoral component is adjusted to
make contact with the 7° intramedullary guide. This sets the femoral
mechanical axis

this, we must lower the prosthetic posterior condyles onto
the tibial plateau. It is uncommon to have to recut the posterior condyles except to change the rotation or if the posterior
femoral offset was too large and contributed to the knee for
revision. This anteroposterior dimension will guide the final
choice of the size of the implant. However, in case of mismatch in the medio-lateral plane, we can discuss the compromise of retaining a femoral component of a smaller size and
a somewhat thicker polyethylene.
Nevertheless, this compromise is limited by changing the
height of the new joint line and by the risk of patella infera.
Rotation is controlled by referencing from the epicondyles, which is often difficult in revision surgery. Therefore,
preoperative measurement of rotational positioning of the
femoral component is useful (Fig. 32.17). One can correct
internal rotation diagnosed preoperatively using a CT or
MRI scan by adding a posterolateral metal augment (or by a
posteromedial bone resection). Some instrumentation has an
angel wing to mark the rotation of the previous femoral component before removing it (Fig. 32.18).
We then check the balance in flexion (Fig. 32.19). A symmetric gap of a few millimeters in flexion is acceptable. In
case of asymmetric laxity, we can use a posterior wedge on
the femoral condyle (be careful of the rotational changes
generated) or tighten peripheral structures, but a more con-

Fig. 32.17 Pre-operative measurement of rotational positioning of the
femoral component. (a) Using CT scan. (b) Using MRI. In both cases,
the prosthesis is parallel to the surgical trans-epicondylar axis

strained implant may be necessary. In case of symmetric laxity, our preferred option is to add a posterior medial and
lateral shim rather than using a thicker polyethylene, but
again, one may have to use a more constrained prosthetic
system in such cases.

Extension Gap Management
The third step is to balance the extension gap.
Again, this is not done through ligament release, but by
filling gaps with the implant. Rarely, it is necessary to recut
the distal condyles except to correct a flexion contracture.
To assess the need for distalizing the distal femoral
component, one uses pre-operative factors (presence of a

32 Revision Total Knee Arthroplasty: Planning and Technical Considerations

Fig. 32.18 Instrumentation can include a slot for an angel wing to help
mark the rotation of the previous femoral component

375

metrical from asymmetrical frontal laxity and detect any
hyperextension. Measuring the level of the new joint line
relative to prior marks, and relative to the meniscal remnant are key to knowing accurately the changes in the level
of the joint line. A change in the joint line of more than
5 mm is associated with an inferior functional outcome
and one must consider a more constrained prosthesis,
allowing one to maintain the femorotibial joint line near to
the original.
In case of significant symmetric laxity in extension, the
femoral implant is distalized using two distal augments.
Using a thicker polyethylene is the other option, depending
on the flexion space. We set to 8 mm the acceptable limit of
lengthening the femur, which is rarely required. We prefer to
increase the tibial component space and the femoral component space in a 2:1 ratio.
In case of asymmetric laxity, one can consider balancing
the laxity by releasing the less elongated structures, more
rarely to tighten elongated structures. We can use the same
medial or lateral releases used in primary TKA (“Pie-­
crusting” or release the tibial insertion of the distal medial
collateral ligament in case of varus and “pie-crusting” the IT
band or osteotomy of the lateral condyle in valgus cases).

 daptation of the Choice of the Implant
A
During the Intervention
Long Keels
The use of wedges due to insufficient bone support necessitates the use of long keels to stabilize the implants. After the
passage of rigid reamers of increasing size, we check the stability of the implants with the keel chosen. Cones and sleeves
are very useful to fill the metaphysis and to obtain the primary stability required before cementing.
Fixation (Fig. 32.20a, b)
We always use antibiotic cement (Palacos Genta®) fixation in
revision TKA. Cementation is most often done in a single
stage for all of the components. The knee is kept in full
extension until the cement is cured. We choose to release the
tourniquet after cement hardening and carry out a selective
hemostasis with electrocoagulation. Currently, tranexamic
acid is used intravenously or topically.

Fig. 32.19 Ligament balancing in flexion

recurvatum or flexion contracture, templates, stress views),
and intra-operative factors. The intra-operative assessment
of laxity is difficult but it is essential to differentiate sym-

Constraint
As we have noted, during the operation the surgeon may be
faced with an inability to balance the gaps, avoid a mismatch
between gaps (flexion-extension), and avoid a significant
change in joint line. In such a situation it is necessary to use
a more constrained prosthesis, usually a rotatory hinge prosthesis (Fig. 32.21).

376

S Lustig et al.

Special Case of Sepsis
We perform most often a two stage revision, with a free
interval of 6 weeks, during which an articulating spacer is
placed in situ. The decision is made in collaboration with the
infectious disease department.
Multiple deep samples (joint fluid, synovium from different areas around the implants, cement) are obtained during
the removal of the implant and are analyzed in the bacteriology and pathology department.
In cases where there is a sinus present, care is taken to
excise it during explantation. The exposure is sometimes difficult because of tissue inflammation, but we try not to perform a tibial tubercle osteotomy during explantation. If it is
needed, we fix the tubercle temporarily at the end of the
explantation stage with absorbable sutures through the screw
holes and corresponding 3.2 drill holes on the tibial shaft
(Fig. 32.22).
Fig. 32.20 Cemented implants with antibiotic cement

Fig. 32.21 Hinged prosthesis

Closure
The closure is done in three layers with the knee flexed to
90°. The deepest layer is closed first with figure of eight
sutures using number 2 Vicryl®. We choose to use a drain
placed in the joint. The subcutaneous layer is closed with 0
Vicryl®, and the skin with staples.

Fig. 32.22 Case of sepsis: articulated spacer of appropriate size. Note
drill holes for stabilization of the tibial tuberosity fragment

32 Revision Total Knee Arthroplasty: Planning and Technical Considerations

It is essential to maintain the joint space during the free
interval between the two stages, including the back of the
condyles and between the patella and femur. We use molds to
obtain articulated spacers of appropriate size. The cement
used contains antibiotics (Palacos Genta®) with extra added
antibiotic. Weight bearing is not generally permitted but
mobilization is possible, depending on the stability of the
implants during surgery.
For the second stage (re-implantation), it is very useful to
check the joint line with the drill hole method described
above.

 ostoperative Course After Re-implantation
P
Weight bearing is authorized from day one. Flexion is limited to 95° for 45 days and then without limitation. An extension splint is maintained for walking until the quadriceps has
recovered. If an elevation of the tibial tubercle has been performed, the splint is retained between rehabilitation sessions
for 2 months.

Special Cases
Proximalization of the Tibial Tubercle
If there is a significant limitation of the flexion preoperatively, it is sometimes necessary to set the tibial tubercle
a

b

377

more proximal (Fig. 32.23a–c). In this case, one can use two
screws and reinforce the attachment with a metal wire, which
is passed through a hole in the distal part of the bone and
fixed with a screw distally. This additional distal fixation is
intended to prevent proximal migration of the tubercle
(Fig. 32.24).
Sometimes, in the case of a massive keel, we must fix the
ATT with metal wires only (Fig. 32.25). In this case, we must
avoid going behind the tibia because of significant ­vascular
risk, but pass the wire just behind the tibial keel itself before
its introduction.
In all cases, one should use a splint in extension for walking and resting between rehabilitation sessions during the
period of consolidation of the osteotomy (2 months).

 sing a Customized “Tulip-Shaped” Implant
U
In case of significant metaphyseal defect, but with preservation of the cortical envelope, we use a tulip-shaped tibial
implant based on a long keel. The interest is to preserve the
insertion of the collateral ligaments even in cases of major
bony defects. It then becomes possible to use a posterior
stabilized prosthesis with minimal constraint. Today it is
rarely necessary to incur the delay required for customized
implants and rather utilize the latest generation of revision
components, including offset stems, cones, sleeves, and
augments.
c

Fig. 32.23 Proximal transfer of the tibial tubercle and fixation by two 4.5 mm screws. (a, b) Intra-operative views before and after fixation. (c)
Post-operative x-ray

378

S Lustig et al.

Fig. 32.25 Metal wires fix the tuberosity fragment, and have been
passed behind the tibial keel before its introduction

Fig. 32.24 Additional distal fixation of the tibial tubercle to prevent
proximal migration

Part III
Surgery for Patellar Conditions

Surgical Management of Episodic
Patellar Dislocation

33

E Servien, P Archbold, P Neyret, and C Butcher

Background

Morphological Abnormalities

Episodic patellar dislocation is also termed objective patellar
instability or occasional patellar dislocation.
This condition is defined by a history of one or more episodes of patellar dislocation confirmed by the patient or the
physician and/or a radiographic abnormality due to a dislocation (e.g. fracture/bone bruise of the medial border of the
patella or lateral condyle).
This terminology avoids the term of instability. H. Dejour
highlighted the different meaning of instability and separated objective “laxity” from subjective “instability.”
Nevertheless, instability is a symptom and not a syndrome.
Moreover, a misunderstanding still exists with the terminology of objective patellar instability in the English-speaking
world. With Dan Fithian, after his visit to Lyon for 4 months,
we came to this suggestion that clarifies the situation:
Episodic Patellar Dislocation (EPD). Our global approach
concerning the predisposing factors was modified during
the 15th Journées Lyonnaises du Genou, coordinated by
David Dejour in 2012.

In the EPD group, we have identified several morphological
anomalies that facilitate or allow patellar dislocation. In
more than 96% of the cases of EPD, the radiographic examination will reveal at least one of the following anomalies:
• Medio Patellar Femoral Ligament (MPFL) insufficiency
• Trochlear dysplasia
• Patella alta
• Tibial Tubercle-Trochlea Groove (TT-TG) distance >20 mm
• Patellar tilt >20° (a consequence of the other morphological abnormalities)

Constant Factor
MPFL Insufficiency
This may be due to an acute rupture from a direct traumatic
event, and/or chronic incompetence secondary to other morphological factors.

Fundamental Factor

E Servien
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
P Archbold
Centre Albert Trillat, Lyon, France

Trochlear Dysplasia
Trochlear dysplasia, according to the literature and our experience, is present in more than 90% of patients in whom a
patellar dislocation has occurred. It is the principal anatomical feature of EPD and consists in a flattening or convexity of
the upper part of the trochlear groove.
Imaging features: crossing sign and a prominence
(“bump,” “boss,” or “eminence”) of the floor of the
groove on the lateral radiograph.

P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

Main Factors

C Butcher
Healthpoint, Abu Dhabi, UAE

These factors are called main factors for several reasons.
They are very often present in the EPD group and absent in a

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_33

381

382

control group (patient without any history of patellar dislocation). We were able to measure them radiographically and a
threshold has been defined, so that they can be corrected.

Tibial Tubercle-Trochlea Groove
(TT-TG) Distance
The TT-TG distance is used to assess rotational alignment of
the extensor mechanism. It is obtained by superimposing CT
images of the summit of the trochlear groove (coronal cut
where the femoral notch resembles a roman arch) and the
tibial tubercle in a fully extended knee. The deepest point of
the trochlear groove and the highest point of the tibial tuberosity are projected perpendicularly on the line tangent to the
posterior condyles. The distance between these points is
defined as the TT-TG.
Abnormal when the TT-TG >20 mm in CT coronal
images.
Patellar Height
The patella engages in the trochlea in the first degrees of
flexion. If the patella is too high in relation to the trochlea,
this engagement will occur too late, with an increased risk
for dislocation.
Abnormal when the Caton-Deschamps index >1.2 on the
lateral radiograph.
Patellar Tilt
The patellar tilt is the inclination of the patella in its transverse plane in relation to a line tangent to the posterior femoral condyles. Several factors may result in patellar tilt:
dysplasia of the quadriceps muscle, dysplasia of the trochlear groove, and patella alta. It can be addressed by a soft
tissue reconstruction, e.g., medial patellofemoral ligament
reconstruction or vastus medialis obliquus plasty.
Abnormal when the patellar tilt >20° on CT images.

Secondary Factors
We call them secondary factors because they are present in
the EPD group at a lower frequency and we were not able to
establish a threshold. They are more common in females. We
must consider them as potential factors and it is uncommon
to propose a surgical act to correct them.
–– Genu valgum
–– Genu recurvatum
–– Excessive femoral ante-torsion

Clinical Examination
The clinical findings are less reliable in the evaluation of EPD.

E Servien et al.

Smillie Test (The Apprehension Sign)
With the patient in the supine position and the knee extended,
the patella is forced laterally by the examining physician. To
be positive, the patient and the physician must have the
impression of imminent dislocation. A negative Smillie is
much more helpful than a positive sign; a negative Smillie
sign rules out a dislocatable patella while a positive Smillie
sign does not confirm a dislocatable patella.

“J” or “Comma” Sign
Lateral subluxation of the patella in terminal active knee
extension due to the non-linear path of the patella during the
first 30° of flexion.

Lateral “Squint” of the Patella
The so-called “grasshopper” sign, due to the appearance of
the high-riding and laterally subluxated patella at the upper
outer corner of the knee at 90° of flexion.

Increased Q-angle
It is also known as “bayonet sign.” The distal insertion of the
patellar tendon is too lateral with respect to the patella itself
and the quadriceps muscle. It is a clinical finding that is difficult to quantify. It is an indication of a potential excessive
TT-TG. To quantify the exact position of the tibial tubercle
with respect to the trochlear groove, a TT-TG measurement
should be performed.
Other aspects of the physical examination, such as effusion and tenderness, recurvatum and lower limb alignment,
are secondary or indirect signs and do not contribute strongly
to treatment decisions.

Imaging Studies
Trochlear Dysplasia
Crossing Sign
On the strict lateral radiograph (the posterior portions of the
femoral condyles are aligned), the floor of the normal groove
is visible in profile as a distinct sclerotic line curving distally
and posteriorly, starting from the anterior cortex and ending
at the anterior end of Blumensaat’s line. In its entire course,
this line should never pass anterior to a tangent line extending down the anterior femoral cortex.

33

Surgical Management of Episodic Patellar Dislocation

383

Patients with trochlear dysplasia have an abnormally
prominent groove which passes anteriorly to the anterior cortex and eventually crosses the medial and lateral trochlea
walls (Fig. 33.1a, b). The more distal the crossing, the worse
is the trochlea dysplasia.

a

Prominence
Also called the trochlea “boss,” “bump,” or “eminence” of
the floor of the groove with respect to the distal 10 cm of the
anterior femoral cortex. Values superior to 3 mm are considered pathological (Figs. 33.2 and 33.3).

b

Fig. 33.1 Lateral view. (a) Normal knee. (b) Trochlea dysplasia with crossing sign (arrow)

a

b

X

c

X

Fig. 33.2 Prominence. (a) No prominence. (b) Positive prominence. (c) Negative prominence

X

384

E Servien et al.

15°
B

A

Y

Fig. 33.4 Trochlear depth (=BA)

Fig. 33.3 Radiological measurement of the prominence on lateral
view X-ray

Trochlear Depth
A tangent to the posterior cortex of the femur and its perpendicular passing through the posterior condyles are traced.
The trochlea depth is measured in a line 15° from the perpendicular, crossing the trochlear groove line, where the trochlear depth is measured. Values below 4 mm are considered
abnormal (Fig. 33.4).

Tibial Tubercle-Trochlea Groove (TT-TG) Distance
The TT-TG distance is used to assess rotational alignment of
the extensor mechanism. It is obtained by superimposing CT
images of the summit of the trochlear groove (coronal cut
where the femoral notch resembles a roman arch) and the
tibial tubercle in a fully extended knee. The deepest point of
the trochlear groove and the highest point of the tibial tuberosity are projected perpendicularly on the line tangent to the
posterior condyles. The distance between these points is
defined as the TT-TG.
Measures greater than 20 mm are considered abnormal
(Fig. 33.5). The TT-TG is the result of both the lateral position of tibial tubercle and the external torsion of the knee. In
some situations, this measurement is therefore not accurate,
for instance, in permanent dislocation of the patella where
the external rotation of the knee is increased.

Fig. 33.5 TT-TG distance measurement on CT scan = 27 mm

Patella Alta
There are several indices to measure patellar height. These
indices can be categorized into those referencing the tibia
(for example, Insall-Salvati, Caton-Deschamps, Blackburne-­
Peel) and those referencing the femur (for example,
Blumensaat, Bernageau).
Theoretically, referencing the patellar height to the femur
is more logical because what matters is how the patella
engages in the femoral groove. But femoral referencing is

33

Surgical Management of Episodic Patellar Dislocation

385
3

2
P

1

A

T

1

Fig. 33.6 Caton-Deschamps index measurement = AT/AP

less reproducible. For this reason, tibial referencing is the
standard method to measure patellar height.
Caton-Deschamps Index: It is measured in a strict lateral
view. It is the ratio between the distance of the inferior border of
the patellar articular surface to the anterior border of the tibial
plateau and the patellar articular surface length (Fig. 33.6). It is
simple to trace and it is not altered by knee flexion on the radiograph. Values greater than 1.2 characterize patella alta.

Patellar Tilt (Maldague and Malghem)
It cannot be well studied from strict lateral radiographies with
the knee flexed to 30° (Fig. 33.7). It is measured on CT in extension, with the quadriceps contracted and relaxed. It is defined by
the angle between the tangent of the posterior condyles at the

2

3

Fig. 33.7 Patellar tilt (Maldague and Malghem). (1) normal, (2) moderate tilt, (3) severe tilt

level of the roman arch, and the long transverse axis of the
patella. Sometimes it is necessary to superimpose two slices in
order to measure this angle, particularly in case of patella alta.
Merchant angle: It is calculated on axial radiographic
views of the knee at 45° of flexion. The bisecting line of the
angle between the lateral and medial trochlea facets is traced.
A second line is then traced between the deepest portion of
trochlea groove and the most posterior (inferior) portion of
the patellar ridge. The angle of Merchant is the angle between
both lines. If the angle is medial to the bisecting line, it has a
negative value; if it is lateral, its value is positive. Normally
the angle is −6°. Merchant considered an angle superior than
+16° abnormal (Fig. 33.8). The Merchant angle is not significantly altered, however, when the dysplasia of the trochlea is limited to its proximal part.

386

E Servien et al.

Angle de
MERCHANT

Fig. 33.8 Merchant angle

Patellar Tendon Length
Measured on MRI, it is more specific and sensitive for the
study of patellar instability than the patellar height index
measured on profile X-rays (Fig. 33.9). The patellar tendon
is excessively long (generally greater than 52 mm) in patients
with patellar instability.

Fig. 33.9 Patellar tendon length after tenodesis measuring 49 mm on
MRI

Treatment
An algorithm helps to plan treatment in a logical way:
Episodic Patellofemoral pain with knee instability

PF Laxity at 0º,30º;
Strict Lateral X-ray
No PF Laxity,No Dysplasia,
Normal radiographs

Laxity, crossing sign,
prominence >4 mm

CT Scan
TT-GT

≥20 mm

Medialize
TT

No CT Required

Patella Alta

Patella Tilt

Index >1.2

>20°

Distalize TT

Balance
Ligaments
MPFL
reconstruction

MR Arthrogram

cartilage flap tear or
chondrosis

Consider Examination
under anesthesia,
scope

D. Fithian, Ph Neyret:Patellar Instability: The Lyon Experience

MR neg

Conservative Tx,
consider other Dx

33

Surgical Management of Episodic Patellar Dislocation

Conservative Treatment
Non-surgical treatment is not the objective of this chapter
and will not be developed here.
Unlike in painful patella syndrome, where conservative
treatment is the rule and surgical treatment usually worsens
symptoms, conservative treatment has a less important role
in EPD management. This is particularly true when a threshold can be established due to a factor such as patella alta,
augmented TT-TG or patellar tilt and the patient presents
with repetitive episodes of patellar dislocation.
In cases of infrequent instability in which no threshold
abnormality has been identified and pain is the most prominent
symptom, a course of physiotherapy can be prescribed. These
exercises consist of quadriceps and hamstrings muscle stretching and quadriceps reinforcement (especially VMO muscle).

Surgical Treatment
Surgical treatment is indicated in the presence of morphological anomalies. These patients should have had at least one or
more episodes of patellar dislocation AND one or more main
factors (patella alta, excessive TT-TG or patellar tilt).
In this chapter, we do not describe the trochleoplasty as
proposed by H. Dejour and G. Walch in 1987. We do not
treat the trochlear dysplasia in a primary setting, the fundamental factor responsible for EPD, for several reasons. Most
frequently, the trochlear dysplasia is mild and well tolerated
by the patient. Deepening of the trochlea groove is in our
hands is only effective in severe cases with abnormal patellar
tracking. It remains a very technical and demanding procedure with a variable outcome. Therefore, in our department a
deepening trochleoplasty is only indicated for severe trochlear dysplasia (with a bump of >6 mm, abnormal patellar
tracking or failure of previous surgery).

387

lateral displacement of more than 9 mm with the knee at 30°
of flexion. After the procedure, the patella must be in a horizontal position and no longer dislocatable. An arthroscopy is
done first to evaluate associated lesions and patellar tracking,
which can be performed using an accessory superolateral
portal. Three small incisions are needed, one for graft harvesting and two for graft fixation, on the patella and on
medial femoral epicondyle (Fig. 33.10).

Harvest of the Semitendinosus Tendon
The next step is to obtain the graft for the MPFL reconstruction. A small (5 cm) longitudinal or oblique incision over the
pes anserinus is made. The conjoint tendon of the hamstrings
is incised in “L” shape, with its angle positioned superiorly
and medially. The semitendinosus tendon is identified and its
insertions to the crural fascia and posteromedial corner are
cut. Absorbable sutures are placed at its free end, and the
tendon is released from the tibial insertion (Figs. 33.11 and
33.12). The graft is subsequently stripped using a closed
stripper.
The graft is prepared on the back table. For an MPFL
reconstruction, the length of the graft should be between 16
and 20 cm. Whipstitches are placed in the other free end, and
the tendon is looped in two. The folded end is sutured
together over a distance of 2–3 (Fig. 33.13).

Techniques
The following procedures are generally easy, but can lead to
significant complications if not carried out with prudence
and for the correct indications. These techniques are not
­indicated for painful patella syndrome, which can be worsened by these procedures.

 edial Patellofemoral Ligament
M
Reconstruction (D. Fithian’s Technique)
This technique is indicated when there is excessive laxity of
medial retinacular patellar stabilizers, specifically the
MPFL. A laterally applied force to the patella will result in

Fig. 33.10 Three small incisions are needed for MPFL reconstruction

388

E Servien et al.

Figs. 33.11 and 33.12 Harvest of the semitendinosus tendon

Fig. 33.13 Graft preparation

Patellar Tunnels
A longitudinal incision of about 4–5 cm is made over the
patella, in between its medial border and the midline. The
medial third of patella is exposed by subperiosteal dissection
(Fig. 33.14). The dissection extends medially between the
original MPFL and the capsular layer.
A 3.2 and subsequently a 4.5 mm drill are used to drill
two tunnels in the proximal 1/3 of the patella. These tunnels

Fig. 33.14 Exposure of the medial third of patella

start on the medial border of the patella, horizontally, and the
exit holes are made on the anterior surface of the patella,
8–10 mm lateral to medial border (Fig. 33.15).

33

Surgical Management of Episodic Patellar Dislocation

389

Fig. 33.16 Guide pin passed proximally to the medial epicondyle

Fig. 33.15 Patellar tunnels

 he Medial Epicondylar Tunnel and Its Dissection
T
A 5 cm longitudinal incision is made over the ridge connecting the medial epicondyle to the adductor tubercle
(AT). Dissection is carried to the bone surface. From the
patellar incision, in between the original MPFL and capsular layer, dissection scissors are driven toward the
medial epicondyle, uniting both epicondylar and patellar
incisions. A guide pin is passed just proximally to the
medial epicondyle, distal to the AT toward the lateral epicondyle, under fluoroscopic guidance (Figs. 33.16 and
33.17). MPFL ligament isometry can be tested now, passing a #5 braided polyethylene suture around the guide pin
through the patellar tunnels. If lengthening occurs in
extension, the pin is placed more proximally, closer to the
AT. If lengthening occurs in flexion, the pin is placed
more distally, closer to the medial epicondyle. A blind
bone tunnel is created on the medial epicondyle, 7 mm in
diameter and 25–30 mm in length (enough to receive the
folded end of the graft). The graft is pulled into the tunnel
by a perforated pin, and then fixed with an interference
screw (Habilis 7 mm, Phusis) (Figs. 33.18 and 33.19). The
two free extremities of the graft are passed under the original MPFL between the two incisions, to enter the patellar
tunnels medially and exit through the anterior drill holes.
Each free extremity is sutured side-to-side onto itself with
non-­
absorbable sutures (Figs. 33.20 and 33.21). The
patella must be centered at the time of tying the sutures,
so an adequate ligament tension is obtained. It remains

Fig. 33.17 The tunnel is too proximal and anterior, and the pin will be
repositioned. We recommend a fluoroscopic control to optimise femoral tunnel positioning

very difficult to define “adequate” tension. Generally, tensioning is performed at 70° of flexion.
Patellar mobility is checked. A good endpoint must be
achieved, with patella lateral mobility of 7–9 mm, the patella
must be in a horizontal position and it should be impossible
to dislocate it laterally. A 3.2 mm drain is placed subcutaneously, and the incision is closed.

390

Fig. 33.18 Femoral fixation using absorbable screw

E Servien et al.

Fig. 33.19 The two bundles are long enough

Fig. 33.20 The two free extremities of the graft are passed into the
patellar tunnels
Fig. 33.21 Suture side to side

33

Surgical Management of Episodic Patellar Dislocation

 edial Patellofemoral Ligament
M
Reconstruction Using Quadriceps Tendon
Autograft
This has become our preferred technique since 2013. The
main advantage is the absence of risk of patellar fracture.
The femoral tunnel, tensioning, and femoral fixation are the
same as the previous technique.
The quadriceps tendon graft is taken through an open
incision, or in a minimally invasive manner using a proprietary harvester (Karl Storz Minimally Invasive Quadriceps
a

391

Tendon Harvesting System) (Fig. 33.22a, b). The quadriceps
graft is 9 cm long, 6 mm wide, and partial thickness of the
tendon (around 3–4 mm thick). The free proximal end is
tubularized with No. 2 Ethibond for 3 cm. The distal end is
dissected no more than 5 mm subperiosteally onto the patella,
to preserve the strong quadriceps insertion. The periosteal
sleeve of the medial 1/3 of the proximal patella is raised
(Fig. 33.23). The graft is folded 90° on itself to point medially, and passed deep to the periosteum and the native MPFL
toward the femoral tunnel site, where it will be tensioned and
fixed as per the previous technique (Fig. 33.24). At the
b

Fig. 33.22 (a) Quadriceps tendon harvester (Karl Storz Minimally Invasive Quadriceps Tendon Harvesting System). (1) Double blade to define
width. (2) Transverse blade to define thickness. (3) Handle. (4) Cutter. (b) Vertical incision for quadriceps harvest

Fig. 33.23 Tubularized quadriceps tendon graft. The periosteal sleeve
of the medial 1/3 of the proximal patella has been raised to allow passage of the graft

Fig. 33.24 The graft is routed deep to the periosteum and native MPFL
to reach the femoral tunnel site

392

E Servien et al.

Fig. 33.25 At the patella, the graft is secured to itself and to the patella
periosteum with absorbable suture

patella, it is secured to itself and to the patella periosteum
with absorbable suture (Fig. 33.25).

Fig. 33.26 Skin incision

Distal Tibial Tubercle Transfer (TTT)
This technique is indicated to correct patella alta. The patient is
prepared and a tourniquet is applied high on the proximal thigh.
An arthroscopy can be performed in combination in order to
verify patellar tracking and look for possible chondral lesions.
The objective of the procedure is to bring the tibial tubercle (TT) to a more distal position in order to obtain a Caton-­
Deschamps index of 1. For example, in a patient with a
Caton-Deschamps index of 1.3, with AT distance of 39 mm
and an AP distance of 30 mm, the distalization necessary is
9 mm to reach an index of 1. Two extra millimeters should be
added due to possible proximal movement of TT during
screw fixation, resulting in a total of 11 mm of distalization.
An 8 cm medial parapatellar skin incision is made, centered on the TT. The subcutaneous tissues are dissected
(Fig. 33.26). The TT osteotomy has a length of 6 cm and is
marked using the electrocautery. The patellar tendon and the
inferior pole of the patella are identified.
Two 4.5 mm holes are drilled in the midline of the TT. A
countersink is then used in each hole in order to avoid prominence of the screw heads underneath the skin (Fig. 33.27).
The osteotomy is done with an oscillating saw and completed with an osteotome. The lateral cut is done first, in a
horizontal direction, followed by the medial cut, in an
almost vertical direction, and finally the transverse distal Fig. 33.27 Tibial tubercle osteotomy: 6 cm length. Two 4.5 mm holes
cut. Distal to this transverse cut, an additional bone block is are drilled

33

Surgical Management of Episodic Patellar Dislocation

removed of which the length corresponds to the amount of
distalization. This allows the TT to be positioned distally
(Figs. 33.28 and 33.29). The free TT is now transferred to its
more distal position as planned and kept in position with a

393

Farabeuf retractor. With the knee in 90° of flexion and the
calf of the lower limb free, two 3.2 mm holes are made in
the posterior tibial cortex through the TT 4.5 mm drill holes,
perpendicular to the tibial shaft. The osteotomy is then fixed
with two 4.5 mm cortical screws. It’s imperative that the
screws are fixed in a strict perpendicular position in relation
to the tibial shaft. The screw length should be 2 mm longer
than the measured length of the drill trajectory to ensure
adequate fixation and to avoid postoperative detachment of
the TT (Fig. 33.30). The screws should not be tightened
excessively, otherwise the TT might be positioned too posteriorly. Care must be taken to keep the TT parallel to its
original bed, otherwise a lateral patellar tilt might occur
(Fig. 33.31).
The incision is closed over a drain.

Fig. 33.28 Lateral cut in a horizontal direction

Med.

Lat.

Fig. 33.30 Postoperative X-ray

Fig. 33.29 Lateral cut in a horizontal direction and medial cut in a
vertical direction

394

a

E Servien et al.

b

Fig. 33.31 Example of medial and distal tibial tubercle transfer with correct fixation (a) and incorrect fixation (b) with a lateral patellar tilt

Patellar Tendon Tenodesis
This is an adjuvant procedure to a distal TTT procedure. It is
indicated in cases where the patellar tendon length is greater
than 52 mm (Fig. 33.32).
After the TT osteotomy but prior to its fixation, two
anchors with sutures are fixed on both sides of the patellar
tendon, about 3 cm distal to tibial plateau level (the normal
insertion level of the tendon) (Fig. 33.33). The TT is subse-

quently fixed at the desired position with two 4.5 mm cortical screws (Fig. 33.34).
After fixation of the osteotomy, the tendon is vertically
incised at 1/3 and 2/3 of its width with a 23 scalpel blade.
The sutures are tied across the lateral and medial 1/3
(Fig. 33.35). Thus the length of the patellar tendon is reduced.
This can be assessed postoperatively with MRI (Fig. 33.36).
The remaining steps of the surgery are the same as for a
distal TTT.

Surgical Management of Episodic Patellar Dislocation

395

6-8 cm

33

Fig. 33.32 Principles of patellar tendon tenodesis associated with distal tibial tubercle transfer

Fig. 33.33 Anchors fixation on the normal insertion level of the
tendon

Fig. 33.34 Tibial tubercle fixation

396

E Servien et al.

Medial Tibial Tubercle Transfer
This technique is indicated in the correction of an
increased TT-TG. A 6 cm medial parapatellar skin incision is made, centered on the TT (Fig. 33.37). The subcutaneous tissues are dissected. Comparable to a distal TTT,
the TT bone block has a length of 6 cm. The patellar tendon, the lateral and medial attachments, the inferior pole
of the patella, and the TT are identified. The objective is
to bring the TT to a more medial position in order to
obtain a TT-TG value of 12 mm. For example, in a patient
with a TT-TG value of 20 mm, it is necessary to medialize
the TT by 8 mm. One 4.5 mm hole is drilled in the center
of the TT. The osteotomy is made with an oscillating saw.
The cut begins laterally and exits out of the cortex medially. Leaving an intact portion of inferior TT attached to
the anterior tibial cortex. Subsequently, the TT is brought
to a more medial position as planned preoperatively
(Fig. 33.38). Different to the distal TTT, only one screw is
necessary for the TT fixation (Fig. 33.39). A 3.2 mm drill
hole is made in the posterior cortex through the TT 4.5 mm
hole, in a slightly more proximal position. The osteotomy
is then fixed with a 4.5 mm cortical screw, 2 mm longer
than the measured drill trajectory. The incision is closed
over a drain.

Fig. 33.35 Final aspect after suture

a

b

Fig. 33.36 (a, b) Pre- and postoperative MRI showing length modification of the patellar tendon

33

Surgical Management of Episodic Patellar Dislocation

397

Fig. 33.37 Skin incision

Fig. 33.39 Postoperative X-ray

Postoperative Care
Prophylactic antibiotics are administered for 24 hours.
LMWH thromboprophylaxis is continued for 10 days. Ice is
generally applied for 5 days. A 30° splint is used at night and
between periods of walking and physiotherapy. Protected
full weight bearing is allowed immediately using crutches.
In the case of a TTT osteotomy, a locked splint in extension must be used for walking until radiographic evidence of
consolidation is found. Daily physiotherapy consists of
active isometric quadriceps contractions with good patellar
ascension and medial-lateral patella mobilization. Passive
flexion is initiated early, but must be limited at 95°.

Fig. 33.38 Medial transfer, distance measured using a ruler

398

In case of MPFL reconstruction, protected (crutches)
full weight bearing is allowed on the first post-surgical day.
Knee flexion is unlimited.
After 45 days or when bone consolidation is obtained, the
patient returns to normal walking, avoiding stairs. Full flexion must be recovered. After 60 days, normal activities of
daily life and driving are started. Open kinetic chain excercises are initiated. Patient can commence sports activities
after 4 months. Forced kneeling and jumping are not
allowed for 6 months. In the case of a combination of procedures, rehabilitation is limited by the most demanding
procedure.

Complications
The most frequent complication, inherent to all kinds of EPD
surgery, is hematoma. It can cause intense pain and can lead
to wound dehiscence and infection. This problem can be
avoided with careful coagulation and the use of a vacuum
drain. As in any surgical procedure, infections can occur in
EPD procedures. Complex regional pain syndromes may
arise after surgery and may result in a patella infera
(Fig. 33.40). A prominent screw head usually causes discom-

Fig. 33.40 Patella infera

E Servien et al.

fort or pain. Countersinking of the screw head is sufficient to
prevent this problem.
Failure to obtain sufficient TT osteotomy fixation can
result in migration, delayed union, or non-union. If this is the
case, a revision operation must be performed. It is of major
importance to always use a screw 2 mm longer than the measured drill trajectory in order to provide adequate fixation
(Fig. 33.41). Fractures of the tibial shaft can occur when the
ATT osteotomy cuts are too aggressive, especially the distal
cut, even several weeks after surgery (Fig. 33.42).
Undercorrection can result in persistent instability and
dislocation. Insufficient distalization or medialization, and
suboptimal tensioning of the MPFL reconstruction or of the
VMO plasty can result in this situation.
Overcorrection can be even worse. Patients usually present with pain and with signs of medial patellar impingement.
Patella baja is further complication of overcorrection, leading to increased patellar pressures and pain. These complications frequently cause more disability than the instability
itself (Fig. 33.43a, b). Procedures that include an ATT osteotomy can lead to non-union. This risk can be minimized
with a TT fragment of at least 6 cm.
The MPFL reconstruction can cause an avulsion fracture
of the medial border of the patella.

33

Surgical Management of Episodic Patellar Dislocation

399

Fig. 33.42 Tibial shaft fracture

Fig. 33.41 Tibial tubercle non-union

a

b

Fig. 33.43 (a) Postoperative TT-TG distance = 10.4 mm; (b) on contralateral side, overcorrection with a TT-TG distance = −9 mm

Deepening Femoral Trochleoplasty

34

E Servien, P Archbold, P Neyret, and C Butcher

Introduction
With the exception of the trochleoplasty, the management of
episodic patellar dislocation (EPD) has been described in the
previous chapter. Although trochlear dysplasia is the primary
problem in EPD, addressing the principal factors of a high
tibial tuberosity to trochlear groove distance (TT-TG) and
patella alta is usually sufficient to obtain stability.
Trochleoplasty is a technically demanding procedure that is
rarely required in our daily practice (once or twice per year),
despite the encouraging recent results in the literature. Its
indications include patients with habitual dislocation of the
patella (Fig. 34.1), abnormal patellar tracking, or in revision
surgery. It can correct grade 4 dysplasia with a femoral
groove prominence of >6 mm. It is always combined with
another surgical procedure to correct the other contributing
factors of the instability.

 he Principals of the Deepening
T
Trochleoplasty

trochleoplasty (Fig. 34.2). This improves engagement of the
patella in the trochlea groove in the early degrees of flexion.

Incision
The trochleoplasty is performed using an anteromedial
approach. Following an anteromedial arthrotomy, the patella
is everted.

Planning
The key to performing a successful trochleoplasty is precise
intra-operative planning. Elevation of the synovium on the anterior cortex of the distal femur is performed to expose the proximal edge of the trochlear groove. A marker pen is then used to
define the center of the new groove, extending from the proximal
edge of the trochlea to the center of the notch. The medial and
lateral facets are then marked (Fig. 34.3).

H. Dejour and G. Walch believed that the primary problem in
trochlear dysplasia is a prominent trochlea floor that causes
it to be flat. The technique therefore consists of a deepening

E Servien
Service d orthopedie de l Hopital de la Croix Rousse, Lyon 69004,
France
P Archbold
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Fig. 34.1 Habitual dislocation of the patella

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_34

401

402

E Servien et al.

Fig. 34.2 The deepening
trochleoplasty

a

b

Fig. 34.3 Intra-operative planning of the trochleoplasty

Deepening
Once the walls of the trochlea have been exposed, periosteal
stripping is performed around the periphery. A 10 mm osteotome is then used to remove cortical bone around the periphery of the trochlea; from the proximal edge and lateral/
medial walls (Fig. 34.3). This forms a 3–4 mm trench which
exposes the underlying cancellous bone.
A power burr (Fig. 34.4a, b) equipped with an adjustable
stylus is now used to complete the deepening trochleoplasty.
It is used to determine the appropriate resection depth, avoid
penetration of the cartilage, and avoid damage to the cartilage due to heat generation. A minimum residual thickness
of about 4 mm consisting of cartilage and a thin layer of subchondral bone is recommended to achieve a bone surface
that will be easy to fashion. The remaining cancellous bone
is removed using a small curette. The cancellous bone bed

Fig. 34.4 (a, b) The power burr

should extend as far as the roof of the femoral notch. Once
proper trochlear depth has been achieved, attention is
directed to the preparation of the medial and lateral facets.

Trochleoplasty
A cut is made in the middle of the trochlear groove using a scalpel (Fig. 34.5). The line of the cartilage incision has been previously marked with a sterile pen, and the cartilage weakened
with a 2 mm diameter drill along the marked line. This allows
bone shell impaction into the new sulcus. The new facets are
then fixed with two metal staples at its upper end (Fig. 34.6).
Patellar tracking is checked again. Fixation continues to be an
area for improvement. We sometimes also use 3 mm AO screws,
and remove them arthroscopically at 6 months post op.

34 Deepening Femoral Trochleoplasty

403

a

b

Fig. 34.5 A cut is made in the middle of the trochlear groove using a
scalpel

Fig. 34.6 Fixation of the trochleoplasty

Closure
The synovial tissue is then sutured back onto the edges of the
trochlea. The staples can be easily removed (usually at least
3–6 months later) under arthroscopy.

Post-operative Care
The rehabilitation protocol is dictated by whether the trochleoplasty was combined with a distal transfer or medialization of the tibial tubercle (TT). In the absence of surgery to
the TT, immediate weight bearing with no restriction of
movement is allowed. In the presence of a distalization or

Fig. 34.7 (a, b) Post-operative X-ray

medialization of the TT, flexion is limited to 95° for 45 days.
Flexion beyond 95° is allowed once consolidation has been
achieved (Fig. 34.7a, b).

404

E Servien et al.

Reflections on the Deepening Trochleoplasty
and Future Directions

Alternative Procedures
An alternative approach to dealing with the femoral groove
prominence typical of severe dysplasia is the “Recession
Trochleoplasty,” described by D. Goutallier and popularized
by Beaufils. A proximally based wedge of bone is removed
from the distal femur, allowing the prominence to be
“recessed” to the level of the anterior femoral cortex. The
hinge point is the distal trochlea. It is fixed with two screws
just lateral to the chondral surfaces (Fig. 34.8a, b).
Although the shape of the trochlear groove is not improved
in the axial plane, accordingly the patellar femoral congruence is not changed, and there is less intra-operative risk of
injury to the chondral surfaces.

The greatest concern in the development and use of this technique over the past 20 years has been the risk of necrosis to
the cartilage of the trochlea. Over this period, the technique
has changed little due to the lack of industrial support and the
little time invested by surgeons to improve it. This is unfortunate as this technique addresses the principal abnormality
found in EPD. Its reproducibility and accuracy could be optimized by computer-assisted surgical techniques, and this
would be an ideal indication for robotically assisted surgery.
The intended location and dimensions of the new trochlea
could be planned with accuracy, and the intra-operative
stages performed to exactly match the plan, minimizing risk
to the chondral surfaces. However, at present due to the technical demands of the conventionally performed procedure,
there remains a reluctance to perform this surgery.

a

M

b

M

P

P

Fig. 34.8 (a, b) A closing wedge osteotomy of the distal femur in the coronal plane allows the trochlear prominence to be recessed to the level of
the anterior femoral cortex

Patellar Tendon Shortening

35

E Servien, P Archbold, and P Neyret

In patella alta associated with an excessively long patellar tendon, and without an abnormal tibial tuberosity to trochlear
groove distance (TT-TG), it may be more logical to shorten the
patellar tendon than to distalize the tibial tubercle. We have
therefore developed a technique to correct this anomaly. It can
be also indicated in patients who are skeletally immature in
whom a transfer would be contraindicated. However, it must be
used with caution. It is not a conventional Z-plasty and has the
advantage of maintaining the integrity of the posterior half of
the patellar tendon, limiting the risk of rupture postoperatively.
It is often combined with an MPFL reconstruction. Jack Andrish
has recently described a similar technique.

Preparation of Tendon
The planned shortening is marked on the tendon to show the
two levels of the partial tenotomy (Fig. 35.2). In this example, the tendon is to be shortened by 25 mm. The upper and
lower boundaries are clearly marked. This tenotomy is performed in the central portion of the tendon (relative to its
patellar and tibial insertions).

Surgical Technique
Incision
A midline or parapatellar incision is made. Medial and lateral full thickness flaps are elevated to fully expose the patellar tendon. The prepatellar bursa and paratenon are incised
and the medial and lateral edges of the tendon are defined to
allow measurement (Fig. 35.1).

E Servien
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
P Archbold
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

Fig. 35.1 Surgical exposure of the patellar tendon and measurement of
its length

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_35

405

406

E Servien et al.

Fig. 35.2 The planned amount of shortening

Fig. 35.3 Elevation of the sheet of patellar tendon

The tendon is cut horizontally along the distal line, perpendicular to the direction of its fibers. It is incised carefully
with a scalpel to a depth of 50% of its thickness. A proximally based tendinous sheet is then progressively raised in
the direction of the fibers over the desired length of shortening (in this case 25 mm) (Fig. 35.3).

the proximal tendon, under the raised 25 mm sheet of tendon. To shorten the tendon, these proximal sutures are tightened and held with a Kocher (Fig. 35.5). The 25 mm raised
sheet of tendon is then sutured onto the front of the distal
tendon with at least three separate passes through the entire
thickness of the tendon (Fig. 35.6). The sutures are tied at
90° of flexion (Fig. 35.7). Patella tracking is checked. The
paratenon is closed with absorbable suture.

Shortening and Repair (Fig. 35.4)
To shorten and repair the tendon, a nonabsorbable suture
(FiberWire®) is used. Two to three sutures are passed from
the proximal tendon to the distal tendon and then back into

Postoperative
Full weight bearing is permitted in an extension brace for
21 days. Flexion is limited to 90° for 45 days.

35 Patellar Tendon Shortening

407

Fig. 35.4 The technique used to shorten the tendon

Fig. 35.5 Sutures placed to shorten tendon

Fig. 35.6 Sutures placed to complete the repair

408

Fig. 35.7 Completed shortening and repair

E Servien et al.

Acute Ruptures of the Quadriceps
and Patellar Tendons

36

G Demey, R Magnussen, C Fiquet, P Neyret,
and C Butcher

Acute Ruptures of the Quadriceps Tendon
After patella fractures, rupture of the quadriceps tendon is the
most common cause of a disruption of the extensor mechanism. Sixty percent of quadriceps tendon ruptures occur
through the tendon and 40% occur due to avulsion of the tendon from its insertion onto the patella. This latter injury was
first described by Albert Trillat and is due to a periosteal
sleeve avulsion at the quadriceps tendon insertion.
The typical history is of a knee injury associated with
eccentric loading of the quadriceps tendon, such as tripping.
Injury can be somewhat subtle and present late due to the
ability of the retinaculum to transmit some load through the
extensor mechanism. Bilateral ruptures often have predisposing factors that lead to tendon degeneration such as the
use of corticosteroids, renal dialysis, or a history or treatment
with fluoroquinolone antibiotics. It is a diagnosis that is often
missed and must always be in the back of the clinician’s
mind particularly following trauma to the knee. An active
extension deficit from a flexed position is the fundamental
symptom and sign. Suprapatellar swelling and a defect can
often be found. The lateral X-ray may show a sagittal tilt of
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
R Magnussen
Centre Albert Trillat, Lyon, France
C Fiquet
Infirmerie Protestante, Lyon, Caluire 69300, France
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

the patella. Both MRI and ultrasound are useful to diagnose
and determine the extent of the tear.

Indication
Surgical treatment is the rule. It is a relative surgical urgency,
as a delay of more than a few days leads to quadriceps contracture and scarring. This scarring makes the repair more difficult, places the repair under greater stress during knee flexion,
and leads to an increased incidence of chronic stiffness.

Surgical Technique
The goal of surgery is to repair the rupture “fiber by fiber.”
The repair must be strong enough to allow early rehabilitation. Reinforcement with a semitendinosus or patella tendon
graft is necessary when the surgical repair is insufficient to
allow 90° of knee flexion.

Tendinous Repair
Patient Positioning and Setup
The patient is positioned on the operating table in the supine
position. A horizontal post is positioned distally on the table to
hold the knee in a 70° flexed position when the heel rests against
it and about 90° when the toes are resting on it. A lateral support
holds the knee in this position. The surgery is performed with a
proximally placed tourniquet. The rupture is easily palpated
with the knee in extension. The knee is then flexed to 90°.
Incision
The technique described is a direct repair without reinforcement. A midline longitudinal incision centered on the tear
and extending distally to expose the upper edge of the
patella is made. If the paratenon is intact, it is split longitudinally to expose the tear and the hematoma is evacuated

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_36

409

410
Fig. 36.1 (a, b) Exposure of
the tear and evacuation of the
hematoma

G Demey et al.

a

b

Fig. 36.2 Sutures placed in a mattress fashion

Fig. 36.3 Sutures crossing the gap; three are shown, four provide
superior strength

(Fig. 36.1a). The ends of the tear are carefully identified and
mobilized by dissecting medially and/or laterally as required
(Fig. 36.1b).

should be placed into both the proximal and distal stumps,
resulting in four strands exiting the end of each stump
(Fig. 36.3). The knee is then placed in extension and the
sutures are tied across the rupture, opposing the torn ends
of the tendon together (Fig. 36.4). Four sutures crossing the
gap are required to provide sufficient strength. Further
strength is added by reinforcing the repair with additional
No. 0 Vicryl sutures around the repair site in an interrupted
fashion.

Repair
Once the tear has been fully exposed, we place sutures in
the tendon as follows: starting in each end of the tear, No. 2
Fiberwire® sutures are placed in a locking whipstitch or
horizontal mattress manner (Fig. 36.2). Two such sutures

36

Acute Ruptures of the Quadriceps and Patellar Tendons

411

Fig. 36.4 Appearance before the sutures are tied across the rupture

Osteotendinous Tear
When the disruption occurs at the osteotendinous junction,
the incision is extended distally to expose the patella
(Fig. 36.5). Full thickness flaps are elevated to expose the
extent of the tear (Fig. 36.6). The quadriceps tendon is mobilized, and in an acute tear, the quadriceps and patella can be
approximated (Fig. 36.7). The proximal pole of the patella is
debrided of necrotic or frayed tissue and the periosteal sleeve
is opened longitudinally for 1 cm and elevated (Fig. 36.8).
2.5 mm transosseous tunnels are drilled in the proximal pole
of the patella (Figs. 36.9, 36.10, and 36.11). These are started
on the anterior surface of the patella 1 cm from its proximal
edge and exit through the midpoint of the proximal pole, taking care not to damage the articular surface. Fiberwire®
sutures are then placed through the proximal tendon stump in
a whipstitch manner as described above and then passed
through the transosseous tunnels and back into the tendon
(Figs. 36.12 and 36.13). These sutures are tied in extension
(Fig. 36.14). The periosteal flaps are closed over the transosseous sutures (Fig. 36.15). Multiple interrupted No. 0 Vicryl
sutures are placed to reinforce the repair and close the medial
and lateral retinacular defects (Fig. 36.16).
Postoperatively
The strength of the repair is tested up to 90° of flexion
(Fig. 36.17). A drain is placed and the closure is achieved at
70° of flexion. The wound is dressed with a compression

Fig. 36.5 Skin incision

bandage, which is removed after 1 h. An AP and lateral
radiographs of the knee are requested and DVT prophylaxis
is provided for 15 days. Prophylactic antibiotics are prescribed for 24 h and skin staples are removed on the 15th
postoperative day.
Two removable splints are used for the first 45 days: A
splint at 30° flexion for rest and an extension brace for mobilization. Physiotherapy is started early with the aim of achieving 90° knee flexion by day 45:
–– 0–45° D0 to D15.
–– 0–70° D16 to D30.
–– 0–90° D31 to D45.
Full flexion is not allowed before 6 months. Caution is
recommended in descending stairs (ramp or step by step) for
4–6 months.

412

Fig. 36.6 The surgical exposure required for the repair of an osteotendinous avulsion of the quadriceps tendon

G Demey et al.

Fig. 36.8 Exposure of the patella: the overlying tissue is split longitudinally and dissected away from the midline to facilitate the exposure

a

1 cm

b

1 cm

Fig. 36.7 The quadriceps tendon is mobilized. In an acute tear, the
quadriceps and patella can be approximated

Fig. 36.9 (a, b) Direction of the transosseous patella tunnels

36

Acute Ruptures of the Quadriceps and Patellar Tendons

413

Figs. 36.10 and 36.11 Creation of the transosseous patella tunnels with a 2.5 mm drill

Fig. 36.12 Horizontal mattress suture passing through a tunnel into
the quadriceps tendon

Fig. 36.13 All the sutures in place before being tied

414

G Demey et al.

Fig. 36.14 The sutures are tied in extension

Fig. 36.15 Periosteal flaps are repositioned for suture over the transosseous repair

Fig. 36.16 The retinacular defects are sutured

Fig. 36.17 Testing the strength of the repair at 90° of flexion

36

Acute Ruptures of the Quadriceps and Patellar Tendons

Acute Ruptures of the Patella Tendon
Patellar tendon ruptures are rare. It is usually a relatively easy
diagnosis to make although the analogous sleeve fractures are
easy to miss in young children. Patients typically present with
a history of a definite knee injury and an inability to walk.
Physical examination reveals a high riding patella with tenderness and bruising at the inferior aspect of the patella, and
a palpable defect in the tendon (Fig. 36.18). Occasionally, the
patient can perform a straight leg raise due to an intact extensor retinaculum; however, in this situation, there will still be
an extension lag apparent when extending from a flexed position. Most commonly, the patellar tendon is avulsed from the
inferior pole of the patella. Radiographs reveal patella alta,

415

and MRI is diagnostic (Fig. 36.19a, b). Disruptions of the
middle or distal insertion of the tendon are rare.

Indication
Surgical repair is the rule. Unlike an acute rupture of the
quadriceps tendon, direct repair of the patellar tendon does
not facilitate early mobilization, therefore we believe that
reinforcement is always required. We achieve this by using a
semitendinosus graft in front of the patellar tendon. PDS
tape can also be used either alone or in combination with the
semitendinosus according to the quality of the repair. We do
not recommend the use of a cerclage wire as it is too rigid. It
can cause sagittal malalignment of the patella and always
requires removal, which carries a risk of re-rupture.

Surgical Technique
Pre-operatively radiographs of the contralateral knee should
be obtained in 30° of flexion to measure patellar height using
the Caton–Deschamps index. This allows an intra-operative
comparison to be made with the aim of achieving an identical index to the normal knee.
The technique described below is for avulsions of the
patella tendon from the inferior pole of the patella. The setup
is the same as for repair of the quadriceps tendon. A ­longitudinal
paramedian incision is made extending from distal end of the
quadriceps tendon to the tibial tuberosity. If possible, the
paratenon should be identified and incised to expose the ruptured patella tendon. The knee joint is frequently visible

Fig. 36.18 A high riding patella

Fig. 36.19 (a) Patella alta
secondary to avulsion of the
patella tendon from the
inferior pole of the patella, (b)
MRI appearance of patellar
tendon rupture

a

b

416

through the rupture and should be irrigated and inspected for
damage. The ends of the frayed patellar tendon are cleaned
and debrided (Figs. 36.20 and 36.21). The knee is placed in
extension and the tendon ends are opposed.

 uture and Mesh Reinforcement Using PDS®
S
Fiberwire® sutures are placed through the tendon. These
sutures are tied in a semi-flexed position. Multiple interrupted No. 0 Vicryl mattress sutures are placed along the
width of the tear to reinforce the repair (Fig. 36.22) and the
medial and lateral retinacular defects are closed. If there is
insufficient proximal tendon tissue, bone anchors are placed
in the inferior pole to facilitate the repair.
If following this repair there is no separation at the repair
site, it is reinforced by a strip of PDS® tape placed on the
anterior aspect of the tendon. Biomechanically, the tape
changes the forces of distraction to compression. It is folded
in half and is fixed to the tibial tuberosity using an Orthomed®
staple. The two strands are then sutured into a “V,” onto the
patellar tendon, patella and quadriceps tendon (Fig. 36.23a,
b). The suturing is done at 60° of knee flexion to prevent
shortening of the patellar tendon and the occurrence of a
patella baja. An intra-operative radiograph ensures the correct restoration of patellar height in relation to the contralateral side (Fig. 36.24).
 uture and Reinforcement with a Tendon Graft
S
In the presence of separation, a semitendinosus graft is used
for reinforcement. The tendon is harvested by extending the
incision 2 cm distally. The pes anserinus is exposed and the
tendon is stripped and prepared (Fig. 36.25). A 4.5 mm trans-

Figs. 36.20 and 36.21 Surgical exposure of the torn patella tendon

G Demey et al.

verse tunnel is drilled in the TT (Fig. 36.26) and the distal
patella (Fig. 36.27). This drill hole should not be made too
proximally in the patella, to avoid tilting it (Fig. 36.28). The
graft is passed through the two tunnels (Fig. 36.29a, b). The
knee is placed in extension and the two strands are tightened
and sutured together. They are then sutured edge to edge
with the patellar tendon.
To help avoid a fracture in a small patella, an alternative
technique is to pass the semitendinosus tendon in front of
the patella, subperiosteally. This transforms the forces of
distraction into compression and stops the risk of tipping
the patella.
If the quality of the repair is still poor following the use of
the semitendinosus graft, we use a quadriceps graft to further
reinforce the repair. A 15 cm long by 15 mm wide quadriceps
graft is harvested from the middle third of the quadriceps
tendon, centered on the proximal edge of the patella. In order
not to breach the capsule, we try to only take the most superficial layer of the quadriceps tendon. To achieve this, it is
often easier to find the correct cleavage plane from the proximal horizontal edge of the quadriceps tendon graft. The
quadriceps graft is not detached distally but is instead elevated with a periosteal hinge, which extends for half the
length of the patella. Once the periosteal hinge has been
elevated from the anterior cortex, the entire graft is flipped in
continuity to cover the patellar tendon. It is then sutured at its
edges into place. This secondary reinforcement is particularly indicated if the length of the semitendinosus graft is
insufficient.
The postoperative management is identical to acute ruptures of the quadriceps tendon.

36

Acute Ruptures of the Quadriceps and Patellar Tendons

417

a

b

Fig. 36.22 Surgical repair of the tear with sutures

Fig. 36.24 Peroperative X-ray. This can be compared with the preoperative X-ray of the contralateral knee

Fig. 36.25 Semitendinosus
graft

Fig. 36.23 (a, b) Reinforcement of the repair with PDS tape

418

G Demey et al.

Fig. 36.26 The tibial tunnel is made transversely with a 4.5 drill

Fig. 36.28 A proximal patella tunnel is avoided to prevent tilt of the
patella

Fig. 36.27 The patellar tunnel is made transversely with a 4.5 drill

a

b

Fig. 36.29 (a) The semitendinosus graft has been passed through the patellar tunnel. (b) After passing though the tuberosity tunnel, it is first
sutured to itself, and then to the edges of the patellar tendon

Chronic Rupture of the Extensor
Apparatus

37

G Demey, R Magnussen, C Fiquet, S Lustig, P Neyret,
and C Butcher

Chronic Rupture of the Quadriceps Tendon
Often, the clinical picture is very suggestive of a chronic
deficiency of the extensor mechanism. Typical symptoms
relate to the extensor lag, episodes of giving way, and disability with stairs and rising from a seated position. Clinical
examination should focus on the degree of extensor lag, the
restriction to passive extension (flexion contracture), and
patellar height. In particular, patellar mobility should be
assessed. If the patella cannot be mobilized proximally, it
indicates that the patellar tendon is retracted.
MRI confirms the diagnosis. It also assesses the feasibility of surgery by measuring the size of the gap and by showing the degenerative change in the quadriceps muscle
(Fig. 37.1). Due to fibrosis and retraction, it is a more complex surgery than that for acute repairs and requires reinforcement. If patella infera is present on comparative
weightbearing lateral radiographs at 30° of flexion, the patellar tendon is retracted. This finding also indicates that it will

be necessary to reinforce the repair. This technique was first
proposed by Pierre Chambat.
This chapter also describes the use of an extensor mechanism allograft although this technique is more suitable for
chronic ruptures of the patellar tendon.

Suture Technique Protected by Metal Framing
Chronic quadriceps tendon ruptures occur through the tendon or secondary to an avulsion of the tendon from its patellar insertion.

G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
R Magnussen
Centre Albert Trillat, Lyon, France
C Fiquet
Infirmerie Protestante, Lyon, Caluire 69300, France
Service d orthopedie de l Hopital de la Croix Rousse,
Lyon 69004, France
Centre Albert Trillat, Lyon, France
S Lustig
Service d orthopedie de l Hopital de la Croix Rousse, Lyon 69004,
France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Fig. 37.1 MRI findings in a chronic tear of the quadriceps tendon

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_37

419

420

G Demey et al.

 atient Positioning and Setup
P
The patient is positioned on the operating table in the supine
position. A horizontal post is positioned distally on the table
to hold the knee in a 60° flexed position. A lateral support
holds the knee in this position. A tourniquet is placed at the
base of the thigh but not inflated, as it can hamper the repair
by preventing full mobilization of the quadriceps. In a chronic
rupture with tendon retraction, it is sometimes unrealistic to
seek to achieve 90° of flexion during the procedure.
Incision
A midline longitudinal incision is made beginning at the
lower pole of the patella and extending 10 cm above the
superior pole (Fig. 37.2).
The dissection is carried down in the midline elevating
subcutaneous flaps. The upper pole of the patella and ends of
the tear are exposed. The ends of the tear are carefully identified and mobilized by excising scar tissue and dissecting
medially and/or laterally as required (Fig. 37.3). It is critically important to preserve as much healthy tissue as possible. It is usually not necessary to perform arthroscopic
arthrolysis or to incise the retinaculum.

Fig. 37.2 Surgical exposure of the chronic tear

Fig. 37.3 Mobilization and
debridement of the ends of
the tear (a) and measurement
of the gap (b)

a

 obilizing the Proximal Quadriceps Stump
M
A 2 mm K-wire is inserted transversely into the stump of the
quadriceps tendon. Contrary to popular belief, there is no
“cheese slicing effect” and the K-wire will not pull out of the
tendon distally. A second trans-patellar 2 mm K-wire is
placed transversally 1 cm below the proximal pole of the

b

37 Chronic Rupture of the Extensor Apparatus

Fig. 37.4 Placement of the 2 mm K-wires and wire loops

patella. A loop of metal wire is then placed on either side of
the K-wires (this arrangement is preferred to a mounting
frame or a figure of eight with a single wire) (Fig. 37.4). By
progressively tightening the wires with the knee in extension, the proximal quadriceps tendon stump is pulled to the
stump at the proximal pole of the patella (Fig. 37.5).
To complete the repair, sutures are placed in the two tendon stumps with Fiberwire suture and the ends are tied over
the tear. The repair is then reinforced with No. 0 Vicryl suture
around the repair site as described in the previous chapter.
The strength of the repair is tested at 60° and, if possible,
90° of knee flexion. Closure is achieved in layers and a suction drain is placed subcutaneously. The skin is closed with
staples (Fig. 37.6).
In cases of osteotendinous avulsion, a similar technique to
that used in acute ruptures of the quadriceps tendon is used
to complete the repair after placement of the wire augment as
described above. The sutures are passed through the quadriceps tendon stump proximally and then through longitudinal

421

Fig. 37.5 Tightening of the wires and closure of the tear

bone tunnels in the patella and tied (see previous chapter on
acute ruptures of the extensor mechanism).

 einforcement of the Tear Using a Patellar
R
Tendon and Semitendinosus Graft
Reinforcement of a chronic quadriceps tendon rupture is
indicated when the direct repair is at high risk of failure due
to poor tissue quality. The reinforcement may be achieved
with both a patellar tendon and semitendinosus graft.
Dependent on the quality of the repair, a single or double
reinforcement can be done.

Incision
A longitudinal paramedian incision is made extending from
10 cm above the superior pole of the patella ending on the
medial side of the tibial tuberosity.

422

G Demey et al.

Fig. 37.6 Postoperative
X-ray

Fig. 37.8 Harvesting of the semitendinosus graft (1)

expose the upper pole of the patella and the ends of the tear
(Fig. 37.7).

Harvesting the Semitendinosus Graft
The tendon is harvested by extending the incision 2 cm distally. The pes anserinus is exposed and the tendon is stripped
and prepared (Figs. 37.8, 37.9, and 37.10). This produces a
graft about 25–30 cm long (Fig. 37.11).
Fig. 37.7 Surgical exposure of the chronic tear

Full thickness flaps are elevated to expose the chronic
rupture. Scar tissue is excised in an economical manner to

Harvesting the Patellar Tendon Graft
A 1 cm strip in the middle third of the patellar tendon is outlined and incised. Its distal insertion is mobilized with a strip
of periosteum from the tuberosity using a scalpel. This strip

37 Chronic Rupture of the Extensor Apparatus

423

Fig. 37.11 Harvesting of the semitendinosus graft (3)

a

Fig. 37.9 Preparation of the tendon extremity

b

Fig. 37.10 Harvesting of the semitendinosus graft (2)

of tendon is then peeled from the front of the patella being
careful to leave it attached to at least half the height of the
patella. A compromise between getting sufficient graft length
and keeping enough of its attachment to the patella must be
made (Fig. 37.12a–c).
Three 2.7 mm transosseous tunnels are drilled in the proximal pole of the patella. These are started on the anterior surface of the patella 1 cm from its proximal edge and exit

Fig. 37.12 (a–c) Harvesting the central third of the patellar tendon. At
least half of its attachment to the anterior patellar is preserved

424

G Demey et al.

c

through the midpoint of the proximal pole, taking care not to
damage the articular surface (Fig. 37.13).
A 4.5 mm horizontal tunnel is made through the proximal
third of the patella. Care must be taken to do this in the right
direction to prevent any weakening of the patella and a possible fracture (Fig. 37.14). Fiberwire® sutures are then placed

Fig. 37.12 (continued)

Fig. 37.14 Transverse transosseous patellar tunnel

Fig. 37.13 Transosseous tunnels in the proximal pole of the patella

Fig. 37.15 Creating a tunnel for the semitendinosus graft within the
quadriceps tendon

37 Chronic Rupture of the Extensor Apparatus
Fig. 37.16 Patellar tendon
and semitendinosus grafts are
placed (a) and sutured in
position (b)

a

through the tendon and then through the transosseous tunnels back into the tendon. These sutures are tied in
extension.
The semitendinosus tendon graft is then passed through
the patella using a guide pin.
Dissection is then carried out to create a tunnel in the
quadriceps tendon in order to create a tunnel for the semitendinosus graft (Fig. 37.15). The graft is then pulled tight
at 60° of flexion and sutured to itself with absorbable Vicryl
at multiple points. The sutures are tied in extension
(Fig. 37.16a, b).
Finally, the strip of patellar tendon is turned over and
sutured onto the front of the repair using No. 2 Vicryl. The
strength of the repair is tested by placing the knee in 60° of
flexion. Closure is achieved in layers and a suction drain is
placed subcutaneously. The skin is closed with staples. The
postoperative instructions are identical to those described in
acute ruptures of the quadriceps tendon.

Chronic Rupture of the Patellar Tendon
Reconstruction of chronic ruptures of the patellar tendon is
difficult due to contraction of the quadriceps and hence the
difficulty of restoring the correct height of the patella.
If the correct patellar height is achieved relatively easily,
reinforcement of the repair can be achieved by using a strip
of PDS tape or a quadriceps tendon graft (see acute ruptures of the patellar tendon). In contrast, if it is difficult to
lower the patella (Fig. 37.17), it is necessary to use an
extensor mechanism graft. The options are a contralateral

425

b

autograft and an allograft, and the technique is similar. The
two major situations in which this is indicated are in the
native knee and after a TKA (particularly after rotatory
hinge TKA). The allograft is nowadays our preferred graft
material, and our results, reported by C. Fiquet, are better
in native knees.

Autologous Transplantation of the Extensor
Mechanism
We presented this technique with Henri Dejour in Toronto
in 1991. In revision surgery or when the quality of the
patellar tendon is insufficient to achieve a satisfactory
repair, we use an autograft taken at the expense of the middle third of the contralateral extensor mechanism. This is a
composite graft: quadriceps tendon, patellar bone block,
patellar tendon, tibial bone block. The contralateral patellar tendon must be healthy and have no previous surgery
(e.g., tibial nailing, tuberosity transfer, harvest for ACL
reconstruction).

 arvesting the Autograft (Contralateral Knee)
H
Both lower limbs are placed in the operative field. A tourniquet is placed at the base of each thigh (Fig. 37.18).
Incision
The incision begins 3 cm below the tibial insertion of the
patellar tendon and extends 5–7 cm above the proximal
pole of the patella. The paratenon is incised vertically
(Fig. 37.19).

426

G Demey et al.

Fig. 37.17 MRI showing a chronic tear of the patellar tendon

Fig. 37.19 Exposure of the extensor mechanism autograft

Fig. 37.18 Patient setup

Delineation of the Transplant
The quadriceps tendon is exposed along its entire length,
extending the exposure to reveal the most distal muscle fibers
of the rectus femoris. The tendon is then incised for 5–6 cm
in line with its fibres. Only the two most superficial layers
are incised to avoid entering the knee joint. The width of the
graft should be 12–14 mm (Fig. 37.20a, b).
The incision extends onto the periosteum of the patella as a
“dovetail,” that is to say, a trapezoidal base of 14 mm proximally with a width of 10 mm distally. The harvest then continues on to the patellar tendon. The middle third is harvested at
a width of 10 mm (Fig. 37.21). The tibial periosteum is then
incised to mark out a 35 mm long bone block that is 10 mm
wide at its proximal portion and 12 mm at its base (Fig. 37.22).
It should be noted that the bone blocks must have a trapezoidal
shape in order to prevent migration of the graft.
A variant of these bone blocks can be harvested, i.e., a
bone block with a narrower waist that is wider proximally
and distally (hourglass shape). This method allows the harvest of a wider patellar tendon graft while still avoiding
migration of the graft, and is especially useful if allograft is
used.

37 Chronic Rupture of the Extensor Apparatus
Fig. 37.20 (a, b) Harvesting
the quadriceps tendon. The
central third is identified and
detached proximally

a

427

b

Fig. 37.21 Harvest of the patellar tendon

Harvesting the Bone Blocks
The tibial and patellar bone blocks are harvested with an
oscillating saw (Fig. 37.23a, b). At the distal end of the
tibial block, the saw has to be tilted to avoid any risk of
fracture. The bone block is then separated using an open
gouge. The patellar tendon is then retracted upward and it
is released from the fat on its posterior surface. It must be
elevated off the tip of the patella in order to assess the
thickness of the bone block. A 10 mm Lambotte osteotome
is then introduced parallel to the anterior cortex. This helps
remove the entire bone block (Fig. 37.24). The osteotome
should not be used as a lever due to risk of breaking the
bone block or fracturing the patella (Figs. 37.25 and 37.26).

Fig. 37.22 Dimensions of the extensor mechanism autograft

In case of osteoporotic bone, or for fine-tuning, a burr can
also be used.

Preparation of Recipient Site
Incision
A paramedian skin incision is made beginning 10 cm above
the proximal pole of the patella and ending 3 cm below the
distal insertion of the patellar tendon. In revision surgery,
previous incisions must be taken into consideration.

428
Fig. 37.23 Harvesting the
bone blocks from patella (a)
and tibia (b)

G Demey et al.

a

a

Fig. 37.24 Careful elevation of the tibial (a) and patellar bone blocks (b)

Fig. 37.25 The extensor
mechanism autograft

b

b

37 Chronic Rupture of the Extensor Apparatus
Fig. 37.26 (a, b) Closure
following harvest of the
autograft

a

Fig. 37.27 (a, b) Preparation
of the recipient knee—
creation of the patellar bone
trench for the corresponding
graft bone block

a

Exposure
The medial and lateral edges of the patellar tendon are identified and the scar tissue is excised to expose the two tendon
stumps. The quadriceps tendon is exposed using the technique described above. The next part of the procedure
involves creating a tibial and patellar bone trench to accommodate the harvested bone blocks (Fig. 37.27a, b).

429

b

b

Preparation of Bone Trenches
The recipient sites are marked out on the periosteum using a
scalpel. The bone trenches are then cut with an oscillating saw.
On the tuberosity, the trench is 10 mm wide proximally,
12 mm distally, and 35 mm long. In order to elevate the block
of bone and make the trench, a gouge is inserted vertically just
above the reflected patellar tendon (Fig. 37.28a, b). On the

430
Fig. 37.28 (a, b) Preparation
of the recipient knee—
creation of the tibial bone
trench for the corresponding
graft bone block

G Demey et al.

a

a

b

b

Fig. 37.29 (a, b) Patellar fixation with transosseous wires tied anteriorly

patella, the trench is trapezoidal measuring 14 mm wide proximally and 10 mm wide at its distal end. In order to elevate the
bone block and from the trench a Lambotte osteotome is
inserted parallel to the anterior cortex at the tip and at the proximal pole of the patella. The match between the transplant and
recipient site is then evaluated. It is often necessary to adjust
the graft using a rongeur.

dovetail of the bone block is placed proximally in the trench.
The bone block is fixed with two separate metal wires, which
pass transversely through the patella and bone block. The
wires are tightened on one side by twisting and then cut short
and buried. Proximally an opening is made in the midline of
the quadriceps tendon and the quadriceps tendon graft is
sutured into this site using No. 2 absorbable suture.

Fixation of the Graft

Tibial Tuberosity Fixation
A wire is passed through the tibial bone block. The tibial
bone block is then positioned in the trench and impacted into
the recipient site on the tibia. This restores the correct patellar height. Fixation is achieved with the wire and a screw
(Hooper, Lepine®). The screw which is placed distally prevents proximal migration of the bone block (Fig. 37.30a, b).

Patellar Fixation (Fig. 37.29a, b)
The extensor mechanism graft is initially fixed proximally.
The patellar bone block is positioned in the recipient patellar
trench. This should be achieved without impact in order to
prevent a fracture or injury to the patellar cartilage. The

37 Chronic Rupture of the Extensor Apparatus
Fig. 37.30 (a, b) Tibial
tuberosity fixation initially
with a transosseous wire
secured around a tibial screw

a

Fig. 37.31 (a, b) Tibial
tuberosity fixation completed
with two staples

a

This fixation is supplemented by two Orthomed® staples,
which can also be used to fix a strip of PDS prepared by the
same technique described in the chapter “acute ruptures of
the extensor mechanism.” In the recent past we do not hesitate, when possible, to fix the tibial bone block with three
3.5 mm AO cortical screws instead of the staples.
The edges of the graft are sutured to the patellar tendon
with a No. 2 absorbable suture (Fig. 37.31a, b). Imaging is
not necessary prior to closure, as the corrected patellar height
is only dependent on the length of the contralateral patellar
tendon (Fig. 37.32a, b).

Closure
Recipient Site (Fig. 37.33)
A suction drain is placed in contact with the graft. Hemostasis
is achieved and closure is achieved in layers. Staples are used
in the skin.

431

b

b

Donor Site
The edges of the tendons are approximated with a No. 2
absorbable suture. A suction drain is placed in the subcutaneous space and closure is achieved in layers. We do not fill the
bone defects on the patella or tuberosity with the bone fragments taken from the recipient knee.

Postoperative
The postoperative regime is identical to the protocol
described above. Prophylactic anticoagulation should be
avoided unless absolutely necessary. The skin should be
monitored closely due to the risk of infection or necrosis.
Weightbearing is allowed on the grafted side, protected by a
cast in extension for 2 months, and the use of crutches.
Passive range of motion of the knee is prudently allowed up
to 45° of flexion for the first 2 months then progressively and
carefully increased in order to permit consolidation of the
bone blocks.

432
Fig. 37.32 Postoperative
lateral (a) and axial (b)
X-rays

G Demey et al.

a

b

Fig. 37.33 Diagram showing the graft in situ

Allograft Transplantation of the Extensor
Mechanism
There are some advantages to the use of allograft tissue for
extensor mechanism reconstruction. There is no harvest site
morbidity on the other extremity such as fracture, extensor
mechanism rupture or pain. Allograft tissue is particularly
useful in patients with a collagen disease or a history of injury
or surgery involving the contralateral knee (contralateral
TKA, fracture or more often, osteoporosis) (Fig. 37.34).
Furthermore, the allograft can be thicker and longer than
autograft and allows shorter surgical time. The width of the
graft can be 14 mm when using allograft without harvest site
morbidity. The shape can be an “hourglass shape” contribut-

Fig. 37.34 Chronic patellar tendon rupture on TKA with failure of
conventional technique. An allograft is required. Another option would
be a technique utilizing synthetic mesh (recently described by Hanssen
and Browne)

ing to primary stability. The soft tissue of all the anterior part
of the allograft can also be preserved, allowing for better fixation of the patellar bone block, and lower risk of migration.
On the other hand, there are a few disadvantages. Allograft
is not available in every country and can be costly. While the
risk of viral contamination is low (estimated to be 1/106), the

37 Chronic Rupture of the Extensor Apparatus

433

Fig. 37.35 Complete
extensor mechanism

Fig. 37.37 Tibial fixation (1) using metal wire passed through the
bone block and fixed using a cortical screw

Fig. 37.36 Shape of the graft (hourglass patellar bone block)

patient must be informed of the risk of this devastating complication. Finally, poor tissue quality is sometimes seen with
allografts relative to autografts.
The length of the patellar tendon and patella should be
matched to the patient when allograft is used. Specific
­measures of these structures must be made during the preoperative radiological assessment.
The surgical technique for reconstruction with allograft is
very similar to the autograft technique described above. The

surgical assistant can prepare the graft while the senior surgeon prepares the surgical site.
There are a few key differences between the techniques.
The allograft typically can be made larger because of the
absence of morbidity at the donor site. The allograft arrives
as a complete extensor mechanism with tibial tuberosity,
patellar tendon, patella, and quadriceps tendon (Fig. 37.35).
The patellar bone block is prepared to be wider proximally
and distally with concave edges medially and laterally (hourglass shape). This method allows for the use of wider patellar
and quadriceps tendon grafts (Fig. 37.36).
The fixation technique is similar. We use a wire wrapped
around a low-profile screw distally with the addition of staples for tibial bone block fixation (Figs. 37.37 and 37.38).
Metal wires are used to fix the patella, and multiple sutures
of the preserved anterior soft tissue of all the donor patella
are performed to help prevent migration of the bone block.
The tendon is secured with absorbable suture and sometimes
Fiberwire® depending on tissue quality. One PDS tape is distally fixed by the staple and sutured throughout the extensor
mechanism with the knee flexed to 90°. This tape protects
allograft stress during bending (Fig. 37.39). The final aspect

434

G Demey et al.

Fig. 37.39 Final aspect after fixation, suture, and reinforcement using
PDS tape

Fig. 37.38 Tibial fixation (2) using Blount staples

of the allograft in situ is schematized in Fig. 37.40.
Rehabilitation is very careful (see protocol cited above). The
range of motion is particularly cautious and progressive to
allow time for consolidation of the bone blocks. Successive
radiographs are done every 45 days to check bone block consolidation before advancing range of motion. Despite the
cautious physiotherapy protocol to prevent bone block
migration and encourage healing, we avoid prolonged extension in a cast or locked brace.

Fig. 37.40 Case of chronic rupture with previous failure of suture plus
augmentation. Excellent result at 6 months follow-up

38

Patella Fractures
G Demey, R Magnussen, P Neyret,
and C Butcher

The aims in the treatment of a patella fracture are:

Non-operative Management

•
•
•
•
•

Tense hemarthroses are painful and can damage the articular
cartilage. Therefore they should be drained. An important
fact to be considered during a patient’s rehabilitation is that
forces applied to the patella are small in extension but
increase dramatically during flexion.

To restore the continuity of the extensor mechanism
To restore articular congruity
To avoid at all costs a patellectomy
To limit the devascularization of the patella
To restore adequate stability in order to achieve early
mobilization of the knee

Indication
Non-operative management:
• Stable, congruent fracture
• Longitudinal fractures with an interfragmentary gap of
≤1 mm
• Transverse fracture without articular impaction (separation and step-off ≤1 mm)
Open reduction and internal fixation (usually indicated):
• Fractures resulting in disruption of the extensor mechanism (Fig. 38.1)
• Articular incongruity (step-off >1 mm)
• Osteochondral fractures
• Open fractures
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher (*)
Healthpoint, Abu Dhabi, UAE
e-mail: c.butcher@healthpoint.ae

Early Rehabilitation
This should be cautious and protected. Its aim is to gently
mobilize the knee, fire the quadriceps and help prevent
venous thrombosis. Knee flexion is commenced after the
third or fourth day during the hyperanalgesic phase. Passive
flexion by a CPM or physiotherapist should not exceed 45°
for the first 3 weeks. After 3 weeks, flexion is gradually
increased to 90° by the 45th day. Mobilization full weight
bearing is allowed in an extension brace. A second splint at
30° of flexion is used at night and prevents the occurrence of
patella baja. Radiographs are repeated at D10, D21, and D45
to ensure that an adequate reduction has been maintained. At
day 45, fracture consolidation is usually sufficient to allow
full flexion. Caution is recommended in descending stairs
(ramp or step by step) for 4–6 months.

Surgical Management
Incision
A midline or paramedian longitudinal incision centered over the
patella is made. Consideration should be made to incorporate
open wounds (Fig. 38.2). The joint is typically visible through
the fracture site. The fracture is exposed and debrided. The periosteum adjacent to the fracture is elevated to allow accurate
reduction (Fig. 38.3). Pointed reduction forceps are used to

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_38

435

436
Fig. 38.1 (a, b) Displaced
patella fracture

G Demey et al.

a

Fig. 38.2 Surgical approach

maintain the reduction while fixation is achieved (Fig. 38.4).
Following reduction and fixation of the fracture, the periosteal
flaps and retinacular defects should be closed with sutures.

Method of Osteosynthesis
• Tension band wiring
This technique is indicated when anatomical reduction is
achieved (Fig. 38.5). The reduction is initially held with two
vertical, parallel 1.6 mm K-wires. The cerclage wire is
passed behind the K-wires and over the front of the patella in

b

Fig. 38.3 Fracture site following exposure and debridement

a figure of 8. This configuration turns the distractive force
into a compressive force during knee flexion and prevents
separation of the fracture fragments. The wire must be firmly
against the upper and lower edges of the patella close to the
emergence of the wires (Figs. 38.6 and 38.7).
• Cerclage wiring
In comminuted fractures, a second cerclage wire around
the periphery of the patella is useful in maintaining the
reduction of the fragments (Fig. 38.8).
• Screw fixation
In transverse fractures that can be anatomically reduced,
osteosynthesis with screws is an alternative to using
K-wires (Fig. 38.9).

38

Patella Fractures

437

Postoperative Rehabilitation
Rehabilitation is started early. The protocol is identical to
that of conservative treatment.

Patellectomy
Partial Patellectomy

Fig. 38.4 Reduction of the fracture fragments

a

This is rarely indicated but may be necessary when there is a
significant defect in the articular cartilage due to the removal
of fragments or when significant comminution prevents a
satisfactory reduction. In partial patellectomies involving the
upper pole of the patella, the quadriceps tendon should be
reattached via transosseous tunnels. This technique is similar

b

c

Fig. 38.5 (a) Reduction and wire stabilisation (b) Tension band wiring in progress (c) Completed fixation

438

G Demey et al.

Fig. 38.6 Diagram showing
the correct technique for
tension band wiring of the
patella

Fig. 38.7 Postoperative
radiographs following tension
band wiring of the patella

a

b

38

Patella Fractures

439

Fig. 38.8 Cerclage wiring of a patella fracture

Fig. 38.9 Screw osteosynthesis of a patella fracture

to that used in the repair of a distal quadriceps tendon
rupture.
When the partial patellectomy involves the distal pole of
the patella, the patellar tendon should be reattached to the
remaining patella fragment on its posterior aspect via transosseous tunnels in order to avoid sagittal tilting of the
patella. This repair (Fig. 38.10) must be reinforced with a
strip of PDS® or a semitendinosus graft rather than a cerclage
wire.
In comminuted fractures, an attempt should be made to
preserve the distal end of the patella which is then fixed to
the remaining proximal fragment. The fixation is achieved
by using a transosseous wire that passes through the distal
fragment into the proximal fragment. This is tightened and
buried at the proximal edge of the patella, thus compressing
the fragments together.
In longitudinal fractures, simple excision of small fragments is usually sufficient. When the vertical resection
involves more than half of the patella, a total patellectomy
should be considered as in this circumstance the joint incongruity results in disordered patellofemoral kinetics and pain.

Total Patellectomy
A total patellectomy can be debilitating and should be considered only as a last resort (Fig. 38.11). When it is necessary, care should be taken when excising the bony fragments,
to maintain the continuity of the extensor mechanism. To
achieve this, we use a medial parapatellar arthrotomy and
evert the patella laterally. The extensor mechanism is then
restored by closure of the remaining soft tissue. Some surgeons, like Trillat, feel that the extra-articular tip of the
patella should be retained.
Following the patellectomy, if the remaining soft tissue is
insufficient to restore the continuity of the extensor mechanism, a flap of the quadriceps in the form of an inverted V
can be turned down to fill the defect. The goal in all patellectomies is to moderately shorten the extensor mechanism,
thus allowing limited flexion. This is because the repair tends
to lengthen gradually over time, improving flexion. One
must also ensure correct centering of the extensor mechanism over the trochlea in order to maximize the efficiency of
the remaining extensor mechanism.

440

Fig. 38.10 Partial patellectomy of the distal pole of the patella

Fig. 38.11 The mutilating effect of a total patellectomy of the right
knee

G Demey et al.

Surgical Management of the Stiff Knee

39

R Debarge, P Archbold, P Neyret,
and C Butcher

Introduction
Stiffness of the knee or, more precisely, limited range of
motion of the knee is an ill-defined term. The reason is that it
is both a functional description and clinical sign. It can evolve
over time and absolute numbers therefore have a limited value.
Stiffness of the knee can be defined by certain variables:
• Evolution over time
• Tolerance (there is a difference between a total knee
arthroplasty and ligament surgery)
• Etiology (e.g., ACL surgery, intra-articular fractures)
The precise range of motion must be clearly recorded in
the patient’s clinical notes with the same care that is taken to
document, for example, the body temperature and arterial
blood pressure during the pre- and postoperative period.
These values should be transferred to the physiotherapist
once the patient leaves hospital. During surgery, it is important to document the range of motion prior to the anesthetic
induction and also at the end of the surgical procedure.
The clinical history should be analyzed carefully, in particular the circumstances of the initial accident, the previous
surgical interventions and the different rehabilitation programs undertaken. The range of motion must be documented
during each of these steps, in order to make it possible to
document the evolution of the stiffness. Some threshold values are known: 90° of flexion is required for stair climbing
and 120° is needed to comfortably perform the activities of

daily living. Three measurements can quantify the range of
knee motion: the first is the hyperextension, the second is the
extension deficit, and the third is the maximal flexion. For
example, a range of motion documented as 5/0/120 represents 5° of hyperextension, 0° of extension deficit, and 120°
of flexion. The clinical examination should always be comparative, therefore the values for the contralateral knee need
to be documented as well. A clinical examination in the
prone position is useful in order to evaluate an extension
deficit (e.g., cyclops of the ACL) (Fig. 39.1). Normal function of the medial and lateral gutters and the supra patellar
pouch are necessary to have a normal range of motion, in
particular for flexion. Stiffness of the knee can be classified
using different criteria.

Classification of Stiffness According to Etiology
• Reflex sympathetic dystrophy—RSD (complex regional
pain syndrome—CRPS): usually conservative therapy is
initiated.
• Post-traumatic (femoral fracture, patella fracture, tibial
plateau fracture, grade III sprain of the medial collateral
ligament, ACL rupture with cyclops lesion).

R Debarge · P Archbold
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

Fig. 39.1 Clinical examination in the prone position to evaluate an
extension deficit

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_39

441

442

• Postoperative (ACL reconstruction, total knee arthroplasty, synovectomy).
• Post-infectious (septic arthritis of the knee).
• Certain muscular diseases.
Albert Trillat illustrated that adhesions between the medial
collateral ligament and the medial femoral condyle could limit
the motion of the medial capsular structures (Fig. 39.2a, b).
These adhesions cause a functional shortening of the medial
collateral ligament, and the center of the rotation moves from

a

R Debarge et al.

the medial condyle to a point close to the tibiofemoral joint
line. These adhesions limit flexion to approximately 60°.

 lassification of Stiffness According to
C
Type of Limitation
Flexion Limitation
In the case of limitation of flexion, it is necessary to release the
adhesions in the supra patella pouch and the condylar gutters.

b

Fig. 39.2 (a, b) Albert Trillat’s drawing: stiffness and MCL lesion. Adhesions limit flexion to approximately 60°

39 Surgical Management of the Stiff Knee

Sometimes retraction of the quadriceps and extensor mechanism necessitates an additional release according to Judet.

 imitation of Extension (Fixed Flexion Deformity)
L
A fixed flexion deformity (FFD) results in a reduction of
the contact area of the cartilage, which can lead to pain and
arthritis. It is very important to address a fixed flexion
deformity in the hip or the contralateral knee since these
can result in fixed flexion deformity of the index knee.
Because of the limited extension it is important to think
about an obstacle situated in front of the intra-condylar
notch, e.g. a free body or a dislocated bucket handle tear of
the meniscus.
Other reasons could be a recent rupture of the ACL (mop
tear) or a cyclops syndrome secondary to an ACL reconstruction. Finally a reconstructed ACL can also cause a fixed flexion
deformity—most frequently secondary to malpositioning of the
femoral or tibial tunnels—and sometimes necessitate a resection of the reconstructed ACL to overcome the deformity.
Capsular and ligamentous scarring are less frequent causes,
but can require a posterior capsulotomy in chronic cases.
Mixed Limitation (Flexion and Extension)
Iatrogenic Limitation Associated with
Tibial External Rotation
This type of stiffness is essentially observed after a Lemaire
extra-articular anterolateral tenodesis fixed in external rotation, which has been described in detail by H. Jaeger.
Reflex Sympathetic Dystrophy
Another reason for mixed limitation of movement, as well as
ACL graft malpositioning, is RSD. In this case the retraction
involves the capsule, but also includes the patellar tendon,
and consequently patella infera is often present.

 lassification of Stiffness According
C
to Anatomy
Articular Stiffness (Fig. 39.3a, b)
Capsular or intra-capsular.
Intra-capsular stiffness can be addressed by arthroscopy.
Extra-Articular Stiffness
Requires open surgery.
In the case of limitation of flexion and extension, multiple anatomical structures are involved. The common
denominator remains capsular retraction. Shortening of
the posterior knee capsule secondary to a fixed flexion

443

deformity (because of an obstacle in front of the intercondylar notch) can perpetuate the deformity. Posterior osteophyte causing a flexion deformity cannot be addressed
arthroscopically.

 lassification of Stiffness According to Degree
C
of Functional Limitation
A clear difference needs to be made between a fixed flexion
deformity observed in the athletic population versus a limitation in flexion after a total knee arthroplasty.

Surgical Options
The following surgical options are available:
• Manipulation under anesthesia (MUA)
• Arthroscopic arthrolysis
• Open arthrolysis with an arthrotomy (anterior and
posterior)
• Arthrolysis according to Judet (not in this chapter)

Manipulation Under Anesthesia
Indications and Risks
The aim of this intervention is to overcome intra-articular
adhesions. Sometimes, these adhesions are present between
the articular surfaces. However, one must be aware that during a forceful manipulation a fracture or damage to the articular surface of the joint can occur.
Therefore:
• Manipulation under anesthesia should be performed after
healing of the skin incision.
• In the case of a non-prosthetic knee (after trauma or ligament injury), the MUA should be performed with extreme
caution and should be performed at an early stage (less
than 30 days from surgery).
• Arthroscopic arthrolysis is indicated and preferable
within 45 days of the initial surgery that caused the
stiffness. During the arthroscopy, the synovial and cartilaginous adhesions can be cut, avoiding a forceful
MUA.
• In the case of a total knee prosthesis, MUA can be done
up to the 90th day after surgery. Risk for injury to cartilage is limited (except in cases where the patella is not
resurfaced or in case of a unicompartmental knee
prosthesis).

444
Fig. 39.3 (a, b) Intra-capsular
stiffness (supra patella pouch,
condylar gutters, anterior
compartment)

R Debarge et al.

a

Surgical Technique
Prior to the manipulation, the full clinical history of the
patient and the most recent radiographs should be available.
The status of the skin and in particular the skin incision
should be examined to avoid complications (Fig. 39.4a, b).
Once the patient is under anesthesia, the initial range of
motion is documented. The mobilization starts gently by
progressively exerting manual pressure with both hands on
the tibial tubercle. The hip should be flexed. Commonly the
adhesions are easily overcome. Sometimes, small cracks can
be heard.
If the abovementioned details are respected, everything
should go according to plan. At the end of the procedure,
the range of motion is documented. Spontaneous flexion
should also be documented. Spontaneous flexion is
defined as the maximal flexion obtained by gravity with
the hip in flexion. This spontaneous flexion is most commonly the flexion obtained at the end of the rehabilitation
period. In the case of a limitation in flexion, the patient is
positioned with a specially designed flexion cushion in
the post-anesthesia care unit (Fig. 39.5). If a “delayed”
MUA is performed, the surgeon has to be aware of the
risks (diaphyseal fractures, rupture of the extensor mechanism). Most importantly, the manipulation has to be done
progressively without excessive force. In case of an MUA
on a non-prosthetic knee, the more important, but fre-

b

quently unrecognized, complication is damage to the
articular cartilage.

Arthroscopic Arthrolysis
Indications
This type of surgery is indicated in cases of stiffness secondary to an intra-capsular cause, most commonly following
ligamentous surgery.
Several surgical procedures can be done:
• Section of the synovial adhesions
• Removal of intra-articular loose bodies
• Meniscectomy or meniscal suture for a dislocated bucked
handle tear
• Treatment of ligamentous lesions (cyclops, mop tear—
positioning of the distal end of the ruptured ACL inside
the intercondylar notch)
• Treatment of the stiffness after prosthetic knee surgery

Surgical Technique
The classic portals are used: anterolateral and anteromedial,
but also superomedial and superolateral. Different surgical

39 Surgical Management of the Stiff Knee
Fig. 39.4 (a, b) Case of
serious complication after
MUA: patellar tendon
avulsion associated with
wound dehiscence

a

445

b

Fig. 39.6 Specifically designed knife blade

It is very easy to handle and it does not necessitate a skin
incision. The procedure can be easily performed under visual
control. Only those adhesions that are under tension will be
cut, thus limiting blood loss.

Fig. 39.5 Flexion cushion

procedures are available depending on the cause of the
stiffness:
• Release of the adhesions in the supra patella pouch and
condylar gutters.
A specifically designed knife blade is used for this procedure (Figs. 39.6, 39.7, and 39.8).

• Removal of foreign bodies (cyclops, anterior osteophyte,
osteochondral fragment)
• Meniscectomy or meniscal suture for a displaced bucked
handle tear
• Stiffness after total knee prosthesis is detailed in another
chapter
Introduced by W. Clancy, we have also observed that scarring of the anterior inter-meniscal ligament can be a cause of
an extension deficit. Due to retraction, both the medial and
lateral meniscuses are pulled anteriorly and can impinge
with the femoral condyle. It is possible to transect this anterior inter-meniscal ligament under arthroscopy. This procedure can be useful in a chronic fixed flexion deformity
(several months).

446

R Debarge et al.

Posterior Arthrolysis

Fig. 39.7 Adhesions in the supra patella pouch

A posteromedial and posterolateral arthrotomy is performed
behind the respective collateral ligaments. In some rare cases,
a posteromedial arthrolysis alone can suffice (after open
medial meniscus suture or retraction of the posteromedial capsule). In all other cases, a posterior arthrolysis should be performed using both approaches. Through the posteromedial
and posterolateral arthrotomies, one can easily release the posteromedial and posterolateral capsule from the posterior femoral condyles with a 15 blade knife. This release has to be a total
release, meaning that “light should be observed between the
incisions.” Full extension is generally obtained. Sometimes
full extension however is somewhat elastic in the final degrees
of extension. In this situation, we prefer the application of an
extension brace postoperatively rather than transection of the
capsule or a transection of the hamstrings. This brace should
be applied for at least five nights and should be combined with
a strict rehabilitation protocol.
Remarks
• Posterior arthrolysis is being performed less frequently.
• Anterior arthrolysis is more frequently performed under
arthroscopy.
• In case of flexion deficit, one can consider the section of
the deep layers of the quadriceps muscle that correspond
to the vastus intermedius. This technique has been
recently recommended by Philipp Lobenhoffer and can
be particularly useful in avoiding the need for Judet
release. The range of motion is not dependent of the flexion of the hip due to the fact that the vastus intermedius
muscle is monoarticular. On the contrary, when the knee
flexion is influenced by the hip position, then the rectus
femoris (biarticular muscle) is involved (Fig. 39.9). In this
case, a section of the rectus femoris is discussed.

Fig. 39.8 Release of the adhesions in the medial condylar gutter

Release According to Judet
Open Arthrolysis
We will not detail the surgical approaches (see chapter on
synovectomies).

Anterior Arthrolysis
Two skin incisions can be used: anteromedial and superolateral. The skin incisions allow an anteromedial arthrotomy
and a superolateral arthrotomy.

This release is beyond the aim of this chapter and is detailed in
another chapter. Release of the extensor apparatus according
to Judet is a very rare surgical intervention since the introduction of arthroscopy and the decrease in road traffic accidents.
Nevertheless, a release of the quadriceps muscle according to
Judet is indicated in very severe cases of stiffness.
The surgical procedure is composed of two essential
steps:
• The first is the arthrolysis.
• The second is release of the quadriceps muscle.

39 Surgical Management of the Stiff Knee

447

a

b

Fig. 39.10 Recent rupture of the ACL with a mop tear

c

Stiffness and the ACL
Although the treatment is similar, one must make a distinction between two diagnostically different situations.

Fig. 39.9 When rectus tightness exists, hip extension will exacerbate
the limitation of knee flexion. The influence of rectus contracture can be
demonstrated by hip extension in either supine (a, right limb) or prone
(b, left limb) positions. Hip flexion reduces the influence of the rectus
tightness (c, right limb)

This intervention is very demanding and the risk for fractures because of devascularization is real. An isolated transection of the deep layer of the quad tendon can be an
alternative if the knee flexion is not influenced by the position of the hip.

(a) Recent rupture of the ACL with a mop tear (positioning
of the distal end of the ruptured ACL inside the
intercondylar notch). Arthroscopic resection should
­
result in full extension (Fig. 39.10). A reconstruction can
be indicated during the same surgical procedure.
(b) Cyclops syndrome after reconstruction of the ACL,
which is a fibrous soft tissue reaction in front of the
reconstructed ACL. Arthroscopic resection usually
results in full extension (Fig. 39.11). In general, we combine this procedure with a notchplasty. The notchplasty
can be done with a curved osteotome or a burr.
Not uncommonly, an osteophyte can be observed in front
of the tibial tunnel. This osteophyte can be easily removed.
Again, we commonly combined this procedure with a notchplasty to create a sufficiently large clearance for the graft.
If extension still cannot be obtained, a complete sectioning
or removal of the ACL reconstruction should be considered.
This decision is of course easier to make if the femoral and tibial
tunnels are malpositioned. A subsequent revision ACL reconstruction is not frequently necessary in this circumstance.

448

R Debarge et al.

The surgeon should follow up his patients in the early-­
postoperative period to observe the progression in the range
of motion. If he is confronted with a stiff knee within
3 months post surgery, he can consider an MUA after the
elimination of potential complications such as complex
regional pain syndrome or infection.
In cases in which the posterior cruciate ligament was
retained, sectioning of it can be an option. Increased constraint
is then advised, with change to an ultracongruent polyethylene
insert, or in some cases, change to a posterior cruciate substituting design for both the femoral component and insert.

Arthroscopic Arthrolysis (Fig. 39.12a–c)

Fig. 39.11 Resection of a fibrous soft tissue reaction in front of the
reconstructed ACL (cyclops syndrome)

Stiffness After Total Knee Arthroplasty
General Information
Stiffness after a total knee arthroplasty is not an uncommon
phenomenon (10–15%). It is important to determine the
origin of the stiffness, its extent and impact on function in
order to institute the appropriate treatment. Stiffness after a
total knee arthroplasty can be defined as a flexion less than
90°, or an extension deficit greater than 10° irrespective of
the type of knee prosthesis.
Four surgical procedures are available:
• Manipulation under anesthesia
• Arthroscopic arthrolysis (attention should be paid not to
damage to the prosthetic surfaces)
• Open arthrolysis
• Revision of the prosthesis
Our therapeutic approach:
• Well-positioned implant
–– between days 15 and 90: Manipulation under
anesthesia
–– between days 90 and 180: Arthroscopic arthrolysis
–– after day 180: Open arthrolysis
• Malpositioned implant
–– revision of the component(s)

Stiff total knee arthroplasties can be treated arthroscopically
by sectioning the adhesions in the supra patella pouch, the
condylar gutters, and the space in front of the knee. The
articulating surfaces of the total knee arthroplasty should not
be damaged. In some cases, it could be necessary to perform
a medial and lateral patellar retinaculum release to obtain a
better flexion.

Adjuvant Treatments and Postoperative
Guidelines
Full weight bearing is allowed with a brace in full extension
(for 3–5 days).
Low molecular weight heparins are not prescribed.
However, strong pain medication and muscle relaxants are
indicated.

Flexion Deficit
At the end of the mobilization, a specifically designed 90°
cushion is applied to the knee for the first postoperative night
and continued every 6 h thereafter (Fig. 39.5).
Continuous passive motion is started on the second or
third postoperative day (Fig. 39.13).

Extension Deficit
Routinely a brace in extension is applied to the knee before
the end of the anesthesia. This brace should be worn initially
until the next morning and should be continued during the
night for another 5–10 nights, depending on the progression
and the result that was obtained during surgery.

39 Surgical Management of the Stiff Knee

449

a

b

c

Fig. 39.12 (a–c) Arthroscopic arthrolysis after TKA

Fig. 39.13 Continuous passive motion

Lengthening of the Patella Tendon

40

G Demey, P Archbold, and P Neyret

Patella Infera
Patella infera with a Caton-Deschamps (C-D) index of 0.6–
0.8 is not uncommon. If symptomatic, a proximal transfer of
the anterior tibial tubercle (ATT) can be considered.
Patella infera (C-D index <0.6) occurs following surgery or trauma secondary to reflex sympathetic dystrophy
(complex regional pain syndrome). It presents with pain
out of proportion to the initial injury. An early diagnosis is
critical to obtaining a good outcome, and its occurrence
should be suspected in patients who are slow to rehabilitate, have reduced mobility of the patella and whose quadriceps fire late. Patients typically complain of a burning
pre-patellar pain, a “vice” like sensation, or sub-patellar
tightness. Descending stairs, prolonged sitting, and rising
to stand increase the pain. Flexion is also limited.
Osteopenia of the patella is seen on radiographs
(Fig. 40.1), and patellar infera occurs secondary to contracture of the patella tendon/retinaculum and quadriceps
hypotonia. The axial view may reveal the classical “sunset” (Fig. 40.2). If diagnosed early, it can be treated by
bracing the knee in 30° of flexion to put the patellar tendon under tension and rehabilitation with active quadriceps contractions.
Once a significant infera has occurred (index <0.6), it can
be treated by lengthening the patellar tendon or by proximalization of the ATT. Preoperatively the patellar height and ten-

don length can be accurately assessed from radiographs by
measuring the C-D index. This information can also be
obtained from magnetic resonance imaging (MRI). Tibial

G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
P Archbold
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

Fig. 40.1 Lateral radiograph of the knee showing patella infera

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_40

451

452

G Demey et al.

Fig. 40.2 Classical “sunset” sign on the axial radiograph

tubercle transfer is an easier procedure, but is less logical
than lengthening the tendon itself if the tendon is short. An
ATT transfer is generally insufficient to treat severe patellar
infera (C-D index <0.6).

Surgical Technique for Lengthening
the Patella Tendon
This technique is based on the principle of the sliding flap
first described by H. Dejour.

Fig. 40.3 Incising the patella tendon

Incision
A midline longitudinal incision is made taking into consideration any previous incisions.

Patellar Tendon Lengthening
The patellar tendon is exposed throughout its width and
length and the tendon incised vertically from the tip of the
patella to the tibial tubercle in the midline. The incision is
continued over the patella and onto the quadriceps tendon for
a distance of 2 cm (Fig. 40.3). The patellar tendon is released
from the retinaculum medially and laterally, and the retracted
fat pad is excised from the posterior aspect of the tendon.
The medial flap remains in continuity with the patella. It is
mobilized by releasing the medial half of the patella tendon
from the ATT with a minimum 2 cm periosteal flap (Fig. 40.4).
The incision medially is continued along the medial border of
patellar tendon up to the patella, completely mobilizing the flap.
The lateral flap is the lateral half of the patellar tendon.
The flap is left attached distally and elevated from the patella
with a periosteal flap and a half thickness strip of the quadriceps tendon. The previously made vertical incision in the
middle of the patellar tendon extends over the patellar and
onto the quadriceps tendon for 2 cm. Two centimeters above

Fig. 40.4 Mobilizing the medial flap—medial half of the patella tendon

40 Lengthening of the Patella Tendon

the top edge of the patella, the lateral aspect of the ­quadriceps
tendon and the vastus lateralis are incised transversely, half
of the thickness of the tendon. The lateral flap is then formed
by elevating half thickness quadriceps tendon and full thickness patellar periosteum, and then mobilizing the lateral strip
of the patella tendon. It is left attached on the tibial tuberosity (Fig. 40.5a, b). The elevation of the lateral flap completes
the release of the patella from the ATT (Fig. 40.6).
An intraoperative radiograph ensures the correct restoration of patellar height (true lateral at 30° of flexion) so as to
plan the position of the medial and lateral flaps in relation to
each other (Fig. 40.7). The lateral flap is fixed on the lateral
part of the proximal pole of the patella and the medial flap is
fixed to the ATT, both with two suture anchors (Figs. 40.8
and 40.9).
The sliding flap is sutured along its length, and reinforced
by a strip of PDS® (Fig. 40.10a, b). It is folded in half and
fixed to the tibial tuberosity using an Orthomed® staple. The
two strands are then sutured into a “V,” onto the patellar tendon, patella, and quadriceps tendon. The suturing is done at

a

453

60° of knee flexion to prevent shortening of the patellar tendon and the reoccurrence of a patella infera.
The closure often requires several small vertical incisions
in the medial capsule close to the medial collateral ligament
in order to lengthen the medial retinaculum.
The lateral retinaculum is left open.

Postoperative
Radiographs should be obtained (Fig. 40.11). There is no
specific physiotherapy protocol for lengthening of the patella
tendon. We recommend a protocol identical to that for acute
ruptures of the patellar tendon, the only difference being that
the knee should be immobilized in 60° of flexion to keep the
lengthened patella tendon in slight tension. Therefore in the
immediate postoperative period, the knee is immobilized on
a cushion at 60° of flexion (Fig. 40.12). We recommend
avoiding the use of anticoagulants in the postoperative period
to prevent hematoma and skin necrosis.

b

Fig. 40.5 Mobilization of the lateral flap. (a) The transverse cut is only 50% of the thickness of the tendon, leaving the deep 50% intact. (b) The
half thickness quadriceps flap is held by the forceps while the lateral periosteal sleeve is being raised from the patella

454

G Demey et al.

Fig. 40.6 The complete release of the patella

a

Fig. 40.7 The relative position of the medial and the lateral flaps is
planned—with the aid of intraoperative X-ray

b

Fig. 40.8 Suture anchors are used to fix the lateral flap to the patella and the medial flap to the tibial tuberosity. (a) Inserting the patellar anchor.
(b) Both anchors in situ and ready for suture repair

40 Lengthening of the Patella Tendon
Fig. 40.9 (a, b) Diagram to
show sliding flap technique
(right knee)

a

455

a

b

b

Fig. 40.10 (a, b) Side to side suture of the sliding flaps, and reinforcement with PDS tape

456

G Demey et al.

Fig. 40.12 Postoperative immobilization at 60° flexion

Fig. 40.11 Postoperative X-ray

Patella Infera Following TKA
Following TKA it is still unknown why patella infera causes
pain. In cases that require surgery, we do not recommend a
simple sliding flap due to the significant risk of rupture.
Instead, we perform a technique involving an extensor mechanism reconstruction as described previously.

Stiffness of the Knee: Release According
to Judet

41

H Hobbs, J Bruderer, G Demey, and P Neyret

We do not aim to accurately reproduce the technique that
was originally and precisely described by Robert and Jean
Judet, but rather to describe the procedure that we perform
when confronted with severe and permanent stiffness of the
knee. This is often secondary to a fracture of the distal
femur.
Stiffness of the knee can be defined as a limitation of the
range of motion of the knee. The absence of spontaneous
resolution is characteristic. An absolute value of the range
of motion should not be considered without looking at the
whole context.
Similarly, the procedure described below is not a “one
size fits all” operation. We find the different parts of the procedure need to be performed in different patients depending
on the etiology of the stiffness being treated. For example, in
cases of stiffness after a fracture, only a localized release of
subsequent scarring under the quadriceps may be indicated,
whereas in cases of chronic dislocation of the patella and
shortening of the entire extensor mechanism, a more complete elevation of the musculature is needed. Similarly, physical examination can inform this decision. Cases in which the
degree of knee flexion obtained does not depend on hip flexion or extension will likely not benefit from a rectus femoris
tenotomy (see Chap. 39).
Several factors should be considered:
• The range of motion and the patient’s activity level. Climbing
stairs requires a minimum of 90° of flexion (the initial goal
when performing a total knee arthroplasty). 120° is usually
sufficient for most everyday activities of daily living.

H Hobbs · J Bruderer
Centre Albert Trillat, Lyon, France
G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

• The evolution of the stiffness over time.
• The etiology (ligament surgery, peri-prosthetic or articular fracture)
• The anatomical structures involved.
• The influence of hip position on the degree of knee flexion that can be obtained.
All these elements will determine what management is
needed:
•
•
•
•
•
•

Conservative treatment
Mobilization under general anesthesia
Arthroscopic arthrolysis
Open arthrolysis via an arthrotomy (anterior and posterior)
Arthrolysis and release according to Judet
Arthrolysis and release according to Lobenhöffer

The release of the extensor mechanism, according to Judet,
is for stiffness of an extra-articular origin and is fortunately an
intervention that is less frequently performed today due to the
introduction of stronger fixation and more aggressive rehabilitation protocols following femur fractures. Nevertheless,
it must be discussed and is indicated in cases of severe stiffness, especially those associated with sequelae of fractures of
the femur or with short extensor mechanism (permanent dislocation of the patella). The release according to Lobenhöffer
is an alternative in cases of stiffness related to fracture or following TKA.

The Release of Quadriceps
Indication
The main indication is severe stiffness, resulting in loss of
flexion. This stiffness is often post-traumatic (following fixation of fractures of diaphyseal or distal femur fractures).
Two mechanisms contribute to the stiffness:

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_41

457

458

–– Intra-articular: it may be capsular or intra-capsular.
–– Extra-articular: adherence due to a previous external fixator, adherence of the quadriceps to the femur and fascia,
muscle shortening due to scarring, or adherence to the
skin.
In light of this, the operative technique has two fundamental components:
–– The arthrolysis
–– The quadriceps release
A second indication is when there is a permanent dislocation of the patella, which is always associated with a short
extensor mechanism. The release according to Judet lengthens the extensor mechanism and allows rotation of the quadriceps muscle to realign the patella.

Risks
This procedure is a technically demanding and painful intervention. Prolonged postoperative rehabilitation is required.
Patients should be warned of this.
We must ensure that there is proper consolidation of the
femur following fracture fixation and that there is no infectious process present. In cases of infection, we recommend
waiting for a period of at least 1 year after the infection has
been cleared.
Any retained hardware in the lower metaphysis may be
removed at the same time as the release. If there is an intramedullary nail present, it is usually left in situ to prevent
femur fracture following treatment.
Finally, we must pay special attention to the condition of
the skin and soft tissues prior to embarking on such a release.
a

H Hobbs et al.

Technique
The patient is placed in a supine position. A side post and a
distal post keep the knee at 60° of flexion (in practice this is
typically the maximum flexion possible). The procedure is
performed without a tourniquet. The leg and hip are free
draped from the iliac crest down (Fig. 41.1a, b).

Incision
The release of the extensor mechanism is performed using
several steps. At each stage, the effectiveness of the procedure is verified. If the flexion is inadequate, the procedure
continues. The objective is to achieve knee flexion with gravity from 100° to 120° with the hip flexed.
Two surgical approaches are needed. They take into
account the previous approaches used. The aim is to achieve
an intra-articular arthrolysis via an arthrotomy (or rarely an
arthroscopy) and an extra-articular release via a lateral (and
sometimes also medial) approach to the quadriceps.
The technique begins with a medial incision from the
medial tibial plateau to the medial edge of the patella, extending 3 cm above the patella to follow the margin of the vastus
medialis. The incision is approximately 10 cm long.
Arthrotomy for arthrolysis is made in the subvastus plane to
keep the continuity of the extensor mechanism.
The lateral approach extends from Gerdy’s tubercle proximally to the greater trochanter in the line of the anterior
fibers of the fascia lata.
An anterior approach to the hip, in line with the anterior
superior iliac spine, dissecting between the sartorius and tensor fascia lata is conducted if a section of the direct head of
the rectus femoris tendon is needed. It is sectioned under
direct vision.

b

Fig. 41.1 (a, b) Patient setup. Two incisions are needed for this procedure

41 Stiffness of the Knee: Release According to Judet

Arthrolysis
The arthrolysis is done before the extra-articular procedure.
We release the supra-patella pouch and both the medial and
lateral condylar gutters. The condylar recesses are released
with a knife following the anatomical insertions until normal
condyle is visible (after A. Trillat). The ligamentum mucosa
and fat pad are resected. Adhesions of the patella in the
trochlea are released with a scalpel. Lastly, a lateral release
of the patella is done.
This procedure alone may sometimes yield flexion
between 100° and 120°; if not, then the quadriceps release is
continued.
Quadriceps Release
There may be adhesions in a number of sites, and each is
addressed as necessary. The first aim is to achieve a release
of adhesions between the fascia lata and skin, remaining as

Fig. 41.2 Vastus lateralis and intermediate are released

a

Fig. 41.3 (a, b) Rectus femoris tendon section

459

close as possible to the fascia lata. Adhesions between the
fascia lata and quadriceps are then released gradually by dissection using a blunt periosteal elevator. Incision of the anterior edge of the fascia lata along its entire length at the
junction with the fascia of the vastus lateralis is made (possible transverse section of the fascia lata initially recommended). The vastus lateralis and intermediate are released
from their aponeurosis and linea aspera by using a knife and
not the blunt periosteal elevator (Fig. 41.2). This release
necessitates a careful dissection and hemostasis of the perforating vessels. The vastus lateralis and intermediate are
released from the femoral diaphysis remaining extra periosteal. The main tendon of the vastus lateralis is then sectioned
and released from the subtrochanteric ridge and the anterior
greater trochanter. With two Homan spreaders, the vastus
lateralis and intermedius are separated. The vastus medialis
is then freed off the femur. At this stage, if stiffness persists
one must look for a contracture at the distal end of the vastus
medialis, a contracture of the fascia lata, or an extremely
tight rectus femoris tendon.
In the presence of a significant medial contracture due to
medialis adherence, the medial skin incision can be extended,
respecting the skin bridge, and the arthrotomy extended
proximally into the quads, to mobilize the medialis. Risks,
however, are of skin necrosis and further quads weakness.
If there is a contracture of the fascia lata, a Z-plasty of the
tensor fascia lata is required.In the event of a retraction of the
rectus femoris tendon, a section is performed under direct
vision to avoid trauma to the femoral nerve (Fig. 41.3a, b).
This completes the Judet procedure (Fig. 41.4). The knee may
then be mobilized and the amount of flexion visualized.
After surgery, it is verified that the flexion obtained is at
least 100°, with the hip flexed. The flexion obtained at the
end of the operation is the maximum flexion that will be
obtained at the end of the rehabilitation with physiotherapy.
b

460

H Hobbs et al.

potential complication in the early postoperative period due
to numerous incisions. Pre-patellar hematoma formation can
worsen this risk and anticoagulation is avoided. Mechanical
deep venous thrombosis (DVT) prophylaxis is initiated and
ultrasound surveillance is used to exclude DVT.

Rehabilitation
Rehabilitation begins the same day of the surgery. The knee
is kept flexed on a pillow at 90° and taken through flexion
and extension exercises several times a day in order to prevent a recurrence of the flexion contracture. The pillow is
kept for 7 days and nights.
We use regular continuous passive motion for 3 weeks.
During the first 3 weeks, passive mobilization with stretching is used. After the third week, active movement with contraction of the quadriceps is begun.
Full weight bearing with crutches is allowed immediately
postoperatively as long as the quadriceps can actively lock
the knee in full extension. Crutches are advised for a minimum of 2 months postoperatively because devascularization
of the femur caused by the surgery weakens the femur, leading to a fracture risk. Contact sports are prohibited for 1 year.
The rehabilitation is long and lasts several months and
one can expect the flexion to continue to improve for months
after the surgery.

Lobenhoffer Algorithm
Fig. 41.4 Judet procedure. Skin incisions are depicted on the right,
and deep releases on the left

Spontaneous flexion of the knee with gravity (leg hanging) is
the final flexion that is most often obtained.
After careful hemostasis, closure is performed at 90° knee
flexion. Two suction drains are placed, one subfascial and
one intra-articular. The patella retinaculum is left open. The
subcutaneous tissue and the skin are closed carefully. If necessary, judicious undermining of the skin and the subcutaneous tissue is done to reduce the tension on the skin sutures.

Postoperative Care
The knee is placed on a pillow at 90° flexion before leaving
the operating room. The position of the knee is alternated
between flexion and extension every 6 h. We avoid periods of
flexion longer than this in order to maintain the perfusion of
the anterior skin. Analgesia should be sufficient to allow
optimal rehabilitation. Epidural anesthesia, femoral and sciatic blocks, and morphine PCA pump are all very important
in the early postoperative period. Skin necrosis is a major

An alternative to the release according to Judet is to follow
the algorithm of P. Lobenhöffer. It also differentiates intra-­
articular from extra-articular stiffness. Treatment of the
intra-articular pathology is performed by arthrotomy or
arthroscopic arthrolysis. The treatment of the extra-articular
pathology is via a resection of the vastus intermedius in cases
of fibrosis or if the patella is low, a lengthening of the patellar
tendon or proximal transfer of the tibial tubercle.

 esection of the Vastus Intermedius
R
According to Thompson
If the stiffness is caused by fibrosis of the vastus intermedius,
the patella cannot move distally during flexion of the knee,
and this finding is not altered by hip position. This fibrosis is
visualized on MRI. The Judet release would be ineffective in
such a case.
An intra-articular arthrolysis is initially done. The patella
retinaculum is then released starting anterior medially. A
long lateral incision is then made to release the vastus lateralis and rectus femoris. Under the latter, the fibrosed vastus

41 Stiffness of the Knee: Release According to Judet

intermedius is identified. The fibrosed muscle is resected,
from the insertion of the muscle on the patella to its origin.
Closure is done in flexion. The vastus lateralis and vastus
medialis are sutured anteriorly to the rectus femoris in maximal flexion. The patella retinaculum is closed, being aware
that you may need to release it again if necessary, so as not to
close the wound under any tension. At least two suction
drains are placed in situ.

461

We have now more experience with this procedure, and
we believe that it does have a place in stiffness of the knee
caused by extra-articular pathology. We have found it particularly useful in addressing significant stiffness due to soft
tissue contracture at least 1 year following TKA or supracondylar fracture.

Principles of Knee Surgery: Case
Examples

42

P Neyret and C Butcher

Although much of the knee surgeon’s workload consists of
a small number of routine case types, occasionally he/she
is faced with an unusual and more complex case for which
the usual treatment protocols do not supply sufficient guidance. There is often a history of trauma, and/or previous
surgical intervention, and in these cases the surgeon will
need to draw on some dependable surgical principles to be
able to form a safe and effective treatment plan. Quite
often these guiding principles have not been written down,
but are acquired with time through the surgeon’s own
experience, or gleaned from others over the years of their
hospital practice.
Fortunate is the one who whilst operating has the eye of his
boss over his shoulder

The challenge may be to decide what to do (if anything at
all), and perhaps equally difficult, when to do it. Many factors
affect our decision-making—cultural, legal or regulatory,
technical and personal (both surgeon and patient). With the
routine cases, there is precedent for a particular treatment
method, familiar to those within the hospital and those without. However, what is appropriate in one religion, jurisdiction, or hospital may not be in another. So judging the
treatment of a routine case performed in another geographical
area often requires some lateral thinking. But the unusual
case provides the greatest potential options and it is with
these cases that we need to think most widely. Sometimes the
result of the intervention is satisfactory, and sometimes not,
but in both cases there will be useful learning points. In this

P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com
C Butcher
Healthpoint, Abu Dhabi, UAE

chapter, we present a series of such cases and outline some of
these lessons and principles which can help the surgeon navigate through what may otherwise appear uncharted territory.
Fools learn from experience. I prefer to learn from the
experience of others (Otto von Bismark)

Case 1. UKA to UKA/HTO. UKA/HTO to TKA
Presentation
A 55-year-old female presented with knee pain and malalignment after multiple knee surgery (Fig. 42.1).
The past history included a closing wedge valgus osteotomy and, 2 years later, a medial UKA.
There was a satisfactory balance after the UKA, but
uncorrected extra-articular valgus.

Management
A decision was made to revise the tibial component of the
UKA with a minimal tibial cut, and perform a medial closing
wedge osteotomy beneath, to correct the extra-articular
deformity (Fig. 42.2). The alternative option was to revise to
a TKA. The advantage would be a more certain result than a
UKA to UKA revision but this intra-articular method of
realignment would create medial laxity.
Resulting alignment after the revision was satisfactory
(with a slightly varus tibial component), and again satisfactory balance and laxity. However, the patient experienced
persistent medial pain, and at removal of the plate, avascular
necrosis of the medial tibia was confirmed (Fig. 42.3).
A conversion to TKA followed, and although this required
augmentation for the medial bone defect, the procedure was
straightforward, as the extra-articular deformity had been

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_42

463

464

P Neyret and C Butcher

corrected in the first revision surgery (Fig. 42.4). A satisfactory clinical result was finally achieved.

Principles and Critique
• The ideal use for a (primary) UKA is as a spacer in a
joint where the malalignment arises from the intra-articular wear.
a

• In the first unicompartmental arthroplasty, the pre-­existing
extra-articular deformity was not corrected. It would have
been difficult to correct this intra-articularly in a subsequent procedure without causing medial laxity.
• Successful surgery requires both mechanical restoration
and good biology. The first unicompartmental arthroplasty failed for mechanical reasons. However, in the
revision unicompartmental arthroplasty, despite correction of the mechanical issue, a biological complication

b

c

Fig. 42.1 (a–c) Plain X-rays at presentation. There is valgus, mostly metaphyseal

a

b

c

Fig. 42.2 (a–e) Revision UKA to UKA, and simultaneous closing wedge medial HTO. The femoral component was not revised. Satisfactory
overall alignment, albeit with a slightly varus tibial component

42 Principles of Knee Surgery: Case Examples

d

e

465

resulted. The small size of the osteotomised segment may
have precluded good blood supply.

Cases 2a and 2b. TKA After HTO
Case 2a
Presentation
A 66-year-old gentleman presented with genu varum, medial
compartment osteoarthritis, and a history of HTO 6 years
previously (Fig. 42.5). There was significant proximal tibial
metaphyseal varus.
Management
A combined TKA and valgus opening wedge osteotomy was
performed (Fig. 42.6).
The deformity correction was made away from the joint
with an osteotomy, rather than with the proximal tibial TKA
cut, and thus a release of the medial soft tissues was not necessary to achieve a good balance; a standard constraint was
possible.

Fig. 42.2 (Continued)

Fig. 42.3 (a) Due to
progressive pain, the plate and
staple were removed.
Avascular necrosis of the
proximal tibia was confirmed.
(b) Alignment was maintained

a

b

466
Fig. 42.4 (a–e) Revision to
TKA with a medial augment

P Neyret and C Butcher

a

b

c

d
e

42 Principles of Knee Surgery: Case Examples

a

b

467

c

Fig. 42.5 (a–c) X-rays at presentation, 6 years after HTO. There is significant metaphyseal varus

The osteotomy was performed distally, both to avoid the
previous more proximal osteotomy, and to allow translation
of the distal fragment and allow a narrow stem without cortical conflict, a problem common after HTO. This more distal osteotomy necessitated an elevation of the ATT.
A locked plate was used for fixation, which is now the
method of choice, and allows early weight bearing.

Case 2b
Presentation
A 70-year-old lady with genu varum and medial compartment osteoarthritis secondary to Paget’s disease (Fig. 42.7).
There was significant overall varus, more in the proximal
tibial metaphysis than the joint.
Management
A combined TKA and valgus opening wedge osteotomy was
performed (Fig. 42.8). The correction did not require the
release of the medial soft tissues, and a standard constraint

was used. Some residual varus was accepted in this elderly
patient (Fig. 42.9).

Cases 2a and 2b Principles and Critique
• It is important to distinguish intra-articular from extra-­
articular deformity. Correction of varus alignment at the
origin of deformity allows for a standard orthogonal tibial
cut without lateral resection laxity, and thus a more easily
balanced and stable knee.
• If correction is made in the joint, there is a maximum limit
to medial lengthening to compensate for the resulting lateral resection laxity (in our experience 8–10 mm).
• Leaving some residual varus reduces the asymmetry of
the resection gap, but may risk loosening. Variables in
deciding the amount of correction desired will then
include the age and activity of the patient.
• Preoperative analysis of the morphology of the proximal
tibia is required when performing TKA after HTO, to
avoid conflict of the keel and tibial cortex.
• A locked plate would be an option now.

468
Fig. 42.6 (a–d) Post-­
operative X-rays show good
alignment. Note lateral
translation of the tibia to
prevent stem/cortex conflict,
and fixation after tibial
tuberosity osteotomy

P Neyret and C Butcher

a

c

b

d

42 Principles of Knee Surgery: Case Examples

469

a

c

Fig. 42.7 (a–c) Clinical and radiological appearance. Paget’s disease and significant tibial varus

b

470

P Neyret and C Butcher

a

b

c

Fig. 42.8 (a–c) Intra-operative pictures showing elevation of the plateau prior to the tibial cut. Fixation with a metal wedge and a staple. A plate
would be used now

Fig. 42.9 (a, b) Post-­
operative X-rays show mild
residual varus. There is no
keel/cortex conflict; a longer
stem might or might not be
possible

a

b

42 Principles of Knee Surgery: Case Examples

471

 ase 3. Unconventional Fixation of Peri-­
C
prosthetic Fracture
Presentation
A 68-year-old lady presented after a fall at home sustaining
a mildly displaced peri-prosthetic tibial fracture (Fig. 42.10).
Past history included opening wedge HTO, and subsequent TKA 8 years later (6 years prior to injury).

Management
The tibial component was considered to be well fixed to the
proximal fragment, but this was too small to accept conventional plate stabilisation. A tension band construct was fashFig. 42.10 (a, b) A minimally displaced
peri-prosthetic fracture of the tibia

a

Fig. 42.11 (a, b) Six years after fixation.
The tibial component appears well fixed

a

ioned, and once the fracture was reduced anatomically, low
viscosity cement was injected around the keel.
The fracture went on to heal, and the prosthesis remained
well fixed 6 years later (Fig. 42.11).
An alternative would have been to revise the tibial component, but significant bone loss would be expected during
the explantation, complicating any further surgery.

Principles and Critique
• Basic principles of orthopaedic surgery and fixation can
be drawn upon to provide an unconventional solution.
• Although this treatment was successful, caution is
required in using unconventional treatment methods.
Equally, however, rigid adherence to common techniques
b

b

472

P Neyret and C Butcher

may not always be in the patient’s best interest. An open
mind is an asset to the thinking surgeon.

The measure of intelligence is the ability to change
(Albert Einstein)

 ase 4. Instability and Stiffness Following
C
Complex Trauma: A Less Invasive Treatment
Presentation
A 16-year-old girl presented with stiffness and subluxation
of the knee 6 months after a motorcycle accident (scooter
versus tractor). The tibial fracture was initially treated with

Fig. 42.12 (a, b) Initial tibial
injury and treatment with
external fixator

a

an external fixator, and secondary fixation of the tibial tuberosity followed some weeks later (Fig. 42.12).
Clinical and radiological assessment on presentation
revealed severe PCL insufficiency, limitation of movement
(0/10/70), healing of the fracture, but also evidence of injury
to posterior tibial artery and both peroneal and tibial nerves
(Fig. 42.13).

Management
An articulated external fixator was applied and a posterior and anterior arthrolysis performed (Fig. 42.14).
Intensive physiotherapy followed. Pin site infection,
common with external fixators, was not a problem on this
occasion.

b

42 Principles of Knee Surgery: Case Examples

473

a

b

c

d

e

Fig. 42.13 (a–e) Plain X-rays, angiogram, CT arthrogram and clinical situation on presentation 6 months after injury. These showed
evidence of chronic PCL insufficiency, and a posterior tibial vascular injury, but reasonable articular surfaces

474

P Neyret and C Butcher

A satisfactory functional result at 2 years post injury was
achieved, with range of motion and stability sufficient for
activities of daily living.
Clinically and radiologically the PCL insufficiency was
improved although patella infera was significant
(Fig. 42.15).

Principles and Critique
• One cannot accept a chronic posterior dislocation - function
is poor.
• Successful PCL reconstruction would require good preoperative movement to avoid recurrent stiffness.
• Simultaneous ligament reconstruction and arthrolysis
would also result in recurrent stiffness.
• Articulating fixators in the knee are an option.
• Consider the soft tissue envelope and perform the least invasive procedure when this is significantly compromised.
• The patellar infera potentially could be addressed to
reduce the risk of late pain, but would not necessarily
increase flexion due to the multiple sites of contracture.

Fig. 42.14 Reduction of the posterior tibial translation, and application of an articulated external fixator
Fig. 42.15 (a–d) Clinical and radiological result
2 years post-operatively. Range of motion has
been increased to functional levels. The posterior
stress view shows posterior translation, but less
than preoperatively. Patella infera is evident

a

There is not a situation sufficiently grave that the surgeon
cannot aggravate

b

c

d

42 Principles of Knee Surgery: Case Examples

475

 ase 5. Deformity and Instability Following
C
Complex Trauma: Osteotomy
Presentation
A 26-year-old gentleman presented with knee pain and valgus deformity 8 years after a motorbike accident. Initially he

Fig. 42.16 (a, b, c) Femoral and tibial fixation
seen on X-rays 8 years after injury. These reveal
5° femoral valgus, 14 mm shortening and neutral
tibial slope

a

c

had a floating knee, and fixation of both the femur and tibia
(Fig. 42.16).
Examination showed valgus deformity, 15° external rotational deformity of the femur, and posterior cruciate deficiency (Fig. 42.17).
X-rays revealed 5° valgus, limb shortening of 11 mm, and
a neutral tibial slope (Fig. 42.16).

b

476

P Neyret and C Butcher

Management
A combined femoral and tibial osteotomy was performed
(Fig. 42.18). The femoral was an opening wedge lateral distal femoral osteotomy correcting the valgus and rotational

Fig. 42.17 (a–c) Valgus,
external rotational deformity,
and posterior sag

a

deformities. The tibial was in the form of an anterior opening
wedge osteotomy.
At 1 year review, he was found to be asymptomatic, and
despite a posterior drawer of around 10 mm, he did not
require PCL reconstruction (Fig. 42.19).

b

c

Fig. 42.18 (a, b)
Radiological picture after
femoral and tibial osteotomy
showing symmetrical coronal
alignment and improved slope

a

b

42 Principles of Knee Surgery: Case Examples
Fig. 42.19 (a, b) Clinical
result 1 year after osteotomy

a

Principles and Critique
• Analysis of multiplanar deformity is challenging. There
are sophisticated computerised methods, but these are not
available to most patients. Analysis of one plane deformity is easy, two plane more difficult, and three plane
deformity even more so. Here, the deformities were dealt
with in one sitting, but staging the procedures is an option
to prioritise one deformity correction and improve the
analysis of each deformity.
• Sagittal deformity analysis is different from coronal. The
dynamic assessment of knee joint movement is key in
deciding the parameters of the correction. For instance, if
joint movement is not considered, appropriate correction
of a sagittal bony deformity may result in an unacceptable
flexion deformity.
• Rotational deformities are key in the development of
osteoarthritis. For instance, a 10° external rotational
deformity of the femur will have a greater effect than a
varus deformity of the same magnitude.
• Options include correction in the diaphysis or the metaphysis (at the origin of deformity or elsewhere), and internal
fixation with plates, intramedullary fixation or the use of
a frame. Each has its advantages and potential complications; in this case the methods of correction and fixation
were chosen to reduce the chance of length discrepancy,
non-union, and infection.
• Opening wedge osteotomies allow deformity correction
at multiple sites without further shortening.
• Alignment and stability of the knee joint are inter-related.
Sagittal stability of the knee is a function of both the soft
tissues (static and dynamic), and the bony/cartilaginous

477

b

architecture. In this case, once the sagittal tibial alignment
had been improved, PCL reconstruction was not required.
• The tibial fixation in this historical case was with staples,
and utilising a cement wedge. Current extra-medullary
fixation devices (locking plates) provide better stability
than previous types—these probably expand both the
indications for multiple corrections, and the role of
metaphyseal osteotomies.

 ase 6. Intra-articular Fracture: Lateral UKA
C
in a Young Patient
Presentation
A 29-year-old, 130 kg gentleman presented with lateral knee
pain and valgus deformity. Three years previously he had
sustained a lateral tibial plateau fracture which was initially
treated with internal fixation (Fig. 42.20). Radiology revealed
significant intra-articular valgus deformity due to articular
depression (Fig. 42.21).

Management
A decision to perform a lateral UKA was made. After the
appropriate tibial cut there was a peripheral defect, and a
plate was used as a direct buttress for the all polyethylene
tibial component (Figs. 42.22 and 42.23). This permitted
early weight bearing whilst allowing bone growth beneath.
At latest follow-up 9 years after surgery, he was symptom
free, with good ROM. Radiologically there was some deformation of the polyethylene, but no loosening (Fig. 42.24).
The mild medial joint narrowing was not symptomatic.

478

P Neyret and C Butcher

Fig. 42.20 (a, b) Initial
injury and fixation

a

Fig. 42.21 (a, b) Plain
X-rays and CT at presentation
3 years after injury, showing
valgus deformity and joint
depression

a

Principles and Critique
• Alignment correction is important in trauma cases, to
unload the involved compartment.
• The indication for UKA was stretched somewhat, with
bone loss beneath the tibial component. Support for a
prosthesis can be provided in a number of ways; through
augmentation with bone, cement or metal blocks, and in
the case of a TKA, a long stem. The method utilised in
this case of UKA was unconventional, but worked reason-

b

b

ably well in a challenging situation. Perhaps the 20-year
follow-up will provide the final conclusion.
• The default option would be to perform an intra-articular
osteotomy alone, and reduce into position the remains of
the joint surface. However, in this case, the joint was considered too damaged to salvage.
• A TKA was another option, but would have the disadvantages of bone loss, more complicated future revision, and
potentially lower function than a UKA.
• Age is an important factor in arthroplasty, but not the only
consideration.

42 Principles of Knee Surgery: Case Examples

a

479

b

d

c

Fig. 42.22 Intra-operative pictures showing (a) depression of the lateral tibial plateau. (b) Trials in place and peripheral defect. (c) Plate
applied at site of defect. (d) Plate directly supporting prosthesis

480

P Neyret and C Butcher

a

b

c

Fig. 42.23 (a–c) Post-operative radiological result

a

b

c

Fig. 42.24 (a–c) Radiological and clinical result 9 years after UKA

42 Principles of Knee Surgery: Case Examples

 ases 7a and 7b. Intra-articular Fracture:
C
Intra-articular Osteotomy

481

Case 7a

(Fig. 42.25). Plain films on presentation showed significant
malunion with depression of the medial condyle and resulting
varus (Fig. 42.26). A CT arthrogram showed some remaining
articular surfaces in the medial compartment (Fig. 42.27).

Presentation
A 38-year-old gentleman presented some years after injury
with pain and deformity. He had undergone ORIF of the tibia,
including a large posterior medial fragment, and the tibial spine

Management
A medial parapatellar approach was performed (using the
previous lateral incision), preserving some of the medial soft
tissue envelope (Fig. 42.28). The medial condyle was osteot-

a

b

c

d

Fig. 42.25 (a–d) Initial injury and fixation
Fig. 42.26 (a, b) X-rays at
presentation, some years after
injury. Malunion with
depression of the medial
condyle and resulting varus

a

b

482

P Neyret and C Butcher

a

b

c

Fig. 42.27 (a–c) CT arthrogram; some articular surface remains

a

d

b

c

e

f

Fig. 42.28 (a–d) Exposure preserving the medial envelope, followed by intra-articular osteotomy, elevation and fixation of the medial condyle.
(e, f) Radiological result

42 Principles of Knee Surgery: Case Examples
Fig. 42.29 (a, b) Clinically
good movement and mild
varus

a

483

a

b

b

c

Fig. 42.30 (a, b) Initial X-rays showing a bicondylar proximal tibial fracture. (c) An ORIF had been performed, but there was significant
malunion

omised, elevated, and stabilised with a locked plate. Despite
the poor articular surfaces, and mild remaining varus, he was
satisfied with the result at 5-year follow-up (Fig. 42.29).

Principles and Critique
• The risk of AVN is reduced by preserving the medial soft
tissue envelope.
• A posterior medial approach may be possible when malunion of the posterior medial condyle is the main problem. In
this case, the whole medial condyle required mobilisation.
• Realignment is the goal, but may be technically difficult
with an intra-articular osteotomy alone.

Case 7b
Presentation
A 49-year-old gentleman presented with pain and deformity
1 year after injury. An ORIF had been performed, but there
was significant malunion (Fig. 42.30).
Management
After anterior tibial tuberosity osteotomy for access, an
intra-articular osteotomy was performed, mobilising and fixing the lateral as well as the medial condyle (Figs. 42.31 and
42.32). The width of the condyles was reduced by excision
of bone in the intercondylar area. At 13-year follow-up,

484

P Neyret and C Butcher

a

b

c

Fig. 42.31 (a–c) Intra-operative pictures showing osteotomy and fixation of both condyles

Fig. 42.32 (a, b) Post-­
operative X-rays 6 years after
osteotomy

a

b

42 Principles of Knee Surgery: Case Examples

a

485

b

c

Fig. 42.33 (a–c) Thirteen years after osteotomy, reasonable joint preservation and good alignment

function and radiology were satisfactory, with good alignment (Fig. 42.33).

 ase 8. Lower Limb Torsion: Bilateral
C
Femoral and Tibial Osteotomy

Principles and Critique

Presentation

• Correct intra-articular deformity intra-articularly if
possible.
• Young patients with intra-articular trauma generally tolerate symptoms from cartilage damage if well aligned.
• Often the tibia becomes wider due to intercondylar malunion, and osteotomy may involve correcting the dimensions of the bone, as well as the alignment and axial
position of the fragments.

A 23-year-old lady presented with anterior knee pain whilst
walking, and especially when on inclines or during sports
such as water-skiing. A tibial tuberosity medialisation had
been carried out on the right side 5 years previously.
Clinical assessment showed mild bilateral varus, and
symmetrical external tibial torsion which was confirmed on
CT as 36° on the right and 40° on the left (Figs. 42.34 and
42.35). The femoral neck anteversion was 19° on the right

486
Fig. 42.34 (a) Clinical
appearance after right leg
surgery, and before left
surgery; varus alignment and
internally rotated patella on
the left. (b) X-ray prior to
surgery of both legs; knees
appear internally rotated,
whilst the ankles appear
anatomical

a

P Neyret and C Butcher

a

b

b

c

Fig. 42.35 (a–c) Axial CT shows external torsion of the tibia, anteversion of the femoral necks, and TT-TG distance of 2 mm on the right, and
5 mm on the left

42 Principles of Knee Surgery: Case Examples
Fig. 42.36 (a–e) Intra-­
operative photographs and
post-operative X-rays of left
knee

487

a

c

and 16° on the left. Patellae were objectively stable, there
was no trochlea dysplasia or patellar tilt (with and without
quads contraction), and TT-TG values were 2 mm on the
right and 5 mm on the left.

Management
A decision to perform staged derotational osteotomies of
both tibia and femurs was made. Femoral osteotomy was
performed first (derotating to neutral anteversion of the femoral neck), followed by tibial osteotomy 2 weeks later (correcting 10° of the internal tibial rotational deformity and

b

d

e

providing slight valgisation) (Fig. 42.36). The previously
medialised tibial tuberosity was lateralised to a more normal
position. The patient returned for the second side surgeries
2 years after the first.
At follow-up 8 years later the patient reported no knee
pain, and regular sporting activities.

Principles and Critique
• The indications for surgery in these cases are not well
defined, and the patient’s attitude towards their symptoms
and to the surgery and rehabilitation are key.

488

P Neyret and C Butcher

• Performing one’s own measurement rather than relying
on the radiologists report will ensure familiarity with the
case, improve accuracy, and provide a backup to avoid
gross errors. Not only the absolute figures must be confirmed, but the relevant direction!
• Performing staged procedures for multiple site correction
has advantages. Simultaneous analysis of the deformity in
more than one bone is challenging, and a re-evaluation of
second deformity can more easily be performed once one
first bone is normally aligned. The rehabilitation is not
affected significantly if the surgeries are performed
2 weeks apart.
• It is tempting to think that valgus osteotomy is always
required. Often it is not.
• The TT-TG is usually low in these cases. To keep the
TT-TG the same during the derotational tibial osteotomy,
the tuberosity is simply kept in same place in relation to
the proximal fragment.

Osteotomy is like salt or pepper in the kitchen. You have to add
enough, but not too much, and measurements are not everything

Fig. 42.37 (a) Ten degrees
valgus, in the distal femur.
(b, c) Absent patella on axial
cuts indicative of patella alta.
Patellar tilt and TT-TG of
23 mm

a

 ase 9. Episodic Patellar Dislocation:
C
Comprehensive Surgery
Presentation
A 40-year-old gentleman presented with a 30-year history of
recurrent right patellar instability.
Clinical assessment showed asymmetric valgus knee
alignment and a positive apprehension test. Long films
revealed femoral valgus of 10°, and CT showed a TT-TG of
23 mm, severe trochlear dysplasia, and patellar tilt (equal
with and without quadriceps contraction) (Fig. 42.37). The
Caton-Deschamps ratio was more than 1.2.

Management
A combined distal femoral osteotomy, trochleoplasty, tibial
tuberosity distalisation/medialisation, and MPFL reconstruction using quadriceps tendon was performed (Figs. 42.38 and
42.39).
Despite the considerable effort required to rehabilitate
from this extensive surgery, the patient returned for the same
surgery of the other limb.

b

c

42 Principles of Knee Surgery: Case Examples
Fig. 42.38 (a) Right knee;
viewed from distal. The tibial
tuberosity has been
osteotomised. The dysplasia
is obvious from this angle.
The trochleoplasty has been
marked out and is in process.
(b) The completed
trochleoplasty

a

Fig. 42.39 (a, b) Post-­
operative X-ray. Note the
staple fixation of the
trochleoplasty

a

489

b

b

490

P Neyret and C Butcher

Principles and Critique

Management

• Soft tissue reconstruction for patellar instability has become
a standard and popular procedure, due to the potential minimally invasive approach and relatively simple rehabilitation.
Correcting the bony contributions is more difficult for the
surgeon, and more so for the patient, due to the extent of the
surgical trauma, and the long period of rehabilitation. This
extreme case is at the far end of the spectrum, but there will
be many patients in between who may benefit from consideration of more than just the ruptured soft tissue restraints.

In the preoperative planning, the entry point for the intramedullary femoral guide was proposed at the exit point of the
original anatomic axis (Fig. 42.41). A 3° valgus distal femoral
cut was chosen (Fig. 42.42), anticipating a medial translation
of the femoral component to centre it on the condyles
(Fig. 42.43). The resulting alignment was only slightly varus,
and allowed for a balanced knee with minimal medial release.

Principles and Critique

• The operative options include extra-articular correction of
the deformity, with combined simultaneous or staged distal femoral osteotomy/TKA, or intra-articular correction
during a TKA. A combined procedure allows more norPresentation
mal bone cuts, and straightforward balancing, but at the
expense of a more extensive procedure with greater
A 78-year-old lady presented with night pain and restricted
potential complications and longer rehabilitation.
mobility. She had a history of femoral fracture treated non-­ • Conventional correction of the deformity during TKA
operatively some years earlier (Fig. 42.40).
using an orthogonal distal femoral cut would remove
Clinical and radiological assessment showed medial femmore lateral condyle than medial, creating lateral resecorotibial osteoarthritis, 25 mm shortening, and mechanical
tion laxity. A large medial release would be necessary to
varus of 12°, but there was no rotational deformity on CT
create a rectangular extension gap. Transferring this rectscans.
angular gap to the flexion gap would require internal rota-

 ase 10. Osteoarthritis After Varus Femoral
C
Malunion: TKA

a

b

c

Fig. 42.40 (a, b) Medial femorotibial osteoarthritis, metaphyseal/diaphyseal varus malunion, and HKA of 168°. (c) An orthogonal distal femoral
cut angle is shown, which would remove significantly more lateral bone than medial

42 Principles of Knee Surgery: Case Examples
Fig. 42.41 (a, b) Planning
and execution of the femoral
intramedullary guide entry
point

491

a

a

b

b

Fig. 42.42 (a) The distal femoral guide is set to 3° valgus, removing slightly more lateral than medial bone. (b) The posterior femoral cut is symmetrical - no rotation is required as the distal femoral cut is also symmetrical

tion of the femoral component, causing a trapezoidal
anterior gap (Fig. 42.44). The more desirable alternative
would be to rotate the femoral prosthesis correctly for the
anterior gap, and to use a constrained prosthesis to control
the stability of the femorotibial joint (Fig. 42.45).

• In this case, medial translation of the prosthesis was used
to reduce the varus limb alignment, allowing for reliable
intramedullary guidance, more symmetrical resection of
distal femoral bone, and less requirement for medial
release.

492

Fig. 42.43 Resulting alignment, mild varus

P Neyret and C Butcher

42 Principles of Knee Surgery: Case Examples

Fig. 42.44 (a) In this example, both femoral and tibial cuts would create lateral resection laxity. (b) MCL release will create a symmetric
extension gap. (c) Subsequent internal rotation of the femoral compo-

Fig. 42.45 Appropriate rotation for the patellar femoral joint. A flexion gap of this sort would require a constrained prosthesis to stabilise
the femorotibial joint

493

nent would be required to match the gaps in flexion and extension, but
the anterior gap would then be trapezoidal

Postoperative Complications

43

G Demey, R Magnussen, and P Neyret

In the postoperative period, the guidance from the surgeons,
anesthetist, physiotherapists, and general physicians should
have one common aim: to obtain the best possible result for
the patient, if possible in the shortest recovery period. Many
choices today are affected by this fast recovery period. The
management of the perioperative period is the result of good
team work, and all efforts are made to keep the patient
central.
We would therefore like to discuss some of our postoperative protocols. We would like to stress that these protocols
are open for discussion.

Thromboprophylaxis
Although the use of anticoagulants is widespread, its use is
‘pushed’ more by the industry itself than by scientific data.
We believe that the risk benefit for thromboprophylaxis is
less clear for knee surgery than it is for hip surgery.
A postoperative hematoma is a considerable complication
both for the patient and for the surgeon. A hematoma can
result in pain (complex regional pain syndrome), necrosis,
infection, and stiffness of the knee. The efficacy of thromboprophylaxis should not only be measured with biological
tests but should also be evaluated clinically.
In France, it is common practice to prescribe low dose
anticoagulants for a limited postoperative period. In our

G Demey
Clinique de la Sauvegarde, Lyon Ortho Clinic, Lyon, France
R Magnussen
Centre Albert Trillat, Lyon, France
P Neyret (*)
Infirmerie Protestante, Lyon, Caluire, France
e-mail: Philippe.neyret01@gmail.com

own experience, we have noted that the prescription period
is getting shorter as our experience with them increases. For
example, 0.3 mL of Fraxiparine (low molecular weight heparin) for 1 month after a total knee arthroplasty or an osteotomy (although 15 days could suffice) and 15 days for
ligament surgery and unicompartmental arthroplasty. In the
situation of a clinically significant hematoma, we do not
hesitate to interrupt the treatment for a couple of days while
we carefully review the situation. Because of the risk for
skin necrosis, we also reduce postoperative exercises when
a clinically significant hematoma occurs. Our preferred
choice for TKA is to prescribe oral anticoagulation for
1 month instead of low molecular weight heparin (we have
a long experience with Xarelto 10 mg a day). It avoids regular platelets control and also the need for injection. It cannot
be prescribed to patients at risk (e.g., disorientation, frequent trauma).
It is important to remember that thromboprophylaxis is
not only medical. Other preventive measures should be
undertaken: elevation of the limb, antiembolism stockings,
calf or foot pumps, mobilization of the ankle, early weight
bearing, and encouragement of ambulation.
If venous thrombosis is suspected, a venous ultrasound
examination is performed. The ultrasound provides valuable
information on the extent of the thrombosis. If the venous
thrombosis is situated at the level of the calf, we do not recommend complete anticoagulation. Postoperative venous
thrombosis should not be treated the same as a spontaneous
venous thrombosis since the etiology (e.g., mechanical, tourniquet, anterior dislocation of the tibia, anesthesia) is not the
same as for spontaneous thrombosis (e.g., hypercoagulation
syndrome, cancer). The treatment should be adapted and
should not cause more harm than the pathology itself. It is not
uncommon to see a recurrent thrombosis due to a hematoma
in the popliteal fossa secondary to the anticoagulation therapy
for the initial thrombosis. Although the risk of embolism

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6_43

495

496

exists, superficial venous thrombosis seldom results in fatal
embolism and severe complications. Therefore, we recommend Fraxiparine 0.3 ml twice a day, independent of the
weight of the patient for a period of 1 month and a repeat
ultrasound. Compression socks are prescribed, physiotherapy
is continued, and the patient is advised to remain ambulatory.
If the symptoms of venous thrombosis persist after a couple
of days, a repeat ultrasound is performed to rule out the extension of the thrombus. The cardiovascular risk is of major
importance and the prescription of anticoagulant medication
has to take this risk into account. The patient is informed of
the increased risk of a hematoma and the postoperative rehabilitation and medications are modified.
Although a proximal venous thrombosis, i.e., femoral or
popliteal, is a rare event (less than 1%), this condition is
potentially lethal. The patient should be fully anticoagulated
and postoperative rehab should be interrupted until full anticoagulation is achieved. The length of therapy is prolonged.

Pulmonary Embolism
Although the diagnosis of a pulmonary embolism (PE) is
frequently made, there should be distinction between an
embolic event due to a deep venous thrombosis and a fat
embolism. This is of critical importance as the latter results
in unnecessary anticoagulation with its potential complications. If a PE is suspected in the postoperative period, a
V/Q scan or spiral CT should be acquired urgently.
Characteristic perfusion defects are visualized. An ultrasound of the lower limb should be performed. If this study
is negative, the diagnosis of fat embolism should be made.
If the study is positive, a proximal deep venous thrombosis
is the more likely cause of the embolism. All investigations
should be considered in combination when making the final
decision on the therapeutic approach. The aim of this paragraph is to remind us of the frequency of fat embolism after
TKR which is frequently misinterpreted as being a deep
venous thrombosis.

Bleeding
As well as transfusions, intra-operative and postoperative
bleeding cause pain and may explain longer hospital stay
and subsequent hospitalization. The rate of blood transfusions after TKA should be below 5%, and if possible
around 1%. Pre-operative hemoglobin should be optimized, and anticoagulants stopped in consultation with
our medical colleagues. In order to reduce bleeding, we
use intra-operative intravenous and/or topical tranexamic

G Demey et al.

acid 1 g, with a further 1 g 3 h postoperatively. Our current preference is to deflate the tourniquet and achieve
hemostasis before closure.

Postoperative Pain
Anesthesiology has made remarkable progress in pain management. Pain management, however, is not the sole domain
of the anesthesiologist. A team strategy allows a multimodal
pain control.
In open surgery, particularly in TKA, a local injection can
decrease postoperative pain and bleeding. In TKA, it consists
of 100 mL injection of the deep soft tissue and one 50 mL
subcutaneous injection (total ropivacaine 300 mg, adrenaline
0.5 mg, and ketorolac 30 mg, made up to 150 mL with
saline).
In the case of postoperative pain resistant to classic analgesia including morphine, the surgeon should always
exclude:
• Hemarthrosis: consider interruption of anticoagulant,
decelerated rehabilitation, aspiration, and cryotherapy
• Excessively tight dressings
But more importantly:
•
•
•
•

Vascular injury
Compartment syndrome
Skin necrosis
Infection

Infection
Infection should always be excluded in the early or late postoperative period and should not go unrecognized. The treatment of an infection also includes a multidisciplinary
approach. This does not mean that the infectious disease
physician should be the sole treating physician of the patient.
The surgeon has to decide whether to perform an arthrotomy
(which we prefer above an arthroscopy) with or without
removal of the hardware (osteosynthesis or prosthesis) and
exchange of polyethylene liner. In our experience, immobilization plays as an important role in the treatment of
infection.
Prevention is the key. We just wish to highlight a few
essential points: no touch technique, regular change of
gloves, control of operating room numbers and circulation,
and of course an adapted flash antibiotic prophylaxis (accord-

43 Postoperative Complications

ing to the weight and the institutional preference).
Intra-­articular antibiotics during the procedure may play a
role. An annual meeting including anesthesiologist, infection
specialist, and surgeon is mandatory to define the protocol.

Skin Problems
The skin incisions and the wounds should be inspected regularly to observe and prevent skin problems. Certain predisposing conditions are well known: multiple incisions and
diabetes.
In these situations, surveillance needs to be performed
more frequently. If there is concern, vasodilators can be
used, flexion should be limited or even stopped. Cryotherapy
should be limited. In case of skin necrosis, the wound

497

should be covered. Skin flap and gastrocnemius flap surgery to cover the wound can be done at an early stage, so
early consultation with our plastic surgery colleagues is
critical.
Besides the more serious complications, unsightly scars
and skin pigmentation problems should not be discarded as
unimportant. In certain patients, a skin problem can sometimes be considered a real complication (large scar, colored
scar, keloid). More frequently observed in a young girl after
surgery for episodic dislocation of the patella, these problems sometimes necessitate referral to a plastic surgeon.
Skin pigmentation is often observed as a consequence of a
hemarthrosis or infiltration of the soft tissues with blood.
This problem catches the attention of the patient although the
surgeon generally has a tendency to neglect it, and a sensitive
approach is rewarded.

Index

A
Acute patellar tendon ruptures, 415
indication, 415
physical examination, 415
surgical repair technique, 415
Fiberwire® sutures, 416
longitudinal paramedian incision, 415
PDS® tape, 416, 417
quadriceps graft usage, 416
semitendinosus graft suture and reinforcement, 416, 417
Acute quadriceps tendon rupture, 409
fiber by fiber repair, 409
Fiberwire® sutures, 410
indication, 409
osteotendinous tear, 411
patient positioning and setup, 409, 410
physiotherapy, 411
prophylactic antibiotics, 411
Anterior closing-wedge osteotomy, 82
Anterior cruciate ligament (ACL), 15
Anterior cruciate ligament laxity
OA secondary to, 32
with pre-arthritis, 32
Anterior cruciate ligament reconstruction
and anterior tibial closing osteotomy, 81–84
arthroscopic exploration, 37
autograft shape and size, 36
bone blocks drilling, 34
classification, 31
closed tendon stripper, 61
diagnosis, 57
double blade scalpel, 34
endobutton device, 62
error
incorrect tunnel position, 71–72
poor fixation, 72
poor graft quality, 72
femoral tunnel length, 66
gracilis and semitendinosus tendons, 60
graft choice, 57
graft pretensionning, 64
hamstring tendon graft, 57
with high tibial osteotomy, 55–56
imaging, 57
incision landmarks, 34
intra-articular position, tibial aimer, 67
and posterolateral deficiency, 31
reefing, posteromedial soft tissues, 84
revision (see Revision anterior cruciate ligament reconstruction)
Rigidfix tunnels, 67
routine set-up, 33

Sartorius fascia, 60
semitendinosus tendon, 59, 60
with six-strand hamstring tendon graft, 57
stiffness of knee and, 447
surface markings, 58
surgical technique
anesthesia, 57
arthroscopic portal placement, 64
closure, 70
complications, 70
endobutton CL length and final graft preparation, 68
femoral tunnel, 65–67
graft fixation, 67–68
graft passage and femoral fixation, 68
graft tensioning, 68
hamstring tendon graft harvest, 58–61
notch preparation, 64–65
patient position, 57
postoperative care, 70
six bundled hamstring tendon graft
preparation, 61–64
tibial fixation, 70
tibial tunnel, 67
technique, 110
tendon harvest, 61
tibial bone block cutting, 35
treatment, 31
triple-bundle gracilis tendon, 63
UKA implant, osteoarthritis, 33
and valgus-producing high tibial osteotomy, 80–81
Anterior tibial closing osteotomy, 84
Anterior tibial tuberosity (ATT) osteotomy, 313, 314
Anterolateral capsular structures, 308–309
Anteromedial portal
inferior, 8
superior, 8
Anti-recurvatum osteotomies, 145
Arthrolysis
arthroscopic, 448, 449
stiffness of knee, 459
Arthroscopic arthrolysis, after TKA, 448, 449
Arthroscopy, knee
complications, 15–16
instrumentation, 6
intercondylar notch, 11–14
lateral tibiofemoral compartment, 11
medial tibiofemoral compartment, 10–11
patellofemoral compartment, 10
patient positioning, 5–6
patient preparation, 5
portals, 7–8

© Springer Nature Switzerland AG 2020
P Neyret et al. (eds.), Surgery of the Knee, https://doi.org/10.1007/978-3-030-19073-6

499

500
Arthroscopy, knee (cont.)
anterolateral portal, 8
inferior anteromedial portal, 8
lateral parapatellar portal of patel, 9–10
posterolateral portal, 9
posteromedial portal, 8–9
superior anteromedial portal, 8
superolateral portal, 8
superomedial portal, 8
postoperative care, 14–15
surgical instruments, 6–7
and unicompartmental knee arthroplasty, 360–362
Autologous chondrocyte implantation, 129
B
Bayonet sign, 382
Bi-compartmental UKA
arthroscopy, 220
CT scan, 220
femoral side, 222–223
indications, 219
laboratory tests, 220
lateral UKA after medial UKA, 223–224
medial UKA after lateral UKA, 221–223
patient selection, 219
postoperative care, 224
radiological evaluation, 219–220
scintigraphy, 220
skin appearance, 225
surgical technique, 221
tibial side, 222
trials and implant fixation, 223
Bicruciate injuries, see Pentades
Bleeding, 496
Blumensaat’s line, 382
Bone balancing, 373
Bone blocks
calibration and adjustment, 36
patellar, 33–34
perforation of, 34
tibial, 35
Bone patellar tendon bone graft, 33, 34, 36–37, 42, 47, 48, 55
Bony tunnel under Gerdy’s tubercle, 50
C
Cartipatch technique, 129
Cartipatch®
calibrated drill bit, 130
graft insertion, 130
14 mm diameter grafts, 129
single 18 mm diameter graft, 130
trial component, 130
Caton-Deschamps (C-D) index, 385, 392, 451
Cerclage wiring, of patella fracture, 436, 439
Chondral and osteochondral lesions
autologous chondrocyte transplantation
associated procedures, 130
chondrocytes cultures, 128
implantation, 128–130
cartilage transplantation, 125
conservative surgical techniques
drilling, 131–132
fixation, 130–131
diagnosis, 125
fibrocartilage, 125

Index
indications, 125
microfracture and abrasion, 125
osteochondral grafting and mosaicplasty
osteochondral allograft, 128
postoperative guidelines, 128
principles, 126
surgical technique, 126–128
postoperative guidelines, 132
preoperative planning, 125
subchondral bone perforation, 126
surgical techniques, 125–126
treatment, 125
Chondrocalcinosis, 311
Chondrocyte cell transplantation, 125
Chronic patellar tendon ruptures, 425
allograft transplantation for extensor mechanism, 432–434
autologous transplantation of extensor mechanism, 425
delineation of transplant, 426
donor site closure, 431
incision, 425–426
patellar graft fixation, 430
postoperative regimen, 431
recipient site closure, 431
recipient site preparation, 427–430
tibial and patellar bone block harvesting, 427
tibial tuberosity fixation, 430, 431
on TKA, 432
Chronic quadriceps tendon ruptures
reinforcement of, 421
incision, 421
using patellar tendon graft, 422, 423, 425
using semitendinosus graft, 422, 423, 425
suture technique protected by metal framing, 419
incision, 420
mobilizing proximal quadriceps tendon stump, 421
patient positioning and setup, 420
Closing wedge high tibial osteotomy, 151, 177
Closing wedge osteotomy, 148–149, 173
Closing wedge osteotomy of distal femur, 404
Computer-assisted total knee arthroplasty, 325–330
anteromedial and anterolateral approach, 326
closure and post-operative care, 331
final implant placement, 330
patient positioning and setup, 325
tibial and femoral sensors, 326
trials, 330
Condylar twist angle, 258
Constitutional varus deformity, 145, 230
Convex side laxity, 241
D
Deepening femoral trochleoplasty, see Trochleoplasty
Derotation osteotomy, 197
Diaphyseal deformity, 155
Dislocations, of knee
anterior, 108
classification, 105
of knee, 105, 106, 110, 112–114
operative timing, 110
posterior, 108
postoperative care, 114
reduction under general anesthesia, 106
rotational, 107
setup, 110
stress x-rays, 108
stripping capsular-periosteal detachments, 108

Index

501

Distal femoral condyle cut, 294–295
Distal femoral cut, 249–250
Distal femoral osteotomy, 193, 195
Distal opening wedge femoral osteotomy
blade, 156
diaphyseal deformity, 155
metaphyseal deformity, 155
post-operative X-rays, 157
preoperative planning, 153
radiological workup, 153
rotational deformity, 154
Schuss X-rays, 153
surgical technique
blade, 154
guide pins, 153
intra-operative control, 155
locked plate, 156–158
osteotomy, 155
post-operative guidelines, 158
Distal tibial tubercle transfer technique, 392–397
Don O’Donoghue’s triad, 105
Double osteotomy
complications, 193
indications, 193
postoperative guidelines, 198
principles
malunion with torsional problem, 195–196
valgus knee, 195
varus knee, 193–195
Dysesthesias, 15

Extra-articular deformity, 230
Extra-articular varus deformity of tibia, 241
Extra-medullary (EM) guide, 252

E
Epicondylar axes, 257
Epiphyseal axis, 144
Episodic patellar dislocation (EPD), 404
bayonet sign, 382
comprehensive surgery, 488–490
conservative treatment, 387
defined, 381
distal tibial tubercle transfer technique, 392, 393
grasshopper sign, 382
“J” or “comma” sign, 382
medial patellofemoral ligament reconstruction
arthroscopy, 387
medial epicondylar tunnel and dissection, 389
patellar tunnels, 388
semitendinosus tendon, 387
using quadriceps tendon autograft, 391, 392
medial tibial tubercle transfer, 396
medio patellar femoral ligament (MPFL) insufficiency, 381
patellar height, 382
patellar tendon tenodesis, 394
patellar tilt, 382
postoperative care, 397, 398
prophylactic antibiotics, 397
secondary factors, 382
Smillie test, 382
surgical complications, 398
surgical treatment, 387
tibial tubercle-trochlea groove (TT-TG) distance, 382, 384
trochlear dysplasia, 381–384, 401
Extra-articular augmentation
Kenneth Jones surgical technique, 33
Lemaire technique, 51–54
postoperative care, 56
variant with short graft, 54–55

G
Gap balancing, 237
Genu recurvatum
asymmetric, 187
bone grafts, 191
indications, 187
post-operative guidelines, 192
radiological measurements, 188
radiological workup, 187
skin incision, 189
surgical management, 187
technique, 187
tibial osteotomy, 190
tibial tubercle osteotomy, 189
Genu valgum, 328, 382
Genu varum, 159
Gerdy’s tubercle, 153
Graft impaction, 43
Graft-sizing block, 36

F
Fascia lata suture, 51
FastFix 360® implant, 22–26
Fat embolism, 496
Femoral osteotomy, 149, 151
Femoral rotation, in TKA, 257–259, 280, 281, 285, 295
Femoral space
creation of, 243–245
definition, 240
in TKA, 243–245
Femoral tunnel, 40, 49
drilling, 42
inspection, 41
Femoral valgus, 243
Femoro-tibial angle, 287
Fiberwire suture, 36
Fibular head avulsion, 112
Fibular head avulsion fracture fixation, 112
Fibular tunnel drilling, 101
Fixed flexion deformity (FFD)
correction, 307
pre-operative, 266, 267
in TKA, 307, 443
Flexion coronal alignment, in TKA, 257, 233, 235, 257
Flexion gap, alignment and balance, in TKA, 257–259

H
Hamiltonian cycle, 228
Hamstring tendons, 47
High tibial osteotomy (HTO), 150
valgus
biplanar osteotomy, 168
blade, 162, 163
bone wedge removal, 165
clinical outcome, 159
completed fixation, 166
cutting guide, 164
distal cut of osteotomy, 164
fibula holes, 162
fixation of osteotomy, 165

502
High tibial osteotomy (HTO) (cont.)
freehand distal cut, 164
guide pin, 161
hole drilling, 163
medial cortex, 165
medial opening wedge, 167–171
osteotomy, 162
osteotomy of neck of fibula, 160
patient set-up, 161
peroneal nerve, 162
postoperative guidelines, 171–172
postoperative X-ray, 166
proximal cut, 164
radiological evaluation, 159
reasons for failure, 159
superficial medial collateral ligament, 167
surgical techniques, 159
tibial cut, 170
tibial osteotomy, 160–161
tibialis anterior muscle, 161
varus (see Varus high tibial osteotomy)
Hip Knee femoral angles, 144
Hip rotation, lateral support, 2
I
Iatrogenic intra-articular injuries, 16
Iliotibial band (ITB)
harvesting, 51, 52
preparation, 52
Iliotibial band friction syndrome (ITBFS), 133
Iliotibial band syndrome (ITBS), 133
dissection, 134
imaging before surgery, 133–134
MRI, 134
posterior fibers, 134
postoperative care, 135
preoperative clinical evaluation, 133
surgical technique, 134–135
treatment strategy and indications, 133
Implant space and prosthetic interface, 235–236
Infection, 496
Inside-out technique using aiming cannulas (Henning), 26
Intercondylar notch, 11–14, 38
Intra-articular deformity, 229
Intra-articular fracture, 481, 483
Intra-articular osteotomy, 483, 485
Intra-medullary (IM) guide, 252
K
Keel impingement, 254
Keelboat, 264
Kenneth Jones surgical technique
arthroscopy, 37–46
bone blocks harvest, 35
bone blocks preparation, 34
bone patellar tendon bone graft, 42–44
harvest, 33–36
preparation, 36
femoral tunnel, 38, 39
graft fixation, 44–45
intercondylar notch preparation, 37–38
KJ modification(KJT/KJG), 46–51
notch plasty, 38
patella tendon part preparation, 34
patellar bone block, 35

Index
postoperative, 46
setup and clinical examination, 33
tibial bone block, 35
tibial tunnel, 40–42
Knee motion, range of, 441
L
Lateral closing wedge osteotomy, varus knee, 196
Lateral collateral ligament (LCL)
dissection, 52, 121
femoral tunnel, 101–102
fibular tunnel, 100–101
graft, 54
vertical arthrotomy, 121
Lateral collateral ligament (LCL) injuries
clinical examination, 110
femoral avulsion, 111–112
fibular head avulsion, 112–113
mid-substance ligament rupture, 113
posterolateral reconstructions, 113
postoperative care, 113
surgical exposure, 111
Lateral compartment OA, with lateral condyle hypoplasia, 281
Lateral condylar hypoplasia, 245
Lateral condyle osteotomy, 341, 343, 345, 347
Lateral epicondyle, 133–135
guided pin placement, 101
Lateral extra-articular tenodesis, 84–85
Lateral facetectomy, 201
Lateral femoral condyle osteotomy, 283
Lateral laxity in flexion, 257
Lateral opening wedge osteotomy, 173
valgus knee, 196
Lateral opening wedge tibial osteotomy
Farabeuf retractor, 182
fibular neck osteotomy, 182
indications, 179
surgical technique
complications, 183
incision, 179
osteosynthesis, 184
osteotomy of neck of fibula, 179–180
osteotomy of tibia, 180–184
postoperative guidelines, 184–186
set up, 179
Lateral osteoarthritis, 149–150
Lateral unicompartmental knee arthroplasty
intra-articular fracture, 477–478
after medial UKA, 223–224
surgical technique
approach, 215
associated procedures, 216
femoral resurfacing, 215–216
tibial cut, 215
Lateral unicompartmental knee arthroplasty revision
cemented implants, 360, 361
femur preparation, 359
ligament balancing with trial implants, 359
postoperative rehabilitation, 360
skin incision, 357
tibia preparation, 357
Leg preparation, surgery, 2
Lemaire extra-articular augmentation
anterior cruciate ligament reconstruction, 51
approach, 51
fascia lata graft harvest and preparation, 51

Index
femoral tunnel preparation, 52
graft passage, 53
lateral augmentation fixation, 54
patient positioning, 51
postoperative care, 54
tibial tunnel preparation, 53
Lemaire extra-articular tenodesis, 55
Lepine HTO plate, 161
Ligament balancing, 237
Lobenhöffer algorithm, 460
Lower limb torsion, bilateral femoral and tibial osteotomy, 485–488
M
Magnetic resonance imaging (MRI), 17, 75, 108, 110, 114
Manipulation under anesthesia (MUA), stiff knee, 443–444
Medial collateral ligament (MCL)
distal release, 309
pie crust, 307
release of, 305–307
in TKA, 289, 365, 375
Measured resection strategy, in TKA, 231, 243, 266, 268
Medial compartment osteoarthritis, 207
in knee with no extra-articular deformity, 274
in knee with varus deformity of proximal femoral origin, 279
in knee with varus deformity of tibial origin, 276
Medial knee structures reconstruction
arthroscopy, 113
bony avulsions repair, 114
MCL reconstruction, 114
mid-substance rupture, 114
type of injury, 113
Medial midvastus arthrotomy, 209
Medial opening wedge high tibial osteotomy
set-up, 167–171
skin incision, 167
Medial osteoarthritis, 147–149, 151
Medial parapatellar arthrotomy, 119
Medial patellofemoral ligament reconstruction, episodic patellar
dislocation
arthroscopy, 387
medial epicondylar tunnel and dissection, 389, 390
patellar tunnels, 388, 389
using quadriceps tendon autograft, 391–392
semitendinosus tendon, 387, 388
Medial tibial tubercle transfer, episodic patellar dislocation, 396
Medial tibiofemoral compartment, 10–11
Medial unicompartmental knee arthroplasty
after lateral UKA, 221–223
surgical technique
approach, 208
femoral resurfacing, 211–214
implantation, 214
set up, 206–214
tibial cut, 208
Medial unicompartmental knee arthroplasty revision
femoral cut, 353, 356
filling bone defects, 355
final implants, 356
ligament balancing, 355
skin incision, 351
tibial cut, 351, 352
trial implant positioning, 355
Meniscal ramp lesion, 85
Meniscal repair
contraindications to, 21
postoperative care, 29

503
Meniscal root tear, 29
Meniscectomy, 7
anterior horn suture, 20
bucket handle tear, 19
degenerative tear, 18
lateral, 19
medial, 17
meniscal cyst, 19
partial, 21
partial medial, 17
peripheral posterior tear, 19
total, 18
Menisci
healing, 21
ISAKOS classification, 21
lateral tear, 21
red-red zone, 21
red-white zone, 21
structure, 21
suturing techniques, 21
white-white zone, 21
Merchant angle, 385, 386
Metaphyseal deformity, 155
Mid-substance ligament rupture, 113
Mosaicplasty
final aspect, 128
first osteochondral plug, 127
graded adjustable plunger, 127
osteochondral plug insertion, 127
osteochondral plug sizing, 127
specific calibrated drill, 126
N
Navigation systems, 325
O
Objective patellar instability (see Episodic patellar dislocation (EPD))
Occasional patellar dislocation (see Episodic patellar
dislocation(EPD))
Open total synovectomy, 118
Opening wedge osteotomy
advantages, 148
disadvantages, 148
distal femoral osteotomy, 149–150
high tibial osteotomy, 151
Osteoarthritis
anatomic factors, 139
clinical parameters, 139
functional envelope, 140
functional outcomes, 140–141
indications, 141–144
lateral, 149–150
medial, 147–149
patient’s satisfaction following surgery, 139
prostheses, 139
Osteochondral allograft, 128
Osteophytes, posterior, in TKA, 260, 267, 270, 289, 294, 295, 299,
307, 311, 326, 345
Osteosynthesis, 169–171
Osteotomy
anatomic factors, 147
clinical and radiological criteria, 150
closing wedge, 148–149
disputable indications, 141
expected postoperative level of activity, 151

504
Osteotomy (cont.)
fixation with staples, 32
goal, 147
indications, 141
opening wedge, 148
opening wedge distal femoral, 149–150
radiological findings, 141
Outside-in technique with a loop (Warren), 27
P
Partial patellectomy, for patella fracture, 437–439
Patella
alta, 384–385, 387, 392, 415
bone block detachment, 35
cut, 296–299
extreme wear of, 311
femoral/anterior gap, 234
height, 382
preparation difficulties, 309
Patella fracture
anatomical reduction, 436
displaced, 436
early rehabilitation, 435
indication, 435
non-operative management, 435
osteosynthesis methods, 436
partial patellectomy, 437–439
postoperative rehabilitation, 437
surgical incisions, 435
total patellectomy, 439, 440
treatment aims, 435
Patella infera, 451
after TKA, 456
Patella tendon lengthening
midline longitudinal incision, 452
postoperative period, 453
sliding flap principle, 452
surgical technique, 452, 453
Patella tendon shortening
FiberWire ®nonabsorbable suture, 406
midline/parapatellar incision, 405
patella tendon sheet elevation, 406
postoperative management, 406
and repair, 406, 407
tendon preparation, 405
Patellar tendon
bone autograft, 31
graft, 37
length, 386
medial and lateral side identification, 34
tenodesis, episodic patellar dislocation, 394
Patellar tilt, 382, 385
Patellofemoral arthritis (PFA)
anteriorization, 201
arthroplasty, 201
complications, 203
distalization, 201
incidence, 199
lateral patellar release, 200
medialization, 201
non-arthroplasty surgery, 200–201
postoperative guidelines, 202–203
radiographic workup, 199
surgical technique, 201–202
tibial tubercle osteotomy, 200–201
treatment options, 199

Index
Patellofemoral arthroplasty, 201
Patellofemoral compartment, 10, 11
Patient positioning
acute quadriceps tendon rupture, 409, 410
anterior cruciate ligament reconstruction, 57
chronic quadriceps tendon ruptures, 420
computer-assisted total knee arthroplasty, 325
knee arthroscopy, 6
Lemaire extra-articular augmentation, 51
posterior cruciate ligament reconstruction, 88–89
posterolateral corner injuries, 99
revision TKA, 366
robotic assisted unicompartmental knee arthroplasty, 333, 334
Judet release technique, 458
and setup, general principals of, 1
Pentades
classification, 107
lateral, 105, 106
posterior, 105
progressive force application, 106
types, 105
Peri-prosthetic tibial fracture, unconventional fixation of, 471–472
Peroneal nerve, 101
Pie crust technique, 307, 320–323, 345, 355, 360, 375
Pigmented villonodular synovitis, 118, 124
PLEOS navigation system, 325
Popliteus/popliteal tendon, 14
grafts, 103
LCL and, 100
Popliteus tendon reconstruction
distal tunnel, 103
femoral tunnel, 102–103
graft, 102
Portals, knee arthroscopy
anterolateral portal, 8
anteromedial portals, 8
lateral parapatellar portal of Patel, 9–10
posterolateral portal, 9
posteromedial portal, 8–9
superolateral portal, 8
superomedial portal, 8
triangulation, 7–8
Posterior and anterior femoral condyle cuts, in TKA, 295
Posterior cruciate ligament (PCL)
excision, in TKA, 299
and femoral entry point in TKA, 294
Posterior cruciate ligament (PCL) reconstruction
indications, 87–88
sequence, 110
steps in, 110
surgical technique
arthroscopic control, 96
arthroscopy, 89–90
Bartlett posterior stress view, 87
examination under anesthesia, 88
femoral tunnel, 94
femoral tunnel drilling, 93–94
femoral tunnel incision, 93
fiberwire, 95
graft passage, 94–97
graft preparation, 89
guidewire insertion, 94
guidewire placement under fluoroscopic control, 92
metal traction wire preparation, 94
patient positioning and initial setup, 88–89
posterior medial portal, 89
posterior tibial translation under fluoroscopy, 88

Index
postoperative care, 97
preoperative positioning, 88
quadriceps tendon allograft, 89
quadriceps tendon autograft, 89
quadriceps tendon harvest, 89
tibial guide, 91
tibial tunnel, 92
tibial tunnel preparation, 90–93
Posterior stabilised box cuts, in TKA, 295
Posterior-lateral lesion, 31
Posterolateral approach and anatomy, 121
Posterolateral corner anatomy, 100
Posterolateral corner injuries
characteristics, 99
classification, 99
clinical findings, 99
initial setup, 99
patient positioning, 99
technique, 99–100
Posterolateral entry portal, 116
Posterolateral synovectomy, 122
Posteromedial approach, 119
Postoperative complications
bleeding, 496
infection, 496
pulmonary embolism, 496
skin necrosis, 497
skin pigmentation, 497
thromboprophylaxis, 495–496
Postoperative pain, 496
Primary (osteo) chondromatosis, 124
Prominence, trochlea, 383
Proximal tibia deformity, 241
Proximal tibial osteotomy, 193
Pulmonary embolism (PE), 496
R
Recession trochleoplasty, 404
Recurvatum
bony, 187
evaluation, 187
femoral, 188
genu (see Genu recurvatum)
pre-operative, 270
tibial, 188
Resection laxity, TKA, 241, 256, 267
Revision anterior cruciate ligament reconstruction
analysis of failure causes
clinical examination, 73
CT with 3D reconstructions, 73–75
initial operative report, 73
radiographic examination, 73
anterior femoral tunnel malposition, 72
anterior tibial translation, 72
causes of failure
associated lesions, 72
biologic failure, 73
poor fixation, 72
poor graft quality, 72
re-traumatic rupture, 73
technical error, 71
double tibial fixation, 78
femoral tunnel, 74
lateral extra-articular tenodesis, 85
osteotomy direction, 81
posterior tibial slope, 73

505
surgical technique
choice of graft, 75
combined procedures, 80–85
fixation, 78
general, 75
joint exploration, 75–76
tunnel placement, 76–78
two-stage reconstruction, 78–80
Revision TKA, 367
anterior tibial tuberosity osteotomy, 368
antibiotic cement fixation, 375, 376
cleaning (cement, granuloma), 372
closure, 376
customized “tulip-shaped” implant usage, 377
extension gap management, 374, 375
femur reconstruction, 373
flexion gap management, 373, 374
hinged prosthesis, 376
Hoffa fat pad resection, 368
implant removal, 368–370
joint line level tracking, 368
long keels usage, 375
medial arthrotomy, 366, 368
medial parapatellar arthrotomy, 368
patellar button, 371, 372
patient positioning, 366
postoperative course after re-implantation, 377
preoperative planning, 365
bone loss evaluation, 365
collateral ligaments evaluation, 365
hinge prosthesis, 366
implant size assessment, 365
joint line assessment, 366
modular implant selection, 366
rotating hinge prosthesis, 366
screening for infection, 365
specific equipment for implant removal, 365
unconstrained prosthesis, 366
rotatory hinge prosthesis, 375
sepsis case, 376, 377
skin incision, 366
synovial biopsies, 368
tibia reconstruction, 372
tibial tubercle proximalization, 377
Revision unicompartmental knee arthroplasty
and arthroscopy, 360–362
bone loss, 351
frontal plane laxity, 351
implants, 351
indications, 349–350
lateral, 356–360
medial, 351–357
posterior stabilized TKA, 351
pre-operative planning, 350–351
surgical correction of revision, 363
surgical history, 349
technical difficulties, 351
Robotic assisted unicompartmental knee arthroplasty
angular correction, 335
bone surface preparation, 336
definitive implants, 338
femoral implant placement planning, 335
mediolateral implant positioning, 335, 336
patient positioning, 333, 334
post-operative guidelines, 338
results, 338
testing, 338

506
Robotic surgery, 333
Rotation, tibial component, 258–259
Runner’s knee (see Iliotibial band syndrome (ITBS)
S
Screw osteosynthesis, of patella fracture, 436, 439
Segond fracture, 310
Semitendinosus tendon, 47
Skin necrosis, 497
Skin pigmentation, 497
Smillie test, 382
Soft tissue envelope, TKA, 237
Stiffness of knee
and ACL reconstruction, 446–447
adjuvant treatments and postoperative guidelines, 448
arthrolysis, 459
arthroscopic arthrolysis, 444–445
articular stiffness, 443
capsular/intra-capsular stiffness, 443
clinical history, 441
definition, 441, 457
etiological classification, 441
extra-articular stiffness, 443
fixed flexion deformity/limitation of extension, 443
flexion limitation, 442
iatrogenic limitation associated with tibial external rotation, 443
indication, 457
Judet release, 446–447
incisions, 458
indication, 458
patient positioning, 458
quadriceps release, 459–460
surgical risks, 458
manipulation under anesthesia, 443–444
open arthrolyis, 446
anterior arthrolysis, 446
posterior arthrolysis, 446
postoperative care, 460
rehabilitation, 460
Synovectomy, 118
anterior compartment, 117
arthroscopic
limited, 115–116
total, 116–117
combined, 123
indications, 115
open (with arthrotomy)
limited, 117
total, 117–123
open limited, 117
posterior compartment, 117
in posteromedial compartment, 120
postoperative care, 123
preoperative planning, 115
reduction, 115
total, 115
Synovitis, pigmented villonodular, 124
T
Tension band wiring, of patella, 436, 438
Three dimensional collagen matrix, 129
Thromboprophylaxis, 495–496
Tibial component rotation, 258–259

Index
Tibial cut
inclination, 252
level, 256
orthogonal, 230
in TKA, 144, 227, 230, 232–234, 240, 243, 247, 248, 252–259,
261–263, 265, 267, 268, 273, 276, 277, 279, 281, 291–294,
304, 307, 308, 309, 310, 313, 316–317, 323, 325, 326, 328,
329, 344–348, 372, 463
Tibial fixation with wire, 56
Tibial guide positioning, ACL, 42
Tibial intramedullary (IM) rod, TKA, 291
Tibial osteotomy, 151, 167–169
ACL reconstruction, 55–56
double osteotomy, 193–195
genu recurvatum, 190
HTO, 160–161
Tibial slope, posterior, 72, 73, 84
Tibial space
creation, 241
definition, 240
TKA, 240, 241, 253, 255–256, 267, 268, 317, 328
Tibial tubercle fixation, 191
Tibial tubercle osteotomy, 200–201
Tibial tubercle-trochlea groove (TT-TG) distance, 382, 384
Tibial tunnel, 42
ACL, 40–44, 46, 53–55, 63, 67, 68, 70, 72, 74–79, 82, 443, 447
drilling, 42
Total knee arthroplasty (TKA)
algorithm, 227
alignment, 230
alignment and balance
in extension, 247–256
in flexion, 257–259
computer-assisted, 239, 325–331
cuts, 231
definition, 229
deformity, 229–230, 270
disputable indications, 143
distractor, 237
femoral cuts, 233–234
femoral space, 240
fixed flexion deformity, 270
gap, 234
gap balance, 260–261
gap balancing, 237
after HTO, 465–467
implant, 235
indications, 143
in lateral arthritis
anterior tibial tuberosity osteotomy, 313, 314
femoral cut, 317, 318
lateral condyle osteotomy, 318, 320
lateral paramedian skin incision, 313
Pie Crust technique, 320, 323
prosthesis usage, 323
tibial cut, 316, 317
Trillat periosteal elevator, 313
lateral patellar release, 307–308
laxity, 270
ligament balancing, 237
osteoarthritis after varus femoral malunion, 490–491
patella infera after, 456
postoperative care, 309–311
prosthetic interface, 262–267
radiologic evaluation, 144

Index
radiology, 270
rehabilitation, 311
rotational deformity, 271
sequence of steps in, 267–269
soft tissue envelope, 237
spacer, 237
stiffness after, 448
synovectomy, 123
tensor, 237
tibial and femoral segments, 236–237
tibial cut, 232
tibial slope, 345
tibial space, 240
UKA revision to, 463
after valgus high tibial osteotomy, 341–348 (see also
Unicompartmental knee arthroplasty (UKA))
Total knee arthroplasty, in medial arthritis
final tibial preparation, 299
implantation, 299–302
patellar cut, 296–299
preoperative planning, 287
rotation of femoral component, 303–305
surgical technique, 287–291
tibial cut, 291–293
trial reduction and assessment of gaps, 295, 296
wound closure, 302, 303
Total patellectomy, 201, 439, 440
Traction injury, tibial vascular structures, 107
Trauma
knee pain and valgus deformity, 475–477
stiffness and subluxation of knee, 472–474
Triple gracilis and triple semitendinosus tendon (TGST), 57
Trochlear depth, 384
Trochlear dysplasia, 401
with crossing sign, 382, 383
Trochleoplasty
anteromedial approach, 401
closures, 403
computer-assisted surgical techniques, 404
deepening, 402
fixation, 402, 403
indications, 401
intra-operative planning, 401, 402
post-operative care, 403
rehabilitation protocol, 403
trochlear groove marking, 402, 403
Tunnel placement, revision ACL
backup fixation, 77
drilling of tunnels, 76–77
femoral tunnel, 77
incorrect prior tunnel position, 77
malposition, 77–78
removal of hardware, 76
tibial tunnel, 76
U
UKA, see Unicompartmental knee arthroplasty
(UKA)
U-Kneetec® prosthesis, 205
Unicompartmental knee arthroplasty (UKA)
and arthroscopy, 360–362
complications, 217
lateral (see Lateral unicompartmental knee arthroplasty)
medial (see Medial unicompartmental knee arthroplasty)

507
post-operative guidelines, 214
post-operative x-rays, 215
radiological workup, 206
revision of
bone loss, 351
frontal plane laxity, 351
implants, 351
indications, 349–350
lateral (see Lateral unicompartmental knee arthroplasty
revision)
medial, 351–357
posterior stabilized TKA, 351
pre-operative planning, 350–351
surgical correction of revision, 363
surgical history, 349
technical difficulties, 351
robotic assisted, 333–339
surgical technique, 205
tibial component revision, 463
Unicompartmental prosthesis, 141–142
V
Valgus high tibial osteotomy
blade, 163
clinical outcome, 159
distal cut of osteotomy, 164
fixation of osteotomy, 165
guide pin, 161
hole drilling, 163
medial cortex, 165
medial opening wedge, 167–171
patient set-up, 161
postoperative guidelines, 171–172
proximal cut, 164
radiological evaluation, 159
reasons for failure, 159
surgical techniques, 159
tibial osteotomy, 160–161
total knee arthroplasty after, 348
Valgus hindfoot deformity, 160
Valgus knees, 195, 245
Valgus malunion, 179
Valgus stress, 6, 210
Valgus stress X-ray, 288
Valgus tibial deformity, 230
Valgus-producing high tibial osteotomy, 80–81
Varus deformity, in joint, 280
Varus high tibial osteotomy
femoro-tibial mechanical angle, 174
indications, 173
intraoperative evaluation, 176, 177
intraoperative fluoroscopic control, 175
patient set-up, 174
postoperative x-rays, 177, 178
radiological workup, 173
skin incision, 174
superficial medial collateral ligament, 174
surgical technique
complications, 177
improvements, 178
patient set-up, 173
postoperative guidelines, 177
tibial osteotomy, 173–176
Varus knees, 193–195, 243

508
Varus stress
view in 90° flexion, 257
x-rays, 100
Vastus intermedius, resection of, 461
Velpeau bandage, 309
Venous thrombosis, 495, 496

Index
Y
Y-graft, 102
Z
Zero degree tibial slope, in TKA, 262