Revision of loose femoral prostheses with a modular stem system basing on the ''press-fit'' principle

REVISION OF LOOSE FEMORAL PROSTHESES WITH A STEM SYSTEM BASED ON THE “PRESS-FIT” PRINCIPLE A CONCEPT AND A SYSTEM OF IMPLANTS A METHOD AND ITS RESULTS

Springer Paris Berlin Heidelberg New York Hong Kong London Milan

REVISION OF LOOSE FEMORAL PROSTHESES WITH A STEM SYSTEM BASED ON THE “PRESS-FIT” PRINCIPLE A CONCEPT AND A SYSTEM OF IMPLANTS A METHOD AND ITS RESULTS

PIERRE LE BEGUEC HANS- PETER SIEBER

INTRODUCTION BY PROF. ERWIN W. MORSCHER

Dr Pierre Le Béguec 11, Galeries du Théâtre 35000 Rennes Dr Hans-Peter Sieber Orthopäd Klinik Spitalzentrun Biel A6 Vogel Sang 84 Postfach 1664 2501 BIEL Switzerland

© Cover illustrations by courtesy of P. Schuster and M. Goldschild ISBN: 978-2-287-39626-7 Springer Paris Berlin Heidelberg New York © Springer-Verlag France, Paris, 2007 Printed in France Springer-Verlag France is a member of Springer Science + Business Media Apart from any fair dealing for the purposes of the research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1998, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms of licenses issued by the copyright. Enquiry concerning reproduction outside those terms should be sent to the publishers. The use of registered names, trademarks, etc, in this publication does not imply, even in the absence of a specific statement, that such names exempt from the relevant laws and regulations and therefore free for general use. Poduct liability: the publisher can give no guarantee for information about drug dosage and application thereof contained in this book. In every individual case, the user must check its accuracy by consulting other pharmaceutical literature.

SPIN: 11854418 Cover design: Nadia OUDDANE

WARNING

This book is a monography with the purpose of telling our personal experience with the use of a cementless modular stem system in femoral revision surgery, which relaies on the “press-fit”principle in order to achieve primary stability. The monography does not address all the problems that could arise while performing revision surgery of loose femoral components and it does not mention the different techniques actually available, e.g. cemented femoral revision. The implants which have been used by the authors are commercialized as a system named REVITAN, which unites different cementless stems showing a large variety of geometric forms and which have only the modularity as a common principle. This way of presentation by the producer could mislead the user by suggesting him an identical way of using those different stems. This is clearly not the case. If a surgeon wants to achieve a primary fixation of a stem by the means of the “press-fit”-principle, the geometry of the stem plays an essential role in doing so. It is evident that the operating principles for a straight stem are quite different from those used for the fixation of a curved stem. The user must be aware of the fact that modularity is not a fixation concept by itself, but the basic step for achieving a correct leg length in order to restore the biomechanics of the revised hip. The fact that implants which require different surgical techniques and strategies are presented under the same name may be interesting from a commercial point of view, but we strongly feel this could give rise to errors during a revision surgery, especially by suggesting to the user a versatility which in fact does not exist. This error of appreciation could lead to poor technical results. The authors want to draw your attention to the fact that this monography is dealing with the technique and the results that apply only for PFM-Revision prostheses of first generation (PressFit Modular Revision prostheses), whose stability is purely based on the “press-fit”-principle, presently called Revitan Straight system or PFM-Revision of second generation. NB. The differences between the stems of first or second generation are minor; they concern mostly the adjustment of the antetorsion on the assembly system of the two prosthetic components. The PFM-Revision stems of first generation have been implanted since 1994 and must be clearly differentiated from the other implants of the REVITAN system, especially the newly introduced (in 2003) REVITAN CURVED STEMS, which should not be used in the same way. We urge those surgeons who want to implant such a curved stem to follow exactly the strategies and implanting techniques that have been developed for those curved stems.

Finally, at a somewhat unfavourable time, when the necessity for mercantilism “at any price” prevails over the age of the “doctrines”, which was sometimes slightly restrained by a narrow dogmatism, it is worth emphasizing that this monography as been written in utter independence and that the author has not been influenced in any way during the process of this work.

INTRODUCTION

PROFESSOR ERWIN W. MORSCHER The objective of the revision of a loose femoral prosthesis is to restore a hip joint as close as possible to normal through the replacement of the deficient implant with a new one which anchorage should be both stable and durable. Among the concepts allowing to ensure a durable anchorage without resorting to cement, none has ever been as effective as the press-fit and this is true for cups, femoral stems, primary arthroplasty or revision of femoral cemented implant. The press-fit can be described as a means to guarantee the stability of a rigid structure and of a conical shape (the implant) which is wedged in an elastic environment (medullary cavity) slightly under dimensioned. If a pre-load is applied axially to the implant, a conical shape will entail stressing forces toward the host bone, which consequence is to improve stability; moreover, the intensity of the forces applied diminishes gradually, from the proximal to the distal region, which results in a better distribution of the loads. This way, the bony atrophy due to a stress-shielding phenomenon can be avoided and in case of a late seating, the stability of the implant can be restored. The press-fit is thus the only concept providing a “second line of defense” because it allows a second stabilization of the implant with a further wedging in case of a secondary seating. But any system of implants can only achieve a good result if the designs, the material, the surface of the stem, and last but not least, the surgical technique, comply with the requirements of the chosen concept to ensure the primary fixation of the implant. These various factors will have to combine in the most favourable manner in order to perform any arthroplasty (or revision of arthroplasty). THE MONOGRAPHY WRITTEN BY DOCTOR LE BEGUEC IS DIVIDED IN 4 PARTS The first part concerns the description of the press-fit concept and of the implants which constitute the PFM-Revision of first generation (Revitan Straight system). At the time when the system of these implants was created and during its development, the author strictly complied with the requirements of the press-fit concept, avoiding some preposterous innovations. This way, the author resorted to well-tried principles: straight stem with sufficient conical area of anchorage to ensure a press-fit effect, as well as a range bearing fins which has been also tried and tested with the revision stems selected by Wagner. Similarly the grit-blasted titanium surface is well-tried as an osteophile substrate highly beneficial to bony ongrowth. The modularity shows some advantages: it facilitates a more anatomical reconstruction and thus a better function of the hip, it enables to guarantee an effective press-fit; but this characteristic can result in a weakening of the implant and the author was aware of the potential dangers of such a system. The construction of the assembly system of the proximal and distal components has been technically dealt with in an ingenious manner and after 12 years of use without encountering any serious incidents, this coupling system can be considered as reliable. The second part concerns the planning of the operation. A satisfactory planning of the operation is decisive for a successful primary arthroplasty or revision and the author puts a special accent on the preoperative planning: “failing to make a preoperative planning is a plan on how to fail”.

During the radiological analysis,the attention of the surgeon is drawn on the causes of complications, especially the presence of a femoral curving when evaluating the morphotype, as well as the difficulties which can arise during the cement removal.When determining a strategy,the author rightly consider the actual classifications as insufficient,which,in daily practice,makes them often useless because each concept has its own requirements and each patient is a special case. Finally, it is reminded that the planning is always completed by a template, necessary to measure the main reference lines useful during the operation. The third part is devoted to the surgical technique. For the operation, 2 options are proposed depending on the femoral approach selected: endofemoral approach or femoral flap. Each option is detailed, and the author, aware of the necessity for a perfect surgical technique in order to achieve good results, complies continuously with the required principles to ensure a stable anchorage with the press-fit effect: creation of a boneimplant contact surface, avoiding isolated contact points, then secure a perfect wedging and he underlines the significant advantages provided by a modular system in order to achieve an effective press-fit effect. Thanks to his extensive experience in femoral revision surgery, the author is also able to make numerous recommendations on how to manage difficult or special situations. The fourth part concerns the analysis of the results. The author discloses his results from a series of 180 operated hips, 152 being meticulously documented both clinically and radiologically (25% of patients were between 80 and 90 years old and 65% were overweighed). It is interesting to note that with the test of time, short stems are increasingly implanted: for the 78 first prostheses, following the learning curve, a short stem is placed in 44% of cases while this figure is of 68% for patients in the second group (74 prostheses). On the whole, clinical and radiological results are outstanding considering the complexity of the operation. It is also worth emphasizing that these results are improving with experience: regarding bone regeneration, the percentage of very good results progresses from 25% for the patients in the first group to 46% for the patients of the second group, and simultaneously, the figure for poor results has decreased from 22% to 4%! The radiographs displayed show the possibilities provided by the method in order to achieve good or excellent results, even in the most difficult cases. Lastly, the different possible complications are mentioned, their number should be considered as reasonable and it should be especially observed that the number of PFM-R stems that required revision is low: 5 among a total of 180 operated hips! In conclusion: one should wish to this PFM-Revision system of implants a deserved success and it is highly recommended to each surgeon to read the monography by Doctor Le Béguec, primarily when this system has been selected for a revision surgery, it is the best way to reduce the learning curve which also exists for these implants.

FOREWORD

The revision of a loose femoral prosthesis causes apprehension among many surgeons because this operation involves numerous pitfalls. To avoid any hazardous surgery leading to failure, which would be, so to speak,“planned” and most of the time immediate, the only possible option for the surgeon is to make the reasonable choice of a concept and an implant, to plan the surgical procedure in a second step and apply the selected method logically and rigorously. The first part of this work is devoted to the press-fit concept and to the system of implant used by the author. Making the choice of a concept in order to ensure primary stability of a cementless implant, with a thorough knowledge of its fundamentals, is the first step the surgeon should take. It is the best way to avoid serious errors when selecting an implant and determining a surgical strategy. The failures of a press-fit stem are often due to an insufficient understanding of the requirements imposed by the method. The second part concerns the planning of the surgery. An extensive experience in surgery, in the field of hip arthroplasties, has taught us that surgery is never compatible with improvisation and that the revision of a loose femoral prosthesis should not be limited to a surgical step. The planning of an operation, consisting in a rigorous radiological analysis of the femur in order to determine a strategy complying with the chosen concept, is the way to achieve a successful surgery; conversely, overlooking this preoperative evaluation is running the risk of any hazard during the operation. The third part is devoted to the surgical technique. Most of the time, the medical documentation related to an implant is merely composed of the description of the surgical technique, which is necessary as well as insufficient. Indeed, if a surgical technique is not associated with the description of the concept and its requirements, numerous decisive technical details are likely to be overlooked (or to be misunderstood) and, in these conditions, the surgeon will hardly be able to carry them out during the surgery. The fourth part is related to the results. Considering the heterogeneity and insufficiency of the methods employed to evaluate these results until now, we have felt the necessity of proposing a new methodology. The way we proceed seems interesting and practical, but to definitely finalize it requires a wider clinical study which has not been performed yet. Between 1994 and 1999, 180 prostheses have been placed by the author. The results concern a series of 152 prostheses, clinically and radiologically analyzed with a mean follow-up time of three years. We are aware that this mean follow-up time, although not negligible, is not sufficient to obtain definitive results. Nevertheless, and considering the fact that it is an already long-established concept and an implant which design is also well-tried, we think we can publish these results, with limited risks, since the innovative character of the implants used only concerns the modular system. This innovation has been employed successfully since 1989 with primary stems and since 1994 with revision stems.

TABLE OF CONTENTS

Warning ..........................................................................................5 Introduction .................................................................................7 Professor Erwin W. Morscher Foreword ........................................................................................9 FIRST PART: THE CONCEPT OF PRESS-FIT A SYSTEM OF IMPLANTS Preliminary Remarks .............................................................15 Choice of a Concept and an implant General remarks Chapter 1......................................................................................19 The concept of “press-fit” Principles – characteristics of an implant with press-fit fixation Chapter 2......................................................................................29 The PFM-Revision System Implants and Ancillary Conclusions of the first part................................................35 Why make these choices? SECOND PART: PREOPERATIVE PLANNING Chapter 1......................................................................................41 Radiological analysis of the femur Chapter 2......................................................................................53 Determining a surgical strategy Chapter 3......................................................................................67 Making a preoperative template THIRD PART: SURGICAL TECHNIQUE Chapter 1......................................................................................73 General considerations Chapter 2......................................................................................83 Option 1: femoral flap Fixation in the diaphysis Chapter 3......................................................................................99 Option 2: endofemoral approach Proximal fixation

12

Revision of loose femoral prostheses

Chapter 4 ...................................................................................113 Postoperative care Chapter 5 ...................................................................................115 Conclusions Appendix 1 ...............................................................................117 Assembly of the two prosthetic components Appendix 2 ...............................................................................123 Removal of a PFM-Revision 2nd generation (Revitan Straight Stem) FOURTH PART: THE RESULTS Chapter 1 ...................................................................................133 Patients and type of operation Chapter 2 ...................................................................................137 Clinical results Chapter 3 ...................................................................................141 Preoperative radiological results Chapter 4 ...................................................................................145 Postoperative radiological results Chapter 5 ...................................................................................173 The complications, limitations and indications of the “press-fit” concept Conclusions ...............................................................................177 Bibliography ..............................................................................179

FIRST PART

THE CONCEPT OF PRESS-FIT A SYSTEM OF IMPLANTS

Choosing an implant system for revision surgery can sometimes be a troublesome task as most authors of a new product have a tendency to underplay the inherent problems or the natural limitations of “their” implant or method. On the other hand, these limitations will be magnified in the presentations of defenders of a different concept. Every surgeon however must acquire information for himself. The information should be presented in a manner which allows a potential user of the system to ask the right questions which will direct his choice and remind him that every concept has advantages and pitfalls and that every implant is the result of some compromise.

PRELIMINARY REMARKS

CHOICE OF A CONCEPT AND AN IMPLANT GENERAL REMARKS

“To choose, decision by which one thing is regarded as preerable above others, among a number of alternatives.” French dictionary Le Petit Robert We will not present here an exhaustive picture of the numerous questions which may arise when one is confronted with the task of making the choice of a concept and an implant. We will however draw the surgeon’s attention to the difficulties which can arise during this first step.

1. DO WE FIRST CHOOSE A CONCEPT AND THEN THE CORRESPONDING IMPLANT, OR SHOULD THE STEPS BE THE OTHER WAY ROUND? The choice can be made in two different manners: ● First manner: The surgeon decides to use a new implant. Thereafter, he studies the biomechanical characteristics of this implant, especially its ability to ensure primary stability as this is the first goal in cementless systems, be it in primary or revision surgery. This is certainly the most frequent procedure, but at the same time also the most dangerous one as the biomechanical characteristics of many implants for revision surgery are poorly defined or worse, they are hardly applicable during surgery. It is advisable to make the choice of a system under exclusion of any “marketing” considerations. ● Second manner: The surgeon proceeds the other way round. In a first step, he studies the different concepts which ensure primary stability of a cementless implant and in a second step, he chooses the implant which best corresponds to the concept he has decided to follow. This method is safer because it ensures a perfect understanding of the topic and of the goals which should be attained. In this way, the choice of the implant will be easier, more precise and finally, safer for both the patient and the surgeon.

16

First part – The concept of Press-Fit

2. CRITERIA FOR THE CHOICE OF A CONCEPT AND AN IMPLANT These criteria apply to all total hip arthroplasties (primary or revision), but some of them carry more weight when dealing with revision surgeries.

2.1 CONCERNING THE CONCEPT In general we believe that a concept should fulfill four requirements: It must be reliable, universal, reversible and economically reasonable. A reliable concept. Here, we would like to remind the reader that especially in uncemented revision surgery, one should only rely on concepts and technologies which have stood the test of time for many decades. We must also emphasize, that any coating, be it osteo-inducing or osteoconducting, is not a concept in itself for ensuring primary stability of an uncemented implant, but it is an effective aid in ensuring secondary stability of the implant. ● A universal concept. This factor requires clarification as no concept can really be universal in uncemented revision surgery. A reasonable choice is to select that concept which is applicable to the greatest number of cases. It is difficult to acquire mastership of a technique if either the strategy or the implant is frequently changed. Monotony is not an inconvenience in surgery. NB. The method proposed by Kerboull (14) with cemented stem and bone graft is probably the method suitable for the greater number of cases, provided that bone grafts are available in sufficient amount. ●

A reversible concept. Every implant should be removable. This is probably the biggest drawback of any uncemented prosthesis system, especially in absence of an abnormal mobility of the implant, the prosthesis is often soundly integrated into the bone. In order to minimize this possible inconvenience, the surgeon should always prefer short stems, since they will be much easier to extract than longer ones. We also think that modular systems should be preferred as they allow the possibility of an easy uncoupling of the proximal modular part, thus giving direct access to the well anchored stem. NB. The difficulties encountered during the removal of the implant do not affect only the cementless stems, and the excision of the cement after a long cemented stem has been implanted is another ordeal for the surgeon to overcome. ●

An economically reasonable concept. This factor must be taken into consideration at a time where financial restrictions are becoming greater and greater. Modularity allows a high number of individualized solutions, with few implants on the shelf and with the added advantage of having them all available during surgery. ●

2.2 CONCERNING THE IMPLANT ANY IMPLANT SHOULD BE ABLE TO ENSURE A PAINLESS AND FUNCTIONING HIP The geometry of an implant should not only ensure a perfect primary stability according to the chosen concept, but at the same time also allow a good function of the gluteal muscle sling (CCDangle, offset).

Preliminary remarks – Choice of a concept and an implant

17

LOOK FOR THE CAPABILITY OF AN IMPLANT TO HELP AND SUPPORT BONY REGENERATION IN THE SHORT TERM AND TO PRESERVE BONE STOCK IN THE LONG TERM In cementless revision surgery, regeneration and preservation of the bone stock depend on many factors, an important factor are the design properties of the implant. One should pay attention to the ability of the implant in preventing rigidification of the periprosthetic tissue, thus preserving the possibility of a load transmission in flexion through the bony femur. Short term remodeling and long term preservation of bone depends largely on compressive or traction forces acting during normal gait and they should not be disturbed or bypassed by the implant. In cases where loosening of a prosthetic stem is followed by a significant loss of bone stock, the question of bone grafting may arise. This aspect represents one of the major advantages of a cementless system when compared to cemented revision stems, since past experience has shown that if primary stability is achieved, important bone regeneration can be observed with the former implants without any additional bone grafting (21, 28 and 29). On the other hand, bone grafting will be mandatory in most revisions with cemented stems. In the eyes of many surgeons, this more pragmatic than scientific argument has sustained the choice of an uncemented revision stem, since bone banks are not always available and the transplants are expensive, especially when large amounts of bank bone are needed.

AS A SUMMARY, The first difficulty that a surgeon will encounter when he is planning the revision of a loose femoral implant is the necessity of making the choice of a clearly defined concept and only then does the surgeon face the problem of choosing an implant which allows the technical realization of that concept. The rapidly growing number of revision stems in the market can make these decisions difficult, especially if the surgeon is young and inexperienced in the field and thus mainly relying on personal convictions.

CHAPTER 1

THE CONCEPT OF “PRESS-FIT” PRINCIPLES – CHARACTERISTICS OF AN IMPLANT WITH PRESS-FIT FIXATION

INTRODUCTION Press-fit is a commonly used fixation principle in industrial technology, the most known example being the fixation of two separate parts using a tapered cone system. The same principle is also applied when a cementless revision stem is firmly anchored in the femur. This principle was the choice of Wagner (20) in 1987. N.B. In the following, the term “press-fit” will exclusively be applied referring to the primary stability of a cementless stem which is the basic precondition for secondary osseointegration (or stability). In order to guarantee primary stability of an implant by press-fit, the pre-load applied to the interface between bone and implant must always be superior to the destabilizing forces, which are the axial and the rotational forces in femoral implants. According to Morscher (18, 19) two prerequisites must be fulfilled: The contact between bone and implant under the form of a surface and a perfect wedging of the implant must be guaranteed. To these two prerequisites, we will add a third condition: The implant should not stiffen the surrounding bone, in order to reduce the risk of subsequent stress-shielding. The design of the implants varies from author to author whilst trying to realize these three prerequisites. A reasonable implant selection is only possible if the surgeon has taken note of the geometry of the different implants and has evaluated the inherent advantages and disadvantages of every single one.

20

First part – The concept of Press-Fit

1. CONTACT SURFACE BETWEEN BONE AND IMPLANT There are two ways to ensure the contact surface between bone and implant: ● Creating an implant with geometry close to the one of the medullary canal, e.g. with a custom-made stem; ● Adapting the inner surface of the femur to the implant. This is the most frequently used concept. Depending on the geometry of the implant surface, different possibilities are available. NB. For a better understanding of this chapter, we request the reader to refer to the pages 46 ff. concerning the description of the different morphotypes of the femur.

1.1 STRAIGHT STEM The straight stem configuration is the safest design in order to ensure a good contact surface, between bone and implant, since it is, by definition, invariable. However, this good contact surface means always to avoid any 3-point contact. Two conditions are required to avoid this unstable configuration: The femur must be straight and an implantation in varus position must be avoided. THE FEMUR MUST BE STRAIGHT If a straight stem is implanted into a curved femur in the frontal plane, the surface of fixation will be reduced to a 3-point fixation which will not withstand either the axial forces (subsidence) or the rotational forces, since the implant will invariably be underdimensioned (fig. 1A). Figure 1 The choice of a straight stem calls for a femoral osteotomy, if the femur is curved in the frontal plane (fig. 1B) We suggest that every curving of the femur in the frontal plane should be considered, even if it does not seem to be very relevant. A marked curving in the sagittal plane can also be an obstacle to the precise seating of a straight stem with a 3-point fixation in this view (fig. 2A): this will lead to the choice of an underdimensioned Figure 2 stem in both views which will further diminish the contact surface between bone and implant, even if the femur was straight in the frontal plane (fig. 2B) A curving of the femur in the sagittal plane must only be taken into consideration if the femur is otherwise straight. Not every curving in the sagittal plane is automatically an obstacle to the implantation of straight stem. AVOID IMPLANTING IN VARUS POSITION When a straight stem is implanted in a varus position, in general a 3-point fixation will result, especially if the femur is straight in frontal plane (fig. 3).

Figure 3

Chapter 1 – The concept of “Press-fit”

21

Thus, as in the prior case, the consequences will be the same: Implantation of an under-dimensioned implant with no surface contact between bone and implant. When a femur is straight in the frontal plane, and the surgeon has made the decision to implant a straight stem, it is of utter importance to create a large opening through at the medial base of greater trochanter. Omission of this step will result in varus implantation.

1.2 CURVED STEM Many surgeons believe that the natural decision is to opt for a curved stem, since the femur itself is curved in the diaphyseal region. In reality, this is a misleading first impression and when a boneimplant contact in the form of a surface should be realized, such implants can create some problems. ● The curvature of the implant can only be of some use in the sagittal plane; in the frontal plane all these implants show a straight configuration (fig. 4). Under the frequent situation of an additional curving in the frontal plane, the surgeon still has to proceed with a femoral osteotomy in order to avoid a 3-point fixation, just as he would have to do when using a straight implant because a 3-point contact in frontal plane results in an insufficient fit and fill of the femur in the frontal as well as in the sagittal plane, which makes any bone-implant surface contact difficult to achieve in this plane, even with a curved stem. ● If we aim for a surface-like contact zone between a curved implant and the femur, a longer zone of coincidence of the two radii is mandatory (fig. 5). No doubt that this ideal situation is rarely seen, and as long as the straight configuration is invariable, on the other side the Figure 4 variability of two radii will be very high. ● Furthermore, the curving of these implants pertains mostly to the diaphyseal region and this will automatically call for longer implants if a sufficient contact zone is to be established. In the frequent situation, where a short implant could be safely anchored, the choice of a curved stem becomes inappropriate and bears more risks than the straight stem option because, in this situation, it is definitely easier to ensure a bone-implant contact surface wih a straight stem. Figure 5 ● The implantation of a curved stem using an endofemoral approach (without a flap) is a very delicate procedure in cases where the femur is straight, and this could force the surgeon to do a femoral flap even when this step would have been unnecessary. A curved stem could be of some interest, when, after having done a long lateral flap, a long stem must be chosen. It is important to note that in the same situation a straight stem can also be implanted if the surgeon makes an additional osteotomy of the medial cortex. This medial osteotomy will anyway be necessary in order to bring the medial cortical bone back into contact with the implant, especially in cases where there is an important varus deformity in the a-pview, which is a frequent pattern. When the goal of a surface shaped contact zone between bone and implant should be attained, the straight stem option is always the safer choice. Inversely, the curved stem has many dis-

22

First part – The concept of Press-Fit

advantages, especially when a short fixation should be realized. In fact, most of these curved implants find their stability through a 3-point fixation and, unfortunately, by means of a much longer stem configuration.

1.3 SECTION(S) OF THE STEMS The cross-sectional shape of a stem with press-fit properties must be carefully taken into consideration since the transmission of rotational load will largely be influenced by this pattern. Different patterns are possible: ● A circular cross-section is the best way of creating a large contact surface, but it calls for a circular preparation of the medullar cavity, which may be difficult to realize. A circular contact zone will however stiffen the surrounding bone and, since the surface of the circular implant presents a flat structure, this design will not offer enough resistance to rotational loading (fig. 6). ● A quadrangular cross-section has proven to be highly successful in providing primary stability for a cementless implant as has been shown by Zweymüller (30). In fact, this implant stabilizes itself not through surface contact, but more through a linear Figure 6 endo-cortical fixation with the four edges of the implant, which allows for an excellent neutralization of rotational torque (fig. 7). The femoral canal is prepared by means of a quadrangular rasp, which cuts 4 longitudinal fins, sparing the adjacent spongious structures between the cuts. ● A stem with longitudinal cutting fins can be considered as an evolution of the quadrangular stem. The medullar cavity is prepared by means of a cutting reamer and the contact between bone and implant will be guaranteed through the cutFigure 7 ting of the sharp stem fins into the bone (fig. 8). This design, which is more and more frequently used for revision stems, has some clear advantages: The longitudinal fins present a cutting edge which eases the penetration of the fins into the cortex, thus enhancing the contact zone between bone and implant, especially if the cutting is deep enough.

Figure 8

Chapter 1 – The concept of “Press-fit”

23

The fins guarantee a perfect neutralization of the rotational forces. Furthermore, they help to correct imprecisions of the preparation and overall, they usually preserve a space between the implant and the cortical bone, which will ease revascularization and thus bony regeneration. At the same time, this space will reduce the rigidity of the implant-bone assembly.

2. WEDGING OF THE IMPLANT This is the second precondition in order to ensure a functioning press-fit fixation. To jam an implant into the bone without the help of cement is in other words, to keep a pre-load on the junction between bone and implant which will result in its primary stability. This goal can either be reached through a cylindrical or conical shaped stem. 2.1 CYLINDRICAL STEM (FIG. 9) If a cylindrical cross-section is chosen for the above-mentioned pre-load, then there is the necessity for an equally cylindrical preparation of the medullar cavity, but the implant will have to be slightly over-dimensioned in relation to the medullary canal (see also fig. 9: D>d). This necessity for over-dimensioning, is not without problems: ● There will be a high risk for fractures if the quality of the cortical bone is poor. ● If the medullary canal is only slightly over-reamed, contact pressure will be too low and it will not be sufficient for the neutralization of rotational torque or subsiding forces. This results in a high risk for instability, especially since the cylindrical design does not allow for further wedging of the implant during its subsidence. ● If, on the other hand, the medullary canal is too narrow, an early blocFigure 9 king of the stem during its insertion could occur, giving rise to a wrong leg length or a local peak load within the intact femur. 2.2 CONICAL STEM (FIG. 10) A conical design of the stem favors the wedging of an implant. In a stem with a decreasing cross sectional diameter, the subsiding forces which present initially as shearing forces will typically be transformed progressively into compressive forces at the bone-implant junction. This automatically enhances the inherent primary stability. This biomechanical characteristic, which has been described by Kerboull (15) for cemented stems, can also be applied to uncemented implants. It is important to realize that in contrast to industrial applications, in the surgical field wedging takes place between two materials of different characteristics, e.g. the visco-elastic properties of the bone. By this, the bone will undergo deformation under load, which is always the case when wedging takes place. The visco-elastic properties can however give rise to a local loss of contact pressure. This could explain the secondary subsidence that is sometimes observed in perfectly seated stems with press-fit fixation. The risk Figure 10

24

First part – The concept of Press-Fit

of instability is however small, since a conical shaped stem will undergo a new wedging during its subsidence if the preparation of the canal was correct (and thus conical). A major secondary subsidence is always due to a technical error during preparation of the canal. One of the main merits of the press-fit stems with a conical shape is the fact, that they offer (according to the terms used by Morscher) “a second line of defense”. The choice of a conical stem calls for an equally conical shape of the medullary canal of the femur. This goal is easy to reach if the medullary canal has a primary cone-shaped aspect in its proximal half and the original cortex is sufficiently large. On the other hand, if the medullary canal is altered to a cylindrical shape or if the cortical thickness is diminished due to atrophy, this goal can become more difficult to achieve. In the latter situation, it is mandatory to proceed (on the preoperative X-rays) with a meticulous analysis of the possibilities of reaming the medullary canal to a conical shape over the length required for a secure fixation.

3. AVOID STIFFENING THE FEMUR TOO MUCH Every press-fit fixation will result in some stiffening of the adjacent bone. Subsequently, the transmission of load may be modified and thus, give rise to a stress-shielding phenomenon since Wolf ’s law of transformation is no longer respected. NB. The term stress-shielding is applied to every modification of the bony structures which arises as a result of its contact with the implant. Mostly, it will present as a simple diminution of the bone density, the most advanced form being cortical atrophy. In order to diminish the risk of a stress-shielding phenomenon, the implant must have some geometric characteristics, which diminish its tendency to stiffen the femur. Furthermore, 2 rules should be strictly observed during the operation: Aim for a proximal fixation whenever this is possible and limit the length of the contact zone between bone and implant if the fixation can only be achieved in the diaphyseal cortex. 3.1 CHARACTERISTICS OF A PRESS-FIT STEM WHICH AVOIDS EXCESSIVE STIFFENING A stem with press-fit properties should present some characteristic features which help to limit the contact zone between bone and implant in the horizontal plane, but at the same time allow for an even distribution of forces in the frontal plane, thereby diminishing the tendency to excessively stiffen the femur. LIMIT THE CONTACT ZONE BETWEEN BONE AND IMPLANT IN THE HORIZONTAL PLANE Limiting the contact zone in the horizontal plane means that an excessive fit-and-fill of the medullary canal is avoided. Blaimont (1) has depicted the zones in the endo-medullary femur, where the press-fit effect should principally take place. He has shown that during hip flexion, the cantilever effect created by the femoral neck gives rise to forces which correspond to traction and compression forces in the cortical bone. At the level of the proximal femur, the tensile forces are concentrated on the lateral side and the compressive forces on the medial side of the cortex. In the diaphyseal region, the tensile forces are

Figure 11

Chapter 1 – The concept of “Press-fit”

25

mostly located in the anterior half, and the compressive forces in the posterior half of the femur. Tensile forces are separated from compressive forces by a plane called the neutral zone, which is situated in a sagittal plane in the proximal femur and in a frontal plane in the diaphyseal zone of the femur (fig. 11). From these findings, it can be deduced that if the contact between bone and implant takes place exclusively in the region of this neutral zone, the ability of the bone to react and remodel according to the acting forces can be preserved at least partially. From this study, it can also be deduced that an implant should feature precise geometric characteristics in order to prevent an excessive stiffening of the femur. ● In the proximal region of the femur: To privilege contact in the neutral zone in its sagittal plane, and diminish the contact in the frontal plane (lateral and medial corticals) means to design an implant with a large or heavily conical proximal body shape in the sagittal plane, but a rather small shape in the frontal plane (fig. 12 A). NB. The goal of avoiding over-filling in the frontal plane can be easily achieved if the different anatomical variations of this region or the loss of bone stock in revision surgery are taken into account. ● In the diaphyseal region of the femur: Avoid a b any circular contact and favor the contact in the Figure 12 frontal plane, which corresponds to the neutral plane in this region. This goal is easily achieved with either a quadrangular stem, or, in the case of a stem with fins, by placing these fins in the frontal plane (fig. 12 B) NB. In the present system, the stems with big diameters (18 mm and bigger) show a flattened zone in the sagittal plane, which also helps to diminish the rigidity of the bone-implant assembly. We are convinced that these features merit particular attention, since it is always easy to create a proximal contact zone in the sagittal plane that will largely be sufficient for the stability of an uncemented implant. This is of high interest not only in revision prostheses, but especially in primary cementless prostheses, when a pure metaphyseal fixation is possible. ALLOW FOR AN EVEN DISTRIBUTION OF FORCES IN THE FRONTAL PLANE We have already emphasized the importance of a conical shaped stem for a safe primary stabilization by means of the wedging mechanism, called press-fit. This configuration also allows for a better distribution of forces in the frontal plane at the contact zone between bone and implant by creating a stabilizing horizontal force and at the same time by a progressive diminution of the axial loading (subsiding forces) from the proximal to the distal zones of the implant (see fig. 10). In contrast, a cylindrically shaped stem will wedge in a too short distance and hence create stress peaks (see fig. 9) or fractures.

Figure 13

26

First part – The concept of Press-Fit

3.2 PROXIMAL FIXATION MODE (FIG. 13) A proximal fixation helps to avoid a diaphyseal fixation; thereby diminishing the potential for stress-shielding phenomena. In revision surgery however, a strictly proximal fixation is seldom possible. Therefore, a fixation mode in the metaphyseo-diaphyseal region should be preferred. Besides, this fixation mode is easier to reach and is overall more reliable. NB. We advocate looking for this mode of fixation every time the femur is straight and the defects are small. Since in revision surgery, pure metaphyseal fixation modes are more coincidental, we think that primary cementless stems are less suitable for revision surgery and hence should be avoided. 3.3 SHORT DIAPHYSEAL FIXATION (FIG. 14) If the primary fixation has to take place in the diaphyseal region, it is advisable to limit the length of the contact zone. This is possible in most of the cases. On this point of view, our opinion diverges from that of Wagner (29). According to Wagner, the primary stability of a press-fit implant depends largely on the length of contact between bone and implant and he therefore proposes a contact zone of 70 to 100 mm. This is in our view, too long and unnecessary in most cases. We believe on the contrary that a good press-fit fixation does not depend on the length of the zone but the quality of the wedging and a contact zone of 30 (in excellent bone) or from 40 to 50 mm is sufficient, when the bone quality is good. This way of proceeding allows for the systematic choice of short revision implants. Moreover, it is true that a Figure 14 surface contact can more easily be achieved, when the length of the fixation zone is restricted and a precise machining of the seating with the conical reamers thus becomes possible.

Chapter 1 – The concept of “Press-fit”

AS A SUMMARY The press-fit fixation as a way and means for achieving primary stability in uncemented revision surgery is probably a not-well-enough known procedure for most of the operating surgeons and, under these circumstances, a wrong and detrimental use of such stems is becoming more likely. It was up to Morscher to have given the best definition of a surgical press-fit fixation, i. e. the contact between bone and implant must correspond to a contact surface and a perfect wedging of the implant must be ensured under the condition that the implant should not stiffen the surrounding bone. To reach these three goals, the implant must present some well defined, characteristic geometric features. We are convinced that the conical straight stem is a good compromise, especially if its design allows for a selective contact in the sagittal plane of the proximal femur and in the frontal plane of the diaphyseal femur. The surface of the implant, with its cutting fins is also of interest in the discussion of the press-fit concept. Even if till today the role of these longitudinal fins has not been sufficiently examined, one can be sure that they play a decisive role in stabilizing and even augmenting the contact surface between the implant and the bone bed; during the wedging process, they ensure the cutting in and thus the anchoring of the stem and finally they contribute greatly to a perfect control of the torque forces arising during normal gait. The last factor is decisive for the success of an uncemented implant. The surface of the implant itself may play a role while a choice of an implant has to be made, however it must be kept in mind that whatever the structure of the surface will be (corunded, grit-blasted, sulmesh, osteoinducing surfaces,…) this is only a factor which helps to achieve secondary osseointegration of the implant. It has strictly nothing to do with the achievement of primary stabilization, which depends only on geometric characteristics of the implant.

27

CHAPTER 2

THE PFM-REVISION SYSTEM* IMPLANTS AND ANCILLARY

INTRODUCTION The PFM-Revision system of first generation consists in a set of femoral stems made of Ti 6A17Nb titanium alloy (Protasul – 100). Each implant is made up of 2 parts: one proximal component and one distal component. Mechanical coupling of the components is ensured by a morse taper connection which guarantees an excellent mechanical stability confirmed by the laboratory tests and clinical study. The ancillary instrumentation is modular as well. It is mainly composed of reamers, rasps, trial prostheses and of a system which allows an implantation in 2 stages of the definitive stem.

* Called presently Revitan straight or PFM-Revision system of second generation.

30

First part – The concept of Press-Fit

1. DESCRIPTION OF THE IMPLANTS 1.1 PROXIMAL COMPONENTS (FIG.1) There are 2 types of proximal components: spout or cylindrical. The spout components are wider on the frontal plane whilst the medial profile of the cylindrical components is thicker. Components of 6 different heights are available for each type, in increasing size by steps of 10 mm, from 55 mm to 105 mm. The CCD angle is 135° and the offset is 44 mm. The lateral side of the proximal components which is wide in the sagittal plane, bears ribs and grooves; it is hollow i.e. featuring the female part of the morse taper connection. The top of the female morse taper is threaded for the impactor and the disassembly instruments for the proximal component. There are two holes in the medial part that can be used to reinsert a flap, or the greater trochanter. Figure 1

1.2 DISTAL COMPONENTS (FIG.2) The distal components are available in three different lengths: 140, 200 and 260 mm. The diameter increases by steps of 2 mm from 14 to 28 mm. The whole range includes a total of 21 components (see fig. 3, p.27). They are straight stems with 8 longitudinal ribs and, from the size of diameter 18 mm, each stem has a flattened anterior-posterior area, with increasing size as the diameter increases. The shape of these implants is conical with a slope of 2 degrees. The conical part of the implant is always in a distal position. The height of the conical area is 100 mm for the 140 mm stems, whereas it is 120 mm for the 200 and 260 mm stems. The lateral fins are also conical in shape. This design seems preferable to us to a vertically grooved design, which would be less effective, or to bladeshaped fins, which would be weaker. In addition to this range, a 120 mm distal component which corresponds to a 140 mm distal compoFigure 2 nent that has been shortened by 20 mm. The conical proximal area is about 45 mm high with lateral ribs. The working area of this implant is the conical proximal area.

Chapter 2 – The PFM-Revision system – Implants and Ancillary

31

SIZES AND POSSIBLE COMBINATIONS (FIG. 3)

Figure 3

The entire set of proximal and distal components is modular allowing the assembly of stems of different sizes ranging from 175 mm to 365 mm.

32

First part – The concept of Press-Fit

2. THE ASSEMBLY SYSTEM The two parts of the prosthesis are coupled together by means of an original and efficient morse taper system, perfected in 1989. 2.1 DESCRIPTION OF THE ASSEMBLY SYSTEM (MORSE TAPER) (FIG.4) The morse taper is made of a wrought alloy with high mechanical resistance (Protasul – 21WF). It is inserted in the distal component made of a titanium alloy. THE MORSE TAPER HAS 4 AREAS: a. Thread for the conical nut. b. Cylindrical area for the centring of the 2 components. c. Conical area, finely grooved, for mechanical coupling. d. Area of a narrower cross-section allowing concentration of flexion stresses at this level. Before the assembly, it is possible to adjust the antetorsion of the proximal component from +40° to -40°. After the assembly, a gap of about 1 mm persists between the 2 components. This gap corresponds to the area of concentration of flexion stresses and enables micromovements without inducing the formation of any metal debris. Figure 4

2.2 TESTS FOR DYNAMIC RESISTANCE Tests for dynamic resistance have been carried out in conformity with ISO 7206/3 or 4 standards. The loads applied were 330N (inferior load) and 3300N (superior load). To perform the study in conformity with ISO 7206/4 standard, the loads were 330N and 2600N. The morse tapers were subjected to an average flexion stress of 760 Mpa. 8 dynamic tests have been performed, 4 went through 5 millions cycles and the 4 others 10 millions cycles, the pulsation frequency was of 6Hz. Results: no fracture due to material fatigue occurred and the study of surface resistance to friction did not show any sign of significant wear. These tests have also confirmed the concentration of stresses (elasticity of the whole implant) at the level of the morse taper of a narrower crosssection, area where there is a gap between the 2 parts of the prosthesis. 2.3 CLINICAL EXPERIMENTATION The clinical study confirmed these laboratory tests. From 1989 to 1994, 350 primary implants, and since 1994 about 10.000 revision stems, made up with this assembly system have been implanted. Until now, no breakage of the morse taper was reported. However, in 3 cases, during a revision for a different reason, failing mechanical stability at the level of the coupling area of the 2 parts of the prosthesis was observed; and in another case, the weakness of a female morse taper was

Chapter 2 – The PFM-Revision system – Implants and Ancillary

33

reported. For these 4 cases, the plausible explanation was a problem during the assembly of the morse taper and the distal component or an error during the coupling manoeuvre of the 2 parts of the prosthesis. Besides, the radiological study of the 152 implants that were a analyzed here did not show any Figures 5 significant radiological abnormaNB. Follow-up radiography lity, in particular, there was no at 51 months after the surgery. presence of punched-out lesions to be found that would lead to suspect a formation of wear debris. Conversely, the presence of osseocondensation at the level of the coupling area has been observed, probably in relation with a better elasticity of the implant at this level (fig. 5 A and B). NB. Since 2003, one case of fracture of the morse taper has been reported, probably due to a manufacturing error. b

3. ANCILLARY INSTRUMENTATION The choice of a modular implant entails a modular instrument set. With respect to the PFM-Revision system of first generation, in addition to a set of conical reamers (diameters 14 to 28 mm), the surgeon can make use of a system of rasps and modular test prostheses allowing an implantation in 2 stages of the definitive implant. NB. For the reamers, refer to the surgical technique: preparing a bone-implant contact surface. Although the rasps and the test prostheses are combined in a single instrument, it is suggested to consider it as two distinct instruments as its function varies depending on the option chosen by the surgeon. 3.1 MODULAR RASPS (FIG.6) The rasp function of the ancillary instrumentation is only used if the endofemoral approach is selected, in the metaphyseo-diaphyseal zone. In this situation, the rasp is also used as test prosthesis. The rasp involves the whole range of proximal components that can be all used in this case. For the distal components, the rasp concerns only the component of length 120 mm and the distal components of 140 mm with the diameter of 14 to 18 mm. NB. The use of an implant with a length of 140 mm and a diameter of 20 mm or larger, means that only a diaphyseal fixation can be achieved. In this case, the preparation of the femur is

Figure 6

34

First part – The concept of Press-Fit

done with a reamer whose diameter is superior to the size of the proximal component, making the rasp ineffective. For the distal rasps of length 120 and 140 mm, the surgeon can also make use of a rasp adaptor (graduated from 55 to 105 mm) that will allow the impaction and the choice of the size of the proximal component; (refer to surgical technique: preparing a bone-implant contact surface). Warning! Although the rasp does include the 200 and 260 mm distal components, we believe that it is of lesser use for these implants. Indeed, when proximal fixation is sought, these implants are too long and they should not be used in that case. The use of a long stem entails the creation of a femoral flap is performed, in most cases. 3.2 MODULAR TEST PROSTHESES (FIG. 7) To each implant corresponds a test prosthesis. When fixation is achieved in the diaphyseal region (mostly after a femoral flap has been carried out) the role of a test modular prosthesis is essential to ensure primary stability with the conical area of the distal component while selecting an implant of a suitable size (refer to surgical technique: ensuring a stable wedging of the implant). In this case, the rasp is no longer necessary, the preparation being performed with the reamers, only the test prosthesis function is of use. The proximal test components can be used as rasps and serve only to adjust the length of the lower limb. Primary stability is ensured with the conical area of one of the distal components; 140 or 200 mm are mostly used. NB.The conical area is demarcated by 1 transversal line. The 140 mm distal components can be used both as rasps and as test prostheses. Only the test prosthesis function is of use when primary stability is achieved in the diaphyseal region, mostly after a femoral flap has been created. The distal components of length 260 mm are seldom used. 3.3 PROXIMAL TRIAL PARTS (FIG. 8) These instruments enable the definitive distal component to be implanted when a femoral flap has been completed. They allow to adjust precisely the antetorsion and mostly to determine the size of the proximal definitive component after having secured a Figure 7 stable wedging of the distal component. There is a proximal trial part corresponding to each definitive proximal component and it is possible to adjust its antetorsion by +/-30° (fig.8 A). It is assembled to the definitive distal component by means of nut screwed to the threaded part of the morse taper without any contact with the morse taper (fig. 8 B). Figure 8 NB. For the instrument allowing the assembly of the 2 components of the definitive implant (torque wrench) or the disassembly of a proximal component refer to appendices 1 and 2 of the surgical technique.

CONCLUSIONS OF THE FIRST PART

WHY MAKE THESE CHOICES?

1. CONCERNING THE CONCEPT When selecting a cementless implant, a careful decision should be made and it is necessary to prefer the well-defined concepts to ensure primary stability, and according to us, there are three of them: the press-fit, the interlocking system and the custom-made stems. 1.1 WHY CHOOSING THE PRESS-FIT? We have made the choice of the press-fit for three main reasons: – This concept has proved its efficiency for many years and we think it is an effective way to ensure primary stability, once the fundamentals of the method are well understood and if they are applied carefully. – This method can be relevant for a great number of cases of loosening, from the most simple to the most complex, which allows to gain a good command of the technique, if one does not change the type of the implants used too often. – This method does not upset too much the biomechanics of the bone (transmission of the loads) if the implant is well designed and if the placement of a too invasive implant (long stem) is avoided each time it is possible. These different reasons should not lead us to ignore that this demanding method has its own limitations (analyzed further down). The requirements of the method should be well known as it is the way to avoid its drawbacks, especially late abnormal mobility, always due to a technical error and easily mastered if one can make good use of modularity. 1.2 WHY NOT CHOOSING A DIFFERENT CONCEPT? The answer is obvious for the custom-made stems: this concept is difficult to apply in daily practice, because it entails major restraints during the assembly of the implants and also economically. Moreover, it is possible that these implants ensure a too tight bone-implant contact surface, sometimes resulting in difficulties during the implantation and, in a longer term perspective, in a poor transmission of the loads. Regarding the concept of distal interlocking bolt stems, proposed by Vives and Picault (28), the answer should be more discriminate, because this well defined method has now been tried and tested. However its does not only bear advantages and several remarks can be made against this concept: there is a real risk of material fracture for the stems with a small diameter; this concept can

36

First part – The concept of Press-Fit

disturb the physiological transmission of flexion forces; in some cases, it entails the placement of a stem longer than necessary; when the corticals are very weakened by an osteoporosis, the quality of the interlocking system is questionable. In conclusion, the concept of interlocking bolt stem, just like every other concept, shows some advantages as well as drawbacks. It also entails some requirements to be complied with, detailed precisely by Van de Velde (24) and it is interesting to note that are not very different from those imposed by the press-fit concept, which leads us to think that the specifically French controversy between the supporters of one or the other method is pointless except for a comparison, as objective as possible, between the results of the 2 concepts.

2. CONCERNING THE IMPLANT AND THE MODULAR SYSTEM 2.1 EVERY IMPLANT IS A COMPROMISE The stems of the PFM-Revision system of first generation (called presently Revitan Straight or PFM-Revision system of 2nd generation) have been designed to ensure primary stability only by a press-fit effect and we consider they bear the main characteristics allowing to achieve the goals set by the method. However these characteristics have not always been optimized, which, most of the time, was to avoid weakening the implant. The following two examples are a good illustration of the necessity for a compromise in most cases. – A 44 mm offset can sometimes seem insufficient but it is useful to know that, the bigger the offset, the bigger the flexion forces are on the le level of the morse taper, which is already located in a fragile area. – The more accentuated the conical slope, the easier the wedging is, but a shape which is clearly conical weakens the implants of a small diameter; conversely, a less accentuated conical slope makes the wedging more difficult to achieve but, in this case, the stems of a small diameter are not too weakened and the risk of fracture is negligible. For a defined concept, every implant is a compromise and the suitable implant is the one which can guarantee the best possible compromise. 2.2 ADVANTAGES OF A MODULAR SYSTEM Though following the release of the study by Wagner (29), the PFM-Revision system of first generation shows significant differences because of the modularity and how it can be use. These 2 systems require a different approach even if they resort to the same concept and to implants, which except for the modularity, bear similar characteristics (stems with fins). The advantages provided by a modular system have been underlined by Essig and Puget (10), who, as far as we know, is the first author, after Bousquet (3), to have made use of the modularity for revision stems. These advantages are of three kinds: Wide range of implants. The first advantage of a modular system is the wide range of stems available with a limited stock of implants. This possibility makes the control easier and can give confidence to the surgeon, because during a revision surgery, it is always difficult to anticipate beforehand which implant will be used. Flexibility in the choice of the implant. The design of the stems and the modularity allow the assembly of different types of implants: ● It can consist in a stem which design and size are not very different from a primary stem, and these implants can be used for “simple” revisions;

Conclusions of the first part – Why make these choices?

37

● It can consist in a longer implant, specifically designed for revisions with significant bone destruction. During the preparation of the femur, the choice of the implant and of the placement of the definitive stem. It is certainly in the course of these three stages of the surgery that modularity provides several decisive advantages (refer to surgical technique). We consider that it is not possible to ensure a real press-fit effect if a modular system is not available (implants or ancillary instruments), and in our opinion, it is also an excellent way to make the choice of a short stem.

3. WHAT SHOULD BE AVOIDED! We think that some caution is necessary regarding two types of implants: those which do not refer to any defined concept to ensure primary stability with a press-fit effect (when this choice has been made) or, conversely, and which is unquestionably more insidious, the implants which combine several concepts in order to ensure “unfailingly” primary stability. The first category of implants is often composed of primary stems which have been made longer. These implants often refer to an indefinite concept and their main characteristic is their size, which is not always an advantage, even for a revision. The longer the stem, the higher the risks of a 3-point contact are, and it is likely that most of these implants will wedge this way, especially when they are implanted only with an endofemoral approach. It is however important to underline that a 3-point contact does not always result in an unstable wedging if we consider the great number of long stems implanted during revisions; but, in most cases, it is a stem longer than necessary, which is in contradiction with our conviction to always prefer a short implant. The second category of implants is becoming more and more usual in the market of revision prostheses and the prevalent example to be found is a stem, at the same time anatomical (i.e. curved), bearing an interlocking system and press-fit. We consider that these implants have more drawbacks than advantages, in spite of the first impression, which can be comforting. Indeed, we have noted that curved stems bear numerous drawbacks, and if a proximal fixation is possible, the choice of a straight stem is preferable. Moreover, if a diaphyseal fixation is required, in many cases, the choice of a curved stem entails a stem longer than necessary because it is easier to guarantee a short fixation with a straight stem than with a curved one. The interlocking system always results in a weakening of the implant, which means that this method can only be applied to implants with a sufficient diameter, and of a cylindrical or conical shape.Additionally,as for the curved stems,the choice of a stem bearing an interlocking bolt system very often implies the use of a stem longer than necessary, especially when a short diaphyseal fixation is possible. Regarding the stems bearing fins, the advantages of which have been seen earlier, the interlocking system can prove ineffective on these implants because it would weaken them too much and the risk of fracture would be high for the implants of a diameter inferior to 18 or even 20 mm. These few remarks are sufficient to demonstrate that it is at least incautious to consider combining several concepts on the same implant and it is likely that these implants will wedge only thanks to their size, which means that the surgeon will often implant a stem longer than necessary. We consider that these implants bear more the potential drawbacks than the advantages of each method, and finally, they provide a false sense of safety. When the press-fit has been selected, it is always preferable to choose an implant which only objective is to comply with the requirements imposed by this method.

SECOND PART

PREOPERATIVE PLANNING

“A reflection which cannot be defined with clear words will be very difficult to put into action.” A. C. Masquelet The preparation for revision surgery is done in three sequential stages: A radiological analysis of the femur, followed by the determination of the surgical strategy and finally the making of a preoperative drawing. In general, the term ‘planning’ is used for these 3 distinct steps which should always precede the implantation of a stem. A planning consists of defining precise goals and is based on the preceding analysis of the xrays, which in itself is mandatory for establishing a strategy for the operation. We therefore use the term ‘planning’ in a broader sense to summarize the different steps which have to been made before the surgery starts.

CHAPTER 1

RADIOLOGICAL ANALYSIS OF THE FEMUR

INTRODUCTION The radiological analysis which has the goal of establishing a strategy for the planned revision should be differentiated from the radiological examination which will help to appreciate the result of that operation in the future. This distinction is important because the factors which are helpful for planning a revision surgery are not directly related to those factors which are examined in order to evaluate the final results of that operation (e.g. curvature of the femur or difficulties in excision of the cement). To establish a strategy for the operation means to perform at least a precise and complete analysis of the whole femur. For this purpose, 4 x-rays are required: An a-p-view of the pelvis, the concerned hip (centered on the loose prosthesis), the hip ap and axially, with a distance of 15 cm distally to the tip of the loose prosthesis. Considering the large amount of incomplete x-ray investigations in daily practice, this point is of utter importance. The responsible surgeon should insist on the quality of these pictures before he starts a revision surgery. The preoperative analysis of the x-rays should not be limited to mere assessment of the importance of the granulomatous lesions or of the cement which has to be excised. If the choice of a cementless fixation with a press-fit straight stem has been made, the appreciation of the morphotype as well as of the osteoporosis is of significant importance. At the end of this chapter, we will discuss the problems arising from the classifications which are proposed in the present literature for grading of defects. We are of the opinion that establishing a hierarchy of bony lesions or of the expected difficulties, as is done by most of these classifications, could be dangerous or misleading in daily practice, as the criteria are not precise enough for the requirements of the press-fit concept.

42

Second part – Preoperative planning

1. THE PREOPERATIVE X-RAYS In any case, four basic x-rays are required for the analysis of the different parameters that will allow the surgeon to establish the strategy of the operation conforming to the concept of press-fit fixation. HIP, AP- VIEW (FIG. 1) Centered on the loose prosthesis, in order to allow the evaluation of the amount and extension of bony lesions. HIP, AP AND AXIAL VIEW (FIG. 2A AND B) Visualizing the femur over a distance of 15 to 20 cm beyond the loose implant, always including the greater trochanter. It is a fact that these long ap- and axial views are often missing in the preoperative analysis. They are of utmost importance for the appreciation of the morphotype, the bone structure outside the loosened zone or for the detection of a distant granuloma formation.

a

Figure 2

Figure 1

b

Figure 3

PELVIS, AP- VIEW (FIG. 3) Needed for the analysis of the equilibrium of the pelvic ring, the location of the center of rotation, both in order to restore the architecture of the hip and correct a leg length difference. What should be avoided! To base an operative strategy on a simple ap-view, with insufficient length of the picture of the femur. A pathological curving could be overseen which could prove to be dangerous, when a straight stem with press-fit fixation is chosen. To evaluate any granuloma formation on long-distance x-rays, because their size will then usually be underestimated. To use digitalized pictures with variable sizing: A planning with templates will be difficult, unless an electronic planning device is available.

Chapter 1 – Radiological analysis of the femur

43

2. APPRECIATION OF DIFFERENT PARAMETERS 2.1 APPRECIATION OF THE SEVERITY OF OSTEOPOROSIS The severity of osteoporosis is always measured outside the loosening zones. 2.1.1 CLASSIFICATION 4 different stages can be distinguished by appreciating the thickness of the cortical bone and the geometry of the medullary canal in the isthmic zone of the femur. Stage 1: Excellent: no signs of osteoporosis; thick cortices and either a narrow, or strongly conical shape of the medullary canal. Stage 2: Good: no osteoporosis. Cortices are normal and the canal is conical Stage 3: Medium: Slight osteoporosis, thinned cortices, but canal of a cylindrical shape. Stage 4: Poor: Severe osteoporosis, very thin cortices and clear cylindrical shape of the wide medullary canal.

1 – excellent

2 – good

3 – medium

4 – poor

CONCERNING THIS CLASSIFICATION: The intermediate stages 2 and 3 are not easily distinguishable. The appreciation of the geometry of the medullary canal is helpful in differentiating these two stages. If the canal is cylindrical, the classification will always be stage 3, even if the cortices are not thinned. The term of conical shape of the medullary canal is not always easy to appreciate in the x-rays. However, it will be also be used, when the canal can be made conical by means of the conical reamers (in a narrow canal with thick cortices). In the same way, the term of cylindrical shape also means that one is dealing with a canal geometry that will be difficult to convert into a conical form during reaming (wide canal with thin cortices). Stage 4 will only be seen in a few patients. It is of great importance to be outlined, because in this stage the risk for complications is high and we here often face the limitations of the press-fit concept.

44

Second part – Preoperative planning

2.1.2 WHY SHOULD THE SEVERITY OF OSTEOPOROSIS BE APPRECIATED? The length of the contact zone between bone and implant will largely depend on it. Excising a thick layer of cement which is tightly adhering to thinned cortices (due to osteoporosis) can be a very challenging and dangerous experience. If there is a severe osteoporosis with thinned cortices and a wide cylindrical canal, the concept of a press-fit fixation is to be questioned critically. The risks of a stress-shielding phenomenon are always higher in presence of osteoporosis. It must be underlined that the appreciation of the degree of osteoporosis is lacking in all the important current preoperative classifications. We believe that this is an error, since the amount of osteoporosis plays an important role in planning the strategy of a revision surgery. 2.2 EVALUATION OF THE BONY DEFECTS This evaluation is made in the zone of the loose implant and deals with all types of bone loss. 2.2.1 POSSIBLE TYPES OF DEFECTS The evaluation should not be limited to granuloma formation. Changes which are the result of a former stress-shielding or the result of a mechanical wear pattern, should be appreciated as well. GRANULOMA FORMATIONS (FIG. 4) They occur as a result of a reaction to wear debris. The amount (quakity of the corticals) as well as the location (in axial and a-p-view) must be noted. A marked granuloma formation can occur at a great distance from the zone of production of these wear debris. Their location beyond the isthmic zone is possible. This makes the curettage difficult when an endofemoral approach is chosen. LOSS OF BONE STOCK DUE TO AN ADAPTIVE REMODELING PHENOMENON (FIG. 5) In general, this loss is summarized under the term of “stress-shielding” and is the result of an insufficient loading of the 䊳 bony structures. It usually indicates a diminution of the mechanical strength of these cortices (reduced bone density Figure 4 and/or thickness) and can occur in cemented as well as uncemented stems. In an advanced stage, it can even result in a loading bypass, with massive cortical atrophy, as it is sometimes observed in uncemented rigid prostheses that rely on a diaphyseal fixation mode.

Chapter 1 – Radiological analysis of the femur

MECHANICAL WEAR OF THE CORTICAL STRUCTURES (FIG. 6) Abnormal motion of the stem within the cement mantle can result in severe inner abrasion of the cortices. This is the third type of bony lesions. The removal of the implant can be difficult in these situations and lead to fractures, especially if the proximal opening in the greater trochanter is not made large enough at the beginning of the operation. 2.2.2 CLASSIFICATION OF THE DEFECTS The following classification is slightly different from other methods, since we have tried to make groups of a large number of patients for each stage. The defects are classified in 4 stages depending on the extension of the lesion, and making reference to the zones of Gruen (11)

Figure 5

Figure 6

Stage 1: no defects, or located in zone 1 and/or 7 (independent of their amount) Stage 2: defects in the metaphyseo-diaphyseal region, corresponding to zone 2 and/or 6 Stage 3: diaphyseal destruction of 1 cortical: lateral (zones 1-2, and 3) or medial (zones 7 – 6, and 5) Stage 4: destruction of 2 corticals of the diaphysis, or pathologic fracture

Stage 1

Stage 2

Stage 3

NB. The examples showed here only deal with granuloma formation.

Stage 4

45

46

Second part – Preoperative planning

COMMENTS ABOUT THIS CLASSIFICATION These lesions can be found simultaneously in the same patient. However it can be said that lesions due to stress-shielding are seldom seen together with lesions due to granuloma formation. ● Concerning stage 1: These lesions are strictly localized in the metaphysis (zones 1 and/or 7) and there is no lesion to be seen in the metaphyseo-diaphyseal region. ● Concerning stage 2: Lesions in zone 2 and/or 6 are often associated with metaphyseal defects (zone 1 and/or 7). Lesions occurring simultaneously in the metaphysis and another diaphyseal zone, and which are restricted to a limited area, should be included into stage 2 (e.g. varus tilting of a stem with subsequent lesion of the calcar and an erosion of the lateral cortex in zone 3). ● Concerning stage 3: These are lesions which extend mostly to one cortical,in most cases the medial one,the other remaining almost intact.In these cases,zones 3 or 5 are always involved.Every granuloma formation in zone 4 (tip) or distant to it, shall be classified as stage 3, if they appear aggressive. ● Concerning stage 4: All lesions which involve both corticals are classified as stage 4, except for those lesions that include the greater trochanter. Pathologic fractures will belong to stage 4, independent of the good quality of the bone stock in other regions. Stress-shielding defects or its equivalent (osteoporosis) which touch both cortices are most times classified as stage 4. A defect will always result in a weakening of the cortices and it will therefore influence the choice of the type of approach to the femur and the fixation-zone of the implant. 2.3 APPRECIATION OF THE MORPHOTYPE To analyze the morphotype means to look for any curving of the femur in the frontal plane or in the sagittal plane. Up to now, this important analysis is not mentioned (except by Vives and Picault, 28), unless it makes reference to sequelae of fractures with axial deformations. HOW CAN A FEMORAL CURVING BE DETERMINED? (FIG. 7) ● In the frontal plane, by means of an x-ray showing a sufficient length of the femur, using then the template of a long straight stem. After an appropriate centering in the diaphysis, the morphotype is analyzed by observing the long axis of the template and its relation to the level of the lesser trochanter. In varus deformity it will be lateralized, and in a valgus deformity of the proximal femur it will be medialized. Only if the line is centered in the whole femur, will it correspond to a straight femur in the frontal plane (ap-view). ● In the sagittal plane, the procedure is the same; however the analysis of the morphotype can be made difficult simply because of the technical problems due Figure 7 to obtaining x-rays of good quality. In loose stems, a pathologic curving of the femur can often be observed, if it is meticulously looked for. 2.3.1 CURVING(S) OF FEMUR IN THE FRONTAL PLANE VARUS DEFORMITIES They are frequent. It is also important to consider a minor curving (fig. 8A) and not to focus only on major ones (fig. 8B). In the frontal plane, a femur is either straight or curved!

Chapter 1 – Radiological analysis of the femur

a

47

b

Figure 8

Figure 9

VALGUS DEFORMITIES (FIG. 9) Although they are less frequent, they also will be an obstacle for the insertion of a straight stem. These deformities are frequent after fractures, angulation osteotomy (Milch- Batchelor type) or in hip dysplasia. 2.3.2 CURVING(S) OF THE FEMUR IN THE SAGITTAL PLANE Slight antecurvation and the double axial curving. Since the normal femur is not straight in the sagittal plane, we will not take into account every curving of the femur: e.g. the physiologic slight antecurvation (fig. 10 A) or the double axial curving (fig. 10 B).

Figure 10A

Figure 10B

48

Second part – Preoperative planning

NB. A double axial curving means that the diaphyseal recurvation is being compensated by a proximal antecurvation, thus resulting globally in a very slight deformity of the femur in the sagittal plane. In general these curvings are not an obstacle to the insertion of a straight stem, because in both cases it will be possible to prepare the femur on a sufficient distance for the press-fit implantation of the straight stem. This proves to be true especially if the stem is short, e.g. 120 or 140 mm (the whole implant length being around 200 mm). We would caution the surgeon to choose the transfemoral approach (with a femoral flap), every time he decides to implant a longer stem, even if the curving of the femur is minor in frontal or axial plane. GLOBALLY ACCENTUATED CURVING IN THE SAGITTAL PLANE (FIG. 11). It should be taken into account since this will be the most frequent single obstacle for a straight stem with press-fit fixation, even if a short stem is planned. NB. Globally accentuated means, that the ventral curving of the diaphysis is not being compensated by a proximal antecurvation in the axial plane. A global curving should only be taken into account, if the femur is straight in the frontal plane. CLASSIFICATION OF THE DIFFERENT MORPHOTYPES Two groups are identified depending on the presence or absence of a curving in the frontal plane, and on the amount of curving in the sagittal plane: 1: Straight femur (in the frontal plane) and a slight curving OR a double axial curving. 2: Curved femur in the frontal plane, of whatever amount and the morphotype in the sagittal plane, OR an accentuated axial curving, even if Figure 11 the femur is straight in the frontal plane. If a straight stem with press-fit fixation has been chosen, the analysis of the morphotype is mandatory. Any curving of the femur be it in a frontal or sagittal plane will end up as an obstacle for the creation of a reliable bone-implant contact surface. In this case, the surgeon should better change to a transfemoral approach by using a pediculated femoral flap.

2.4 EVALUATION OF THE CEMENT- BONE INTERFACE Having made the choice of an uncemented stem for the revision means always having to perform a complete extraction of the cement mantle. It is therefore important to be aware of the possible difficulties associated with this task. The evaluation should not only include the thickness of the mantle but also the quality of the corticals.

Chapter 1 – Radiological analysis of the femur

49

2.4.1 WHAT SHOULD BE APPRECIATED ● The thickness of the cement mantle in the intermediate zone of the femur. If excision of the proximal cement mass is usually not difficult, this is not the case for the intermediate zone, especially if a thick mantle is adhering to fragile cortices. There is a major risk of either perforation or fracture; therefore, vision on the working zone between cement and bone must always be good. ● The distal cement plug, its density and length, the adjacent bone and its quality are to be evaluated. Needless to emphasize the difficulty to excise the cement, mostly if the corticals are thinned. The excision of this plug should be complete, since remnants could misdirect the reamers and lead to a perforation. ● The region of the tip of the former stem. If there is no plug, appreciate the quality of the adjacent corticals. Be aware that any eccentric position of the old stem can misdirect drills, burrs or reamers and be the cause of a perforation. If in such a situation the surgeon does not do a femoral flap, a local femoral window will often be necessary. ● In removal of an uncemented stem, a medullar bony plug or an H-console may be present, which sometimes proves to be difficult to perforate and that could also be the origin of an accidental perforation of the femur.

2.4.2 CLASSIFICATION The classification should summarize the foreseeable difficulties for the step of cement extraction, by taking into account the strength of the cortices. 2 groups of patients can be observed:

1 – Absence of any difficulties

2 – Presence of difficulties

No cement plug OR plug < 4 cm and stable cortices

Plug > 4 cm, even if stable cortices

Plug < 4 cm, eccentric tip position with weak cortices OR thick intermediate cement mantle and thin cortices

50

Second part – Preoperative planning

The radiological analysis of the cement bed should not be limited to the eventual presence of a plug; it should also evaluate the quality of the surrounding bone. The predictable difficulties to remove the cement should be taken into account when opting for the femoral approach.

3. CONCLUSIONS: WHAT ABOUT THE EXISTING CLASSIFICATIONS? DO WE NEED A GLOBAL CLASSIFICATION FOR DEFECTS OF THE BONE STOCK? 3.1 THE EXISTING CLASSIFICATIONS There are many classifications and every new one seems more complex than the former, which means that the ideal one that would adapt to all the concepts and that would be easy to use by the surgeon is not yet found. In the U.S. the two classifications of Paprosky (23) and the A.A.O.S. (6) are those currently used. France has its own called the classification of the SO.F.C.O.T. (27). Some remarks should be made concerning these different classifications. In general those of the U.S. look rather similar which is not surprising, as all are established under the view of similar implants and revision techniques. The classifications are often incomplete. The classification of Paprosky (23) focuses exclusively on the defects of the bone stock. Other classifications are slightly different but are also mainly based on the bony defects encountered. This way of proceeding can prove to be dangerous and is certainly incomplete. It is dangerous because evaluating only the bone defects by measuring the amount of granulomatous lesions leads to underestimation of the real defect, since this is difficult to detect before the operation, even on very good quality pictures. Moreover, these classifications concentrate only on the proximal femur and tend to ignore distant lesions. It is incomplete because bone defects are not the only obstacles for the stable implantation of a straight cementless revision stem. It is therefore mandatory to take into consideration any curving of the femur, be it for the implantation of a stem with press-fit fixation, or for the use a long stem of another system. To be complete, there should also be an evaluation of the bone quality, especially if the uncemented implants are expected to be at least partially osseointegrated. In those classifications which unite more criteria, as the A.A.O.S or the SO.F.C.O.T classifications (1999 version ) (lit. 5), the ranking of the criteria is not easy to define, which gives rise to confusion and makes their practical use difficult. Finally, some studies have shown that none of these classifications gives reproducible results, the least being those of A.A.O.S. and of Paprosky (lit. 12 and 5). In our opinion, a pre-operative classification can only be of interest if the surgeons who are using it will adopt the same concept for their revision surgery. The transfer of one single classification to different concepts is probably not feasible.

3.2 DO WE NEED A GLOBAL CLASSIFICATION? The preoperative radiological evaluation of the femur which should help in outlining a strategy for the operation cannot be a global classification, since it could mislead the surgeon to a choice which is not in accordance with the goals that are fixed by the particular method he had chosen, e.g. a press-fit fixation mode. Global classification systems are either incomplete or too rigid in their application; they even can be dangerous in some situation, as the choice of the parameters

Chapter 1 – Radiological analysis of the femur

51

and their individual importance varies according to the chosen concept. The best example to illustrate this point is the presence of an accentuated curving of the femur which could, by itself, justify the need for an additional transverse femorotomy if an uncemented straight press-fit stem had been chosen. As such, every concept will need its own evaluation. From our point of view, the preoperative analysis of the x-rays will first induce the choice of the parameters, from which the strategy will emanate, always in accordance with the concept of the chosen implant and it will then help to establish a classification of the lesions for every single parameter. As a next step, a ranking of these parameters will be established and pondered according to their importance and influence on the goals that have been fixed by the adopted method. If a press-fit fixation has been chosen, 2 main parameters have to be considered: First the morphotype of the femur, because it will be impossible to create a bone-implant contact surface for a straight stem in a curved femur; then the extent and localization of the defects, in order to decide about the approach to the medullar cavity (as to avoid any aggravation of the defects) and finally to look for the best contact zone between bone and implant, where to ensure primary stability. Possible difficulties for the excision of cement, as well as an pre-existing osteoporosis are of lesser importance, especially if the surgeon hesitates between the two approaches, or less frequent, if the amount of osteoporosis endangers basically the concept of press-fit. The preoperative analysis of the x-rays should have only one goal: A precise analysis of the femur in order to establish the strategy of the operation in accordance to the chosen implant and its special concept of fixation. The method proposed in this book has been established under this point of view and we feel that the criteria we used correspond to the concept of a press-fit fixation. We believe that in the future, a more important place will have to be given to axial x-rays in the general evaluation of the bone defects.

CHAPTER 2

DETERMINING A SURGICAL STRATEGY

INTRODUCTION Determining a surgical strategy is only possible if the surgeon is perfectly aware of the objectives to be reached and after having performed a careful analysis of the various parameters, which, according to A.C. Masquelet, allow to “base one’s deduction on facts and events”. These “facts and events”, considered as a whole, are often in contradiction with the prior objective of every cementless arthroplasty, which is the necessity to achieve, in any case, a perfect primary stability. There are two options: ● It is possible to opt either for the femoral flap, which will always be a pediculated one, with a medial corticotomy, if this is carried out, primary stability can only be diaphyseal. ● If an endofemoral route has been chosen, primary stability in the metaphyseo-diaphyseal area or in the diaphyseal area is possible. NB. A femoral approach using a trochanteric osteotomy can be associated to an endofemoral route, since the way of achieving primary stability is similar. Determining a surgical strategy, is first of all, to choose a femoral approach which will be adapted to the anatomical situation and this choice will depend on the area of femur where primary stability will be achieved.

54

Second part – Preoperative planning

1. INDICATIONS FOR A FEMORAL OSTEOTOMY The indication for a femoral osteotomy, by doing a pediculated flap (in the following section called “femoral flap”) is frequent, since it greatly facilitates the revision surgery. 1.1 INDICATION FOR A FEMORAL FLAP There are 3 principal situations which give the indication for doing a femoral flap: Presence of a femoral curvature, fragile corticals or difficulty of removing the cement. All 3 conditions can be found in the same patient but one is already enough to justify for the choice of a femoral flap. It may also be indicated in two other less frequent situations: The revision of a septic case and the removal of a solidly fixed cementless stem. 1.1.1 FEMORAL CURVATURE(S) Due to the fact that the press-fit concept needs a contact surface between bone and implant, the presence of any femoral curvature will call for a femoral flap, in most cases. CURVING IN THE FRONTAL PLANE ● If the curving is important, a femoral flap is mandatory, whatever the circumstances. ● In these cases, an osteotomy of the medial cortex can be added to the lateral flap, bringing the medial cortex back in contact with the stem (fig. 1), in order to complete primary stablity. ● If there is a less accentuated curving, the need for a femoral flap can be questioned. An endofemoral approach or a trochanteric osteotomy are possible, if there are reasons not to do a flap or if a proximal fixation must be achieved.

Figure 1

Figure 2

Chapter 2 – Determining a surgical strategy

55

Most times there will be a varus curvature. If the curving was in valgus, the procedure would be the same. IMPORTANT CURVATURE IN THE SAGITTAL PLANE If the sagittal curving is important, it will be prudent to do a femoral flap even if the femur was straight in the frontal plane. If a short stem can be implanted, there is not always the need for adding a medial cortical osteotomy (fig.2). If a long stem was implanted, the danger of a 3-point fixation would have to be feared in the sagittal plane. To prevent this, a medial osteotomy of the cortex is often needed. If the axial curvature is important, one should remember that a 3-point fixation could take place in the case where the lateral flap is not wide enough. Furthermore, excision of cement on the convex side is most difficult in those cases where a lateral flap is absent. 1.1.2 WEAKENED CORTICAL STRUCTURES The reason and the amount of the weakening of the cortical structures are to be studied. – If the weakening is due to granuloma formation and the defects are classified stage 4, there is invariably an indication for a femoral flap, regardless of other conditions (fig. 3). If the defects are in stage 3 (only one cortical touched), most time a femoral flap is recommended. However, in some cases an endofemoral approach is possible if the configuration of the femur is a straight one. If the defects are stage 2 (e.g. in the metaphyseo-diaphyseal zone) and the femur is straight in ap-view, the choice will usually be an endofemoral approach. In the case of large defects in the metaphyseo-diaphyseal zone, a flap is recommended in order to avoid any unnecessary enlargement of the bony defects and for a precise curettage of granuloma. – If the weakening of the cortical structures is due to an advanced osteoporosis (osteoporosis stage 4), the weakening and especially the enlargement of the whole femur theoretically would call for a femoral flap. This choice however makes a long stem necessary which is clearly undesirable. The authors prefer in those cases to implant a short, stem with a very proximal press-fit fixation, using an endofemoral Figure 3 approach (for straight femur in the frontal view) or by means of a trochanteric osteotomy, with an additional longitudinal osteotomy of the proximal femur in order to reduce its diameter with the help of circular wire cerclage. As an alternative, local impaction grafting may be helpful too in order to ensure a tight epiphyseal contact zone with the prosthesis (see also fig. 13, p. 61).

56

Second part – Preoperative planning

1.1.3 CEMENT REMOVAL The complete removal of the cement mantle is of outstanding importance in stem revision and this step can turn out to be very stressful for the surgeon (as it is expressed by the high number of different tools for cement removal actually on the market). To avoid this, the surgeon should mind the 3 following principles: To keep a direct vision into the medullary space, remembering that any curving of the femur in the sagittal plane will be the first obstacle to cement removal. Avoid any unnecessary damage of the bone stock and especially devascularization of bony structures. It is always better to do a vascularized flap only for the cement removal than to change the strategy in the middle of the operation. Complete removal of the cement mantle helps to avoid an eccentric reaming, which could be induced by an overseen island of cement within the medullary canal. DISTAL CEMENT PLUGS (FIG. 4) If such a distal cement plug is very long, the cutting of a femoral flap is recommended for the only purpose of plug removal. The problem lies in the fact that the flap should not be too long and hence damage the isthmic region where the seating of the stem is to take place. A flap reaching only half the length of the plug is usually sufficient for the removal of the distal cement piece as this will allow for an excellent view onto the working area thus avoiding any cortical perforation with drilling tools. The cutting of a distal femoral window is an alternative to the flap formation, but only if there is the possibility of a proximal fixation of the new stem through an endofemoral way. THICK CEMENT MANTLE WITHIN THIN CORTICALS (FIG.5) Sometimes the distal cement plug is rather small, but the cement mantle proves to be very thick in the intermediate or distal region of the femur and the corticals are thin. If there is an additional oblique position of the stem within that canal, the endofemoral excision of the cement can be strenuous and risky. In those conditions a femoral flap should be preferred, instead of running the risk of an incomplete cement excision or having to change the strategy in the middle of the operation. NB. Most frequently this situation is encountered with big granuloma formation. But if the fragility is due to an osteoporosis, we prefer to make an endofemoral approach, if the femur is straight in the frontal view. Remember that even if there are only few problems to be expected with cement excision, the presence of any additional obstacle will immediately call for a transfemoral approach with flap formation.

Figure 4

Figure 5

Chapter 2 – Determining a surgical strategy

57

1.1.4 SPECIAL CASES Transfemoral approach for septic revisions. The femoral flap, advocated by Vielpeau will allow for the complete removal of the cement mantle which is required in any septic revision surgery. It will also allow for a meticulous cleaning of the medullary canal. Removal of a stable uncemented stem. Basically, every implant should be removable. This can prove to be difficult with uncemented stems, especially if they show a structured surface. A femoral flap can be very helpful under these circumstances.

1.2 CHOICE OF THE IMPLANT AND ITS FIXATION MODE IN THE DIAPHYSIS After having made a femoral flap, the zone of primary stability will always be in the diaphysis. Whenever possible, a short fixation surface should be reached and, as a rule, it will always be better to prefer the diameter to the length of the stem. 1.2.1 SHORT DIAPHYSEAL FIXATION MODE There is a good quality of the cortical structures that will allow for a deep conical reaming. If a flap is done, it is important to avoid any damage to the isthmic region. A contact zone of about 40 to 50 mm length is enough in order to ensure a primary stability of the implant (fig.6). Usually the distal part is short, having a length of 140 mm. The dimensions of the proximal part will only be chosen during the operation. A straight stem with longitudinal fins and a conical shape is a good compromise for achieving a good primary stability using the press-fit principle on a very short contact surface.

Figure 6

Figure 7

58

Second part – Preoperative planning

1.2.2 LONG DIAPHYSEAL FIXATION MODE This fixation mode will be considered when there is a sufficient quality of the corticals. The medullary canal is rather wide, of cylindrical shape and it will be difficult to create a conical seating on a sufficient distance within the bone (thin corticals). In these cases, the contact surface between bone and implant will be longer (usually 5080 mm) (fig.7) and it is advisable to use longer stems (distal piece of 200 mm). Diameter of the distal part and length of the proximal part will be determined during the operation. NB. An additional osteotomy of the medial cortex to the former femoral flap may be needed if the curving in the axial view becomes an obstacle to the seating of a long straight stem and /or with the goal of bringing the medial cortex back to the surface of the prosthetic stem (see also operation technique: option 1, femoral flap, fig. 17) Often the anatomical findings are not as clearly distinct as has been described and intermediate cases are encountered, especially in long patients. In those cases the corticals are most time of good quality and the medullary canal can be conical. Nevertheless the choice of a longer stem will be mandatory in order to reconstruct the correct leg length.

Chapter 2 – Determining a surgical strategy

59

2. INDICATIONS FOR AN ENDOFEMORAL APPROACH OR TROCHANTERIC OSTEOTOMY When a femur is straight in the frontal view and only slightly curved in the axial view an endofemoral approach should be considered, except when the defects are classified stage 4. The fixation mode will depend on the site of the defects and the presence or not of any osteoporosis. The goal is always to use a short stem and to aim for a proximal fixation. Intermediate cases are better treated with a trochanteric osteotomy if there are problems with an endo-femoral approach and one does not want to create a femoral flap for other reasons. 2.1 INDICATIONS FOR AN ENDOFEMORAL APPROACH 2.1.1 STRAIGHT FEMUR AND ABSENCE OF BONE DEFECTS (FIG.8) If the femur is straight in the frontal view, axially only slightly curved and the bone defects are only situated in zones 1 and/or 7, and there is no osteoporosis. Once the cement mantle has been excised, this situation will correspond to a primary implantation. The goal in this situation will be to look for a proximal fixation mode in the metaphyseo-diaphyseal junction, mostly on the level of the lesser trochanter (fig.9). We suggest the use of a short distal piece available (e.g. 120 or 140 mm) and a spout proximal part. The only obstacle which could be encountered would be a distal cement plug, necessitating a local femoral window cutting for its excision.In this case the longer distal piece of 140 mm Figure 8 Figure 9 is helpful (see fig. 8). 2.1.2 STRAIGHT FEMUR, BUT PRESENCE OF BONE DEFECTS (FIG. 10) This situation differs from the former one by the presence of bony lesions,mostly granuloma formations in the zone 2 and/or 6 of the femur (stage 2). In those cases, the goal will be to find a metaphyseo-diaphyseal fixation,sometimes with the help of a bone transplant into the medullary canal. If this option is not feasible, a short diaphyseal fixation will be the better choice, especially when the corticals are of good quality. The same attitude can be adopted when the defects are classified stage 3 (lesion of one medial or lateral cortical), especially when the corticals are not too fragile. A short stem of 140 mm and a spout proximal piece are then the most frequent choice. A femoral flap should be considered when there is an addition of 2 obstacles (mostly extensive granuloma formation and a distal cement plug), especially when the lesions in zones 2 and 6 are important.

60

Second part – Preoperative planning

Figure 10

Figure 11

2.1.3 STRAIGHT FEMUR, BUT OSTEOPOROSIS STAGE 3 Typical for this situation is the finding of a clear, but not very advanced osteoporosis (stage 3) and a more cylindrical shape of the medullary canal. An endofemoral approach would be possible, but the mode of fixation will vary according to the presence of any defects and their extent (a fact that can only be seen during the operation). If these defects are small, a metaphyseodiaphyseal fixation should be aimed at, with the help of endo-medullar bone-grafting and a cerclage-wiring of the femur, if necessary (fig. 11). Again, in this case a spout proximal piece should be chosen and a short distal stem of 120 or 140 mm, which should not fit too tight into the femur (by one diameter less than the medullary canal). If this option is not possible, there will be only the way of seeking an additional diaphyseal fixation simply by using a distal piece of 140 mm with the same size as the last reamer of the canal. This will result in a sort of global fixation of the implant (fig. 12). We feel that this way of global fixation bears the risk of a stress-shielding phenomenon but our controls showed only a mild reduction of cortical bone density of the proximal femur in those cases. Figure 12

Chapter 2 – Determining a surgical strategy

61

2.1.4 STRAIGHT FEMUR, BUT ADVANCED OSTEOPOROSIS STAGE 4 (FIG. 13) A heavy osteoporosis with thinned corticals and a cylindrical canal are the predominant elements in this situation. The shape of the femur is straight or slightly curved in both views. A straight stem with press-fit locking can only be selected if there is the possibility of a strictly proximal fixation. A diaphyseal fixation should be avoided. In such situations, a trochanteric osteotomy can be discussed in order to facilitate cement removal, using then an endomedullary bone grafting after having done a prophylactic cerclage-wiring, if necessary. Fig. 13 shows an example of a strictly proximal fixation without major transformation of the cortical structures within 2 years after the implantation, and this despite of the presence of a heavy osteoporosis (Case from Prof. J.P. Levai) Theoretically, a fixation could be realized with a long finned stem too, finding its stability in the diaphyseal region. This option bears in itself a heavy risk of later stress-shielding, since it will call for a long and thick stem. Figure 13

2.2 INDICATIONS FOR DOING A TROCHANTERIC OSTEOTOMY There are conditions where a trochanteric osteotomy could be considered as being helpful for the operation. In our view, these situations are not frequent since we advocate for an extended use of the femoral flap for revision surgery. 2.2.1 CURVED FEMUR These conditions are marked by a moderate, but very proximal femoral curving, often as sequelae after a previous varus osteotomy. A metaphyseo-diaphyseal fixation can be reached with a spout proximal piece and a short distal spout (120 or 140 mm) (fig. 14). If a diaphyseal fixation is to be done, a longer distal piece of 140 or 200 mm can be chosen, an option without too many risks if in absence of additional severe osteoporosis (stage 4). An endofemoral approach can also be chosen, but the opening in the greater trochanter will have to be very wide laterally, in order to avoid any varus implantation. This solution can be chosen when the bony structures of the trochanter are strong and wide enough.

Figure 14

62

Second part – Preoperative planning

2.2.2 SPECIAL INDICATIONS If there is a straight femur with a sunken loose prosthesis, the tip of the greater trochanter will be in a relatively proximal position, resulting in a very tight hip situation (fig. 15). A trochanteric osteotomy gives an easier and safer approach to the stem, since a weakening or even fracture of the greater trochanter can be avoided, and furthermore, it will still allow for a proximal fixation mode. We urge less experienced surgeons to do this osteotomy every time where an approach to the articulation could be difficult or a varus implantation could occur.

3. CONCLUSIONS AND SUMMARY If a surgeon decides to use a cementless stem with press-fit fixation mode, it is important he makes no error about the approach to the femur. It is less a matter of being dogmatic than logical during the preparation time of the operation; which means for the case of the press-fit implant, to always ensure a perfect contact surface between bone and implant and by this a stable wedging of the prosFigure 15 thesis. In daily practice, achievement of these 2 goals means very often that a femoral flap has to be done, since, without raising a pediculated femoral flap, it is not possible to create a contact surface between bone and implant using a straight stem in a curved femur. Furthermore, only this strategy makes the preparation of a short fixation surface possible, thus allowing for the perfect diaphyseal wedging of the implant in the bone (see operation technique). An endofemoral approach is possible whenever the femur is straight in frontal view and the defects are small. A proximal, metaphyseo-diaphyseal fixation is then achieved, always by means of a short stem. Even though not frequent, the trochanteric osteotomy keeps its limited but clear indications and it must be stressed that this approach has the merit to ease the implantation of a straight stem in cases where a femoral flap is not desired or a proximal fixation mode must absolutely be achieved for other reasons. Final sizes of the definitive pieces (diameter of the distal part, and length of the proximal part) are only fixed during the operation. The following table of summary deals with the radiological analysis of the femur and the choice of a strategy. 4 parameters are used for the radiological analysis (amount of curving of femur and difficulties with cement removal) and are combined in a binary mode with 4 stages according to their degree of severity (bone defects and amount of osteoporosis), which allows individualizing 6 main strategic options: Options 1, 3 and 5. These 3 options are fundamental: Option 1 with endofemoral approach and proximal fixation mode; options 3 and 5 with a femoral flap and distal fixation mode.

Chapter 2 – Determining a surgical strategy

63

Options 2 and 4. These are intermediate options, where either the endofemoral or the transfemoral approach can be discussed (and the presence or absence of difficulties for cement removal can be decisive for the one or other way of approach). The option 6 which is special and concerns only few patients. It deserves to be individualized, since most of the contraindications to the use of a press-fit stem can be found in this group, basically because a proximal fixation mode is not possible.

64

Second part – Preoperative planning

RADIOLOGICAL ANALYSIS MORPHOTYPE

DEGREE OF OSTEOPOROSIS

DEFECTS

SYNTHESIS: DESCRIPTION OF CASE STRATEGY OPTIONS

CEMENT

OPTION 1 Propitious situation: femur straight in the frontal plane and slightly curved in the sagittal plane. Absence of bone defects or localized onto zones 1 and/or 7 (no defects onto zones 2 and 6). The only obstacle could be the possible difficulties in removing the cement: plug or eccentric stem or thick cement layers (sagittal) adhering to fragile cortex (osteoporosis stage 3 or stress-shielding).

●

No Frontal plane femur straight (1)

Stages

Stage 1 Yes

OPTION 2 Intermediate option characterized by a femur straight in the frontal plane with defects onto zones 2 and/or 6, often combined with defects onto zones 1 and/or 7, but no lesion onto zones 3 or 5. In that case, evaluate the extent of the defects onto zones 2 and/or 6; take into account the quality of the corticals and the possible presence of difficulties in removing the cement: plug or thick cement (sagittal) or eccentric stem.

●

No 1 – Excellent

Stage 2 Yes

OPTION 3 This situation is characterized by a straight femur but with defects leading to a weakening of 1 or 2 corticals, with in any case, lesions of the cortex onto zones 3 and/or 5. It is often a granuloma formation; but also can be, abrasion of one or two cortical(s) due to a loosening bone-cement. When this weakening is the result of a stress-shielding or of osteoporosis, most of the time, both corticals are affected. Beware! In this situation, the femoral isthmus is preserved, otherwise (severe osteoporosis or bone degeneration) refer to option 6 .

●

2 – Good No Stage 3 Yes (1) Sagittal plane: femur straight or slightly curved

3 – Mediocre

No Stage 4 Yes

●

OPTION 5

OPTION 4 Intermediate option which applies only to slight curvatures in the frontal plane. In that case, evaluate the difficulties in removing the cement: plug or thick layers of cement adhering to fragile cortex (osteoporosis stage 3) or eccentric stem. Beware! If the curvature in the frontal or sagittal plane is accentuated carry out a femoral flap in almost any situation (see option 5).

●

No Stages

Stage 1 Yes

Femur curved in frontal or sagittal plane (2)

No 1 – Excellent

Stage 2 Yes No

(2) Global sagittal curvature = accentuated diaphyseal curvature

2 – Good

Stage 3 Yes

This option concerns either an accentuated femoral curving (frontal or sagittal plane), or a less accentuated curving but always combined with another obstacle, which is in the best case, limited to defects stage 2, but it can also be osteoporosis stage 3 and/or presence of difficulties in removing the cement and/or defects st. 3 or 4. Reminder. When the femur is straight in the frontal plane, any curvature must be considered and in the sagittal plane, only an accentuated curvature (global curving) has to be considered. Beware! In this situation, the femoral isthmus is preserved, otherwise (advanced osteoporosis or bone degeneration) refer to option 6.

No 3 – Mediocre

Stage 4 Yes

Femur straight or curved

Osteoporosis stage 4 or isthmic zone damaged (fracture or defects > stage 4)

Yes or no

● OPTION 6 Particular situation characterized either by the presence of an advanced osteoporosis (stage 4) with very thinned corticals and large and cylindrical medullary canal; or by the destruction of the isthmic zone of the femur that can be due to a fracture of the stem or to a late loosening of a long stem.

Chapter 2 – Determining a surgical strategy

65

STRATEGY: FEMORAL APPROACH, FIXATION TYPE(S) AND IMPLANTS FEMORAL APPROACH

FIXATION TYPE(S) AND IMPLANTS

● Endofemoral approach with a window (wide lateral and posterior opening of the greater trochanter) or trochanteric osteotomy if the greater trochanter is fragile (granuloma formations) or if it is an obstacle for the preparation of the femur (coxa vara or previous osteotomy) or if stiff hip or subsidence of the implant. Preserve the insertions of M. vastus lateralis (digastric osteotomy). In both cases, it is possible to create a femoral window to remove the thick distal cement and/or adhering to more or less thinned corticals.

Proximal fixation, in metahyseo-diaphyseal zone, over a height of 2 to 3 cm, adding, if necessary some bone chips into the medullary canal. If the proximal fixation proves to be unstable, aim at a short diaphyseal fixation to achieve a more a loss global fixation (avoid a tight diaphyseal fixation especially if osteoporosis stage 3). ● Implants: short distal component L.120 mm if metaphyseo-diaphyseal fixation or L.140 mm if diaphyseal fixation or window. Proximal spout component. A 200 mm distal component is neither advisable nor necessary, in this situation. Femoral flap = diaphyseal fixation (in isthmic zone) ● Short fixation mode with bone-implant contact over 3 to 5 cm. Always prefer

●

Femoral flap, usual choice, for removal of the cement, curettage of granulo-

●

2 possible choices depending on the presence or not of difficulties in removing the cement – Endofemoral approach or trochanteric osteotomy: if no difficulties in removing the cement and to look for a proximal fixation mostly if osteoporosis is stage 3. In that situation a femoral flap is possible and above all a wide lateral and posterior opening of the greater trochanter should be carried out. (refer to option 1) – Femoral flap: if major difficulties in removing the cement and, in that case, select a short diaphyseal fixation (refer to option 5) NB. In the presence of a cement plug and if a proximal fixation is possible, an endofemoral approach + window can be carried out if small defects are found onto zones 2 and/or 6. matous cysts and avoid any aggravation of the bone lesions. It is a semi-circular lateral flap and, generally, it is not necessary to combine it with an osteotomy of the medial cortex as the femur is straight in the frontal plane and slightly curved in the saggital plane. An endofemoral approach is possible if defects are stade 3 and if the cortical (lateral or medial) is not too thinned in zone 3 or 5. This option should be selected in case of osteoporosis st.3 to avoid a strictly diaphyseal fixation and look for a proximal fixation more or less global. In that situation, it is very often necessary to add some bone into medullary canal (refer to option 1) . NB. If a trochanteric osteotomy is discussed or necessary, it is often preferable to choose the option of a femoral flap. ● Femoral flap in any case, for it is either an accentuated curving, or at least 2

this option mostly if the cortex is good and the medullary canal more or less conical. In that case, use a 140 mm distal component with a proximal component mostly spout (cylindrical only if a long proximal component is necessary). ● Long fixation mode, with bone-implant contact over 5 to 8 cm if the medullary canal is cylindrical or depending on the morphotype (tall patient). In that case, use a 200 mm distal component combined with a proximal component mostly spout or cylindrical (the 260 mm components are limited to the treatment of a complication). NB. If the corticals are not too thinned, presenting a contact surface with the implant, it is possible to ensure a proximal fixation of the implant on the flap. This option is interesting to use in any case and even convenient when the diaphyseal

2 possible choices depending on the presence or not of difficulties in removing the cement – Endofemoral approach or trochanteric osteotomy: if no difficulties in removing the cement and to look for a proximal fixation mostly if osteoporosis is stage 3. In that situation a femoral flap is possible and above all a wide lateral and posterior opening of the greater trochanter should be carried out. (refer to option 1). – Femoral flap: if major difficulties in removing the cement and, in that case, select a short diaphyseal fixation (refer to option 5). NB. In the presence of a cement plug and if a proximal fixation is possible, an endofemoral approach + window can be carried out if small defects are found onto zones 2 and/or 6. obstacles are combined for the placement of a straight press-fit stem and one has to consider every curvature in the frontal plane, even the slightest one. In any case, a flap is safer: it allows to avoid a varus implantation with a 3-point contact in the frontal plane (which is always an issue with straight stems) and eases the complete removal of cement, the curettage of granuloma formations and avoids fractures when the cortex is fragile due to granuloma formations, osteoporosis, stress-shielding or any other cause.In that situation, it is often required to combine the laterla flap with an osteotomy of the medial cortex either to restore the bone-implant contact surface (significant varus position) or when there is an accentuated sagittal curvature. Additionally, restoring the contact between corticals and implant is very beneficial for bone regeneration and secondary osseointegration, especially when a long stem is required. A trochanteric osteotomy is a possible option when the femur presents an accentuated curvature and an osteoporosis st. 3 without major defects (st. 1 or 2). In that situation, a trochanteric osteotomy can ease a proximal fixation and allows to avoid a disphyseal fixation, which is always advisable. In that case (refer to option 1).

Press-fit is unstable or to avoid a too long diaphyseal fixation. Femoral flap = diaphyseal fixation (in isthmic zone) ● Short fixation mode with bone-implant contact over 3 to 5 cm. Always prefer this option mostly if the cortex is good and the medullary canal more or less conical. In that case, use a 140 mm distal component with a proximal component mostly spout (cylindrical only if a long proximal component is necessary). ● Long fixation mode, with bone-implant contact over 5 to 8 cm if the medullary canal is cylindrical or depending on the morphotype (tall patient). In that case, use a 200 mm distal component combined with a proximal component mostly spout or cylindrical (the 260 mm components are limited to the treatment of a complication). NB. If the corticals are not too thinned, presenting a contact surface with the implant, it is possible to ensure a proximal fixation of the implant on the flap. This option is interesting to use in any case and even convenient when the diaphyseal press-fit is unstable or to avoid a too long diaphyseal fixation.

In that case, the implantation of a press-fit stem should be carefully considered or could prove to be impossible (if femoral isthmus is destroyed) – If osteoporosis is advanced the risk for a stress-shielding are high when implanting a long stem of a large diameter. In that situation, it is mandatory to ensure a proximal fixation and avoid a diaphyseal fixation (Refer to option 1). The technique described by Exeter is another possible option. – When the femoral isthmus is destroyed (fracture or late loosening) the press-fit concept becomes inapplicable (a press-fit effect is impossible on the level of distal 1/3 of the femur). In that case, it is required to switch to a different concept and an unterlocking bolt stem can be recommended in this situation.

CHAPTER 3

MAKING A PREOPERATIVE TEMPLATE

INTRODUCTION The making of a preoperative template is the third step in the preparation of a revision surgery. The preoperative template helps in establishing a strategy and in realizing it. The template should be made on an anterior-posterior X-ray which shows the femur over a sufficient length about 15 to 20 cm below the distal tip of the loosened implant to avoid any off-axis errors. An axial view is always needed for the detection of a global axial curving of the femur, which must be taken into consideration if the femur is straight in the frontal view. The template has two goals: to mark every obstacle seen during the analysis of the x-rays (bone defects, curving, and cement plug) and to depict the reference lines which are helpful during the operation. It is not a goal of the template to set the sizes of the implants, since these are only determined during the operation by means of the test prostheses (diameter of distal part and length of the proximal part). The template may be prepared in 5 successive steps, but these steps apply only to the cases where a femoral flap is indicated.

68

Second part – Preoperative planning

1. TRACING THE CONTOURS OF THE FEMUR Trace the contours of the femur on the a-p-view, highlighting the zones of defects, the distal end of the loosened implant and the cement plug, if any. ● Mark the center of rotation of the loosened prosthesis and draw its relation to the tip of the greater trochanter. Warning! The axial view of the template only helps in determining any global axial curving in the case of a straight femur in the frontal view. It is not of any use for the preoperative template, because in case of an axial curving, rhe procedure is the same as for a curved femur in the ap-view: first, a lateral flap has to be performed, with an additional osteotomy of the medial cortex, if necessary. ●

2. TRACE THE AXES OF THE FEMUR ● Medullary axis: carefully center the template of a long stem (200 or 260 mm long distal component) in the diaphyseal region, trace the centromedullary axis and control its position relatively to the centro-medullary axis of the proximal femur, and its relation to the tip of the greater trochanter in order to avoid any off-axis errors. This is an important step of the preparation of a template, since it enables to evaluate the extent of any existing curvature. The curvature is often more apparent on the templates than on the X-rays.

Axis of the center of rotation: trace a line perpendicular to the centromedullary axis at the level of the tip of the greater trochanter. In principle, the center of articular rotation of the revision implant should lie on this horizontal axis. At this moment, the current leg length can be measured, but this will only be a helpful indication, since the correct leg length has to be determined during the operation, after the step of cup implantation and after having done some test-reductions. ●

3. DETERMINE THE LENGTH OF THE FLAP ● Position the template on the centro-medullary axis; so that the tip of the greater trochanter coincides with the center of rotation of the revision stem (choose a medium-sized proximal component). ● Determine the length of the flap, which has to overcome the obstacles (mostly the femoral curvature) and which respects the isthmus of the femur at the same time. ● Trace the distal end of the flap.

Chapter 3 – Making a preoperative template

69

The mean length of the flap is 15 cm +/- 2 cm. It must always respect the isthmic zone of the canal. A too long flap should be avoided when the sole indication of the flap is the need to remove a long distal cement plug. In dysplasia or short stature, a shorter flap of 10 to 12 cm is possible.

4. CHOOSE THE IMPLANT LENGTH This involves determining the length of the bone-implant contact zone (zone of primary stability). Whenever possible, a short distal component (140 mm) should be selected, preferably with a bone-implant contact length of 40 to 50 mm. Trace the contours of the proximal component (in particular the shoulder of the implant, and the center of rotation) as well as the distal end of the possible distal component. Generally, it is possible to determine the length of the distal component, however the other references provided by the template are only indicative (length of the proximal part, diameter of the distal part) and very often the definitive components will be of a different size that those selected during the preoperative planning.

5. VERIFY THE LENGTH OF THE FLAP AND THE DEPTH OF PENETRATION OF THE IMPLANT Determining the length of the flap requires the measuring of the distance between the tip of the greater trochanter and the distal end of the flap. This is a particularly important reference point during surgery, as once the flap is reclined the tip of the greater trochanter can no longer be used as a reference to evaluate the depth of penetration of the implant. It is only the distal end of the flap that can still serve as a referencing guide to evaluate the penetration of the stem. The depth of penetration of the stem is calculated, starting from the distal end of the flap, considering the shoulder of the implant in situ as reference. Since the distance between the center of rotation and the shoulder of the implant in situ is about 20 mm and the length of the flap is known, it is possible to determine whether the implant is in the correct position or not. The distance between the shoulder of the revision stem and the distal end of the flap must be equal to the length of the flap – 20 mm (or 10 mm for the proximal component of height of 55 mm). The depth of penetration calculated on the basis of the template is only indicative and the final choice is always made during surgery, after completing several trial reductions. However when there is significant shortening of the leg, it is often preferable to avoid restoring the exact leg length. ●

70

Second part – Preoperative planning

CONCLUSIONS Making a template is simple and quick if the necessary documentation is available, in particular an X-ray in the frontal plane with sufficient length of the femur visible. Very often it enables identification of a slight frontal curving that might otherwise remain unnoticed until a centromedullary axis has been drawn. Lastly, the length of the flap is the only dimension that the surgeon must always keep in mind during surgery and the final size of the implant will always be determined intraoperatively.

THIRD PART

SURGICAL TECHNIQUE

“Success or failure (of a revision surgery of the hip) depends on the prosthesis design, the chosen surgical technique and the patient. Probably, the surgical technique is the most important factor...” Rik Huiskes This textbook may appear to be too long and many colleagues may not find the time to read the whole book. Nevertheless, we urge those surgeons who have chosen to use the press-fit concept to read this chapter which deals with the surgical technique. This may spare them from sorrows during their surgery and... finally, help them to gain a lot of time. A surgical technique which may seem to be complex at the first look does not always imply that the surgery will be long and complicated. Last, but not least, if a failure has to be explained, mostly, it is the method of application which has to be questioned and not the implant.

CHAPTER 1

GENERAL CONSIDERATIONS

Prudence and perseverance are essential while gaining experience. Any surgeon using a new implant will inevitably go through a learning curve, regardless of the prosthesis that has been selected. It is a good reason not to change the concept or the implant too often. Every concept demands its own specific surgical technique. The surgeon must familiarize himself with the imperatives of a concept before sing it. Any technical error often results in immediate failure. When a cementless implant has been chosen, the surgical technique has to be followed rigorously. The two main goals to be achieved are: sparing the existing bone stock and ensuring perfect primary stability of the implant. Sparing the existing bone stock depends, first and foremost, on the choice of the approach to the femur. We are of the opinion that it is never good to change the strategy during the course of the surgery. In order to ensure primary stability through the press-fit concept, it is advisable to comply with the principles defined by Morscher, that is, ensure a contact on the form of a surface between bone and implant in order to ensure a firm wedging of the prosthetic stem, without stiffening the femur too much. The surgical technique with the PFM-Revision (Revitan straight) will vary depending on the selected approach to the femur: a femoral flap with diaphyseal fixation (option 1) or the endofemoral approach with proximal fixation (option 2). Before describing these two techniques, a few general comments should be made. Among these considerations are: ● a rational use of the various implants ● a practical application of the press-fit concept (or how to prepare a bone-implant contact surface and ensure that the prosthetic stem is properly wedged) NB. Both surgical techniques are presented in such a way that they can be used separately, which explains certain repetitions.

74

Third part – Surgical Technique

1. RATIONAL USE OF THE IMPLANTS A rational use of the implants of the PFM-Revision System (Revitan straight) is based on the knowledge of the principal possibilities which the various implants offer.

1.1 THE PROXIMAL COMPONENTS The “spout” components are larger in the frontal view as compared to the “cylindrical” components. The indication for their use is an endofemoral approach with the so called proximal or metaphyseo-diaphyseal fixation mode. In these cases, the length of the proximal component is mostly within 55 to 85 mm. The “cylindrical” components are narrower in the frontal plane. They can be advocated for the reconstruction of the leg length after having made a flap and a diaphyseal fixation and we recommend not to use the 95 or 105 mm component too often. NB. This choice can easily be avoided if one makes good use of the modularity of the ancillary instruments when selecting the implant.

1.2 THE DISTAL COMPONENTS All distal components show the same shape: Straight stems with longitudinal fins, a conical slope of 2 degrees and a conical zone of 100 mm in the stems 140 mm and a zone of 120 mm in the stems of 200 and 260 mm. The 260 mm distal component is rarely used. Its use is restricted to the treatment of complications such as fractures on long prostheses. The 200 mm distal component always necessitates a femoral flap or a trochanteric osteotomy. It is indicated in the cases where a long diaphyseal fixation is needed (contact zone boneimplant of about 50 to 80 mm) or for reconstruction of the correct leg length in tall patients. The 140 mm distal component is the most frequently used one. It allows for a short diaphyseal fixation, either through an endofemoral approach or a femoral flap. The (new) 120 mm distal component is used for the easiest cases, where there is no destruction of the bone stock. The conical proximal zone (height 45 mm) is the “working area” of this stem. This stem is mostly used together with a “spout” proximal component.

Chapter 1 – General considerations

75

FREQUENTLY USED COMBINATIONS OF COMPONENTS As a general rule, combining long proximal components (95 or 105mm) with short distal components (length 120 or 140 mm) should be avoided.

Length of the Height of proximal component distal component Total length 120 mm

140 mm

55 65 75 85

䊳

95 105

䊳

55 65 75 85 95

䊳

䊳 䊳 䊳

175 185 195 205

mm mm mm mm

Endofemoral approach and fixation in the metaphysealdiaphyseal region In this case: the “working” part of the distal component is the proximal conacal area (height 45 mm). Choose a proximal component, which will usually be of the “spout” type, although it also possible to use a “cylindrical” component.

215 mm 225 mm

It is frequently preferable to choose a 140 mm distal component with shorter proximal component.

Short diaphyseal fixation by the endofemoral approach or after a femoral flap

䊳

195 205 215 225 235

105

䊳

245 mm

Avoid, if posssible, and prefer a 200 mm distal component

55 65 75 85 95 105

䊳

255 265 275 285 295 305

Long diaphyseal fixation with femoral flap

䊳

䊳 䊳 䊳

mm mm mm mm mm

In this case, the “working” part of the distal component is the distal conical area (height 100 mm). Choose a proximal “spout” component in most cases. NB. This implant (L 140 mm) can also be used for a proximal fixation with an endofemoral approach, especially when a femoral window must be bridged.

200 mm 䊳 䊳 䊳 䊳 䊳

mm mm mm mm mm mm

In this case, the “working” part of the distal component is the conical area (120 mm high), remembering that the 200 mm components always have an intermediate cylindrical area that is not adapted for wedging. If a high component is necessary, choose a proximal component of the “cylindrical” type. Reminder: The 260 mm distal components are used very rarely, and mostly for treating complications with multiple bone lesions. In these cases, they are used as a medullary nail.

76

Third part – Surgical Technique

2. PREPARING A BONE-IMPLANT CONTACT SURFACE When the choice of a press-fit concept is made, the major objective which has to be achieved in order to ensure the primary stability of the uncemented implant is a good bone-implant contact surface. The way of preparation, performed with reamers or rasps, depends on the area of the femur where the primary stability will take place. 2.1 PREPARING THE FEMUR WITH THE REAMERS If a diaphyseal fixation is required, the femur is prepared with reamers which will give a conical shape to the medullary canal. The references given by the reamers (diameter and length) are mere indicators for the choice of the definitive components; these are always chosen according to the test prostheses used during the surgery. In order to obtain a bone-implant contact surface with the reamers, three rules have to be observed: A. STAY CLOSE TO THE FIXATION AREA (FIG. 1) If the press-fit area has to be in the middle diaphyseal area (isthmic zone of the femur), it is often preferable to do a femoral flap. Preparing a conical contact surface without having direct insight onto the working area has proven to be difficult. Moreover, reaming the isthmic zone through an endofemoral approach is dangerous since it will often create an undesired 3-point fixation. If the press-fit area can be in the proximal diaphyseal area, preparation through the endofemoral approach is possible. In these cases, the reamers are also used for the regularization of the opening in the greater trochanter (see also option 2: endofemoral approach). B. REAM A STRAIGHT SEGMENT OF THE FEMUR (FIG.2) Conical reamers are only effective if there is a certainty of working on a straight segment of the femur. Conical reamers cannot transform a curved femur into a straight one! C. REAM ONLY A SHORT SEGMENT OF THE FEMUR (FIG. 3) It is easier to make a segment of the femur “conical” if the length of this area is not too important. We warn about trying to prepare a conical segment of 80 to 100 mm.

Figure 1

Figure 2

Figure 3

Chapter 1 – General considerations

77

2.2 PREPARATION OF THE FEMUR WITH THE RASPS It is only possible to prepare a bone-implant contact surface with the rasps if an endofemoral approach is chosen and if the primary stability can be achieved in the metaphyseo-diaphyseal area of the femur. It is important to take advantage of the modularity of the rasps to perform a two-stage preparation of the bone-implant contact zone. A. PREPARATION OF THE METAPHYSEO-DIAPHYSEAL AREA (FIG. 4) (BONE-IMPLANT CONTACT AREA) In a first stage, start to “find” the area of primary stability using the distal part of the rasp of 120 mm. It is impacted with the means of the cylindrical impactor sleeve which shows a graduation that will allow determining the size of the proximal component used in the second stage. It is recommended to use the distal rasp of 120 mm, but it is also possible to use the rasp of 140 mm (wide medullary canal). Distal components of 200 mm or more should not be employed in such a case. NB. In revision surgery the area of primary stability is rarely to be found in the metaphyseal area. Most times, this area will be situated in the meta-diaphyseal junction at a level which is difficult to predict. B. PREPARATION OF THE METAPHYSEAL AREA (FIG. 5) (SELECTION OF THE COMPONENTS) The two components of the rasp are assembled: The distal part which was used during the first step with the proximal part, whose size has been determined in the previous stage. NB. For this operative step, the spout proximal rasp should be the first choice. Impact the assembled rasp; when its depth of penetration corresponds to the one achieved during the first stage, a perfect bone-implant contact in the metaphyseo-diaphyseal area is guaranteed. An assembled “monobloc” rasp can remain stuck in the upper metaphyseal area. In this case, there will no longer be a contact surface but featuring only local point contacts, which are insufficient to guarantee the required primary stability.

Figure 4

Figure 5

78

Third part – Surgical Technique

3. ENSURING A GOOD WEDGING OF THE IMPLANT In addition to the necessity of an exact perception of the area where the implant will be wedged into place, it is important to comply with three rules in order to ensure a good fixation: ● Use the conical part of the implant ● Keep some conical fixation area in reserve ● Complete the implantation in two stages

3.1 USE THE CONICAL PART OF THE IMPLANT The conical area which can be used for the wedging fixation is found in the distal region of the straight stem. And for a given conical slope, its height is constant for the same implant lengths. For the PFM-Revision (Revitan straight) stems with a conical slope of two degrees, the length of the conical zone is 120 mm for the distal components of 200 mm (Fig. 6A) and 100 mm for the distal components of 140 mm (fig. 6B). For the new distal piece of 120 mm, the proximal “working area” is of 45 mm length only. NB. In the distal parts of 140 mm length, in the diameters 14 –18 mm, the distal conical zone is prolonged proximally by a second conical zone of 9 degree slope, which means that they are conical allover. This feature is lost on the 140 mm distal parts with a diameter of 20 mm and bigger, because the diameter of the conical distal zone is superior to the diameter of the conical proximal zone. Every implant shows a well defined conical area and it should be kept in mind that a long stem will always bear an intermediate cylinFigure 6 drical zone which is not suitable for a wedging fixation. Contrarily, short stems ( L. 140mm) of smaller diameter (14 to 18 mm) are conical over the whole length, which is a good reason to prefer this type of implant each time it is possible. The notion of a defined “working area for a conical fixation” must be kept in mind during the preparation of the femur with the conical reamers. This area will vary according to the length and diameter of the selected implant (since the bigger the conical slope, the shorter the area of conical fixation will be).

3.2 KEEP SOME CONICAL FIXATION AREA IN RESERVE Keeping some reserve of the conical anchorage area means ensuring that the implant will be wedging into place with the distal part of the conical area of the chosen implant (bone-implant contact). This notion of a reserve of the conical fixation area applies to all sizes of straight stems, but it is extremely important when using a 200 mm distal component in a diaphyseal fixation mode. WHY SHOULD SOME CONICAL AREA BE KEPT IN RESERVE? (FIG. 7) If a 200 mm distal component has been chosen, and it wedges into place with its proximal half of the conical area, there will be a risk of loss of stability following the later seating of the implant. During this seating, the fixation zone will leave the conical area of the stem and migrate to the

Chapter 1 – General considerations

Figure 7

79

cylindrical part of the stem which is not suitable for a re-wedging of the implant. On the contrary, if the surgeon aims for a wedging in the distal part of the conical area of the stem, there will remain some reserve of the conical anchorage area, allowing for some millimeters of subsidence and secondary re-wedging. It will always be easier to keep a reserve of conical area with a short distal component of 140 mm; this is a sufficient reason to privilege this size of implant. On the other hand it will be difficult to know about the presence or absence of any reserve of the conical anchorage area if there is no precise knowledge of the area of the femur where the wedging will take place. This is a good reason for realizing a femoral flap as soon as the fixation is intended to be in the isthmus of the femur.

HOW CAN SOME CONICAL ANCHORAGE AREA BEST BE KEPT IN RESERVE? By augmenting the diameter of the implant without increasing the diameter of the reaming of the cavity, the anchoring within the distal half of the conical area can be effectively realized. However, this maneuver bears the inconvenience of creating an over-length of the leg. (fig. 8). If the surgeon has selected an uncemented monobloc implant, made following a homothetic production process, this situation can only be mastered by over-reaming the medullary canal. As this step has its natural limitations, an over-length of the leg must be sometimes accepted. Under this aspect, the modularity of the implant offers a real advantage. MODULARITY AND RESERVE OF CONICAL ANCHORAGE AREA With a modular system it is easy to increase the diameter of an implant in order to obtain the mentioned reserve without inducing a difference in the length of the two lower limbs. The surgeon has two options: If a short stem is selected (L 140 mm), by increasing its diameter and choosing a proximal component of less length. If a longer stem had been selected (L 200 mm), by replacing it with a stem of larger diameter Figure 8 (+2 mm) but one size shorter in length (e.g. L 140 mm). If this results in any over-length, it can easily be corrected by selecting one of the shorter proximal components (fig. 9). This situation is frequent in daily practice. NB. Both of these implants have a similar area of anchorage which lies in the distal half of the conical area in the component L 140 mm, thus providing this implant with a superior reserve compared to the longer implant of L 200 mm. For practical purpose, it means that the choice of the implant should not be made with the help of the reamers, but with test prostheses. It also means that if one intends to choose an implant with some reserve of the anchorage zone, avoiding to create a difference of leg length, the test Figure 9 prosthesis inevitably must be modular, too. Furthermore, a modular system makes the selection of shorter stems easier, which is always to be preferred. In press-fit stems, the diameter should always be privileged to the length of an implant.

80

Third part – Surgical Technique

3.3 ENSURING AN IMPLANT PLACEMENT IN TWO STAGES The test prosthesis and the definitive implant do not always wedge into place at the same height. A difference of +/ – 5 mm (sometimes even more) is possible under the following two circumstances: ● Thin corticals and a wide medullary canal (osteoporosis): The longitudinal fins cut deeply into the cortical bone, hence the implant will stabilize at a lower level than the test prosthesis (fig. 10A). ● Thick corticals and a thin medullary canal (no osteoporosis): The fins will cut less into the bone and the implant will wedge into position at a higher level than the test prosthesis (fig. 10 B). In order to correct this problem, a modular system allows for an implant impaction in two stages: ● Fixing first a proximal test component (fig.11 A) onto the definitive distal component before wedging it into place (fig.11 B); ● Then fixing the definitive proximal component, whose size had been selected according to the depth of the seating of the distal part. Placing an uncemented stem with sharp longitudinal fins jeopardizes the stable wedging if one attempts to equalize the length of the two lower limbs by hammering the stem into place. Unfortunately the only criterion for the appreciation of the stable wedging is the “cortical tone”, and we till today do not dispose of any help for quantifying the impaction forces needed for a perfect wedging.

Figure 10

Figure 11

Chapter 1 – General considerations

SUMMARY In revision surgery with press-fit stems, the choice of the right femoral approach is of primary importance and often the creation of a peliculated lateral flap proves to be the best way to achieve a safe press-fit fixation. Creating a good wedging-in means to ensure the primary stability of the implant. An effective press-fit does not depend on the length of the bone-implant contact surface but on how well the implant is wedged into place. A reserve of the conical anchorage zone must be kept in order to guarantee for a permanent wedging in place of the implant, which often calls for the selection of a short stem one size bigger than the diameter of the reamed medullary canal. The wedging into place of a conical implant is a delicate step of the surgery if one does not dispose of a modular stem. This is true as well as for the time of the selection of the implant as for the implantation of the definitive components.

81

CHAPTER 2

OPTION 1: FEMORAL FLAP FIXATION IN THE DIAPHYSIS

MAIN OBJECTIVES Making a femoral flap osteotomy during revision surgery is a good way to avoid incidents and to ensure a perfect primary stability, which will always take place in the diaphyseal region of the femur when the femoral flap is performed. Both of the two articular approaches, anterolateral or posterolateral, may be used. It is however recommended to prefer the posterolateral access if the cutting of a lateral flap is planned. In all cases, the femoral flap osteotomy should be pediculated, with M. vastus lateralis. If necessary, the lateral flap may be combined with an additional osteotomy of the medial cortex, in the course or at the end of the operation. The creation of a too long flap should be avoided, e.g. for the removal of a long distal cement plug. Remember that the isthmic area of the femur has to be preserved, as this will be the main zone of primary stability for a stem with a press-fit fixation mode. When selecting the prosthesis: take full advantage of the modularity of the test-prostheses and, whenever possible, use a short stem (140 mm) with a bone-implant contact over a height of 4 to 5 cm (favour the diameter over the length of the implant). Implantation of the stem is done in two steps. After wedging the distal definitive stem, in-situ assembly of the proximal component with the distal component is carried out.

84

Third part – Surgical Technique

1. CREATION OF THE FEMORAL FLAP The transfemoral approach to the medullary canal, as described by Vielpeau (25) and Wagner (29), under the form of a pediculated lateral flap offers many advantages to the surgeon: ● An excellent view of the articulation and of the medullary canal, ● An easier way of excising the cement and granuloma formations, ● A realignment of the femoral curving, which is mandatory when a straight stem is used. The operation technique described here refers to a loose prosthesis, with the femur in its normal position, before luxation of the prosthetic articulation. The posterolateral approach is described whereupon the flap is raised in a ventral direction (patient in lateral position).

1.1 CHARACTERISTICS OF THE FLAP The portion to be used as the pediculated flap, is located on the lateral side with the muscular insertions of the gluteal muscles proximally and the muscular and aponeurotic attachments of M. vastus lateralis in distal diaphyseal zone. Due to the normal torsion of the femur, the cutting plane will be in the lateral aspect of the metaphyseal area and in the anterolateral plane in its diaphyseal area (fig. 1). The length of the flap, best determined before surgery, will be about 15 +/- 2 cm, the reference point being the tip Figure 1 of the greater trochanter. The distal cut should always preserve the isthmic zone of the femur, which is the zone for primary stability of the revision implant. The flap should be 3 to 4 cm wide in the diaphyseal area and describe a semi-circle (fig. 2). Avoid cutting a too narrow flap distally in order to be able to correct a curving in the sagittal plane later and to avoid any obstacle in the anterior half of the femur which could hinder an exact axial reaming. Figure 2 When calculating the flap length, beware of calcifications on the tip of the greater trochanter. Under some circumstances, longer flaps (18 to 20 cm) may have to be performed. Shorter flaps (of 8 to 10 cm) are possible, but their indications are less frequent.

1.2 PATIENT POSITIONING AND APPROACH (FIG. 3) The patient lies in a fixed lateral position, his pelvis being stabilized dorsally with the help of a sacral post and ventrally with a symphysic post, avoiding here any compression of the femoral vessels. The approach is a classical posterolateral one: Skin incision, longitudinal dissection of fascia lata and of the gluteal muscles. Soft tissues are spread apart with the help of a (Charnley-type) distractor, both stable and inflexible. The horizontal position of the lower limb is secured by means of a cushion that can be easily displaced for the adduction movement during the luxation Figure 3 maneuver and the following femoral exposure.

Chapter 2 – Option 1: Femoral flap

85

1.3 FEMORAL EXPOSURE (FIG. 4) Localize the tip of the greater trochanter; enlarge the skin incision if necessary to the planned level of the distal end of the flap. Place the leg into slight internal rotation, and then transect the M. quadratus femoris and the aponeurotic extension of the greater gluteal muscle if necessary. Separate the dorsal insertion of M. vastus lateralis from intermuscular septum and linea aspera, taking care to ligate the perforating vessels. The distal extent of this dissection corresponds to the planned Figure 4 length of the flap. At this stage of the surgery, the short external rotators and the dorsal capsule are still intact. Distally, the medial aspect of the femur should be easily accessible. The creation of a flap after the complete dissection of the vastus lateralis from the femoral surface is more convenient, but it will result in an avascular flap which will have a poor reconstruction capacity of the bony structures, especially if they were already fragilized by granuloma formation or osteoporosis.

1.4 CUTTING THE FLAP The posterior osteotomy lies slightly anterior to the external bifurcation of linea aspera. Proximally, it will curve to the dorsal aspect in order to protect the insertion of the short external rotators and the dorsal fibers of the M. gluteus medius (fig. 5). The distal transverse osteotomy is at a right angle to the diaphyseal axis, about 3-4 cm long and limited by two drill- holes of 4.5 mm. The anterior osteotomy is done in two steps in order to avoid any damage to the M. vastus lateralis. ● In the first step, the proximal and distal initial cuts are made. Distally with the help of the oscillating saw after the transverse cut is done; proximally with a chisel after having dissected the plane between Vastus medialis and lateralis, and after resection of scarred soft tissues from the prior primary approach to the hip joint (fig. 6A) ● In the second step and after checking that the angles of transverse osteotomy are finalized, the anterior osteotomy is completed by cutting with a small chisel along the planned medial line from distally, under the intact M. vastus lateralis, (fig. 6B) on the anterior aspect of the femur, using the stab incisions previously made with the oscillating saw.

Figure 5

Figure 6

86

Third part – Surgical Technique

The most frequent error is making the flap too narrow and not medial enough in the diaphyseal area. In this situation the anterior cortical becomes an obstacle to the correct axial insertion of the straight stem, especially if the femur previously showed a marked curving in the sagittal plane (fig. 7). An additional medial osteotomy may be helpful at this moment. The posterior and the transverse osteotomy should be made with an oscillating saw. Beware Figure 7 of over-cutting the edges of the flap beyong the transverse osteotomy, since it may give rise to distal fracture lines.

1.5 RAISING THE PEDICULATED FLAP At this stage of the operation, the flap is not ready yet and, before trying to raise the flap, make sure that all the cuts are complete and free from adhering soft tissues. In most cases, the inner flap surface is adhering to the cement and the anterior osteotomy may be incomplete. Avoid brisk movements before separating the distal cement layer from the flap surface with a cement chisel and if necessary, complete the anterior osteotomy in distal and proximal direction. All these movements are repeated gently until the flap is free in the distal attachment area (fig. 8). Once the distal part is freed, proceed with the intermediate and proximal areas of the flap, making sure that the cement separates completely from the inner flap surface. Often, the cement layer is abundant at the level of the greater trochanter laterally (fig. 9)

Figure 8

Figure 9

Usually, after these maneuvers, the flap can be elevated in one piece. If this is not the case, it is often due to an incomplete anterior osteotomy, which has to be finalized during the next surgical step, after having freed the articular cavity.

1.6 FREEING OF THE ARTICULAR CAVITY AND FINAL FLAP ELEVATION Always free the articulation from the pseudo-capsule and scar tissues before trying to elevate and rotate the flap. The elevation is done with the 2-3 parallel chisels, stepwise inserted at the level of the greater trochanter. Cut the short external rotators at the level of their insertion to the trochanter together with the posterior capsule then proceed to the cutting of the attachments inside the cavity on the under surface of the gluteal muscles. This will allow to preserve the short external rotators and a portion of the posterior capsule, which will be helpful for the closure of articulation (fig. 10).

Chapter 2 – Option 1: Femoral flap

87

Once all the adhering fibers are cut, gently turn the flap around the anterior osteotomy, which will act as an anterior hinge (fig. 11). If once the articular cavity has been freed, the flap can still not be elevated, check the ventral osteotomy line. Most times however, even an incomplete osteotomy line will break at the right place during anterior tilting and lifting maneuvers. After having completed this step, freshen the endomedullary aspect of the flap: all granulomatous cysts, pseudo-membranes and remaiFigure 10 ning cement layers are cleaned away with curettes. It is important to extend the cleaning to the greater trochanter; in some cases, it must be deepened or reconstructed and the removal of the cement at this level can sometimes prove laborious. A distal transmuscular stitch can help in preventing the avulsion of the vastus muscle Figure 11 from the flap. The most frequent complication will be the fracture of the flap. This however has no consequences, if the flap has not been devascularized previously. 1.7 REMOVAL OF THE LOOSE FEMORAL STEM The proximal femur should be freed from adhering scar tissue before trying to remove the stem.With the flap being held in an everted position by a hook, excise the fibrous scar tissues and granuloma formations around the articular cavity, proceeding from the center to the periphery (fig. 12). The soft tissues around the proximal femur are freed stepwise by rotating the femur internally and externally, always keeping the instruments close to the periosteal surface of the femur. By doing so, a posterior capsular portion can be preserved and there is no need for any additional visualization of the sciatic nerve. Now the stem can be removed, if the size of the flap is sufficient (fig. 13). If the stem does not disengage from the diaphysis, it may be removed by gentle axial taps after putting the leg into internal rotation - adduction –and flexion position. Caution: All the luxation maneuvers must be Figure 12 done with great precaution. Always make sure that there is no tight aponeurotic insertion of M. gluteus maximus pulling tearing on the posterior aspect of the proximal femur. Remember also that the medial cortex of the femur is not always fully freed from all attachments at this stage of the operation. Figure 13

88

Third part – Surgical Technique

1.8 EXPOSURE OF THE FEMUR AND CEMENT REMOVAL In order to allow a complete exposure of the femur, remove all the adherent scar tissues on the medial side. The femur should not remain fixed deep in the operation wound. End by cleaning the surroundings of the acetabulum in its caudal and anterior part. Then remove the cement adhering to the intermediate area of the medullar cavity. Clean this area thoroughly from every granuloma and pseudo-membranes. A flap will greatly facilitate this step, giving access to all surfaces. Copious rinsing will improve the differentiation of the bone-cement interface. A remaining distal cement plug will only be removed after having cleaned the intermediate area. First, drill a hole in the plug with the 6 mm drill bit, after making sure that the latter is correctly centered (fig. 14A). After verifying that there is no via falsa, enlarge the opening in a second step to about a size of 11 mm, so as to pass a wide reversed cement extractor through it (fig. 14B). If the cortical structures are fragile (osteoporosis), we advocate the use of a prophylactic cerclage wiring, before Figure 14 attempting to remove the cement plug. It is preferable to carry out cement removal after the acetabular surgical step in order to avoid excessive bleeding.

1.9 TECHNICAL VARIATIONS AND POTENTIAL ADDITIONAL STEPS ANTEROLATERAL APPROACH If the patient is in a lateral position, the flap is rotated dorsally. However, there is a risk of avulsing the muscular pedicle. Advocated by some authors, this procedure depends on the habits of the surgeon. CREATING THE FLAP AFTER LUXATION AND REMOVAL OF THE PROSTHESIS (FIG. 15) It is possible to do the cuts for the flap after having removed the prosthesis. The corticals are osteotomized with the reciprocating saw, moving from the lateral to the medial cortex, going straight through the medullary canal, always after having previously carried out the distal cut. Even if this technique may sound interesting, it is based on a previous luxation of the Figure 15 prosthesis and its removal without a fracture. This procedure should be reserved for flexible subjects and surgeons experienced with this variation. We do not advocate the technique of passing a reciprocating saw from the lateral to the medial cortex behind the shoulder of a prosthesis, still in place since there is a clear risk of creating a too small flap, especially in the distal area.

Chapter 2 – Option 1: Femoral flap

89

ENLARGED TROCHANTERIC OSTEOTOMY (FIG. 16) In the case of an enlarged trochanteric osteotomy, the muscle fibers of vastus lateralis adhering to the last centimeters of the proximal femur are removed, preserving the attachment of the muscle at the base of the greater trochanter. The technique of cutting the flap remains basically the same. The flap is then turned ventrally. RAISING A SHORT FLAP (10 TO 12 CM) If the decision is to raise a short flap, (squat patients of femoral dysplasia) the attachment of the M.vastus lateralis should carefully be preserved. A possible indication for an enlarged trochanteric osteotomy is the approach to the articular cavity for a revision of an acetabulum cup or the mere exchange of a proximal component in a modular stem system. Figure 16

ADDITIONAL OSTEOTOMY OF THE MEDIAL CORTEX (FIG. 17) This is the crucial step in order to realign a proximal femur showing a significant varus curving on a straight stem. It should be noted, that if there is a major curving in the sagittal plane, the raising of a lateral flap cannot suffice to overcome the anatomical obstacles. This is particularly true when a long straight stem must be implanted. In both cases the solution is to cut an inverse chevron osteotomy in the medial cortex, about 3 to 4 cm proximal to the transverse distal cut. OSTEOTOMY OF THE LINEA ASPERA This type of osteotomy is a regular technique advocated by Vives and Picault (28).We only use this technique in stiff hips that present a shortening of the lower limb. The freeing of the whole linea aspera from all muscular attachments can facilitate the lengthening of the leg.

Figure 17

2. PREPARING THE FEMUR After the flap has been made,the procedure continues with the preparation of the femur and the final choice of the implant, the two being clearly separated steps. The femur is prepared with the help of the reamers alone and the rasp surface of the test-prostheses is of no utility in this type of approach.

2.1 CALIBRATION OF THE FEMUR AND VERIFICATION OF THE AXIS IN THE SAGITTAL PLANE In a first step, use the cylindrical drill-bits for the enlargement of a narrow medullary canal in the area of the tip of the removed prosthesis. The diameter will help to determine or “calibrate” the canal (fig. 18A). It is of no use trying to enlarge the diameter of the whole distal canal, since the natural isthmic area could be damaged. In a narrow canal, a reamer of a 12 mm diameter should be able to pass through the isthmus.

90

Third part – Surgical Technique

In a second step, make sure that the anterior cortex is not an obstacle for the progression of the reamer due to a major curving in the sagittal plane. Use a conical reamer of one size inferior to the last calibrating drill-bit for this step. If the curving of the femur proves to be an obstacle, then it is better to make an additional osteotomy of the medial cortex, especially if it is fragile (fig. 18B). If the quality of the medial cortex is very good, a “biting” on this structure may be possible, in order to find the correct axial alignment, mostly if a short stem is possible. 2.2 REAMING OF THE FEMUR (FIG. 19) The primary role of the conical reamers is to regularize the canal, so as to create a conical seating in the area of the isthmus. The process of “conisation” should only begin when there are no more obstacles to the even progression of the reamer. Figure 18 At this moment, the diameter of the reamers is increased progressively till a steady resistance to the progression of the reamers indicates that a safe seating has been reached. We advocate securing the femur with the help of a clamp against torsion forces. A motorized reamer may be used only if the cortices are not severely damaged. In the latter situation, manual reaming procard with a gentle. Evaluate the depth of penetration of the reamer by aligning the marks on the handle with the line passing from the tip of the greater trochanter to the center of rotation of the implant. NB. Remember that after the flap has been raised, the greater trochanter is also displaced and no more available for measurements. But it can be positioned, with the help of a sterile centimeter ruler, taking the distal transverse cut on the femur as reference and add to that the length of the whole flap. Example opposite (fig. 19): For a 15 cm long flap, the tip of the greater trochanter will correspond to the mark 2-65. This reference corresponds to an implant of the same diameter as the reamer in place and 265 cm long. i.e. a 200 mm long distal component (the digit 2) coupled with a 65 mm high proximal component (the digit 1 corresponds to the implant size of 140 mm, the digit 3 to that of 260 mm). Very often the reamer will seat in sector 2 (corresponding to a distal Figure 19 component of 200 mm). It is advisable to avoid the choice of a reamer seating in zone 3 (e.g. a length of 260 mm). In such a situation, it is better to increase the diameter of the reamers till zone 2 is reached, corresponding then to the desired length of 200 mm. If the tip of the greater trochanter is situated in between the digits 2 and 3 (or 1 and 2), then proceed to one or more test reductions before trying any additional reaming. Remember that the digits on the reamer are only indications for the choice of the implant. Finally, you must keep in mind that the references given by the reamer only have an informative value for the choice of the definitive implant.

Chapter 2 – Option 1: Femoral flap

91

3. CHOICE OF THE TRIAL IMPLANT Selecting the correct trial implant is a very important and decisive step of the operation, especially when a conical stem fixed with the press-fit technique is chosen. Several trials are needed using the modularity of the stem to its maximum. When a flap has been raised the surgeon should always proceed in two distinct steps: ● In a first step, choose the distal test piece with a reserve of the conical seating area. This will mean that the seating is made with the help of the distal or intermediate region of the conical area of the stem; ● Then, in a second step, select the proximal test component which will establish the correct length of the lower limb and allow for an easy reduction of the hip. It should be remembered that in revision surgery, the classical references (e.g. tip of the greater trochanter- center of rotation) are not absolute rules and that the final size of the components will be decided only after doing repeated trial reductions. NB. In practice, the distal trial component and the length of the proximal test component are selected simultaneously during the course of the surgery. However, in order to clarify those steps, the two stages of the surgery will be described separately below.

3-1 SELECTING THE DISTAL COMPONENT (TEST PROTHESIS) According to the measures made with the conical reamer, the two test components are now assembled. This assembly is introduced by hand into the canal and gently impacted into place by applying light hammer blows, while controlling its penetration. During the seating of the test prosthesis, uncontrolled and forcible hammering must absolutely be avoided, since this could result in a jamming of the assembly, especially when a large stem has been selected or if there is a narrow canal. Now evaluate the stability and compare the position of the conical area of the implant with the anchoring area. This bone-implant contact area is investigated by measuring the distance from the proximal cut delimiting the conical area of the implant to the transverse distal cut of the flap. Even though this procedure gives an approximate measurement, it seems to us to be the safest way. But it is only possible if the flap has a correct size of about 15+/- 2 cm length. For the authors, the possibility of an evaluation of the position of the conical area is an important reason for the raising of a flap, which always should be of an adequate size and never made too short. THE SURGEON MAY THEN BE CONFRONTED WITH THE FOLLOWING TWO SITUATIONS: THERE IS SOME RESERVE OF THE CONICAL ANCHORING AREA (FIG. 20) With the short distal component of 140 mm this situation is frequent and primary stability is ensured with the distal part of the conical area of that implant. The proximal line shows the limits of the working area of the stem in place. Mostly this line will be situated clearly (4 to 5 cm) above the distal end of a flap with the normal size of about 15 cm. Reminder: The conical working area is 120 mm for stems of 200 mm length, and 100 mm for stems of 140 mm length.

92

Third part – Surgical Technique

Figure 20

Figure 21

If the primary stability is achieved in the isthmic area of the femur, it becomes evident that the position of the final anchoring area could not have been evaluated in such a precise manner if an endofemoral approach had been selected. THERE IS NO RESERVE OF THE CONICAL ANCHORING AREA In this situation, the primary stability is ensured by the proximal part of the conical area. The proximal line limiting the working conical area of the stem is situated at the level of or below the distal cut of the femur (fig. 21A). In this case, change the option by selecting another distal component sizing up its diameter: ● If the length of the distal component was 200 mm, this stem should be replaced by a component with the length of 140 mm, but one bigger diameter of + 2 mm without any additional reaming (fig. 21B). Only if the diameter is larger than 4mm, then a prudent reaming may be needed. If the length of the distal component is 140 mm, one has to simply select a diameter of a bigger size and an additional reaming may be required. Reminder: In order to restore some reserve of the conical anchorage area, it is necessary to increase the diameter of the implant without increasing the diameter of the medullary canal. The situation is frequent, where a surgeon has to replace a stem of 200 mm (without any reserve of conical anchoring area) with a shorter but larger stem of 140 mm (these two implants having a similar anchoring area).

3.2 SELECTING THE PROXIMAL COMPONENT (TEST PROSTHESIS) Preliminary remarks. In revision surgery, the correct leg length can only be determined with the help of several trial reductions. If there is a significant leg shortening before the surgery, a reconstruction to the desired anatomical leg length could give the patient an impression of the leg

Chapter 2 – Option 1: Femoral flap

93

being too long. An under-correction may sometimes be better tolerated by the patient. The same could apply to a very stiff articulation before the revision surgery. In such a situation, a relative shortening will be better accepted. Finally, the position of the cup should be considered, because, in revision surgery, it is not always in anatomical position. Here, one should avoid selecting either the longest (105 mm) or the shortest (55 mm) proximal component, since this will reduce the liberty of decision when placing the final components. Test the implantation will always be made in two steps if a flap had previously been raised. First, select the proximal test component which will reestablish the correct leg length, respecting the usual reference points (tip of the greater trochanter- center of rotation with a ball head size of a medium neck) and perform now a first trial reduction. The surgeon may now be confronted with any of the three following situations: THE CORRECT CHOICE HAS BEEN MADE (FIG. 22) The length of the lower limb has been restored using one of the average-sized proximal components and the reduction can be carried out without any difficulty. NB. To calculate the depth of penetration, measure the distance between the shoulder of the test prosthesis and the distal end of the flap (this distance corresponds to the flap length – 2 cm) This situation may be true if the hip has preserved its natural elasticity and the shortening of the lower limb was not significant. THERE IS A NEED FOR A SHORT PROXIMAL COMPONENT (55 MM) In this situation, the leg length can be corrected only by selecting the shortest proximal component of 55 mm (in those the distance between the shoulder and the height of the center of rotation is – 1 cm). Here it will be necessary to choose a higher proximal test component and two decisions can be made: ● Either the trial reduction is still easy to perform and the length of the leg is correct. If the reserve of the conical anchoFigure 22 ring area is sufficient, then the size of the distal component can be left unchanged and an additional reaming will be carried out in order to increase the depth of penetration of that distal component, allowing for the use of a higher proximal component (fig 23 A and B). The reaming is done with a reamer of the same diameter (or + 2 mm under the condition of refraining from any aggressive reaming). The alternative will be to decrease the diameter of the distal component, if the reserve of the conical anchoring area is largely sufficient. ● Or, the trial reduction proves to be difficult (mostly because of articular stiffness). In this case, it is advisable to select a higher proximal component and to diminish at the same time the length of the limb in relation to the Figure 23 usual references. An additional reaming is possible,

94

Third part – Surgical Technique

reducing at the same time the diameter of the distal piece, but in our eyes, a widening of the medullary canal is to be preferred. If a 140 mm long distal component has been selected, it is possible to decrease its diameter without running any risk, since this implant has a conical proximal area that can take over the function of the distal conical area in a situation of a significant subsidence. This is not true for the components of 200 mm length, as these have a cylindrical intermediate area that is not suited for the wedging effect. Special case: The seating of the prosthesis is at a correct height and it is not reasonable to select a shorter implant or make a deeper implantation. If then the reduction proves to be difficult or even impossible, an osteotomy of the linea aspera in addition to the classical intra-articular freeing may help to gain additional length for an easier reduction. THE PROXIMAL COMPONENT IS SIZE 105 (HIGH) If the leg length can only be achieved with the help of the longest proximal component, it will be necessary to decrease the degree of penetration of the distal component in order to then be able to select a lower proximal component. In this case, the distal component has to be changed to a bigger diameter, especially when there was a limited reserve of the conical anchoring area, which is often encountered. If the distal component is 140 mm long (fig. 24), it will be necessary to increase its diameter without changing its height; the same applies to the 200 mm long component. In both cases, additional reaming is not always necessary. NB. If the distal component is 200 mm long, it is often preferable to shorten its length while increasing its diameter, if necessary by + 4 mm. Then, an additional reaming may be needed. Special case: In the same situation, the hip may sometimes reduce in a too easy way and then prove to be unstable. The origin of the problem lies in a weak muscle sling and it will be preferable to select a locking cup than to run the risk of an excessive lengthening of the leg, which usually is badly tolerated. Figure 24 Beware from omitting to measure precisely the depth of penetration (distance shoulder of the prosthesis - distal cut of the flap) if this step has been modified subsequently.

4. IMPLANTING THE DEFINITIVE PROSTHESIS IN TWO STAGES Sometimes, there is no direct correlation between the stable wedging of the test-prosthesis and the definitive implant. The modularity allows for an exchange of the proximal component, if in the previous step of the surgery, a safety range had been observed. Moreover, the modular assembly allows for a precise selection of the correct antetorsion of the proximal component. We warn against using this two-stage implantation with a modular proximal test component through an endofemoral approach. Without raising a flap, the endofemoral approach is likely to induce a wrong assembly of the two components.

Chapter 2 – Option 1: Femoral flap

95

4.1 PLACING THE DEFINITIVE DISTAL COMPONENT (FIG. 25) ● The proximal temporary part is fixed onto the definitive distal component; the antetorsion is in neutral position. ● Then the implant is manually introduced into the medullary canal, with the antetorsion that is thought to be the correct one. The definitive distal component is firmly hammered into place with gentle taps, the progression of the implant being constantly controlled with the help of a sterile centimeter rule. When the wedging reaches completion, the sound of the bone becomes clearer (cortical tone). Then it is important to wait some seconds before some more gentle taps are added. Measure again in order to make sure the progression has stopped. During this most important time of the operation any impaction of the stem without measuring the penetration should be strictly avoided. Don’t try to change the angle of antetorsion during this maneuver, since the fins of the stem will not allow for any rotation.

The depth of impaction is controlled by comparing the actual measure to that fixed during the test trial. If the depth is not in accordance to the previous measurement and if the difference exceeds +/- 10 mm, then replace the proximal temporary component with a (longer or shorter) one. We recommend making a trial reduction with different trial necks before the exchange is made. ● Once the proximal component is chosen, make a final trial reduction in order to control the orientaFigure 25 tion of the articular surfaces and to fix the final antetorsion angle which will be given to the definitive proximal component. The wedging will take place within a distance of about 2 to 4 cm after the manual introduction of the stem into the canal. If the implant progresses too fast, a fissure in the femoral diaphysis may have occurred. ●

4.2 PLACING THE DEFINITIVE PROXIMAL COMPONENT (FIG. 26) ● Carefully rinse the morse taper, which must always be cleaned from blood or fat before the assembly is performed. ● Position the definitive proximal component by hand giving the required antetorsion that was defined during the previous steps. ● Gently push the proximal component onto the distal part in order to block it. Tighten the assembly with the help of the torque wrench and screw in the conical nut (see appendix 1 for the assembly of the 2 components of the implant).

96

Third part – Surgical Technique

● Carry out the trial reduction and place the mediumsized neck ball head. ● Before deciding about the definitive neck length to be implanted, make sure that the implant has stopped progressing. This is done by a second slight hammering on the shoulder of the implant. It is not infrequent that the implant progresses a further 2 to 3 mm at this time. This can be easily corrected if trial reductions have been performed with a ball-head featuring a medium-sized neck. While assembling the two parts with the help of the torque wrench, large torsion forces are applied on the distal component. In order to prevent a torsion fracture or the Figure 26 rotation of the implant, we recommend holding the handles of the stem tensioner very firmly, preferably with the help of an assistant.

5. INCIDENTS Incidents and problems may occur at the stage of selecting the implant and implanting the definitive prosthesis. An insufficient preparation of the implantation area, due to a misjudgement of the femoral axis is the most frequent reason for these incidents. Incidents can be avoided by observing one rule: in case of any doubt, never try to hammer forcibly but remove the implant and redo the preparation of the medullary canal. 5.1 FISSURES IN THE DIAPHYSEAL FEMUR (FIG. 27) This complication is liable to happen when the (test of definitive) implant progresses too deep in a too easy manner and when the wedging (the moment when the progression of the implant stops) is not taking place in the area planned during the test trial. A fracture may happen at the height of one of the two edges of the distal end of the flap. In this situation, remove the implant and proceed to a cerclage wiring (or two) of the femoral diaphysis, then proceed by impacting the implant back into place, constantly verifying that the crack remains reduced. In practice it has proven to be the safest if the surgeon routinely delineates the distal end of the flap with 2 drill holes and makes a preventive (and definitive, in many cases) cerclage wiring, prior either to the reaming of the canal or to the final impaction of the implant, which is especially true when the corticals are weakened. Figure 27

5.2 DEFINITIVE IMPLANT IS TOO HIGH After proceeding to the assembly of the proximal component and the implantation of the definitive prosthesis, the implant is too high and the reduction proves to be difficult, or even impossible. In this case disassemble the proximal component with the help of the special disassembly instrument (see also appendix 2, surgical technique) and select a smaller component. If this maneuver is impossible because the smallest proximal component is already in place (55 mm), there is no

Chapter 2 – Option 1: Femoral flap

97

other possibility than to remove the implant from its seating and carry out an additional reaming or select another distal component. In routine surgery, one should not be confronted with this situation. If it occurs, the main reason is the poor concentration of a surgeon in hurry who has not complied with the protocol. Unfortunately, it will usually be observed that this surgeon has a tendency to try to correct his error by hammering on the stem in the hope of gaining some millimeters of depth. This will inevitably lead to a jamming of the stem within the femoral canal, which can make an extraction impossible. In such a situation, the only way out will be by cutting a longitudinal slit with the oscillating saw into the femoral diaphysis. 5.3 ROTATIONAL MOVEMENT OF THE IMPLANT AT THE TIME OF ASSEMBLY This incident happens if the stem tensioner is not held firmly while assembling the proximal component, then mobilisation in rotation of the distal part can occur. Remove the implant from its seat, correct the rotation and then wedge it back into place. After doing so, the implant may wedge into place in a deeper position than before. If the resulting overlength cannot be equalized with a longer neck, then the size of the proximal component must be changed. If the size of the component in place (e.g. 105 mm) does not allow for this step, then the distal component has to be exchanged. Sometimes, a second cerclage wiring during the reattachment of the flap on the proximal area of the femur can add some fixation area and help to save the situation. This incident can be avoided if the assembly of the two prosthetic components has been carried out with caution. However if this happens, it implies in most times that the wedging of the stem was poor.

6. PUTTING THE FLAP BACK INTO PLACE Putting the flap back into place, at the end of the operation, is an important, and sometimes delicate, surgical step. It must be performed with great care, to prevent any secondary gaps in the flap, due to traction effect of the gluteal or the iliopsoas muscles, and in order to improve the primary stability of the implant in the proximal region of the femur. When putting the flap back into place, the obstacles which prevent a good contact at the level of the osteotomy cuts, are often located at the height of the greater trochanter (corticalisation of the formerly cancellous bone) and at the distal end of the flap (endomedullary ossification) (fig. 28A). If the femur is straight in the frontal plane and slightly curved in the sagittal plane, a good preparation of the endomedullary surface of the flap is sufficient to reduce the gap. If the femur is curved, it is often necessary to carry out an osteotomy of the medial cortex in order to restore the contact between bone and implant (fig. 28B). Osteosynthesis of the flap is completed with two cerclage wires or more if required. We prefer metallic wires or bands to threaded cables, since those may break and leave behind a lot of metal debris which are difficult to extract. Figure 28

98

Third part – Surgical Technique

Bone defects may be ignored if they are not very significant. If they are more extensive, especially on the anterior cortex, it is suggested to add some bone in the form of autologous bone grafts. Please note that in the case of a weak diaphyseal fixation, the quality of the primary stability of the whole proximal femur can be augmented by a meticulous osteosynthesis of the flap, so as to restore a large bone-implant contact zone. PERSISTING DISTANCE OF THE FLAP Even after a correct preparation of the flap and a secondary medial osteotomy of the cortex a gap may remain between the flap and the prosthesis. This problem originates in a proximal femur being narrow in the frontal plane (gap at the level of the anterior and/or posterior corticals)or in an excessive traction of the gluteal or the iliopsoas muscles, if an important lengthening of the leg had taken place (gap at the level of the distal cut of the flap and/or of the osteotomy of the medial cortex). These risks can be diminished by avoiding to choose a spout proximal part when a long component is required (95 or 105 mm), and select instead a cylindrical part, which is less wide in the frontal plane. Heavy traction forces can also be neutralized by means of a lateral plating, if the cortical structures are thick enough. Finally, if the gaps are large, a filling with autologus bone grafts is advised. NB. In case of an osteosynthesis with a plate, make sure there is no contact between screws or drills and the femoral implant in place. The visualization of a gap between flap and the proximal femur on X-rays is more an inconvenience than a complication, since in most of the cases there will be a secondary filling of an astonishing amount, especially when the vascularity of the flap was well preserved. FRACTURE OF THE FLAP In case of weakened cortices, a fracture of the flap during its creation or later during the cerclage wiring is a frequent event. This too is more an inconvenience than a true complication if the M. vastus lateralis remains well fixed on the flap. If however the fracture lies at the base of the greater trochanter, a complementary fixation with the help of a tension band wiring, using the proximal flap cerclage, should be performed. NB. For postoperative care refer to pages 113 ff.

CHAPTER 3

OPTION 2: ENDOFEMORAL APPROACH PROXIMAL FIXATION

MAIN OBJECTIVES This is not the most frequently used option. It may be considered whenever the femur is straight in the frontal plane. In this situation, the objective is to achieve fixation in the metaphyseo-diaphyseal region or in the proximal diaphyseal area. Reminder: A curvature in the frontal view is always a contraindication for an endofemoral approach, especially if it is excessive or localized in the diaphyseal area of the femur. Ensure a good anterior exposure the femur. Irrespective of the chosen approach, it is very important to achieve a perfect anterior exposure of the femur. It is not possible to prepare the medullary cavity properly if the femur remains attached deep in the articular cavity. Open the greater trochanter widely in order to be sure of staying in the axis of the femoral diaphysis and avoiding implantation of the stem in a varus malposition. The cement must always be removed completely. A good visualization of the medullary canal is required for this step of surgery. Sometimes a femorotomy, as a distal femoral window, may be indicated. At the time of selecting the implant, the advantages of the modularity should be exploited to the maximum. The objective is to ensure fixation in the metaphyseo-diaphyseal area of the femur. If this is not possible, try to achieve a short diaphyseal fixation. In any case, use a distal component of length 120 or 140 mm. We never recommend the implantation of a 200 mm distal component when the endofemoral approach has been selected. The definitive stem is placed in a single stage, after assembling the two components of the prosthesis outside the femur.

100

Third part – Surgical Technique

1. ARTICULAR APPROACH(ES) Both approaches, the antero-lateral as well as the postero-lateral can be used; however we draw attention to the fact that the postero-lateral approach provides a more direct vision on the main obstacles that may be encountered during revision surgery with a straight stem.

1.1 POSTERO-LATERAL APPROACH A. POSITIONING OF THE PATIENT AND SKIN INCISION (FIG. 1) ● Place the patient in the lateral decubitus position. The pelvis is fixed with a posterior post and an anterior post on the pubic symphysis, making sure not to compress the femoral vessels. The leg is held in a horizontal position with a cushion that can be easily removed, so as not to hinder the adduction movement in the hip joint. ● Make a skin incision centered on the greater trochanter and curved slightly backwards at the level of the pelvic region. In every case, try to avoid either a too anterior or too posterior skin incision. The proximal half of the incision should Figure 1 be long enough in order to avoid any pressure on the soft tissues while preparing the medullary cavity (remember that all revision instruments are very long). B. APPROACH TO THE ARTICULATION (FIG. 2) ● Split the fibers of the fascia lata and the M. glutaeus maximus longitudinally. A Charnleytype retractor both stable and inflexible may be helpful. ● Identify and retract the posterior edge of the M. glutaeus medius before carrying out the posterior capsulotomy, then detach the pyramidal, obturator and gemelli tendons at their femoral insertions. ● Frequently an incision of M. quadratus and the aponeurotic extension of M. glutaeus maximus may become necessary. ● Before performing the capsulotomy, the insertion of the M. glutaeus medius may be dissected free at the level of the tip of the greater trochanter. Alternatively, a transverse cut of the posterior fibers of the gluteaus medius at its femoral insertions can be performed (fig.3).

Figure 2

Figure 3

Chapter 3 – Option 2: Endofemoral approach

101

This partial transverse cut in the dorsal tendon of M. glutaeus medius prevents the muscular fibers from being lacerated during the preparation of the medullary canal; moreover, it allows for incision of the capsule and the short rotator tendons in one piece at the level of the dorsal trochanter, which facilitates their transosseous reinsertion at the end of the operation. This gives an unrestricted sight on the dorsal tip of the greater trochanter which is the biggest obstacle during the implantation of a straight stem. C. LUXATION OF THE PROSTHESIS (FIG. 4) – This is performed by simultaneous flexion-adduction and internal rotation with traction applied on the femoral neck with a hook. In tight hips this maneuver has to be made diligently, especially when the femur is fragile. D. REMOVAL OF THE PROSTHESIS AND EXPOSURE OF THE FEMUR – The prosthesis is removed with the help of an extractor that seats at the inner part of the neck. Before removal, open the greater trochanter widely in order to remove the cement seating in the former region of the stem shoulder (fig. 5). If there is a loosening between bone and cement, the Figure 4 prosthesis will stick in one piece together with the surrounding cement. Make a large lateral and posterior opening in the greater trochanter in order to allow for the extraction of this block. – Exposure of the proximal femur (fig. 6): All capsular and fibrous attachments on the medial cortex are removed, taking care to stick close to the bone. – Now, excise the fibrous scar tissue and granuloma formations around the articular cavity, proceeding from the center to the periphery. This will allow preservation of a dorsal capsular flap and thus, a prophylactic visualization of sciatic nerve is not necessary.

Figure 5

Figure 6

It is not possible to have an unrestricted sight on the medullar cavity if the femur remains attached deep in the articular cavity. Beware of trying to expose the proximal femur by forcibly adducting the leg since this will cause an avulsion fracture of the proximal femur, if it has not been preliminarily freed from all attachments.

102

Third part – Surgical Technique

1.2 ANTERO-LATERAL APPROACH A. POSITIONING OF THE PATIENT AND SKIN INCISION (FIG. 7)

Figure 7

– Place the patient in a dorsal decubitus position. Stabilize the other hip with the help of a post. – Make a lateral skin incision centered on the greater trochanter, or off-centered slightly backwards. The incision is either linear or it may be angled slightly upwards and forwards in its proximal half. The operated hip should slightly protrude over the edge of the table. Try to avoid using a too anterior incision.

Figure 8

Figure 9

B. APPROACH TO THE ARTICULATION – Make a longitudinal incision in the fascia Iata and M. glutaeus maximus in the direction of the muscular fibers, and then continue with the transgluteal incision of the vastus lateralis and M. glutaeus medius preserving the continuity of the muscles in order to create a digastric sling. (fig. 8). – The structures are held apart with a Charnley-type distractor, both stable and inflexible. – Muscles and tendons are detached at the level of the bone, taking care not to approach the superior gluteal nerve. The vastus lateralis and M. glutaeus medius are held apart with the extractor placed previously (fig. 9). Proceed then to the opening of the articular capsule. Avoid making the transgluteal incision too far anteriorly in order to respect the anatomical continuity between the M. glutaeus medius and M. vastus lateralis; and remember that the point of insertion of the instruments which is slightly off-centered backwards, will allow for a better exposure of the tip of the greater trochanter.

Chapter 3 – Option 2: Endofemoral approach

103

C. LUXATION OF THE PROSTHESIS (FIG. 10) – It is carried out by simultaneous flexion-adduction and external rotation and at the same time pulling on the femoral neck with a hook. In tight hip, remember to free first the proximal femur from its capsular attachments in order to allow for a decoaptation of the articular surfaces before trying the luxation maneuver.

Figure 10

Figure 11

D. REMOVAL OF THE PROSTHESIS AND EXPOSURE OF THE FEMUR – The prosthesis is removed with the help of an extractor that seats at the inner part of the neck. Before removal, open the greater trochanter widely in order to be able to remove the cement seating in the former shoulder region. If there was a loosening between bone and cement, the prosthesis and the surrounding cement will stick together in one piece. Make a large opening in the greater trochanter in order to allow for the extraction of that block (see fig. 5, p.99) – Femoral exposure: excise the fibrous tissues and granuloma formations at the level of the articular cavity, and then free the proximal femur from fibrous attachments incising the pyramidal and posterior capsule, always in tight contact with the bone (fig. 11). Femoral exposure is an especially important surgical step when selecting an antero-lateral approach because the point of insertion of the instruments (rasps and reamers) will be at the height of the trochanteric fossa.

2. FEMORAL APPROACH(ES) 2.1 TROCHANTERIC OSTEOTOMY The trochanteric osteotomy will allow for an articular and a femoral approach at the same time. REMEMBER: We only suggest the trochanteric osteotomy to be done when the femur is straight but the hip is very tight and the prosthesis is wedged too far (loosening with secondary migration of the implant) or when there is a proximal curvature of the femoral metaphysis (e.g. after a former femoral osteotomy). The osteotomy should be a digastric one, preserving the insertion of vastus lateralis, since it plays an important role in later stabilization of the trochanter (fig. 12). Indications for a trochanteric osteotomy are infrequent, since we recommend making a femoFigure 12 ral flap in most cases.

104

Third part – Surgical Technique

2.2 OPENING OF GREATER TROCHANTER (FIG. 13) If the endofemoral approach is selected, a wide opening of the greater trochanter will be necessary. One must remember that a straight revision stem has to follow the main direction of the medullary canal as does an intramedullary nail. Such, a wide lateral and posterior opening in the area of attachment of M. piriformis will be made, once the proximal cement mass has been removed and always before continuing to remove the more distal cement layers in the intermediate and then distal areas of the femur. Often the bone will be sclerotic and dense at this level and this stage of surgery will be best completed with the help of forceps and hollow chisels, trying as much as possible to stay within the spongious area of the bone. In some cases, a resection of the tip of the greater trochanter is required, above the trochanteric fossa, when the GT is on the top Figure 13 of the medullar cavity. Complete the opening of the greater trochanter with conical long reamers before starting the preparation of the femur. 2.3 FEMORAL WINDOW (FIG. 14) If an endofemoral approach has been selected, a femoral window may be indicated for the removal of a cement plug and to allow the possibility of a proximal fixation of the implant. The window may be made either lateral or antero-lateral; if the cortical bone is thick, it will be cut in the form of a wedge, which will make it easier to Figure 14 put it back into place without an additional osteosynthesis. A femoral window will make the removal of a long distal cement plug easier, but this is not an easy operative step at all. On the other hand the window will also enable a direct visual control of the good centering of the drills, chisels and extractor.

3. REMOVAL OF THE CEMENT In revision surgery with an uncemented stem, the complete removal of the cement layer is mandatory.Remnants of cement can induce eccentric reaming or obstruct the osseointegration of the stem. This stage of the surgery is often long and laborious, many surgeons will anticipate that loss of time and prefer to make a femoral flap straight from the beginning. If however the decision is to make an endofemoral approach, then it should be remembered that a direct vision of the medullary canal is essential, thus making a wide opening of the great trochanter mandatory, taking any sagittal curving of the femur into account, even a slight one. After the removal of proximal cement, which is an easy step in most cases, the cement in intermediate area of the femur has to be excised, and only then, can the surgeon proceed to the removal of the distal cement and of the cement plug, if any.

Chapter 3 – Option 2: Endofemoral approach

105

Some authors propose to use mechanical or ultrasonic extraction devices; others use centering devices for drill bits. Neither of these devices prevents the surgeon from trying to have a direct insight into the medullary canal and the later area of fixation of the prosthesis. They especially are not able to prevent from creating a via falsa in a much curved femur!

3.1 CEMENT REMOVAL IN THE INTERMEDIATE AREA (FIG.15) The cement layer of the intermediate area is removed with the help of chisels which are designed for this purpose. The mantle is extracted in pieces whilst carefully controlling the interface at all times. A continuous cleaning from granuloma and fibrous tissue, cold-light and a rinsing-cleaning device will be required for this step. We prefer this procedure to other methods, especially when the cortical structures are weakened.

3.2 REMOVING THE CEMENT IN THE DISTAL AREA (PLUG)

Figure 15

Proceed to the removal of the distal cement only after complete excision of the intermediate cement layer (until the cement plug is reached). – First, drill a hole into the plug with a 6 mm drill bit, making sure that it remains well centered (fig. 16A). – If a via falsa has been excluded, proceed stepwise to the enlargement of the hole with reamers of 10 or 11 mm, in order to pass a large retrograde cement extractor through it, which is very effective in removing even thin layers of remaining cement (fig. 16B). Beware of residual fragments of cement (especially when seated in the convex side of the femur), as these could mislead to an off-the-axis reaming creating eventually a via falsa. The same is true for an endomedullary ossification, frequently present in the area of the former tip of the stem (fig. 17). The risk of a perforation of the corticals is particularly high when the former stem was seated Figure 16 in an oblique position in frontal as well as in sagittal plane. A good centering of the drill bit before drilling is therefore very important. If the bone is weakened by osteoporosis, the fragmentation of the cement mantle through a window should be made in a cautious manner. Never hammer on a chisel at a right angle to the bone surface, since the risk of a fracture is high.

Figure 17

106

Third part – Surgical Technique

4. PREPARATION OF THE FEMUR AND FIRST SELECTION OF THE IMPLANTS PRELIMINARY REMARKS Preparation of the femur. The choice of an endomedullary approach primarily means that a proximal metaphyseo-diaphyseal fixation should be realized. Only when this type of fixation fails, should a secondary option such as a diaphyseal fixation mode be aimed for. In the case of a metaphyseo-diaphyseal fixation, the femur is prepared with the help of a rasp which will also be used as test prosthesis. In such a case, a spout proximal component of the rasp is chosen (mostly 55 to 85 mm high) and is mounted on the distal component of 120 mm. However if the proximal fixation fails and a diaphyseal fixation mode has to be prepared, then preparation of the femur with conical reamers becomes mandatory. Now the rasp functions mostly as test prosthesis and it is composed of a spout proximal part mounted on a distal part of 140 mm. Choice of the implant. Make a preliminary choice by measuring the penetration of the prosthesis using the usual landmark: The center of rotation of the articulation should be located at the level of the tip of the greater trochanter. The final implant sizes are chosen during the preparation steps through repeated trial reductions. The spout proximal components should be used.

4.1 CALIBRATE THE FEMUR AND COMPLETE THE PREPARATION OF THE OPENING OF THE GREATER TROCHANTER ● In a first step, calibrate the femur with a cylindrical reamer in order to reduce a narrowing in the area of the tip of the removed prosthesis and calibrate the medullary canal. (fig. 18A) Beware of remaining cement layers or of an endosteal H-console, both of which can result in an eccentric reaming and subsequent perforation. ● In a second step, verify with the help of a conical reamer of one size smaller than the former calibration of the medullary canal (usually of 14 mm diameter) the alignment of the proximal femur in relation to the diaphyseal femur and adjust the posterior and lateral opening of the greater trochanter (fig. 18B). NB. At this stage of the revision, do not use the conical reamers for the preparation of a diaphyseal fixation. This step is not necessary if a trochanteric osteotomy has been performed previously.

4.2 PROXIMAL FIXATION (METAPHYSEO-DIAPHYSEAL) If a proximal fixation in the metaphyseo-diaphyseal area is to be realized, the preparation of the femur is done with a rasp that will Figure 18 also be used as test prosthesis. A modular rasp allows for preparation of the femur in two stages: ● In the first step, impact the distal rasp (120 mm or 140 mm) with the graduated cylindrical hand-piece, until a perfect primary seating is achieved. Evaluate the depth of penetration in order to choose the appropriate proximal part of the rasp, always referencing to the tip of the greater trochanter (fig.19A). ● In the second step, assemble the distal part of the rasp together with a spout proximal part,

Chapter 3 – Option 2: Endofemoral approach

107

the size of the latter having been determined in the previous step. Impact the assembled rasp with gentle taps down to the level determined in the first step. Verify the degree of impaction by referencing to the tip of greater trochanter and the center of rotation (medium neck length of the rasp which is used as a test prosthesis).(fig. 19B). If it is difficult to impact the assembled rasp, a separate preparation of the metaphyseal area with a proximal rasp of one size smaller can be performed without using its distal component. A fixation in the metaphyseo-diaphyseal area by means of a distal piece of 140 mm is realized in those cases where a distal femoral window has been made in order to extract a cement plug.

4.3 DIAPHYSEAL FIXATION If it is not possible to ensure stability in the proximal area, a distal fixation mode must be aimed for. In this case, the preparation of the femur is carried-out in two steps: Preparation of the diaphyseal femur with the conical reamers, and then the choice of the implant by means of the rasps-test prostheses. Figure 19

PREPARATION OF THE DIAPHYSEAL FEMUR WITH REAMERS In any case it is necessary to increase the diameter of the reamers until sector 1 of the distal reamers with 140 mm of length is reached. (fig. 20). – The diameter of the reamer and its depth of penetration is noted, using the tip of the greater trochanter as a reference. The transverse referencing line on the handle of the reamer will help in selecting the height of the proximal rasp used in the next step of the preparation. Example on the opposite drawing: The reference line 1-75 corresponds to the following test prosthesis: A distal piece having the diameter of the reamer, length 140 mm (e.g. digit 1), which will be Figure 20 coupled with a proximal piece of 75 mm height. In endofemoral preparation of the femur, the distal component should always be of 140 mm length. A component size of 200 mm must be avoided since a safe diaphyseal press-fit fixation is uncertain with long components. In the situation where the reamer line lies in sector 2 (corresponding to a distal trial component of 200 mm), the diameter of the reamer should be augmented, or a distal trial component of 140 mm length with a diameter + 2 mm should be chosen.

PREPARATION OF THE PROXIMAL FEMUR AND THE CHOICE OF THE IMPLANT SIZE WITH THE HELP OF A RASP-TEST-PROSTHESIS The rasp-test-prosthesis fulfills two different functions in this situation: completing the preparation of the fixation area and allowing for the first choice of implant size using the usual reference line, e.g. the center of rotation is at the level of the tip of greater trochanter.

108

Third part – Surgical Technique

– In the first step, prepare the proximal femur with a modular assembly that is composed of a distal component which, if possible, is one diameter inferior to the last conical reamer and a proximal component which should be a one size smaller in height than the one determined by the reamer (fig. 21A). – In the next step, the two components, whose size had been defined during the previous reaming of the diaphyseal femur, are assembled together (diameter of the distal piece of 140 mm length and height of the proximal piece). The assembled rasp is then introduced into the medullar cavity and driven forth by means of gentle taps, always controlling the correct antetorsion. Control the level of the center of rotation medium neck in relation to the tip of the greater trochanter (fig. 21B). WARNING! The sizes and references which are produced by the assembled rasp are only an indication and the height of the proximal component might differ from the one predicted by the reamer. Besides, we recommend avoiding the choice of extreme heights for the proximal component, e.g. 55 or 105 mm, in order to maintain a margin of freedom for Figure 21 the next steps of the operation. In the case of height 55 mm, additional reaming should be performed in order to sink the rasp 10 mm or more. Then select a proximal component of one height bigger, always without changing either the diameter or the length of the distal component. In the case of a height of 105 mm, select a distal component of unchanged 140 mm length, but one diameter bigger. Some times an additional gentle reaming is needed in order to avoid the choice of a too short proximal component.

5. CHOICE OF THE FINAL IMPLANTS During revision procedures, it is not mandatory to respect the usual reference point (tip of the greater trochanter-center of rotation with a ball head, neck length M) in order to calculate the correct leg length. If there was a significant shortening of the limb before the revision, a slight shortening will be better tolerated than an over-lengthening. Sometimes, severe stiffness of the joint makes an ideal restoration of the leg length impossible. Finally, the position of the cup should also be taken into account. Hence it is recommended to carry out several trial reductions before selecting the final prosthesis which is to be implanted. During this stage of the operation the surgeon controls the planned length of the leg and the possibility of an easy reduction. He should remember that with the endomedullary approach, the implantation cannot be done in two steps. This narrows the liberty of maneuvering, since the only way to correct any discrepancy of the leg length is reduced to the different sizes of necks. During the same trial reduction, the anteversion of the neck should be controlled and adjusted if necessary by rotating the proximal component. Once the trial head with medium neck (M) has been placed, another trial reduction is made.

Chapter 3 – Option 2: Endofemoral approach

109

The surgeon may be confronted with any of the three following situations: 5.1 THE REDUCTION IS EASY AND THE TENSION APPROPRIATE The articular surfaces are stable and the correct length has been restored, the selected implant is the right one, and if the definitive prosthesis subsides at a different level from the one reached with the trial prosthesis, this difference can be corrected when choosing the definitive femoral head, provided that the trial reductions have been performed with a medium neck femoral head.

5.2 THE REDUCTION PROVES TO BE DIFFICULT A shorter neck will not be sufficient to correct this situation. The height of the proximal component must be reduced, even if the leg does not seem to be exceedingly long. If the shortest proximal component is already in place, then the whole prosthesis has to be removed and an additional reaming must be performed, keeping the same sizes of both the distal and proximal components. NB. A change in the dimensions of the distal component is infrequently necessary and we advise to enlarge the diameter of the medullar cavity, first using the same diameter, sometimes even one size bigger using extreme caution in these cases. Exceptional case: The seating of the prosthesis is at the correct level and a shorter implant (or a lower seating) is not an option. If the reduction proves however to be difficult or even impossible, then an extensive release of the periacetabular structures and sometimes an osteotomy of the linea aspera may be helpful in order to free the articular movement.

5.3 THE REDUCTION IS TOO EASY AND THE HIP IS UNSTABLE Weakness of the muscular structures may be the main reason for this situation. Instead of running the risk of an ill-tolerated excessive leg length, a locking cup may be preferred. Remember to measure the depth of penetration before starting the final implantation, especially if any changes have taken place during the above mentioned steps.

6. IMPLANTATION OF THE DEFINITIVE IMPLANT When an endo-femoral approach has been selected the assembly of the two components is made outside the femur with the help of a dynamometric wrench (see appendix 1 for detailed description of this technique). If the previous reaming has been done with precision, the assembled implant will wedge into place without difficulty. It is introduced into the medullary canal by hand. Repeated gentle taps with a hammer result in the seating of the implant 2- 4 cm distally. We caution against using forcible hammer blows when the seating takes place too proximally. This problem usually arises after an insufficient preparation of the wedging area. The implant should be wedged by hand while setting the correct antetorsion. In the case of an inappropriate antetorsion of the stem, the implant should be removed and reinserted manually in the right rotational direction. Avoid applying rotational torque on an already implanted stem.

110

Third part – Surgical Technique

The impactor is screwed into the threaded end of the proximal component (fig. 22). Gentle taps are repeated until a “cortical sound” arises from the bone. The depth of impaction is measured with the help of a sterile ruler. It is required to pause for some minutes before before verifying once again if the progression of the implant has stopped. The reduction is now carried out using a test head with medium neck length. Proceed to the usual control of leg length and stability of the hip. Now, the final modular head is fitted after having selected the correct neck length. If the trial reduction was made with a medium head, a deeper wedging of the prostheses can be corrected by using an L or XL sized Figure 22 Figure 23 neck; on the other hand a shallow wedging can be countered using a neck S or XS. If the stem penetrates deeper than planned, check whether a longitudinal fracture in the diaphysis has occurred. If this proves negative, then adding bone chips in the medullary canal will help to control the excessive penetration and improve its stability as well. These cortico-spongious chips must be added before final wedging of the prosthesis takes place or after a few centimetres of the implant have been taken out from its seating (fig. 23). Avoid using strong hammer blows and if the implant progresses in a too easy manner, make sure that there is no cortical fracture which remained unnoticed.

7. PROBLEMS AND INCIDENTS Incidents may occur while preparing the femur; these usually consist of difficulties in impacting the rasps. It is mandatory to avoid any forcible introduction because of the high fracture risk. We advise redoing the femoral preparation before returning to a smaller sized implant which could result in a varus positioning of the implant and the preparation of the femur should be completed. In most cases, it is due to an insufficient preparation of the greater trochanter; and more infrequently, the obstacle is found in the meatphyseo-diaphyseal or diaphyseal zone.

7.1 AT THE LEVEL OF THE GREATER TROCHANTER An insufficient lateral or posterior opening of the greater trochanter will turn out to be the main reason for the malpositioning and the blocking of the rasp. An incomplete lateral opening will result in a varus malposition of the rasp/stem (fig. 24A), or in an increased anteversion, if the opening is insufficient on the posterior aspect (fig. 24B). In both cases, forceful hammering could lead to a fracture of the greater trochanter and, if this incident is avoided, there is a risk for a varus implantation and/ or selecting an implant of one size smaller than the diameter of the medullary canal.

Chapter 3 – Option 2: Endofemoral approach

Figure 24

111

Figure 25

These situations can be overcome by completing the preparation of the femur and introducing a conical reamer of one diameter smaller than the actual diameter of the medullary canal and doing a wide opening in a dorsal and lateral direction, at the level of the trochanteric fossa, in order to make sure to remain in the axis of the femoral diaphysis. 7.2 AT THE METAPHYSEO-DIAPHYSEAL JUNCTION (FIG. 25) It is often found in the femurs which show an accentuated double curving in the sagittal plane, and which are often narrow in the antero-posterior plane. These situations can result in serious obstacles to the penetration of the rasp or can lead to an excessive antetorsion. In these cases, an additional reaming with a large cylindrical reamer can be helpful, taking care to keep close contact with the lateral cortex. 7.3 IN THE DIAPHYSEAL AREA If the diameter of the medullary canal proves to be too narrow (in CDH), the obstacle to the penetration of the rasp can be located at this level. It has to be noted that a reaming in the diaphyseal area with a diameter of 11 mm, or even 12 mm, is required with this type of implant.

8. CLOSURE OF THE JOINT 8.1 POSTERO-LATERAL APPROACH Reattach the tendon with 2 or 3 transosseous sutures to the tip of the greater trochanter if the posterior fibers of the medial gluteal muscle had been cut. The posterior capsule and the short rotators are carefully reattached, whenever possible, utilizing the posterior edge of the greater trochanter and the posterior portion of the medial gluteal muscle.

112

Third part – Surgical Technique

Figure 26

Figure 27

8.2 ANTERO-LATERAL APPROACH (FIG. 26) Reattach, if possible, the posterior capsule utilizing the posterior fibers of the medial gluteal muscle. Then reattach the digastric tendon portion of the anterior gluteal sling and the M.vastus lateralis, by doing transosseous sutures. We also advise to secure these sutures by an additional tension-band wiring using a nonabsorbable suture in the form of 8-sling utilizing the posterior edge of the greater trochanter and the tendons of the gluteal muscles. 8.3 FIXATION OF THE GREATER TROCHANTER (FIG. 27) Reminder: Before reattaching the greater trochanter, make sure that all the bone-cement has been removed. Sometimes an excavation is needed in order to create enough space for the lateral shoulder of the prosthesis. If a digastric osteotomy has been used, reattachment should be performed with the tensionwiring technique taking care to create a sound dorsal and lateral figure of 8 wiring that has a perfect base on the proximal femur. This reconstruction must be sufficient to withstand the simultaneous traction of the gluteal sling and the vastus lateralis (proximal and ventral traction). A classic trochanteric osteotomy should be reattached with the help of 3 metal wires which provide a stable tension wiring, completed by a tension-band wiring. The same principles are applied in the case of a trochanteric fracture.

CHAPTER 4

POSTOPERATIVE CARE

Generally speaking, a too cautions or confusing attitude should be avoided and at first, a physiotherapy which is only functional should be preferred. The too cautions attitude consists in keeping the patient in bed with suspension and traction frame for several weeks. If the strategy chosen by the surgeon entails such an attitude on a regular basis, not only the reliability of the method should be questioned, but also its consequences both economically and for the patient, especially when this surgeon has to recurrently perform revisions! The confusing attitude consists in giving permission to the patient to return home while allowing immediate weight bearing, or even more hazardous, partial weight bearing. It is very likely that this patient once relieved from his immediate postoperative pains, would go about testing the reliability of the assembly made by the surgeon, which in most cases, in absence of pain, leads to an early return to non-weight bearing. Concerning physiotherapy. In the immediate postoperative period, it is always preferable for the physiotherapy to be limited to a gentle and only functional one. An active physiotherapy, often painful or even dangerous, should be avoided. Subsequently, the question that should be asked is when is the right moment to start an active physiotherapy. Usually, it is prescribed after a first follow-up examination, about 8 weeks after the operation, underlining that, if it is not always required, in many cases it is nevertheless necessary to improve a specific motion area and to strengthen the gluteus muscles. Instructions given to the patient for the period immediately following the surgery should be kept simple and pragmatic. It is possible to distinguish between two different situations: The prosthesis is stable as it has been wedged perfectly in a femur which features cortical bone with high mechanical strength and a conical shaped medullary canal. In this situation, immediate weight bearing is allowed using two forearm crutches that have a dual role: taking excessive weight off the hip and avoiding incorrect movements, in the expectation of complete healing of the soft tissues. Early physiotherapy is only supportive, and it aims to teach the patient which movements must be avoided in order minimize rotational stresses on the prosthesis, in particular when standing up from a seated position or when going up- or downstairs. The patient will undergo a follow-up examination, including an X-ray control, two months after the surgery. At this time more active physiotherapy may be prescribed. Use of any aid will be abandoned progressively depending on the recovery of muscular strength, with the awareness that, generally speaking, there should be no hurry to cease using the crutches.

114

Third part – Surgical Technique

The prosthesis is not judged to be perfectly stable since the surgeon has some doubts on the quality of the wedging of the implant. Whatever the reasons may be for this concern, it is recommended to be cautious and to allow only partial weight bearing for a period of 6 to 8 weeks. During this period of non-weight bearing, it is preferable to keep the patient under supervision and, if she/he is admitted to a specialized center, order that she/he does not undergo any active physiotherapy throughout this period. At the end of this period, and after a follow-up X-ray, weight bearing may be allowed. In principle this should be gradual, however it is in practice nearly always complete and immediate. This cautious attitude is recommended in the early stages of experience with the implant. Moreover it should be stressed that each patient is a unique case and that the period for which weight bearing should be avoided can often be shortened to about 4 weeks.

CHAPTER 5

CONCLUSIONS

WHAT TO DO! 1. Have suitable X-rays available before the operation, in order to be able to perform an x-ray analysis which will allow the right approach to be selected, keeping attention to the requirements of the press-fit principles. 2. Do not hesitate to perform a transfemoral approach by creating a pediculated flap. It will allow for a safe revision operation and ease the goals of a good press-fit fixation. 3. Excise the cement mantle without damaging the bone stock. This step calls for a perfect vision of the medullar cavity. 4. Start the preparation of the press-fit area only when you are sure to be in a more or less straight segment of the femur and when all obstacles have been ruled out. 5. Perform a large opening of the trochanteric area when you are using an endofemoral approach. Do not switch to a distal fixation mode unless you really cannot perform a proximal metaphyseo- diaphyseal wedging. – First prepare the proximal area of fixation with the help of the rasps; make use of the reamers only if a diaphyseal fixation mode becomes inevitable. – Select the size of the implant with the help of the rasp, which will be used for the trial reduction. Make good use of the modularity during these crucial steps. – Select the definitive implant only after having performed some trial reductions and keep a reserve by doing the reduction with an M neck length. – Implant the previously externally assembled prosthesis in one step, by adjusting the necklength of the definitive head, or, in case of an early jamming of the prosthesis, by removing the prosthesis and redoing the last steps with the rasps again. 6. When a femoral flap has been selected, the fixation of the implant can only take place in the diaphyseal area and, in this situation, the reamers are only useful to make the medullary canal conical, but they are not used to make the choice of the implant, which is always done with the test prosthesis. – Ream (make conical) a straight segment of femur over a short distance – Use the conical area of the implant to ensure primary stability and during the wedging, remember to keep a reserve of the conical anchoring area, by preferring a shorter stem with a one size wider diameter than the medullary canal. In press-fit stems, always prefer the diameter to the length of the distal component. – If possible, do not perform trial reductions with the extreme heights of the proximal component (55 or 105 mm) in order to keep some liberty of decision during the wedging of the

116

Third part – Surgical Technique

definitive prosthesis, which anyway should always take place in two steps. 7. If there a femoral flap has been made, carefully fix it back into place which is especially true, when the area of fixation in the diaphysis is limited. If the trochanter has been fragilized during the procedure, add a tension-band wiring in every case. 8. During the postoperative period avoid any compromise when instructing your patient and when a full weight bearing is not possible or hazardous, it is preferable to keep the patient under close control during the period of non-weight bearing.

WHAT SHOULD BE AVOIDED! 1. Starting the surgery without having the correct x-rays at hands, allowing to detect the main obstacles to the implantation of a straight press-fit stem, especially the presence of a femoral curving. 2. Being convinced that every layer of bone cement can be excised without aggravating the bony lesions, when the corticals are weakened by a stress-shielding phenomenon or by osteoporosis. In such a situation, there is a risk for an incomplete removal of the cement. 3. Insisting on implanting a straight prosthesis in a curved femur using an endofemoral approach when the need for a femoral flap is obvious. On the other hand, do not try to transform a curved femur into a straight one by excessive reaming with the conical reamers. 4. Running the risk of an insufficient press-fit fixation by trying to introduce a long and thin stem through an endofemoral approach. A good press-fit fixation has nothing to do with the length of the implant, but depends on the quality of the wedging into a short and well prepared conical area of the bone. Try to select a shorter stem one size bigger than the diameter of the canal, if the seating is to take place next to the proximal anchoring zone. 5. Selecting the implant on the basis of the references provided by the reamer. This will in most times lead to the choice of a longer stem than necessary. 6. Not keeping a reserve during the choice of the height of the proximal component or the neck length can lead to problems during the seating of the definitive prosthesis and may sometimes require redoing all the steps of the trial reduction. 7. Trying an implantation in two steps if an endofemoral approach has been chosen. 8. Using strong hammer blows without measuring the depth of penetration in case of an early jamming of the test prosthesis. It will lead to a fracture or an early blocking of the implant. 9. Allowing an early return to home in a situation where the patient must unload the leg.

APPENDIX 1

ASSEMBLY OF THE TWO PROSTHETIC COMPONENTS

The assembly of the two components of a PFM-Revision 2nd generation stem (Revitan straight stem) stem is realised with a torque wrench providing a constant tightening force. It is performed in a different manner depending on the strategy selected by the surgeon. Assembly in one stage (extrafemoral) if an endofemoral approach was chosen or an intrafemoral assembly in two stages in situ, after implantation of the distal component with the help of a proximal temporary component if a femoral flap was performed. NB: The torque wrench is also used for the tightening and untightening of the safety nut. PRINCIPLE OF OPERATION AND DIRECTIONS FOR USE OF THE TORQUE WRENCH

The operating principle of the torque wrench key is based on using a cutting blade to split a polyethylene pin, the diameter of which has been determined such that it is always necessary to exert a torque of about 10 Nm to achieve this splitting. Thanks to the new system composed of PE shear pins, the torque wrench doesn’t need to be recalibrated and offers a high level of safety to the user.

118

Third part – Surgical Technique

b

a

1. Loading of the torque wrench. – Remove the cover (a) of the torque wrench (fig. 5, 6 and 7). – Unpack the loader (b). The polythelene loader holds 6 shear pins. In practice, during each maneuver (torque), it is recommended to split 2 pins. With a shear pin loader it is thus possible to apply a torque three times (under normal conditions, 2 times are required during the assembly).

NB: A shear pin loader is packed with each proximal component of the implant. The shear pins should not be resterilised.

2. Place the pins into the recesses of the torque wrench and turn the loader to free the shear pins.

3. Remove the lock of the loader.

Appendix 1: Assembly of the two prosthetic components

119

4. Put the cover back in place and lock it in position “Lock”.

5. After use, unlock the cover by pushing the button.

6. Turn the cover to the “Open” position.

7. Remove the cover and extract all the “remaining” shear pins from the recesses.

120

Third part – Surgical Technique

ASSEMBLY OF THE STEM IN ONE STAGE (EXTRAFEMORAL) 1. Position the definitive proximal component Before starting with the assembly of the two implant components, position the proximal component onto the morse taper of the distal component by hand and set the desired antetorsion of the proximal component. This step must be done before any assembly force is applied to the stem. Once the antetorsion is chosen, push the two parts together by hand to give them stability before continuing with the assembly.

2. Screw on the stem tensioner Screw the threaded rod of the tensioner onto the threaded part of the morse taper. To screw on the tensioner, hold it in the hand so that the threaded rod protrudes from the tensioner. Alternatively, the threaded rod can be removed out of the tensioner, screwed onto the threaded part of the morse taper and eventually the tensioner is reassembled. Tighten by hand the nut of the stem tensioner.

3. Assembly of the proximal component Hold firmly the stem tensioner and tighten the assembly of the two components with the torque wrench. For this process the request of assistance is strongly recommended. Further, don’t use the stem holder to maintain the implant in order to keep control of the antetorsion of the proximal component.

4. Screw on the conical nut Finally the conical nut is screwed onto the threaded part of the morse taper with the help of the setting instrument and tightened with the torque wrench. For the tightening, the implant is placed into the stem holder for an easier control of the rotational stresses.

Appendix 1: Assembly of the two prosthetic components

121

ASSEMBLY OF THE STEM IN TWO STAGES (INTRAFEMORAL) 1. Assembly of the temporary proximal part and implantation of the definitive distal component After wedging in the definitive distal component, further trial reductions can be carried out if necessary by changing the sizes of the proximal trial part and changing its antetorsion (up to +/–30°).

2. Position the definitive proximal component Before starting with the assembly of the two implant components, position the proximal component onto the morse taper of the distal component by hand and set the desired antetorsion of the proximal component. This step must be done before any assembly force is applied to the stem. Once the antetorsion is set, push the two parts together by hand to give them stability before continuing with the assembly.

3. Screw on the stem tensioner Screw the threaded rod of the tensioner onto the threaded part of the morse taper. To screw on the tensioner, hold it in the hand so that the threaded rod protrudes from the tensioner. Alternatively, the threaded rod can be removed out of the tensioner, screwed onto the threaded part of the morse taper and eventually the tensioner is reassembled. Screw tight by hand the nut of the stem tensioner.

4. Assembly of the definitive proximal component and screw on the conical nut Hold firmly the stem tensioner and tighten the assembly of the two components with the torque wrench. For this process the request of assistance is strongly recommended. Finally the conical nut is screwed onto the threaded part of the morse taper with the help of the setting instrument and tightened with the torque wrench. When tightening the conical nut, neutralise the torsion stresses caused by the torque wrench exerting counter-pressure on the neck in the opposite direction to the tightening by hand or with the specially provided handle positioned over the neck of the implant.

APPENDIX 2

REMOVAL OF A PFM-REVISION 2nd GENERATION (REVITAN STRAIGHT STEM)

The possibility of extraction must be provided for every implant. Although this statement is true, it is nevertheless often difficult to fulfil it. In particular, a cementless stem can be very demanding to extract when it is not loosened. In the PFM-Revision 2nd generation (Revitan Straight system), the development of disassembly instruments for the proximal component allows to some extent to overcome this disadvantage as its removal is always possible without difficulties. On the other hand, the extraction of the distal component is often demanding, especially if it is well osteointegrated or has a long stem.

INDICATIONS The removal of a PFM-Revision 2nd generation (Revitan Straight stem) can involve the proximal part or the entire implant.

REMOVAL OF THE PROXIMAL PART The distal part of the implant is stable but it is necessary to remove the proximale part in order to: ● Change the proximal component to replace it mostly with a higher one for reasons of a secondary subsidence of the implant (fig. 1). Less frequently the antetorsion has to be modified (recurrent dislocation). ● Allow articular access either to carry out cleansing of the joint cavity (sepsis) or to work on the cup (changing it or placing an antidislocation device).

Figure 1

Example attached. Bipolar loosening: revision with PFM revision stem and St. Nabor cup. Radiography after 6 months: significant subsidence of femoral stem. Expanded trochanterotomy and replacement of the proximal component with a component of greater height with extra-long neck. The distal component, which had not loosened, was left in place.

124

Third part – Surgical Technique

REMOVAL OF BOTH COMPONENTS Two different situations can be identified: ● There is a loosening with abnormal mobility of the implant: in this case, removal of the stem does not pose any problems. The stem can be removed in one piece using a simple extractor (fig. 2). ● In absence of a loosening, the implant should be extracted in two stages: removal of the proximal part and then removal of the distal part. Example attached. Revision of PFM revision stem and dualmobility cup, for repeated bipolar loosening (radiography after 3 years). Revision, pseudarthrosis of trochanter major: simple removal, monobloc, of the femoral stem (distal component, 200 mm in length) which was replaced by a shorter stem with a large diameter. Replacement of dualmobility cup with a cementless St. Nabor cup.

NB. The stem is not loosened but it is necessary to remove it in order to carry out cleansing of the medullary canal (sepsis) or because of secondary subsidence, which cannot be compensated by changing the proximal part. Figure 2

FEMORAL APPROACH(ES) Expanded trochanterotomy. If it is intended to remove only the proximal component, articular access is strongly recommended in the form of an expanded trochanterotomy at the lateral cortex up to the point at which the proximal and distal components join (fig. 3). Perform a digastric trochanterotomy, preserving the insertions of the musculus vastus lateralis at the base of the greater trochanter. Figure 3

Appendix 2: Removal of a PFM-Revision 2nd generation (Revitan Straight Stem)

125

If it is intended to remove both components of the prosthesis (proximal and distal components), apart from any loosening, two options are available to the surgeon: ● Lateral flap. If the stem is short (length of distal component 140 mm), create a lateral flap, with a length of between 15 and 20 cm (fig. 4B). ● Expanded trochanterotomy or flap and diaphyseal partition line. If the stem is long (length of distal component 200 mm): after having performed an expanded trochanterotomy or a flap, use an oscillating saw to create a diaphyseal partition line (fig. 4A). Creation of a femoral flap is the most reliable option and it is worth bearing in mind that if the implant has not loosened it is always easier to remove a short stem.

Endofemoral route. If the stem is loose, endofemoral access alone is possible with care being taken to eliminate obstacles at the level of the greater trochanter. Choosing endofemoral access requires wide lateral and posterior opening of the greater trochanter.

126

Third part – Surgical Technique

REMOVAL OF THE PROXIMAL COMPONENT (2nd GENERATION) N.B. The disassembly of the PFM-R 2nd genration (Revitan Straight) is performed following an analogous process as with the first generation stem (PFM-R), with the sole difference that the instruments changed somewhat due to the presence of a thread at the level of the shoulder. For the disassembly of a proximal component of the first generation different instruments are required and available in a separate instrument set.

DISASSEMBLY INSTRUMENTS

a) Disassembly instrument with distal thread

b) Threaded rod with T-handle

c) Disassembly threaded sleeve

1. The torque wrench is in the “Lock” position.

2. Removal of the conical nut with the torque wrench.

3. Screwing the disassembly sleeve onto the distal implant component, using the setting instrument (see fig. 6b).

Appendix 2: Removal of a PFM-Revision 2nd generation (Revitan Straight Stem)

127

4. Screw the disassembly instrument into the thread on the shoulder of the proximal component.

5. The threaded rod with T-handle is screwed into the disassembly instrument (a) which is positioned over the threaded sleeve (c). By firmly turning, the threaded sleeve, and thus the distal implant component is pressed downward; and pressure is exented simultaneously on the proximal component with the disassembly instrument.

NB. If the decoupling cannot be achieved by this simple tightening, hit a sharp blow with a mallet on the handle of the threaded rod.

6a. Decoupling of the proximal component. 6b. Removal of the disassembly sleeve using the setting instrument.

NB. Protect the morse taper with a compress if the removal of the distal part is not envisaged.

128

Third part – Surgical Technique

DISASSEMBLY OF THE PROXIMAL COMPONENT (1st GENERATION) Warning! For the implants of 1st generation, the absence of a thread in the proximal zone of the female morse taper, entails the necessity for a specific ancillary instrument in order to disassemble the proximal component. Placement of the threaded sleeve (fig.7) Unscrew the security nut with the wrench suited for this purpose or with the torque wrench, from the new range. Position the cylindrical threaded sleeve of one diameter smaller than the security nut.

NB. The threaded rod, which will be later used for the disassembly, is based on this threaded sleeve (see fig. 9a). During its placement, it is not necessary to lock it.

Figure 7

Placement of the disassembly instrument (fig.8) The disassembly instrument is a connection taper positioned over the neck, which should be oriented in order to direct the engraved pointer to the shoulder of the implant. Warning! The disassembly instrument should correspond to the implanted proximal component and there are 3 different disassembly instruments: P1 is for P1 former model (old A1.P1) Ref. 0100079013 P1 is for P1 new model Ref. 0100079007 P2 –P2,5 – P3 for the 3 other instruments Ref. 0100079008

Figure 8

Appendix 2: Removal of a PFM-Revision 2nd generation (Revitan Straight Stem)

129

Disassembly of the proximal component The threaded rod with handle (ref. 0100079009) is screwed into the disassembly component and is positioned over the threaded sleeve (fig.9a). The tightening has a decoupling effect on the proximal component by means of the connection taper, which is positioned over the neck. During the decoupling, it is sometimes required to control the rotational forces with the help of the sleeve based on the morse taper (fig.9b). If the decoupling cannot be achieved by this simple tightening, hit a sharp blow with a mallet on the handle of the threaded rod. Once the proximal component has been removed, proceed Figure 9 with a careful cleansing of the medullary cavity and protect the morse taper with a compress if the removal of the distal part is not envisaged.

Removal of the distal component The removal of the distal component is done after having removed the proximal component. It is performed with the extraction instrument (01.00079.011) (fig.10a) that is screwed onto the connection taper and the slap hammer. This allows striking with sharp taps in the axis of the distal component (fig.10b). If the distal stem cannot be extracted from the femur with a few hammerblows additional measures have to be taken, like longitudinal osteotomies, a window, a flap or the introduction of flat chisels along the stem to facilite the extraction of the distal component.

a

Figure 10

b

FOURTH PART

THE RESULTS

FOREWORD The study of the clinical results allows forming a clear picture on the indications and the limitations of a concept or of an implant. It also helps to validate a methodological proceeding, e.g. defining a strategy first and then perfecting the surgical technique. Whereas consensus exists on the methods used for the appreciation of clinical results, this is not the case with the radiological analysis. Only an experienced and critical examiner will be able to distinguish between the characteristics of different types of implants and their radiological implications. This is the reason why we decided to present a new method for evaluation of postoperative radiological results.

CHAPTER 1

PATIENTS AND TYPE OF OPERATION To evaluate the results of an implant, one must have a good knowledge of the patients, of the method used and of the implants placed during revision surgery; the whole of these data is usually called, in a somewhat mundane way, “the material”.

1. PATIENTS 1.1 MATERIAL This is a retrospective study of a series of 180 total hip prostheses that underwent a revision procedure in the years 1994 to 1999. 28 patients were excluded for the following reasons: – 9 patients died before they could be controlled for a reasonable time after the operation – 16 patients refused or were lost to follow-up – 2 patients underwent revision surgery with a different method (1 for major subsidence, 1 deep infection) – 1 patient presented with of the proximal component and could not be evaluated 152 prostheses were available for the present study; 4 of these patients died during the study period but they were included in this study as they had undergone a sufficient follow-up. There were 148 patients with 152 revision procedures (three underwent a bilateral revision and one patient was revised for a second time with the same type of implant). 63 left hips and 89 right hips were controlled. The medium follow-up interval was 3 years (min. 1 yr; max. 7 yrs). There were 76 female and 72 male patients; the mean age was 75 years (min. 31; max. 95 yrs). 71 patients (corresponding to 47%) were classified as “active pensioners” and 64 patients (42%) as “sedentary pensioners”.

1.2 ETIOLOGY AND REASON(S) FOR THE REVISION PROCEDURE Osteoarthritis was the main reason for the primary hip replacement (76%). Other reasons were: Secondary arthritis in 13%; rheumatoid arthritis in 3%; necrosis in 3%; posttraumatic sequelae in 3% and other reasons in 2%. Reasons for the revision procedure were: aseptic loosening of the femoral component with abnormal mobility of the stem, often followed by a distinct bony lesion in 53% of the cases; in 29%, the loosening was not evident on the preoperative x-rays, however local lesions due to granuloma formation made the operation mandatory. In 2% (3 cases) there was a fracture of the implanted material and finally, in 15% of all cases, the stem revision was justified by the presence of a loose cup and the necessity to implant a new tribological coupling giving better results in young patients.

134

Fourth part – The results

1.3 PREVIOUS OPERATIONS 23 patients had undergone previous hip surgery other than implantation of prosthetic implants. 24 patients (15%) had already undergone previous prosthetic revision surgery due to a loosening: 20 were revised for the second and 4 for the third time. 139 patients got a cemented stem (91%) during the previous surgery. In 131 cases the revision was performed on both components, but in 21 cases the cup was left unchanged (which turned out to be an error in some of these cases). 40% of the controls had a total hip replacement on the other hip. COMMENTS Our series is more or less comparable to many other series presented in literature. It differs however in the sex distribution which is almost equal and the large number of patients in the age group between 80 and 90 years (39 patients, equaling 25%). Finally, 64% of the controls were overweight. We find it necessary to make some comments here on the term “learning curve” and on the minimal follow-up interval for a valid appreciation of our results: The learning curve. Although a learning curve is sometimes cited in some publications, an appreciation of the final results in the light of this learning curve is mostly lacking. This is an error in our view. The reader is indeed interested in information on the different stages the author went through during the evolution of the strategy and the operation technique. Thanks to a critical appraisal of his own initial failures or complications, the author may contribute to a better understanding of the potential dangers and pitfalls of his procedure. Unfortunately, distinct errors that occurred to the first author are likely to repeat themselves in the hands of the followers if they are not fully informed of the requirements of the method. That is why we think that every learning curve should be included in the final results, even if this affects the quality of the global outcome of a new technique. Nothing equals the personal experience, especially when it comes from the first author of a method. We therefore evaluated some of the results under this important aspect and hence distinguished between 2 groups of patients: ● Group 1, including 78 prostheses, with patients who were operated on before Jan. 1998. It comprises our own learning curve, as the strategy that we subsequently developed was not yet fully applied to these cases. ● Group 2, including 74 prostheses comprised those patients who were operated on after Dec. 1997 and where the strategy proposed in this book was strictly applied to every case. The medium minimal follow-up time needed. A minimal time of two years is needed for a valid appreciation of the clinical and radiological results. Generally, in the absence of any serious complication, the postoperative result will not evolve further after the first year, with exception of some degrees of a better ROM or a slightly improved gait when the other hip is not impaired. The same can be said about hip pain which is low or absent after this time. Continuous or increasing pain is probably linked to a loosening of either the stem or the cup. If no prosthetic loosening can be found, such painful conditions are sometimes attributed to the presence of an uncemented stem. Concerning the radiological analysis, two years seem to us to be the shortest lap, since in revision surgery this time interval is needed for reconstruction of the bone stock or osseointegration.

Chapter 1 – Patients and type of operation

135

Even more time will be necessary in order to evaluate bone remodeling phenomena or changes which are inherent to any arthroplasty (wear, etc.). As for the long term results (10 years and more), it must be noted that the usage of this notion is sometimes questionable and we consider that the necessity for such results should not prevent from publishing more recent ones, since in the meantime some improvements may have occurred.

2. OPERATION This series of 152 prostheses consists of patients operated on by the same author. Two points are of particular interest: The femoral approach and the selection of the implants.

2.1 APPROACHES TO THE FEMUR The following table shows the different femoral approaches used for to the two groups of patients as described above. The evolutions observed during these two periods clearly show that the notion of a “learning curve” is confirmed by the reality when a strategy is to be established. Evaluation in number of implants and percentage Approach to the femur

Endofemoral Femoral window

Group 1 78 prostheses

Group 2 74 prostheses

34 (15)

43,6%

8 (4)

10,8%

Trochanteric Osteotomy

13

16.7%

4

5,4%

Pediculated flap (vascularized)

6

7,7%

62

83,8%

Non-pediculated flap

25

32,0%

0

0%

Group 1 shows a nearly equal distribution of endofemoral approaches and femoral flaps. An endofemoral approach was often combined with a femoral window. During this period, the femoral flap was usually a devascularized one. In group 2, however, a smaller variation of approaches is seen since 84% were operated through a femoral flap, which preserves the blood supply to the flap. At the same time, the number of femoral windows or simple trochanteric osteotomies was reduced. We nevertheless think that the number of endofemoral approaches performed in this group is too small and that some femoral flaps were performed in excess. It must be noted that 3 patients out of 4 receiving a trochanteric osteotomy already had a pseudarthrosis from a previous operation. N.B. The evolution that took place between the operation modes of group 1 and 2 is the result of the earlier described strategy. The reason for this development is our own complications when preparing the bone bed (fractures, wrong ways), the early appreciable results of an insufficient technique (bone necrosis after devascularizing the bone flap) as well as the relatively high number of cases with cement remnants that were overseen, which occurred with excessive use of the endofemoral

136

Fourth part – The results

approach. Simultaneously, a better understanding of the press-fit concept (the perfect fit of the two contact surfaces) and the role of the femoral curving led to a global understanding which finally evolved in the creation of a femoral flap in the majority of the cases.

2.2 IMPLANTS USED AND MODE OF FIXATION A PFM-Revision of 1st generation stem was used, in the way it has been described earlier. The following table shows the total number of implants used in both periods and the type of primary fixation (bone-implant contact zone). Implants/Mode of fixation Implants

Fixation mode

Proximal components (Height in mm)

Distal components (Length in mm) Proximal Global Short diaphyseal Long diaphyseal

55 65 75 85 140 200

Group 1 78 prostheses 17 (22%) 26 (33%) 0 35 (45%) 34 (44%) 44 (56%) 19 (24%) 20 (26%) 9 (12%) 30 (38%)

Group 2 74 prostheses 35 (47%) 24 (33%) 3 (4%) 12 (16%) 50 (68%) 24 (32%) 3 (4%) 6 (8%) 41 (56%) 24 (32%)

– The long distal component of 260 mm was never used in a first revision situation. It was used only once following a fracture of the femoral shaft that occurred during the impaction of a short revision stem. In our opinion, the indication for these long stems is seldom required should be reserved for the treatment of a complication. – The infrequent use of the proximal component of 75 mm (old: P 2.5) is due to the fact that they were produced at a later stage; the components of 95 and 105 mm are available only since 2003 and in any case, we advocate for a limited use of these implants. – We would like to draw your attention to the fact that the short implants were mostly used in the second group: The percentage of long proximal parts (85 mm) is reduced from 45 to 16% and the percentage of the short distal components (L 140 mm) increases from 44 to 68%. At the same time, the primary fixation mode (location and height of the bone-implant contact zone) also shifted, so that in group 2 a short diaphyseal fixation was largely preferred and 2/3 of these patients received a short implant.

CHAPTER 2

CLINICAL RESULTS

METHODS USED In the literature, most authors use the Harris (13) or the Merle d’Aubigne and Postel (MAP) score systems (17). These so called “universal scores” allow for a good appreciation of the clinical results, since they both rely on criteria that are easily reproducible and show few errors of appreciation between the different authors (which however is only true for the non-modified scores, which, in our opinion, should exclusively be applied). We rely on these two original methods for our own evaluations. The statistical method ● The descriptive statistical method allows the study of a quantitative item with its distribution (percentage, mean value, standard deviation, confidence intervals). ● Deductive statistical method: Using the descriptive bases, this method helps to approach the reality of a collective, such as to confirm or invalidate the relation between 2 parameters with the help of two tests; the chi-square test and the Student t-test. All calculations were made with Excel calculation table.

1. GLOBAL RESULT The following table reveals a significant (arithmetical mean value) global improvement of the functional state of our patients: Mean preoperative MAP value is 10,45 (+/- 0,87) and the postoperative value is 14,70 (+/- 0,9), showing an improvement of 4,25 (p< 0,001); the HARRIS hip score is preoperatively 45,20 (+/- 12,5) and postoperatively 81,04 (+/- 13,3), equaling an improvement of 35,84 points (p<0,001). HARRIS Pain 19,96 40,84

PreOp. PostOp.

Function 28,24 40,19

Total 45,20 81,04

MAP PreOp. PostOp.

Pain 2,74 5,54

Mobility 4,26 4,68

Walking 3,46 4,48

Total 10,45 14,70

138

Fourth part – The results

2. PATIENTS DISTRIBUTION ACCORDING TO THE CRITERIA OF THE MAP SCORE (152 PROSTHESES) Score 0 1 2 3 4 5 6 Average

Pain PreOp. 1 13 33 88 14 1 2 2.74

PostOp. 0 1 0 1 7 48 95 5.54

Mobility PreOp. PostOp. 0 0 0 0 1 2 25 14 75 47 36 57 15 32 4.26 4.68

(+/- 0,89)

(+/- 0,72)

(+/- 0,87)

(+/- 0,95)

Walking PreOp. PostOp. 2 0 0 0 13 8 62 13 62 47 12 66 1 18 3.46 4.48 (+/- 0,88)

(+/- 0,99)

This table shows that major improvements are made for the pain scores. The mobility scores have been preserved but not improved; the same is true concerning the walking and stability scores. (This table is nearly identical to the one published by Boisgard (2) for his series of patients and the scores provided by this author show little difference concerning the pain, are slightly better with regard to the mobility and poorer concerning the walking performance). It is interesting to observe that the 9 patients who complain about severe persistent pain show signs of a cup instability, either with a clear loosening or a probable loosening. This illustrates the fact that a cementless straight stem is not the only culprit for persistent disabling pain.

3. COMPARATIVE STUDY A comparative study of our results with those obtained with other uncemented stem systems, especially with distally interlocked stems (Vives and Picault) (28) or modular stems (Chouteau) (4) did not show significant differences. Implants Pain Mobility Walking

PFM-R 152 cases 5.54 4.68 4.48

Modula 60 cases 4.7 5.1 4.6

Picault 104 cases 4.5 4.7 4.5

NB. The value of this comparative study is doubtful since there is no way to compare the amount of radiological destruction encountered in the 3 patient collectives.

Chapter 2 – Clinical results

139

CLINICAL RESULTS – COMMENTARY Two preliminary comments must be made regarding the following clinical study: ● The criteria that are applied for the functional evaluations (walking capacity – mobility) in hip revision surgery must be more discriminate than those used for the study of primary hip prostheses. In the case of a loosening, the souvenir of a normal functioning hip /limb is much more impaired and these patients have got accustomed to lower levels of mobility and activity. Their most important request is to stop suffering. ● The second comment concerns the methods which include clinical and radiological studies into their global results. This system makes the interpretation of results difficult since there is not always a parallelism between clinical and radiological outcomes; the question may even be raised if it is permitted to present a good clinical outcome with a doubtful radiological result in the same patient or the other way around! Moreover, for some authors, especially Kerboull (14), a loosening which is essentially a radiological diagnosis, does not always mean a clinical failure of the arthroplasty. This is why we decided not to use such methods for this study but to consider the clinical and the radiological outcomes separately. Since the clinical results do not vary much from one author to the other, they will in no way tell about any differences between the several types of prostheses proposed for the uncemented revision of the femoral component. Some comments however can be made about limping and pain.

A. PERSISTENT LIMPING A general remark may be allowed: A persisting limping is a frequent observation after revision surgery and it may even be more frequent when an uncemented component is used. Even if few patients complain spontaneously about this embarrassment, this limping gives rise to some questions. A secondary subsidence resulting in an unequal length of the lower limb is thought to be one of the least satisfying characteristics of uncemented press-fit fixation. This might be true regarding to the 17% of our series presenting an unequal leg length, but it is not the only reason for explaining a limping. If this were true, the distally interlocked stems would give much better results regarding limping. Looking at the published results, this is not the case (28). In reality, the reasons for limping are multiple. On one side, design plays an important role (offset), there is also the fact that we are dealing with secondary surgery and finally in some cases, a strict alignment of both legs is not desired (e.g. in order to avoid the risk of lumbar pain). Persistent limping is somewhat a neglected topic when results of revision surgery are to be analyzed (maybe because patients do not weigh this problem much, once they have become pain free). Moreover, this is a problem which should be better understood (e.g. link between subsidence and limping) and better controlled, especially when an uncemented stem with press-fit fixation is used (we believe that the best way to control any secondary subsidence is to allow for a perfect wedging of the stem in the canal and in some situations, to avoid a too-early full weightbearing). In summary: The absence of a limping corresponds often to an excellent clinical as well as radiological outcome and in general a good long term result can be expected; on the other hand a severe limping is probably the expression of a persistent problem or some other insufficiency, which may impair the long term outcome of this prosthesis. Finally any progress in this field will not only improve the mobility of the revised articulation but also have an impact on the global result of the revision surgery itself.

140

Fourth part – The results

B. MID FEMUR PAIN A persistent mid-femur pain should be considered as a sign of loosening until the contrary has been proven. Before incriminating the press-fit fixation, any abnormal mobility of either the stem or the cup should be excluded, which sometimes proves to be difficult. It is also important to distinguish between immediate postoperative pain, due to further wedging of the implant, and the femoral pain lasting more than 1 year. Only after an abnormal mobility has been excluded, should the press-fit fixation be held responsible for this mid femur pain. In our series, the MAP score is 5.54 points, a good number that is comparable to other published results. In the 9 patients with a score of 4 points or less (1 case is scored 2, another 3 points) the reasons are the following: In one case a femoral loosening, in 5 cases a loosening of the cup, so that the persistence of femoral pain could be associated to a pure stem problem in only 3 patients. Diaphyseal press-fit may be the reason for femoral pain. The number of patients presenting a disabling pain level is however too low to condemn the method, especially when dealing with revision cases and regarding the fact that in literature a MAP score of 4 points is a frequent finding (in our series these are 7 patients, equaling 4,6%). Like it was said for limping, the reasons for persistent femoral pain are multi-factorial (excluding the loosening cases); when a design or concept of a stem is reconsidered, the same attention should be paid to the mechanical strength of the corticals and finally the fact that one is dealing with problems of repetitive surgery.

CHAPTER 3

PREOPERATIVE RADIOLOGICAL RESULTS

The preoperative radiological analysis includes those 4 parameters which we have defined in the chapter on preparation of the operation (see second part, chapter 1): Amount of osteoporosis, loss of bone stock or local bone defects, morphotype of the femur and possible obstacles for cement extraction. At the same time, we also applied both the Paprosky and the SOFCOT classifications of bone defects. It must be underlined that these analyses were made simultaneously by the author and a second surgeon who was not involved with the operations. The following radiological examinations have all been made based on the authors’ series of 152 prostheses, the results of the analysis being expressed in numbers of prostheses and percentage.

1. ANALYSIS OF THE SEVERITY OF OSTEOPOROSIS 152 prostheses Number %

Stage 1 34 22%

Stage 2 66 44%

Stage 3 40 26%

Stage 4 12 8%

The number of cases that were classified in stage 4 is low. Possibly the criteria for this group were too severe and the use of a more objective graduation (e.g. a cortico-medullary index) could have led to a more homogenous distribution between the stages 3 and 4. One third of the patients presented a pronounced osteoporosis and this impressive number must be taken into account during the operation and remembered for the analysis of the results.

2. ANALYSIS OF THE LOSS OF BONE STOCK (DEFECTS) 152 prostheses Number %

Stage 1 48 32%

Stage 2 42 28%

Stage 3 31 20%

Stage 4 31 20%

There is a relatively high percentage of patients in stage 4, which is explained by the fact that the applied criteria for this stage are less severe than those used for other classifications.

142

Fourth part – The results

3. MORPHOTYPES OF FEMUR 152 prostheses

Straight in frontal plane

Curved in frontal plane

slightly curved in sagittal plane

Number

%

70 46%

Curved in sagittal plane straight in frontal plane

70 46%

12 8% 54%

We have made a very meticulous analysis of the femoral morphotype in the frontal plane according to a previously established scheme, which is: a femur in the frontal plane can only be straight or curved. Under these conditions we have analyzed also the slightest deviations in the frontal plane. The deviations in the sagittal plane however were only taken into consideration if these were accentuated and the femur was at the same time straight in the frontal plane. Please note the important amount of curved femora and that these showed in a great amount a varus curving in the frontal plane. Even if we took into consideration some reasons for a possible error of estimation (wrong ap- position due to an external rotation of the hip) this curving of the femora seem to be very frequent and we feel, that it is under esteemed during the preoperative planning in the presence of a loose stem. Remember that the slightest curving of the femur in the frontal plane must be taken into consideration when the revision with a straight stem using press-fit technique is planned. The same is also true for the planning of a primary implantation of any cementless straight stem.

4. THE CEMENT 152 prostheses Number %

Difficulties for cement excision 91 60%

Absence of difficulties for cement excision 61 40%

If the analysis of the cement layer is not restricted to the distal cement plug only, the number of patients presenting such difficulties for the excision of the cement mantle is probably much higher than it is generally accepted in other series (in our series 60 % of all patients). For many surgeons these evident difficulties are a reason enough for the choice of a transfemoral approach, even if there is no other obstacle to the implantation of an uncemented straight stem. This attitude is understandable under the light of the difficulties that one can encounter when trying to excise a thick cement layer out of a weak femoral shaft with thinned corticals and if one always keeps in mind, that the complete excision of the cement mantle is one of the prerogatives for the successful anchoring of a cementless stem.

Chapter 3 – Preoperative radiological results

143

5. SOFCOT CLASSIFICATION 152 prostheses Number %

Stage 1 66 43%

Stage 2 34 22%

Stage 3 38 25%

Stage 4 14 10%

The classification of the defects that we proposed differs from the one of the SOFCOT (Vives, 27).Comparing our results to those of other series using the SOFCOT classification,we see a higher number of patients in stage 4 and an inversion of the percentages between stage 1 and 2. This is explained by the fact that we used to classify the defect in zone 1 and/or 7 in stage 1. The SOFCOT classification seems difficult to understand since, some authors did not classify one single patient in stage 1 (Boisgard) (2).

6. THE PAPROSKY CLASSIFICATION 152 prostheses Number %

Stage 1 44 29%

Stage 2A 58 38%

Stage 2B 6 4%

Stage 2C 15 10%

Stage 3 29 19%

When our results are compared to those of other authors using this classification (Nourissat) (22) and (Picault) (28), we find several important differences which in our view seems to prove that the Paprosky classification lacks reproducible results, as shown by Hamon (12) in his thesis. For our purposes, this type of classification is hardly usable. N.B. The reproducibility of different classifications has been studied by Courpied and Migaud (5) for the SOFCOT symposium in 1999. It is not satisfactory for the majority of the current classifications, the less unsatisfactory of them is still the one of the SOFCOT.

CHAPTER 4

POSTOPERATIVE RADIOLOGICAL RESULTS

INTRODUCTION The postoperative radiological analysis is a decisive step for the appreciation of results of an implant. It is also the most delicate step since the available preoperative classifications are not well suited for this evaluation; on the other hand, the diversity of the evaluative methods proposed for the radiological analysis as well as the incongruent criteria explain the difficulties encountered, when we tried to make a global appreciation of a result and to compare this to the results of other authors. In order to find a solution to this problem we thought that it could be helpful to develop a new method for assessment and appreciation of the radiological results: It consists of establishing an immediate postoperative analysis based on an a-p- X-ray picture which will then be compared to the most distant as possible similar X-ray. This will allow comparing the bony regeneration and the ossintegration or the secondary stabilization. We have added to these 2 items a third parameter to study: The secondary adaptation of all structures which in itself is a synthesis of the two prior analyzes. It will allow for a global appreciation of the overall result and facilitate further long term controls.

PROPOSAL OF A NEW EVALUATIVE METHOD FOR THE POSTOPERATIVE RADIOLOGICAL RESULTS CRITICAL APPRAISAL OF THE EXISTING METHODS Among the methods which have been proposed for the evaluation of postoperative radiological results are the classification of Engh and Massin (8) which is the oldest; in France the “A.R.A.” score proposed by Epinette (9), and more recently the methods of Boisgard (2) or from Picault and Vives (28). In a critical appraisal of these methods, three points must be discussed: ● The methods proposed by Engh and Massin or by Epinette concern primary cementless femoral stems. ● They all base mostly on a single objective which seems insufficient to us for the appreciation of a global result. Boisgard (2) is only interested in the bony regeneration; and Engh and Massin (8) focuses mainly on the osseointegration or secondary stabilization, whereas the “A.R.A.” scores (9) analyze in a general manner the secondary bony adaptation. ● None of these methods references on the initial state of the bone, which makes the analysis of the subsequent results difficult, especially in the field of bone regeneration. The preoperative classifications that are proposed in the literature are difficult to use since they do not rely on an

146

Fourth part – The results

intra-operative assessment of unexpected modifications and none of them refers to an immediate postoperative assessment of the x-ray pictures. ● The interpretation of some evaluative criteria varies from one author to the other (e.g. secondary subsidence), not to speak of unclear defined descriptive items such as “line of condensation” or “reactive lines”. PRINCIPLES OF THE PROPOSED NEW EVALUATIVE METHOD In order to reduce the risk of errors and to make the results easier comparable, we propose the following evaluative method for the postoperative radiological analysis, based on the 4 following principles: ● Establish an initial state (baseline), on which all further appreciations are based. This first appreciation takes immediately place after the operation and allows inclusion of unforeseen patterns during the operation or unexpected results in the first postoperative x-rays (e.g. insufficient reduction of a bone flap). ● Renounce use of unclear defined descriptions such as condensation lines or reactive lines. ● To adapt the criteria to the concept of the chosen implant because the same criteria can have a different significance from one concept to the other. This point seems crucial to us, and is well illustrated by the example of the postoperative migration which is considered by many authors to be a part of the appreciation of the osseointegration. This point is arguable! It is not deniable that any secondary migration of a distally interlocked stem or a cemented stem should be considered as a sign of instability, which is not the case at all for a stem with press-fit fixation: In those stems, a secondary migration even of several millimeters is mostly without any impact on the final outcome (especially not for osseointegration), because we are dealing here with a simple phenomenon of a rewedging of the stem. ● Analyze separately the regeneration of bone stock and the quality of secondary osseointegration. In reality there is no direct correlation between those two items of appreciation: An implant may be perfectly osseo-integrated and stable in the presence of a poor bone quality (stress shielding) and the other way around, an implant may be poorly stabilized in a good bone stock. Complying with these 4 principles, we propose to make an evaluation of the initial state based on the immediate postoperative result, and then to make three separate radiological evaluations: Study of the regeneration of the bone stock; of the osseointegration or secondary stability; and of the secondary adaptation of the implant to the surrounding bone, the least being a synthesis of the two preceding studies. The selection of the criteria for these different studies has been made according to the principles defined hereafter. With exception of the term of spongious transformation of the endosteal bone, they have already been proposed by different authors. The main difference consists in the way they are quantified. Warning. Since 2003, we have improved this method of radiological evaluation while keeping the same principles. Some (minor) modifications have been made and we have mostly exploited the contribution of computer systems.

Chapter 4 – Postoperative radiological results

147

1. EVALUATION OF THE INITIAL STATE This first analysis will establish an initial state that will serve in the future as a reference for the evaluation of the bony regeneration as well of the osteo-integration or the secondary stabilization of the implant. It is made with the help of an a-p- x-ray that is taken immediately or within the first days after the operation (radiological evaluation before discharge) and it will produce a numeric valuation that enters into a global classification according to the severity of the lesions. (The numeric result helps to establish a better hierarchy of the several criteria that make the evaluation possible and thus allow for a more precise classification. The latter will be helpful for later investigations and permit comparative studies). This evaluation concerns on the one hand, the cortical femur and, on the other hand, the endomedullary cavity of the femur. A. THE CORTICAL FEMUR The analysis of the cortical femur is the most important for the global result. On the one side there is an evaluation of the fragility of the corticals by measuring their thickness; on the other side there is the complete loss of cortical substance, hereafter called defects, which are included into the valuation as a weighting factor. ● Fragility of the cortical femur: Every weakening of the cortices is registered irrespective of its origin (granuloma formation, stress shielding, eccentric reaming). This analysis takes the thickness of the cortical bone outside the implant area or of the other femur as a reference, which allows us to use the term of “bone terrain (osteoporosis)”. The extension of these lesions is measured in the zones according to Gruen (11). ● Bone defects: They are registered in a separate manner under the aspect of a weighting factor. This allows for an easier and more precise valuation of the cortical femur. This evaluation is carried out with the immediate postoperative x-ray and it takes into consideration any defect, whatever its causes may be (persisting distance of the flap). B. THE ENDOMEDULLARY CAVITY OF THE FEMUR The valuation of these structures is less important for the global result. Lost endomedullary bone structures, which may correspond to a spongious defect at the metaphyseal level (greater trochanter) and/or a gap between implant and bone at the diaphyseal level are assessed. The whole femur around an implant is analyzed (revision stem) but the endomedullary space outside the zone of primary stability is of spacial interest because, by definition, it is the location of a tight contact between bone and implant. The gaps are measured by zones.

148

Fourth part – The results

EVALUATION CHART FOR THE INITIAL STATE A. CORTICAL FEMUR (FRAGILITY OF THE CORTICALS) 15 12 9 6

● ● ● ●

STAGE 1: no defect or metaphyseal location (zones 1 and/or 7) STAGE 2: proximal femur (zones 2 and/or 6) STAGE 3: diaphyseal femur 1 cortical (zones 2+3 or 6+5) STAGE 4: diaphyseal femur, 2 corticals Stage 1:

Stage 2:

Stage 3: Stage 4:

In presence of an important lesion in calcar region, valuation with a weighting factor is needed. Spongious defects of greater trochanter are evaluated in the endomedullary level. Lesions of the zones 2 and/or 6 are often associated with changes in zones 1 and/or 7. Special case: a lesion in zone 3 is evaluated stage 2 when it may be associated with a lesion in zone 7 (stem tilted in varus position). One cortical must be healthy or only slightly altered; and the defective cortical will be always found in zone 3 or 5. Both cortical are damaged, even of one of both is +/- damaged (do not weight alterations in zones 1 or 7)

CORTICAL DEFECTS (LOSS OF BONE STOCK: WEIGHTING COEFFICIENT) 0 ● STAGE 1: none or only minimal defects. -3 ● STAGE 2: +/- important metaphyseal and/or diaphyseal defects. -6 ● STAGE 3: important metaphyseal defects and/or diaphyseal defects. Cortical bone grafts are not taken into account.

B. ENDOMEDULLARY FEMUR (SPONGIOUS DEFECTS OR GAPS BETWEEN IMPLANT AND BONE) 5 4 2 1

● ● ● ●

STAGE 1: no spongious metaphyseal defects or minimal and no gap bone/implant or < 1 zone. STAGE 2: minor metaphyseal spongious defects and/or a gap bone/implant in 1 zone. STAGE 3: medium metaphyseal spongious defects and/or a gap bone/implant in 2 zones. STAGE 4: extended metaphyseal spongious defects and/or a gap bone/implant in > 2 zones. The gaps between bone-implant in the proximal metaphyseal part are not taken into account. All gaps are noted according to the zones of Gruen. Local lysis of bone = gap bone-implant.

● Very good (18 – 20)

Total: ● Good (14 – 17)

/ 20 ● Sufficient (11 – 13)

● Poor (10 and <)

Chapter 4 – Postoperative radiological results

149

EVALUATION OF THE INITIAL STATE: GLOBAL RESULT The mean arithmetical value of the bone stock immediately postoperatively is 13.44 points. The various groups of patients are classified according to the table below: Values in number of prostheses and percentages (BONE STOCK)

Initial state

Very good (18 to 20 pts)

Good (14 to 17 pts)

Sufficient (11 to 13 pts)

Poor (10 and < pts)

152 prostheses Number %

21 14%

66 43%

31 20%

34 23%

When these results are compared to the valuation of the preoperative defects, it can be noted that the global percentage of very good and good cases (57%) correspond to the defects classified as stage 1 and 2 (60%); in the same way, the percentage of sufficient and poor cases (43%) corresponds to the defects classified as stage 3 and 4 (40%). There is no significant difference (p < 0,3) and one may doubt whether establishing an initial state immediately after the operation is really necessary when there is nearly no difference to the preoperative state. In reality,the global initial state of the patients is however not exactly the same,as one can easily verify in the following table. If the numbers are recalculated according to the defects encountered as registered before operation, then it can be shown that only 76 out of 90 patients having no or only slight prior defects, remain in class Very good or Good after the operation. Moreover, 51 out of the 62 patients classified stages 3 and 4 remain in class Sufficient or Poor. This proves that there is a transfer of patients from one group to the other. This leads us to conclude that the evaluation of loss of bone stock after the operation is more precise than the evaluation before the surgery and that the former evaluation corresponds more to the reality. The same statements cannot be made concerning osteoporosis. This is consistent since we have to take the “bony terrain” into account when making the immediate postoperative appreciation of the bone stock (initial state). Stage 1 (48)

Defects Osteoporosis

Stage 2 (42)

Stage 3 (31)

Stage 4 (31)

Stage 1 (34)

14

1

7

1

4

4

0

3

Stage 2 (66)

20

2

17

6

4

8

0

9

Stage 3 (40)

6

1

8

2

2

6

0

15

Stage 4 (12)

3

1

1

0

1

2

0

4

– Very good and Good results (87) –

– Sufficient and Poor results (65) -

This table shows that 68 patients (44%) have a good pre-disposition for reaching a good final result since they present only few defects and no osteoporosis. On the other hand, the conditions of 30 patients (20%) are poor since they have important defects and advanced osteoporosis as well.

150

Fourth part – The results

2. EVALUATION OF THE BONE REGENERATION This evaluation concerns the regeneration of the bone stock and it does not take into account the quality of the secondary fixation or osseointegration (clear line bone-implant interface). It is carried out with a radiograph in ap-view, compared with the initial state (immediate postoperative period) following the same principles: measured evaluation of the femoral cortex (with a weighting factor for the defects) and of the medullary cavity.

A. THE FEMORAL CORTEX Regeneration of the corticals. This evaluation concerns the cases presenting an alteration of the bone stock in the first place but also the cases without any bone degeneration at the start, which implies to take into account the quality of the corticals evaluating the stress-shielding phenomenon (thickness and density of the corticals, presence or absence of trabeculations). NB. Since it is a comparative evaluation, the reference to the zones of Gruen (11) is of a lesser importance. Bone defects. Their development is evaluated as a weighting factor. It is necessary to take into account the filling or not of the defects. The presence of a secondary lysis is evaluated according to its extent. B. THE MEDULLARY CAVITY The medullary cavity is of minor importance in the global result. Nonetheless, we think that the filling of the gaps in the medullary cavity contributes to globally improve the bone stock. This evaluation concerns the filling or not of the gaps (spongious defects and gaps between bone-implant). On the other hand, and the same way we took the stess-shielding phenomenon into account for the femoral cortex, it is required at this level to evaluate any spongious transformation of the endosteal structures, which is considered in the absence of any stress-shielding.

Chapter 4 – Postoperative radiological results

151

EVALUATION CHART FOR THE BONE REGENERATION A. CORTICAL FEMUR (REGENERATION OF THE CORTICES) 15 12 9 6

● ● ● ●

STAGE 1: complete regeneration or good bone at first and no stress shielding phenomenon. STAGE 2: incomplete regeneration and no stress shielding. STAGE 3: low signs of regeneration or mild stress shielding. STAGE 4: absence of regeneration or worsening or important stress shielding. Mild stress shielding:reduced bone density or thickness Important stress shielding: reduced bone density ands reduced cortical thickness In case of stress shielding and lytic lesions of the corticals, evaluate the stress shielding phenomena on the cortical femur and the lysis in the zones of the cortical defects.

CORTICAL DEFECTS (FILLING OF CORTICAL DEFECTS AND LYTIC LESIONS) 0 ● STAGE 1: Disappearing or important improvement and no or only minimal lytic lesions -3 ● STAGE 2: Incomplete improvement or lysis in less than one zone. -6 ● STAGE 3: persisting or worsening defects or lysis in more than one zone. Lysis of bone analyzed according to the zones of GRUEN

B. ENDOMEDULLARY FEMUR (FILLING OF GAPS BETWEEN IMPLANT AND BONE) 5 ● STAGE 1: No spongious defects or complete filling and spongious transformation of cortex < 25%. 4 ● STAGE 2: Clear improvement but incomplete filling or spongious transformation of 25 to 50%. 2 ● STAGE 3: Slight improvement or spongious transformation of 50 to 75%. 1 ● STAGE 4: No improvement or worsening or spongious transformation of > 75%. The spongious transformation of the endosteal structures is measured in % of the whole implantation area. Do not take in account clear lytic areas. In the case of a stress shielding, do not take in account any spongious transformation of the endosteal structures. Cortical lysis = spongious transformation. In the case of a long stem: Do not take into account the endosteal structures at the level of the isthmus

● Very good (18 – 20)

Total: ● Good (14 – 17)

/ 20 ● Sufficient (11 – 13)

● Poor (10 and <)

152

Fourth part – The results

2.1 REGENERATION OF BONE STOCK: GLOBAL RESULTS The mean value of the bone stock is 13.44 (+/- 4.37) immediately after the operation and 15.34 (+/- 4.48) at the last control, which means an augmentation of 1.90 pts. (p < 0,005). Comparing with the initial state (values in numbers and percentages)

Bone regeneration 152 prostheses

Very good (18 to 20) 1 2

Good (14 to 17) 1 2

Sufficient (11 to 13) 1 2

Poor (10 and <) 1 2

Number

54

21

51

66

27

31

20

34

%

36%

14%

33%

43%

18%

20%

13%

23%

1: regeneration

2: initial value

There is a global improvement of the very good and good results from 57% to 69%. Nevertheless, the understanding of the 13% of bad results seems of great importance to us: They are due to the appearing of a stress shielding phenomenon and/or a secondary osteolysis.

2.2 REGENERATION OF BONE STOCK: CORRELATIONS In order to understand the reasons of the insufficient results or, in a general way, to improve our knowledge about those factors which contribute to a good or bad result regarding the regeneration of the bone stock, we searched for any correlation between the preoperative “bony terrain” and the strategy chosen for the operation. Regarding the “bony terrain”, we also looked for a correlation between the amount of osteoporosis and the extent of the bony defects. For this purpose, we controlled those 30 patients graded as “poor” after the operation (evaluation of the initial state) looking for an insufficient result at the final control. Concerning the strategy of the operation, we searched for a correlation between the chosen femoral approach and the type of primary fixation (which is, in a way, an analysis of the amount of bony regeneration according to the length of the chosen implant). Finally we established a correlation between the bone regeneration and the clinical outcome. N.B. The term of “stress shielding” has been applied to those modifications of the corticals which appeared secondarily to the direct contact of an implant. In its most common aspect this means a slight diminution of the bone density of the femur. Those changes are not only seen in cementless press-fit stems (or other cementless stems) and Figure 1 Figure 2 they seem basically not worrying to us.

Chapter 4 – Postoperative radiological results

153

In its severe form however, the stress-shielding phenomenon is synonymous to an important reduction of the whole bone density which is accompanied by a cortical atrophy (fig. 1) or an absence of reconstruction of the proximal femoral corticals. We restricted the term of “osteolytic lesions” to a local loss of bone stock without additional signs of a stress shielding (fig. 2). BONE REGENERATION AND AMOUNT OF OSTEOPOROSIS Global result (arithmetical mean value and deviations)

Stage of Osteoporosis Stages 1 and 2 100 prostheses Stages 3 and 4 52 prostheses

Initial state (bone stock) 14.4 (+/- 4.1) 11.6 (+/- 4.2)

Bone regeneration 16.7 (+/-3.5) 12.8 (+/- 5.2)

Deviation 2.3 (+/- 4.3) 1.2 (+/- 6.4)

The deviations are an indication for the fact that the presence of osteoporosis does not favour the bone regeneration, which is not surprising (p< 0.1, no statement allowed), but in the reverse, the absence of osteoporosis allows the strong hope for a good regeneration of the bone stock, which is quite encouraging (significative deviation p< 0.00005). The following table gives a confirmation to this previous statement, since 27 of the 52 patients with osteoporosis classified stages 3 and 4 (52%), show sufficient or poor mean bone regeneration, whereas this number is 20 % in the absence of osteoporosis. Bone regeneration correlated to the degree of osteoporosis (amount and number of patients) Bone regeneration Stage of Osteoporosis

Stage 1 (34) Stage 2 (66) Stage 3 (40) Stage 4 (12)

Very good (18 to 20) (54) 16 30 8 0

Good (14 to 17) (51) 12 22 13 4

Sufficient (11 to 13) (27) 4 10 9 4

Poor (10 and <) (20) 2 4 10 4

Osteoporosis is probably a risk factor: 6 of the 8 cases with secondary osteolysis are classified stage 3 and 4, whereas for stress shielding, 17 out of the 24 cases with clear signs of a stress shielding are classified stage 3 and 4.

154

Fourth part – The results

BONE REGENERATION AND DEFECTS Global result (arithmetical mean value and deviations)

Defects 152 prostheses Stages 1 (48) Stages 2 (42) Stages 3 (31) Stages 4 (31)

Initial state (bone stock) 17.2 14.6 12 7.5

Bone regeneration 16.9 15.7 14.6 13.2

Deviation - 0.3 + 1.1 + 2.6 + 5.7

The more severe the defects are, the more the bone regeneration will prove to be. This information is neither a surprise nor real information. However, when the numbers of the mean bone regeneration are compared, a clear and significant deviation (p< 0.01) can be detected between stages 1 and 4 (relative value +3.7 pts). This means that there is an unfavorable condition for bone regeneration whenever severe bone defects and a pronounced osteoporosis are found in combination. The 30 patients who initially showed this pattern will probably be the same with a poor result as regarded to the bone regeneration. BONE REGENERATION AND FEMORAL APPROACH(ES) We use the same two patient groups that were described earlier for the appreciation of the influence that the chosen approaches to the femur have on the bone regeneration. We know however that the strategy of the operation for the patients in group 1 (learning curve) was not clearly defined concerning the femoral approach and the flap, which was non-pediculated in most cases. Global appreciation of the bone stock (arithmetical mean value and deviations)

Bone stock 152 prostheses Group 1 78 prostheses Group 2 74 prostheses

Initial state (bone stock) 13.1 (+/- 4.2) 13.8 (+/- 4.6)

Bone regeneration 14.1 (+/-4,3) 16.6 (+/- 3.2)

Deviation +1 + 2.8

The mean value of the bone stock in the immediate postoperative evaluation is nearly the same for both groups but there is a significant difference (p< 0.005) concerning the bone regeneration in these two patient collectives. We deduce that the principle of choosing a well defined strategy and strictly making only pediculated femoral flaps is beneficial for the bone regeneration. This conviction is underlined by the fact that 6 of the 8 patients who showed a secondary cortical osteolysis were found in group 1 and had therefore devascularized bone flaps, which clearly is detrimental in an osteoporotic femur (a lesson learned in the retrospective view and which is not surprising after all!). BONE REGENERATION AND MODE OF FIXATION (OR LENGTH OF THE IMPLANT) The following table casts light on the influence of the mode of anchoring the prosthesis on the later development of the bone regeneration. Remember that a long diaphyseal fixation always means that a distal piece of 200 mm has been inserted and in a short diaphyseal fixation a distal piece of 140 mm (whenever a proximal or a global fixation was made, the choice usually fell on the short stem).

Chapter 4 – Postoperative radiological results

155

Valuation in number of prostheses

Regeneration Mode of anchoring

Very good and Good (105)

Sufficient and Poor (47)

16 20 37 32 (30%)

6 6 13 22 (47%)

Proximal (22) Global (26) Short diaphyseal (50) Long diaphyseal (54)

22 of the 47 cases which are classified Sufficient and Poor, that is 47% - c.i. 5% (0.47 +/- 0.14) – were operated with a long stem which seems to prove that these implants may hinder bony regeneration. Unfortunately this statement is weakened by the fact that some of these patients were operated with a non-pediculated flap. On the other hand, some 70% of all patients – c.i. 5% (0.70 +/- 0.09) – that were in the group of the Very good and Good results were operated with a short stem. This underlines the fact that short stems are generally better for good bone regeneration compared to long stems. BONE REGENERATION AND CLINICAL RESULTS By comparing the bone regeneration to the clinical results (MAP- score) it can be shown that both patient groups have similar clinical results, but the differences regarding bone regeneration are in a significant manner rather different. Numbers in arithmetical mean values and deviations in points

Bone regeneration/MAP score

Bone regeneration

152 prostheses

Group 1 78 prostheses

Group 2 74 prostheses

Clinical score (MAP, preop./postop.)

13.1/14.1 +1 13.8/16.6 + 2.8

10.1/14.5 + 4.4 10.8/14.8 +4

These findings strengthen our view that the existing radiological/clinical classifications are a poor help for the purpose of a study: There is no correlation between the clinical and radiological results and the importance which is given to the clinical results only contributes to an erroneous interpretation of the global result.

156

Fourth part – The results

2.3 REGENERATION OF BONE STOCK: COMMENTS The evaluation of the bone stock regeneration is an essential step in analyzing the results of an implant, especially when the implant is a cementless stem without additional bone grafting. Many cofactors play a role and specific conditions are required in order to expect regeneration of the bone stock. BONE REGENERATION AND OSTEOPOROSIS An advanced preoperative osteoporosis must be considered not only as a factor which hinders bone stock regeneration but also as a precondition and a risk for later stress shielding phenomena or the apparition of lytic lesions. On the other hand, it can be said that absence of osteoporosis will always allow for very good regeneration, even if important bone lesions were present in the preoperative conditions, provided that the surgeon respects the vascularity of the cortical bone. BONE REGENERATION AND THE SURGERY ITSELF Comparison of the amount of bone stock regeneration in the 2 groups (see the following table) shows a significant improvement in the patients of group 2: A clear diminution of the poor results and at the same time an important rise of the very good results. This probably means that surgical technique is decisive for the later amount of bone stock regeneration, on one side by respecting the vascularity of the bone during the approach (pediculated flap), and on the other side the choice of a short stem (in 2/3 of all patients in group 2) is better for the quality of the bony regeneration. Comparison of the regeneration of the bone stock in the 2 groups (numbers and percentages)

Bone regeneration Global result

Very good (18 to 20 pts) 54 (36%)

Good (14 to 17 pts) 51 (33%)

Sufficient (11 to 13 pts) 27 (18%)

Poor (10 and < pts) 20 (13%)

20 (25%)

28 (36%)

13 (17%)

17 (22%)

34 (46%)

23 (31%)

14 (19%)

3 (4%)

152 prostheses

Group 1 78 prostheses

Group 2 74 prostheses

Ensuring a good regeneration of the bone stock depends on several surgical conditions: ● First condition: Independently of the chosen concept, a good regeneration can only be expected if there is a perfect primary stability of the revision implant.

● Second condition: The vascularity of the corticals must be respected in all situations. We can but insist on the importance of the surgical technique, especially during the crucial step of preparing the approach to the medullary canal (choice of the femoral approach). If a femoral flap has to be raised for this step, the flap must always be peliculated, as described by Wagner (29) and Picault (28). By this procedure, the surgery will be more structured and many obstacles which present during the operation can easily be overcome, without endangering the vascular supply of the cortical bone too much.

Chapter 4 – Postoperative radiological results

157

● Third condition: Always try not to stiffen the femur by carefully selecting the length and the design of the implant.

The longer the implant, the more it will stiffen the femur and we feel that a too frequent use of such implants in revision surgery will not only hinder a good regeneration of the bone stock but will also be a serious factor for a future deterioration of it, especially in the presence of osteoporosis prior to the revision operation. The design of the implant cannot be neglected either. In a first step, the regeneration of the bone stock is linked to biomechanical factors. If it is recognized that a bone must undergo mechanical stresses in order to regenerate, it must also be admitted that every ideal implant should avoid an excessive stiffening of the femoral bone, in order to allow for a load transfer during flexion, essentially under the aspect of traction and compression forces. Similar to many other authors, we have found a faster regeneration on the medial corticals, which is not always the case for the lateral corticals of the femur. This can be explained by the fact that most revision stems are very good in allowing compression forces to act which, according to the laws of WOLF, will result in a faster reconstruction of the medial side. On the contrary, few prostheses permit the transmission of traction forces resulting in a lower reconstruction of the lateral side, especially if the vascularity of the lateral corticals was damaged during the approach. Moreover, the design of a stem has an important effect on the long term outcome of the bone stock due to the same biomechanical reasons. If we disregard the effects of the normal aging of the bone structure, the appearing of an important stress shielding phenomenon is to be considered as a major result of an altered transmission of loading forces and a natural sign of an insufficient adaptation of the stem to the surrounding bone. This can either be the result of a wrong choice of the implant or of a wrong strategy for the surgery. The paradigm for such a condition is the use of a long stem with a large diameter in an osteoporotic femur. Fourth condition: The interest of adding bone grafts. If these ideal conditions for a living bone are respected, we often can see an important, sometimes even spectacular regeneration of the bone stock. But this is not always the case if there was an important bone defect in zones 1 and 2, and we advise to remain cautious in these cases and add some grafts there (preferentially autologous). N.B. For radiological examples, please refer to pages 169 ff. ●

3. EVALUATION OF OSSEOINTEGRATION This chapter deals with the appreciation of the secondary stability of the implant, which is done mainly by X- ray analysis of the interface between bone and implant. Osseointegration is analyzed by comparing the last a-p. X-ray with the first postoperative picture. There are 2 zones of interest: zone A, which corresponds to the area of primary fixation and which can either be proximal or diaphyseal (where there is always a tight contact between bone and implant), and zone B, which corresponds to the femoral areas outside that zone. If there is a proximal as well as a distal (e.g. global) mode of fixation, the analysis describes the proximal femur in zone A and the diaphyseal femur in zone B. The used criteria are the same for both zones, and there will be a numeric result. This will result in a global analysis of the quality of osseointegration, and moreover, allow for an analysis of the changes which have taken place in the zone of primary stability of the stem.

158

Fourth part – The results

Criteria used for evaluation Bone regrowth in the medullary canal, which sometimes looks like a cortical densification of the spongious tissue, or of trabecular bone, which is more frequent. Clear gaps are the expression of a fibrous tissue formation in the space between implant and bone and thus indicate an absence of osseointegration. They are probably a precursory sign of a later loosening, especially if these clear lines are increasing with time. A spongious transformation of the endosteal bone can either be seen alone or together with the appearance of clear gaps. Often found in the absence of a stress shielding, and although its presence is not equivalent to a clear line, a spongious transformation of the endosteal bone is, according to us, the sign of an insufficient osseointegration, or of a change in the fixation mode (which is less worrying). Excluded criteria A distinct transformation of the femoral structures in zone 4, mostly under the aspect of an incomplete pedestal formation. This phenomenon is not taken into consideration for the evaluation of the osseointegration because we think that it is more a sign of an insufficient adaptation of the implant than a sign of instability, except in presence of a bone plug, which is very often an indicator of a loosening. Reactive lines (or increased density liners adjacent to the implant itself). These are in our view the expression of a reaction of the spongious bone that is poorly understood. Secondary migration, which is analyzed separately. Subsidence of some millimeters (5 to 10 mm, or more) is not always worrying, since it can be compatible with a perfect secondary osseointegration of the implant. N.B. There must be a clear distinction of these migrations from a secondary subsidence due to instability of the stem; in these cases the subsidence is often seen together with a complete clear line surrounding the implant.

Chapter 4 – Postoperative radiological results

159

EVALUATION CHART FOR OSSEOINTEGRATION 1. ZONE A: ZONE OF PRIMARY FIXATION, OR PROXIMAL FEMUR IF GLOBAL FIXATION 10 ● STAGE 1: complete endosteal regeneration and clear lines or spongialization of less than 25% and small local bone defects. 6 ● STAGE 2: incomplete endosteal regeneration or clear lines or spongialization of 25 to 50% or local bone defects in one zone. 2 ● STAGE 3: low or no signs of endosteal regeneration or clear lines or spongialization of more than 50% or local bone defects in more than one zone.

2. ZONE B: ZONES OUTSIDE THE AREA OF PRIMARY FIXATION, OR DIAPHYSEAL FEMUR IF GLOBAL FIXATION 10 ● STAGE 1: complete endosteal regeneration and clear lines or spongialization of less than 25% and small local bone defects. 6 ● STAGE 2: incomplete endosteal regeneration or clear lines or spongialization of 25 to 50% or local bone defects in one zone. 2 ● STAGE 3: low or no signs of endosteal regeneration or clear lines or spongialization of more than 50% or local bone defects in more than one zone. – Clear lines or a spongious transformation are measured in % of the given zone. – Cortical defects (cortical lysis) are regarded the same as clear lines. – In the presence of a stress-shielding, only the clear lines are taken into consideration. – A spongious transformation of the endosteal bone is only taken into consideration if there is no stress-shielding. Summarizing table Fixation zone A

Stage 1 (10) None/few clear lines

Stage 2 (6) Clear lines 25-50%

Stage 3 (2) Clear lines > 50%

Stage 1 (10) None/few clear lines

VERY GOOD: 20

GOOD: 16 Few changes +/–

SUFFICIENT: 12 Mixed fixation and + changes

Stage 2 (6) Clear lines 25-50%

GOOD: 16

SUFFICIENT: 12 Mixed fixation

POOR: 8 Fibrous fixation

Stage 3 (2) Clear lines > 50%

SUFFICIENT: 12 Mixed fixation

POOR: 8 Fibrous fixation

BAD: 4 Loosening

Fixation zone B

160

Fourth part – The results

3.1 OSSEOINTEGRATION: GLOBAL RESULTS Number of patients and percentage

Osseointegration 152 prostheses Number %

Very good (20 pts) 37 24%

Good (16 pts) 64 42%

Sufficient (12 pts) 44 29%

Poor (8 and -) 7 4%

A quarter of the patients showed a very good osseointegration, and one third had a sufficient or poor result. The small percentage of very good cases (24%) may be explained by the fact that we took into account the spongious transformation of endosteal bone, too. This is not done by the majority of the other authors. The 5% of bad results show a “fibrous fixation” according to Engh (8) and, apart of the patient that were revised again, they have a poor prognosis concerning their stability. The following table is established according to the proposed method and allows for a better classification of the patients, describing their evolutionary aspects of their secondary stability. Analysis of the osseointegration in numbers of patients (152 prostheses) Fixation zone A

Stage 1 None/few clear lines

Stage 2 Clear lines 25-50%

Stage 3 Clear lines > 50%

Stage 1 None/few clear lines

VERY GOOD: 37

GOOD: 21 Few changes +/–

SUFFICIENT: 9 Mixed fixation and + changes

Stage 2 Clear lines 25-50%

GOOD: 43

SUFFICIENT: 22 Mixed fixation

POOR: 3 Fibrous fixation

Stage 3 Clear lines > 50%

SUFFICIENT: 13 Mixed fixation

POOR: 3 Fibrous fixation

BAD: 1 Loosening

Fixation zone B

It must be underlined that we found a change in the primary zone of anchoring in 30 cases. Mostly, this was a transfer of the fixation to a proximal zone of the femur, a fact which is in our opinion rather securing as this is a sign of a good proximal load transfer. On the other hand, if the predictions of Khalily (16) should come true, a transfer of the proximal to the distal zones of the femur is a sign of insecurity regarding to the future of these implants since in such a situation there is a higher revision rate to be expected.

3.2 OSSEOINTEGRATION: CORRELATIONS In this chapter we looked for those factors which could influence the osseointegration. We remind the reader that all our implants have neither an osteoinductive nor osteconductive coating.

Chapter 4 – Postoperative radiological results

161

OSSEOINTEGRATION AND OSTEOPOROSIS OR BONE DEFECTS Osseointegration (in arithmetical mean values) Osteoporosis Defects

Osseointegration

Stage 1 16.2

16.4

Stage 2 15.9

15.4

Stage 3 14.2

14.4

Stage 4 14.7

14.8

The difference between stage 1 and 4 seems not quite significant (p > 0, 2) with respect to the quality of osseointegration. This is also true for the osteoporosis (+ 1.5 points) and the defects (+1.6 points). It can be deduced that these two factors have no important impact on the quality of the osseointegration. OSSEOINTEGRATION AND THE SURGERY The following table aims to analyze the quality of osseointegration in the 2 groups of patients defined earlier. Remember that there are quite important differences in those two groups as to the type of the selected implants and the type of approach to the femur as well. Looking at the results in the two groups, there is a clear augmentation of very good and a distinct diminution of sufficient and poor results in the patients of group 2. Considering the fact that in 84% of the patients in group 2, a pediculated flap had been made, and that two thirds of those patients had a short stem implanted, we can consider this technique as highly beneficial for the osseintegration as well as for the bone regeneration. Numbers and percentages of prostheses

Osseointegration

Very good

Good

Sufficient

Poor

152 prostheses

(20 pts)

(16 pts)

(12 pts)

(8 pts and <)

Group 1

13

17%

34

44%

26

33%

5

6%

24

32%

30

41%

18

24%

2

3%

74 prostheses

Group 2 74 prostheses

OSSEOINTEGRATION AND SECONDARY MIGRATION OF THE STEM Though we had excluded this criterion from our evaluation of the osseointegration, there was an interest to analyze the phenomenon of secondary migration for two reasons: On one hand in order to be able to show our results with the press-fit technique (since subsidence is the point where critics of the method always start) and on the other hand to justify our decision, not to take this criterion into consideration for the analysis of osseointegration or secondary stabilization. N.B. The amount of subsidence was measured according to the method proposed by Boisgard (2). There, the difference between the distances of two parallel lines is calculated: The first line being at the level of the shoulder of the prosthesis (easy to be shown), the second line on the level of a distinct morphological point of the femoral bone or often a cerclage wire on the flap. We remind the reader that there must be an identical magnification of the pictures and that the difference must be greater than 5 mm between the two measurements in order to give significant result.

162

Fourth part – The results

GLOBAL RESULT (Number of prostheses and percentages)

Subsidence

Global series

Group 1

152 prostheses

78 prostheses

74 prostheses

0 to 5 mm 5 to 20 mm > 20 mm

127 (84%) 23 (15%) 2 (1%)

65 (83%) 12 (16%) 1 (1%)

Group 2 62 (84%) 11 (15%) 1 (1%)

Both patients having a subsidence superior to 20 mm were revised. One prosthesis was loose, in the other case there was a loosening followed by a secondary re-stabilization of the distal component so that only a longer proximal component was fixed on that stem (in addition to these 2 cases of a subsidence superior to 20 mm, we have to mention 2 other cases which arose among our first patients, these 2 cases were revised with a different method and have not been included in this series because they were lost to follow-up). Otherwise there was no difference between the two groups which shows, that the analysis of the secondary migration is of relative interest, except if the is a clear instability. We could see that the results are generally better for the patients in group 2, regarding the bone regeneration as well as the osseointegration. This hypothesis is confirmed by the following table which analyzes the link between secondary subsidence and the quality of the osseointegration: The percentage of patients with a significant subsidence (5 to 20 mm) is nearly the same in the patients with either very good, good (14.8%) on one side, and those with sufficient or poor results (15,6%) on the other side. Number of prostheses and percentages (152 prostheses)

Osseointegration Subsidence 0 to 5 mm 5 to 20 mm > 20 mm

Very good

Good

Sufficient

Poor

(20 pts)

(16 pts)

(12 pts)

(8 pts and <)

30 6 1

55 9 0

36 7 1

6 1 0

Chapter 4 – Postoperative radiological results

163

3.3 OSSEOINTEGRATION: SUMMARY AND RADIOLOGICAL EXAMPLES In contrary to its role regarding the bone regeneration, osteoporosis does not play an important role in osseo-integration. The most important factor that influences the quality of osseointegration seems to be the surgeon’s choice of surgery and the type of implant he selected: A short stem, even if its diameter is large, will be better osseointegrated if the vascular integrity of the femoral bone is preserved. But this is not the only factor to be taken into consideration: In certain cases the necessity of any additional bone transplant or of bone substitutes must be questioned. They can be beneficial and sometimes improve the prognosis of the long time survival since a perfectly osseointegrated implant will always show a better outcome. The methods proposed for the analysis of osseointegration need a more precise definition of the criteria used, as far as the real meaning of reactive lines or the spongious transformation of endosteal structures are concerned, especially if these phenomena are described outside of a stress-shielding. Female, 69 y. Revision of a THR on the right side, with short stem and pediculated femoral flap. X-ray picture immediately after operation and at 4 yrs. Bone stock is preserved, no stress-shielding. Secondary subsidence > 10 mm, which has not hindered a good secondary osseointegration (clear line in zone 1).

For more radiological examples please refer to pages 169 ff.

164

Fourth part – The results

A

B

A

B

Male, 72 y. Loose right THR. Curved femur in the frontal view, important weakening of lateral cortical structures. Fractured lateral flap, with preserved vascularity. Long stem with diaphyseal fixation. Result at 3 yrs: Good bony regeneration, especially of the lateral cortical bone and filling of defects. The analysis of osseointegration and secondary stabilization shows a proximal migration of the site of anchoring with a perfect proximal osseointegration and a spongious transformation (A) of the endosteal cortex in the diaphyseal region (B).

Female, 73 y. Loose right THR. Curved femur in the presence of medium osteoporosis. Non-pediculated (devascularized) femoral flap. Diaphyseal fixation of a long stem with large diameter. Result at 4 yrs: Medium stressshielding, atrophy of lateral cortical bone, sufficient osseo-integration: Endosteal spongious transformation of diaphyseal areas and gap formation proximally.

NB. A devascularized flap and a too big implant are the clear reasons for that result. This is the result of an erroneous choice of the strategy: Regarding the curving of the femur, a trochanteric osteotomy would have been better in such a case, with the aim of a proximal fixation, sometimes augmented with bone transplants and secured by cerclage wiring.

Chapter 4 – Postoperative radiological results

165

4. APPRECIATION OF THE SECONDARY ADAPTIVE TRANSFORMATION OF BONE This appreciation is a synthesis of the two previous evaluations, and will thus, also reflect the global results. The bone regeneration is evaluated on the level of the cortical femur by analyzing the transitional zone (zone 4 of Gruen) as a weighting factor. Osseointegration (or the secondary stability) is analyzed in the endofemoral bone. A. THE CORTICAL FEMUR The same criteria are used for the appreciation of the cortical femur as were applied for the analysis of the bone regeneration, but this time by integrating the evolution of the defects and cortical lytic zones (which themselves are evaluated as a weighting factor for the bone regeneration). Moreover, we also try to give a more important place to stress-shielding phenomena or adaptive bone remodeling. The transitional zone (zone 4 of Gruen) is analyzed as a weighting factor. This particular zone is of special interest when a secondary adaptive transformation is to be analyzed, since any abnormal load transfer will, in the long term, result in a visible modification of the bone structures in this area. Besides looking for the absence or the presence of a pedestral formation, the cortical structure also has to be evaluated for thickness or density. B. THE ENDOFEMORAL BONE STRUCTURES Clear gaps and endosteal bone formation are measured, since they are representative for the quality of the secondary stability of the implant. In the same way, we think that any spongious transformation of endosteal bone should be taken into consideration. We would stress the point that this criterion probably does not have a negative meaning since it is the expression of an ongoing transfer of the area of secondary stability from the diaphyseal to the proximal areas of the femur.

166

Fourth part – The results

EVALUATION CHART FOR THE SECONDARY (ADAPTIVE) TRANSFORMATION A. CORTICAL FEMUR (REGENERATION OF THE CORTICALS, STRESS-SHIELDING OR CORTICAL LYSIS) 15 ● STAGE 1: Complete regeneration or good bone at beginning and no stress shielding and no lysis. 12 ● STAGE 2: Incomplete regeneration or minimal lysis or minor stress shielding. 9 ● STAGE 3: Low signs of regeneration or lysis < = 1 zone or mild stress shielding. 6 ● STAGE 4: Absence of regeneration or worsening or lysis > 1 zone or important stress shielding. – Stress shielding ±: Slightly reduced cortical density – Stress shielding +: Reduced bone density or reduced cortical thickness – Stress shielding ++: Important reduced bone density and/or cortical thickness

Transitional zone or zone 4 (weighting coefficient) 0 ● STAGE 1: No changes of bone or slight pedestral formation -3 ● STAGE 2: Medium pedestral formation, and no cortical changes or increased cortical density. -6 ● STAGE 3: Large pedestral formation or important thickening of corticals. Pedestral formation to be analyzed on frontal and, if necessary, on axial X-ray pictures.

B. ENDOMEDULLARY FEMUR (FILLING OF GAPS BETWEEN IMPLANT AND BONE, CLEAR GAPS OR SPONGIOUS TRANSFORMATION OF ENDOSTEAL BONE) 5 4 2 1

● ● ● ●

STAGE 1: Complete endosteal filling and if clear gaps or spongious transformation < 25%. STAGE 2: Incomplete filling or clear gaps or spongious transformation of 25 to 50%. STAGE 3: Minimal endosteal filling or clear gaps or spongious transformation of 50 to 75%. STAGE 4: No endosteal filling or clear gaps or spongious transformation of > 75%. – The spongious transformation of the endosteal structures is measured in % of the whole implantation area. – If cortical lysis = clear gap. – In the case of a stress-shielding, only measure extension of clear gaps.

● Very good (18 – 20)

Total: ● Good (14 – 17)

/ 20 ● Sufficient (11 – 13)

● Poor (10 and <)

Chapter 4 – Postoperative radiological results

167

4.1 ADAPTIVE TRANSFORMATION: RESULTS In the whole series, 3/4 of the patients (77%) are classified very good or good and 1/4 (23%) show a sufficient or poor result. GLOBAL RESULT AND DISTRIBUTION ACCORDING TO THE 2 GROUPS OF PATIENTS Number of patients and percentages

Adaptation Global result

Very good (18 to 20 pts) 57 (38%)

Good (14 to 17 pts) 59 (39%)

Sufficient (11 to 13 pts) 22 (14%)

Poor (10 and less) 14 (9%)

22 (28%)

30 (38%)

12 (15%)

14 (18%)

35 (47%)

29 (39%)

10 (14%)

0

152 prostheses

Group 1 78 prostheses

Group 2 74 prostheses

Comparing the results of the 2 groups, the same observations can be made as for the bone regeneration, which is not surprising since there are only slight differences in the criteria used (with exception for the weighting coefficient, which also explains the absence of a poor result for patients in group 2). This evaluation confirms the presence of a significant difference (p < 0,01) between the two groups of patients. The results in group 2 are better: A clear increase of very good and good results, going from 28% to 47% now and, in the same manner, the poor results have disappeared in that group 2. The conclusions from the study of the adaptive transformation are the same as for the bone regeneration.

4.2 ANALYSIS OF THE TRANSITIONAL ZONE 4 We made a comparison between the several modes of fixation and the results of the analysis of zone 4. Values in number of prostheses Fixationmode Fixation zone A Fixation zone B Analysis zone 4

Proximal

Global

Short diaphyseal

Long diaphyseal

(22 cases)

(22 cases)

(50 cases)

(54 cases)

17

23

41

51

Stage 2 pedestral formation +

4

3

8

2

Stage 3 pedestral formation ++ and modified corticals

1

1

1

Stage 1 None or pedestral formation +/-

168

Fourth part – The results

This table shows a great number of patients classified stage 1, which leads us to conclude that this analysis should be made in a more precise way, distinguishing between patients that show an unchanged bone and those, who present even a very slight pedestral formation. The most frequently seen alteration is an incomplete pedestral formation, mostly under the aspect of a lateral or anterior bone plug. It is always a sign of some conflict between the distal tip of the prosthesis and the inner cortical bone. In a few cases, this conflict is the consequence of a varus position of the stem, more frequent is a frontal curving of the femur in the varus sense. Why will you ask, is the knowledge about the low frequency of a bone plug formation of any interest when a long stem is implanted? It is so because when a femoral flap is used for the approach, this diminishes the risk for a conflict at the distal tip. Please note the low number of cases classified as stage 3. It signifies a nearly total absence of cortical transformations in transitional zone, this despite the fact that a short fixation with a very tight contact between bone and implant over a very short distance has been made. Moreover, a thickening of the cortices at the level of the fixation was seen in only 2 cases. Hence, the significance of any cortical thickening can be questioned and it seems that Epinette (9) might be right when he says that any thickening of the cortical structures announces an increasing loss of stability, especially if this thickening occurs in a global manner (circular) or only in zones 3 and 5 of Gruen (11). An absence of transformation in zone 4 is to be considered as a good sign and means that the area is “calm”, even if a secondary subsidence with re-stabilization has taken place before. With exception of a bone plug formation (thick pedestral formation), any console formation is the sign of an adaptive process and not of a progressive loosening. We feel that this study of the transitional zone is of some interest since it reflects the behaviors of the bone-implant coupling; if this zone is “calm”, the future of the prosthesis is good in the long term. Finally, the chart for the evaluation of the secondary transformation can be used in two different ways: To allow for a long-term follow-up of an implant. By comparing the results obtained from the analysis of the X-rays in regular intervals (every year or every 2nd year), an evaluation of the ability of any implant to conserve the bone stock in the long term is possible (and by this, also the quality of a design). Primarily, we did not use the chart for this purpose and this is why we cannot (yet) give a result for those patients who have not undergone a sufficient follow-up. This study has to be done in the future and it will be facilitated if the patients are seen at regular intervals. To make an instant evaluation at any defined moment. In this situation the chart will not give more information than those already obtained in the two prior evaluations, but it will allow for a more precise analysis of the actual global state of the revision hip.

Chapter 4 – Postoperative radiological results

169

RADIOLOGICAL EXAMPLES The very good results (18 to 20 pts.) These patients show no or only slight signs of osteoporosis, and the bony defects are sometimes rather important. They were operated with a short stem and the surgeon respected the vascular supply of the cortical femur. There is a good regeneration of the bone stock, which is maintained over the time, there is also an excellent osseointegration and in the most cases no visible conflict at the distal tip of the stem.

Male, 62 y. Right THR in the presence of a DDH in 1986 (12 yrs.). Loosening with appearance of fragile medial corticals; large cement plug. Vascularized femoral flap; short diaphyseal fixation. Result at 2 yrs.: Very good bone regeneration and excellent osseo-integration.

Male, 52 y. Cementless right THR 1990 (10 yrs.). Very important bone lesions. Vascularized femoral flap; short diaphyseal fixation, no bone transplant. Result at 3 yrs: Very good bone regeneration (minimal lateral cortical defect).

170

Fourth part – The results

The good results (14 to 17 pts.) These patients fulfill the same criteria as the former group of the very good results. The difference lies in some (not alarming) details, be it an incomplete osseointegration or regeneration of the bone stock.

Male, 75 y. Loosening with severe varus deformity, long bone defects, cement plug. Operation through a femoral flap, long stem and no bone grafting. Additional osteotomy of the medial corticals, in order to bring the femur back into close contact with the implant. Result at 5 yrs: Very good bone remodeling but incomplete osseo-integration in the area of the proximal femur. Beware: This mode of fixation can lead to an excessive loading of the intermediate zone of the stem (risk of stem breakage with small diameters). In this case, a shorter but thicker implant would have been a better solution.

Male, 76 y. Right THR in 1984 (12 yrs.). Granuloma formation in medial cortex, cement plug, straight femur. Endofemoral approach with additional femoral window for the excision of the cement plug. Proximal fixation mode. Result at 3 yrs. 8 mths: Good osseointegration but incomplete regeneration (due to insufficient curettage of the cystic lesions?)

Chapter 4 – Postoperative radiological results

171

The sufficient results (11 to 13 pts.) In some cases there is a stress-shielding with an important reduction of bone density, but without real atrophy of the cortical bone, often in patients presenting the signs of an important general osteoporosis. In others, there is an incomplete regeneration of the bone stock or osteolytic lesions which extend over not more than one zone (and which is often the result of an insufficient surgical technique). The abstention of bone grafting can predispose to failure in those patients.

Female, 75 y. Left THR in 1977 (19 yrs.). Loosening without bone defects, but osteoporosis. Endofemoral approach, metaphyseo- diaphyseal fixation mode, no distal fixation. Intramedullar grafting with HAP chips. Result at 4 yrs: Good osseointegration but clear presence of a stress shielding. Decreased density of the cortical bones but (at the moment) nearly no change of the cortical thickness.

Female, 38 y. Revision for a loose cup. Stem revision due to the wish for a hard-hard coupling related to the young age of the patient, but also for the varus implantation. Vascularized flap, for a varus curved femur and long distal cement plug, short diaphyseal fixation. Result at 3 yrs: No stress-shielding, but cortical lysis in zone 1. N.B. If there is a weakening of the greater trochanter during cement excision, a bone grafting is advisable in the proximal area.

172

Fourth part – The results

The poor results (10 and less pts.) are associated with an extended loss of bone stock: They show secondary osteolysis or a heavy stress-shielding. They were always related to an erroneous surgical technique (devascularized flap formation) or a wrong indication (long stem in an osteoporotic femur). The worst results could be seen when both errors were added!

Female, 70 y. Secondary loosening of a THR using a cemented stem made out of a Titanium alloy. Diffuse granuloma formation, straight femur, medium osteoporosis. Devascularized femoral flap, long stem, no bone grafting. Result at 3 yrs: Massive lysis of lateral corticals. N.B. In such a case the non-observation of the vascular supply to the lateral flap was particularly detrimental to the femur and it would have been preferable to make an endofemoral approach, using a short stem with additional bone-grafting.(alternatively, a lateral flap with a short stem, diaphyseal fixation, plus additional bone grafting can be discussed).

Female, 78 y. Secondary loosening of a cemented Ti- stem. Endofemoral approach, but change of surgical strategy due to a perforation with secondary fracture of the femur during cement excision. Devascularized femoral flap, long stem with large diameter, no bone grafting: Poor result at 3 yrs. 7 mths. Important stress-shielding and lysis of lateral cortex. N.B. To change the strategy during the operation is a risk for the vascular supply of the proximal femur, especially in an osteoporotic bone.

CHAPTER 5

THE COMPLICATIONS, LIMITATIONS AND INDICATIONS OF THE “PRESS-FIT” CONCEPT

There is no operation without any complication (this is particularly true for revision surgery). Every concept has its own limitations which must be known to the surgeon using the implant only for those circumstances for which it has been designed and that good results can be expected here if the basic techniques are applied.

1. COMPLICATIONS We describe the complications in the series of 152 prostheses which make the contents of this publication. However we will also refer to complications that occurred in the 28 patients excluded from this series for various reasons (totalling 180 patients).

1.1 PERIOPERATIVE COMPLICATIONS These complications mostly occurred during the initial learning curve. We will only discuss those complications that influenced the further surgery or the postoperative sequelae. We have excluded from this survey any general complications inherent to surgery itself (e.g. haematoma, phlebitis and so on). Femoral fractures: 4 cases. Three of these cases needed a change in the strategy concerning the choice of the implant and resulted in a longer stem than originally planned. One patient needed an additional internal plate fixation. All 4 fractures healed uneventfully and without influencing the final result of the surgery. Incomplete excision of bone cement: 6 cases. 5 cases occurred during the initial learning curve when a femoral flap was not routinely done. In 4 cases there were cement remnants on the medial side of the medullary canal which leaded to a malposition of the stem; in 2 of these cases the cement remnants were the cause for a subsequent eccentric reaming and hence weakening of the opposite cortical, resulting in a varus implantation of an under-dimensioned stem. Cortical perforation (via Falsa): 7 cases. All these cases occurred when a femoral flap was not performed. This complication sometimes resulted in longer implants than originally planned. Neurological complications: 2 cases. One temporary crural nerve injury left no sequelae after 18 months and one sciatic nerve traction injury evolved to a good final function.

174

Fourth part – The results

1.2 IMMEDIATE OR EARLY POSTOPERATIVE COMPLICATIONS INFECTIONS: 4 CASES 2 patients showed an early septic loosening. One patient was revised with a cemented implant; the other patient was revised with the presented technique but refused further controls. Both patients were excluded from the series. In 2 other cases, the implant was left in place after a debridement of the hip, since the implant was stable. Both cases evolved positively regarding the infection, but one patient developed later a bone lesion on the lateral cortex (which was then treated with bone-grafting, that unfortunately was not successful). The only infection that occurred in our series was revised in a one-stage surgery and healed completely. DISLOCATIONS: 8 CASES (5%) Simple closed reduction was sufficient in 6 cases; in 2 patients a gluteal sling reconstruction was necessary. In our view, this incidence of dislocations appears too large. The utilisation of larger head diameters or possibly double cup heads may help to reduce this complication which can prove to be difficult to treat successfully. COMPLICATIONS RELATED TO THE GREATER TROCHANTER: 8 CASES 6 patients had a non-union of a classical trochanteric osteotomy.Nevertheless 4 of these cases already had a pre-existing pseudarthrosis prior to the revision surgery. In 2 cases the weakened trochanter fractured at its base at the level of the circular cerclage wiring. This complication can be avoided by respecting the insertions of the vastus lateralis muscle and by making an additional bone grafting if the greater trochanter is weakened after the excision of cement or granuloma formations, and last but not least by making a stable lateral tension band wiring in order to protect the reconstruction. 1.3 LATE COMPLICATIONS We will only speak about the revisions of the femoral stem, since the other secondary complications (e.g. stress-shielding or osteolysis) have been discussed earlier. REVISION OF THE FEMORAL STEM DUE TO LOOSENING: 5 CASES (OF 180 REVISION CASES) The 2 cases of septic failure have already been analyzed in the previous text. They both consist in an early septic loosening: a female patient was revised with a cemented stem, and the male patient was revised with another method and lost to follow-up. Two patients presented with an early instability that occurred within the first month: One major subsidence had to be revised with a cemented femoral stem, another case showed a rotational instability of the stem with an irreducible dislocation of the hip in the first days after revision surgery and it had to be revised with a longer stem of the same type). In reality, all these 4 cases were due to primary instability of the stem. Only one late instability of a stem occurred at 3 years after implantation. The first revision surgery had been complicated by a femoral perforation. The second revision was made with the same type of prosthesis, resulting ultimately in a good clinical outcome and an average radiological picture. Finally, there was one case of secondary subsidence of more than 20 mm, which re-stabilized spontaneously and which was managed by exchanging only the proximal component with a longer implant.

Chapter 5 – The complications, limitations and indications of the “Press-Fit” concept

175

2. LIMITATIONS OF THE PRESS-FIT CONCEPT What are the circumstances which make the choice of a press-fit stem impossible or detrimental for the future of the bone by causing a stress-shield phenomenon? 2.1 PRIMARY STABILITY CANNOT BE ACHIEVED This case can often be seen after loosening of a long revision stem with extended osteolytic lesions, or a fracture around a long implant. The common pattern is a destruction of the isthmic region of the femur which makes the press-fit concept almost inapplicable or applicable only with difficulties. We advise prior planning in such situations so as to be able to switch to another type of prosthesis having a different fixation concept, e.g. a stem with distal interlocking bolts is especially suitable for such cases if the distal cortices are stable enough. 2.2 RISK FACTORS FOR STRESS-SHIELDING PHENOMENON We have already drawn the attention of the reader to the risk of severe stress-shielding when a long stem with a large diameter is implanted into an osteoporotic femur (thin cortices and at the same time a wide medullary canal resulting in a cylindrical shape of the femur that proves difficult to be “conizised” by the reamers). We urge the reader to be cautious with such cases. We are of the opinion that these cases are the real contraindications to the press-fit concept unless the possibility of making a proximal fixation of the implant can be ensured (and a diaphyseal fixation can be avoided), which also means to preserve the vascular supply as much as possible. In the situation of a wide cylindrical femur with thin cortices, alternative techniques as impaction grafting, Exeter type or the double-sleeve technique of Kerboull (14) may find their place.

3. INDICATIONS FOR THE PRESS-FIT CONCEPT In the absence of advanced osteoporosis, even a significant weakening of the proximal cortices by extended bony lesions is not a contraindication to the impaction of a straight stem with press-fit technique. Only a stable primary fixation by wedging the stem in the isthmic zone must be guaranteed in any case. Generally speaking, we can distinguish two groups of patients:

3.1 STRAIGHT FEMUR WITH NO OR ONLY SLIGHT DEFECTS The preferred technique is an endofemoral approach (sometimes with a distal femoral window) in these situations. The primary stability by means of the press-fit technique takes place either in the metaphyseo-diaphyseal junction (only small proximal defects) or in the proximal diaphyseal region. A short stem can be usually implanted (sometimes augmented with bone grafting for better primary stabilization) and the choice of the press-fit concept will not hinder the regeneration or maintenance of bone stock.

176

Fourth part – The results

3.2 CURVED FEMUR, MAJOR DEFECTS OR DIFFICULT BONE CEMENT EXCISION In all other cases, the cutting of a vascularized (pediculated) femoral flap is the choice. This makes the further evolution of the operation easier. The primary stability in these cases will take place only in the diaphyseal area; mostly a short stem is utilized. Nevertheless, if necessitated, the use of a longer stem is without any consequences for bone regeneration or preservation of bone stock if the surgeon has preserved the blood supply to the femur and especially to the flap and has thought to avoid using this technique in cases of severe osteoporosis (stage 4 of our classification).

CONCLUSIONS Choosing a concept requires awareness of its inherent needs and selection of the appropriate implant; make a precise planning and preparation of the surgery and finally, adapt the surgical technique to the needs and constraints of the chosen method, these are the principles which have to be respected by any surgeon who wants to revise a loose femoral stem. ● The concept of press-fit is based on the direct contact between bone and implant in the form of a surface and by a perfect wedging-in of the conical stem in a medullary cavity which has been remodeled in an equally conical shape. In our opinion, the straight stem is the safest manner of creating a contact surface, but it calls for a straight femur, too. If this basic principle is not given, then a femoral osteotomy is mandatory. This is a measure which is not an inconvenience by itself as a pediculated flap bears many advantages for a fast and safe revision operation. The perfect wedging-in of the stem can be guaranteed by two measures: working with reamers having a good insight on the zone of contact and choosing a conical stem. One must exploit the conical area of the implant and a reserve of conical surface during the impaction of the stem must always be kept. Both of these principles are easy to realize if a femoral flap is made at the right time (diaphyseal fixation) and if finally, the choice of a modular system has been made. The implant must be perfectly adapted to the chosen goal of a press-fit fixation and the concept of its realization. We underline the choice of an implant which avoids an unnecessary stiffening of the femur. ● Making a detailed preparation of the operation is the basic step for successful revision surgery; the surgeon must have the right X-rays available, and he must look for a sufficient technical quality of these pictures (this is very often not the case). During the stage of meticulous planning, there should be a systematic analysis of all the steps and of the obstacles that could jeopardize the result, especially taking into account any curving of the femur in the frontal plane (often found in cases with loose prostheses). With a press-fit stem, determining a strategy, which is simple in the end, consists in selecting the appropriate approach for the later fixation mode and adapting it to the anatomical situation and to the needs of the chosen concept. In daily practical use it usually means making a femoral flap and using an endofemoral approach only with a straight configuration of the femur. Finally, a technical drawing will help to establish the length of the femoral flap and the location of the centre of rotation, but not the final size of the implant. The sizing is only established during the operation itself. REGARDING THE SURGICAL TECHNIQUE We took care to show the technical details of every stage of the operation since we believe that the quality of the results relies in a great manner on the quality of the surgical technique itself and on the observation of the following two principles: respecting the vascular supply of the cortical bone especially when a femoral flap is cut (a cortical flap must be pediculated with a muscle attached on it) knowing that dead bone cannot undergo regeneration, and preferring the diameter to the length of the implant by choosing the shortest and thickest implant whenever possible. METHODS FOR THE EVALUATION OF THE RESULTS The best way for the validation of an implant or a method is the analysis of the X-rays of the patients operated with this method. At the end of this book, allow us some remarks concerning

178

Conclusions

the methodology that has been used in our study: It seems to us inevitable to make a clear distinction between the analysis of preoperative and postoperative X-ray pictures. The main merit of the herein proposed method for the analysis of the postoperative X-rays is the one that, for the first time, it is clearly included into the evaluation process. This step must be validated (and improved) by a larger clinical study and its interobserver /intraobserver reliability should be measured. Even if the majority of the principles used for our method can be applied to a majority of revision implants on the market, it is clear that some of the criteria used by us have to undergo some modifications depending on the concept of these other prostheses; this is true not only for the analysis of the preoperative X-rays, but also for the implants themselves. Finally we think that a plain frontal X-ray picture is no longer sufficient for a postoperative radiological evaluation and that bone densitometry or D.E.X.A. as described by Nehme and Puget (21) in one of their latest publications will become an indispensable tool for the appreciation of the bone around an implant. ABOUT THE RESULTS The short and medium term results of the 152 prostheses presented in this series seem globally satisfying. Since any comparison of our results to those published by other authors is nearly impossible, it cannot be said if they are better or worse. A longer observation interval is needed in order to make any distinction possible. To us it seemed important to recognize that the good quality of the results (especially in the group 2) occurred after a long learning curve: The poor results were progressively eliminated and replaced by an increasing number of very good results. If the reading of this book helps surgeons not to repeat the same mistakes and pitfalls made by us, then the great amount of work undertaken for this writing will not have been unnecessary.

BIBLIOGRAPHY

1.Blaimont,P,Halleux P,Jedwab J.,Distribution des contraintes osseuses dans le fémur, Rev. Chir. Orthop., 1968, 54 : 303-319. 2.Boisgard S,Moreau PE,Tixier H,Levai JP.,Reconstruction osseuse, inégalité de longueur des membres, taux de luxations de 52 prothèses de Wagner, en révisions d’arthroplasties totales de hanche au recul moyen de 44 mois,Rev.Chir.Orthop.,2001,87 :147154. 3. ousquet G, Argenson JN et al., Reprises après descellements aseptiques des arthroplasties totales de hanche par la prothèse sans ciment de Bousquet, Rev. Chir. Orthop., 1985, 72 : 70-74. 4. Chouteau J, Vaz G, Chavanne H, Tayot O, Carret JP., Reprise fémorale âr prothèse modulaire non cimentée. Analyse rétrospective d’une série de 60 patients, Communication, journées Lyonnaises de chirurgie de la hanche, 1999, 243-247. 5. Courpied JP, Migaud H., Reprise fémorale dans les arthroplasties itératives aseptiques de la hanche, Symposium S.O.F.C.O.T., Rev. Chir. Orthop., 2000, 86 (Suppl. 1), : 39-90 6. D’Antonio J, MC Carthy J, Bargar WL, et al., Classification of abnormalities in total hip arthroplasty, Clin. Orthop., 1993, 296 : 133-139 7. David Blaha J, Sportorno L, Romagnoli S., CLS press-fit total hip arthroplastie, Techn. Orthop., 1991, 6 (3) : 80-66. 8. Engh CA, Massin P, Suhers KE., Roentgenographic assessment of the biologic fixation of porous surfaced femoral component, Clin. Orthop., 1990, 257 : 107-128. 9. Epinette JA, Geesink R, et le grope AGORA, Proposition d’un nouveau système d’évaluation radiologique des prothèses fémorales non cimentées : le score ARA, Cahiers d’enseignement de la S.O.F.C.O.T., 1994, 50 : 107-210. 10.Essig J.,Puget J.,et al.,Résultats à 5 ans d’une série hoùogène de 74 prothèses de reprise (système PP). Analyse de la stabilité mécanique de l’implant et de la reconstruction osseuse. 11. Gruen TA, MC Neice GM, Amstutz HC., Mode of failure of cemented stem type femoral components. A radiographic analysis of loosening, Clin.

Orthop., 1979, 141 : 17-27. 12. Hamon JM., Intérêts d’une nouvelle méthode radiologique dans la prise en charge thérapeutique des descellements aseptiques fémoraux des arthroplasties totales de hanche.À propos de 180 cas,thèse, Rennes, oct. 2002. 13. Harris WH., Traumatic arthritis of the hip after dislocation and acetabular fracture. Treatment by mold arthroplasty. An end result study using new method of resukt evaluation, J. Bone Joint Surg., 1969, 51A, 737-754. 14. Kerboull M., Traitement des descellements fémoraux aseptiques des prothèses totales de hanche, Cahiers d’enseignement de la S.O.F.C.O.T., 1996, 55 : 1-17. 15. Kerboull M., L’arthroplastie totale de hanche Charnley-Kerboull. Bases mécaniques et techniques. Résultats à long terme, Maîtr. Ortop., 1999, 83 : 6-10. 16.Khalily YC,Witeside LA,Predictive value of early radiographic findings in cementless total hip arthroplasty femoral components. An 8to 12 year followup, J. Arthrop., 1998, 13, 768-773. 17.Merle D’Aubigne R.,Cotation chiffrée de la fonction de la hanche, Rev. Chir. Orthop., 1970, 56 : 481486. 18.Morscher E.,Experience with the press-fit gliding stem,In Kuswetter W.,Noncemented total hip replacement, International symposium Tubigen, 1990, Inc. 221-231, New-York, Thieme Medical Publishers, 1991. 19. Morscher E, Moulin P., L’arthroplastie sans ciment : technique d’avenir ousimple mode ?, Communication, journées Lyonnaises de chirurgie de la hanche (prothèses sans ciment), 1990. 20. Morscher E, Mathys R., La prothèse isoélastique de hanche fixée sans ciment,Acta Orthop.Belg.,40 : 639-647. 21. Nehme A, Puget J., Evaluation de l’état osseux périprothétique dans la chirurgie de reprise de hanche,Communication,Toulouse Hip,2002 : 213218. 22. Nourissat CH et le groupe scientifique ABG, Résultats à plus de 5 ans de la tige ABG HA, dans les

180

Bibliography

descellements fémoraux,Communication,journées Lyonnaises de chirurgie de la hanche,1999,225-230. 23.Paprosky WG,Lawrence JW,Cameron H.,Femoral defects classifications, Clin. Appl. Ortop. Rev., 1990, 75(Suppl. 9) : 9. 24. Van de Velde D, Deroche PH, et al., Approche théorique du verouillage,Communication,journée Lyonnaise de chirurgie de la hanche, 1999, 249-252. 25.Vielpeau C,Locker B,Van Neverbelde T,Heuguet V., Le risque infectieux en chirurgie orthopédique, Encycl. Med. Chirg. Tech. Chir., 1989, 44005, 1-18. 26. Vives P, Mertl P, De Lestang P, Jarde O., Reprise des 69 descellements fémoraux par la tige verouillée

Ultime, Communication, journées Lyonnaises de chirurgie de la hanche, 1999, 285-294. 27. Vives P, De Lestang M, Paclot R, Cazeneuve JF., Le descellement aseptique, définitions, classifications,Rev.Chir.Orthop.,1989,75 (Suppl.1) : 29-31. 28.Vives P, Picault CH.,Voie d’abord transfémorale et tige à verouillage distal, Sauramps Médical, 1999. 29.Wagner H.,Prothèse de révision de l’articulation coxo-fémorale, Orthöpade, 1989, Springer Verlag, Berlin. 30.Zweymüller KA,Litner FK,Semlitsch MF.,Biologic fixation of a press-fit titanium hip joint endoprothesis, Clin. Orthop., 1988, 235 : 195.

Achevé d’imprimer sur les presses de l’Imprimerie BARNÉOUD B.P. 44 - 53960 BONCHAMP-LÈS-LAVAL Dépôt légal : janvier 2007 - N° d’imprimeur : 701017 Imprimé en France