|
 |
|||||||||||||||||||
A. RAY and the ESOP HA Group
Polyclinique Orthopédique de Lyon - 86 bd des Belges F-69006 Lyon France
First of all, I would like to thank those who have been with me from the start or who joined the exercise a few years later. In particular, my thanks are due to M. Philippe, H. Hourlier, G. Gacon, J. Hummer, A. Dambreville, P. Lecestre; to my associates at the orthopaedic centre in Lyon, L. Barba, J.J. Lalain, M. Laurençon, J.Y. Coillard, E. Milon; and to all who have believed in me.
The great story of hip arthroplasty began, for me, back in 1967, when I was starting in my Senior Registrar's post under M. Creyssel and G. de Morgues, where I witnessed the revolution in orthopaedic surgery. Since then, I have listened a lot; I have followed many other people's advice and opinions; I have sought to establish my own register; I have studied many case notes: And now that retirement is approaching, after a career that has had its ups and downs but which has always been rewarding, I would like to tell this story of hip arthroplasty as I have seen and lived it. It has not been a straight road from A to B, and I have not been faithful throughout to any one implant type or design. This has its drawbacks, but it has also been important, because it allowed me to perform a very critical review ten years ago. There are some points that should, I think, be highlighted.
Without wishing to reject cementing of the femoral components, I thought that cementless fixation could be recommended in certain cases, depending on the patient's age and activity level. I had been following the development of uncemented implants over a great many years, and had noted their inadequacies and their complications.
What, then, were these complications?
Mainly, they were things that had gone wrong because surgeons had ignored the physiology of bone and the elasticity of the femur. Poor stress transmission patterns, and a poor match between the femur and its implant had led to fractures and subsidence, either immediately or very early after surgery; also, there had been major bony changes (osteolysis or hypertrophy), which were responsible for disappointing clinical results in a large number of cases. Very disabling thigh pain was seen in a worryingly large percentage of the patients. I therefore went on to use more appropriate components, which allowed me to improve my results; however, things were still not totally satisfactory. I felt that the time had come to grow up, and to have the courage of my convictions: In other words, it was time to start designing an implant which, however imperfect it might be, would reflect my own ideas on the subject.
MODULARITY
The device was to be modular, with a diaphyseal and a metaphyseal part, so as to be able to cater for a wide variety of femoral patterns. It was to come in a Right and a Left version. The junction between the two parts making up the stem was, theoretically, a weak point that should be stressed little, if at all, to avoid breakage.
We did a study using a finite element technique, which showed that the stress level at the junction was very low, anyway.
STRESS TRANSMISSION IN THE PROXIMAL METAPHYSIS
All the stresses were intended to be transferred via the proximal metaphysis, so as not to stress the junction, and so as to transmit the stresses into the proximal femur without putting load directly onto the shaft. The femur was to lose as little as possible of its native elasticity; therefore, it was decided not to go for perfect fit in the middle and the distal metaphysis. The diaphyseal part was to be smooth, cylindrical, and slightly smaller in diameter than the canal of the host bone. It was also to be short, so as not to run into problems with distal wedging. The two ideas were linked, because one cannot conceivably get good wedging in the metaphysis without having a modular implant system.
THE SAFETY TRIANGLE
Femoral fissures were always in the medial arch; and I knew that little fissures produce big fractures. Therefore, wherever possible a little triangle of cancellous bone would have to be preserved on the medial side, between the calcar and the implant, with the implant itself propped against the inside of the trochanter. This triangle came to be called the safety triangle.
LATERAL AND ANTEROPOSTERIOR BULGE
Stability was intended to come from the "crouching" of the implant in the metaphysis; therefore, the metaphyseal part had to be wider laterally and in the a.p. direction. The "back" of the implant had to be rounded, with a smooth line, so as to allow the prosthesis to be inserted in a downward and slightly sideways direction.
IMPLANT SURFACE COATING
The material chosen for the prosthesis was titanium. In order to obtain osseointegration in the metaphysis only, the metaphyseal part was grit-blasted and coated all over with hydroxyapatite (HA), while the diaphyseal part was left uncoated.
Those, then, were the design features of the ESOP femoral component. Once the design work was done, all that remained was to convince others; to do arthroplasties; and to refine the technique. That was not all that easy. We had to go out of the design office into the operating theatre, and carefully study all our patients.
(Figs. 1a, 1b, 1c)
  |
  |
| Figure 1a and b: Metaphyseal part, lateral and a.p. views. Figure 1c: Diaphyseal trials; and modular metaphyseal broaches used to determine the ideal size of the definitive metaphyseal part; the neck length is established at the trial stage. Figure 1d: Assembly of the two definitive parts, just before insertion. There is still a choice of three neck lengths, to be used with 28 mm and 22 mm heads. |
WHY ESOP?
The name was chosen for two reasons: Firstly, because it was the acronym of our first group, and the name reminds us of old times.
Secondly - and that was the main reason - because I am very fond of Aesop, the Greek fabulist. He was a genius, and a very clever teller of fables, even though he was an ugly hunchback with a stammer. Unfortunately for him, he loved truth too much - he spoke the truth, he was right, and that upset an awful lot of people. So finally he got thrown off a rock by the priests of Apollo, who had a somewhat warped idea of truth. Let's hope that things have changed in the meantime!
The technique and the instruments must be such as to permit the easy and accurate insertion of the prosthesis.
The first point that I would like to stress is the importance of pre-operative planning. The drawings are made on the template the day before surgery. Like M.E. Müller, I think that I have never once missed this ceremony. Allow me to explain why I have always attached so much importance to templating: First of all, this exercise makes one think, and it allows one to see where there is a possible risk, of lengthening rather than of shortening. When the ideal centre point of the implant has been established, the templates (a.p. and lateral, if possible) can be used to determine the actual size of the two parts required, and the neck length that will need to be used. (At surgery, one frequently finds that the metaphyseal part will be a little smaller than predicted.) The distance between the tip of the greater trochanter and the horizontal portion of the metaphyseal implant part should be noted. This measurement must always be observed, because it is very reliable.
(Diagrams 3 and 4)
 |
Diagram 1: So-called CL8 bone index. It is measured at a level 8 cm below the midpoint of the lesser trochanter, on the a.p. film. The index is the ratio of the width of the cortices to the width of the shaft. We consider this index to provide good evidence of the condition of the bone. It is easy to obtain both before and after surgery. An index > 0.5 is good. |  |
Diagram 2: The femoral axis. The axis is represented by a line that runs through the centre of the medullary canal, on the a.p. film, at 8, 9, and 10 cm distal to the midpoint of the lesser trochanter. This line may be used even in curved femurs. The object is to make the axis of the implant follow the axis of the femur. When this is achieved, the implant is said to be in line with Axis 1. Axis 2 means an implant in valgus; while the more troublesome varus positioning of the implant would be in line with Axis 3 - this malposition must be avoided if at all possible. |
 |
 |
| Diagram 3: The distance from the tip of the greater trochanter to the horizontal portion of the metaphyseal part (GTPM) is determined before surgery. This distance must be kept once the centre of the implant has been established. The medial cut is then made in such a way as to insert the metaphyseal part completely in the host bone. The femoral resection line must not be too oblique on the medial side. | |
| Diagram 4: All the information is recorded on the pre-operative template. The required implant sizes are noted. The template is preserved, and used in the follow-up of the patient. |  |
The level of the medial resection is defined. It must allow the implant to be inserted flush with the bone on the medial side, without leaving any HA coated portion standing proud, and making sure that the cut is not too oblique, which would lead to a rapid increase in the rotational stress on the implant (Diagram 8). It is better to do a little trimming afterwards than to cut too obliquely from the word go. All the details are recorded on the template, which goes into the patient's notes. I also record the cortical bone index (CL8) 8 cm below the midpoint of the lesser trochanter (Diagram 1).
After the hip has been dislocated and the femur resected, a special chisel is used to enter the metaphysis. The chisel must be introduced as far laterally as possible; this is to ensure that the "back" of the implant can be accommodated correctly (Diagram 5). A safety triangle is left standing on the medial side. The medullary canal is identified, and a series of reamers is introduced - not for reaming, but to gauge the internal diameter of the canal (Diagram 6; Fig. 1c). Successive reamers of increasing calibre are introduced. The canal diameter is found when the reamer makes a grating noise as it is introduced. It should be noted that the reamers are used for probing, not for actual reaming: The technique described here does not involve any widening of the femoral canal. The reamers provide accurate information on the diameter of the canal, and thus allow the surgeon to select a trial stem that is 1 mm less in a diameter than the final reamer, and which has a long flexible rod with a button tip, used to show the axis of the canal.
 |
 |
 |
| Diagram 5 : Surgical technique. After the resection of the neck, a special chisel is used to enter the metaphysis as far laterally as possible and in the femoral axis, leaving a little cancellous bone on the medial side. | Diagram 6 : After identifying the medullary canal with reamers, the internal diameter of the canal is established, without any actual reaming. | Diagram 7 : The lateral bulge of the metaphyseal part "crouching" against the trochanter gives good stability; medially, a small safety triangle should be preserved. |
 |
| Diagram 8: Rotational stability is excellent if the metaphyseal part is inserted flush with the neck resection level. If the cut is too oblique, rotational stresses will rapidly increase, which will have an adverse effect on stability (Whiteside). |
Next, a succession of metaphyseal broaches are fitted onto the stem, and advanced into the femur, until a sufficiently stable seating of the broach is obtained. The distance between the tip of the greater trochanter and the broach is checked. It must correspond to the value measured on the template. As pointed out above, it is desirable to have a safety triangle. A check is therefore made to ensure that cancellous bone has been preserved medially. Also, rotatory stability is checked (Diagram 7). A trial reduction is performed, and the definitive components are selected.
All that remains to be done is to screw the two parts of the femoral component together. The thread will automatically lock when the stem has been screwed home. The implant is inserted into its prepared bed, and lightly tapped into place to give perfect stability. Once the implant is firmly seated, tapping should be stopped since it would not make the component go down any further. A check is made for primary stability in the three planes (coronal, sagittal, rotatory). The implant's shape, which makes it fit well into the metaphysis, is yet another factor that helps to produce primary stability. There is no need to use the largest possible component: The first size that achieves stability is the right one.
At the medial resection level, a careful check is made to see if there is any HA-coated portion standing proud. (Ideally, no coating ought to be visible.) Also, the GTPM distance is checked (see Diagram 3). An X-ray check of the femoral axis is made; this, too, is entered in the patient's notes.
|
- After neck resection, a chisel is introduced into the lateral part of the metaphysis, following the femoral axis. - The canal is identified and reamers are used to gently probe the canal. - The diaphyseal part is selected; next, successive broaches are used, to find the size that is stably seated in the metaphysis. - The distance from the implant to the greater trochanter is checked. - The medial cut must be such as to ensure flush seating of the implant, to prevent the rapid build-up of rotational stresses (Whiteside). - The definitive implant is inserted, and a trial made to find the correct neck length. - Pre-operative template |
Two questions are asked whenever this implant is discussed:
(1) Is there really a need for the diaphyseal part?
Questioners always want to know whether there really is a need for a smooth, uncoated, underdimensioned part that transmits little if any stress. In fact, I was once told about a somewhat unique case in which only the metaphyseal part of the prosthesis was inserted, with a good result at two years. The case involved a revision in a patient with a sound metaphysis, but with an insuperable problem regarding the total removal of the cement left in the femur from the primary arthroplasty.
The diaphyseal part has an important rôle to play in guiding and stabilizing the femoral component. The part has to be inserted as outlined above.
(2) Does one really need a modular system?
Modularity is important, as I learned from a study of the pairing of the two parts in the series of ESOP arthroplasties.
As far as the diaphyseal part is concerned, all the sizes provided have been used in patients; the medium sizes (10,11,12) were the ones most frequently employed. Each diaphyseal part size has been used in conjunction with a wide range of metaphyseal parts (with a maximum of 8) (Fig. 1c)
As regards the metaphyseal part, all the sizes provided have been used, with a large number of diaphyseal parts (between 2 and 4, depending on the size involved).
A large metaphyseal part may be used on a small diaphyseal part, in the management of flared femora. Conversely, large diaphyseal parts have been safely used in wide and somewhat cylindrical femoral shafts. With increasing experience, it has been possible to extend the range of indications for treatment with an ESOP prosthesis to the very old.
An implant must be easy to insert, and equally easy to remove. In an earlier series, we had to revise three patients for deterioration of the spigot. At revision, we found that the periprosthetic bone stock was of excellent quality, and that bone had grown closely onto the entire metaphyseal surface of the implant. By simply inserting a thin flexible blade between the device and the host bone, removal was readily accomplished. The diaphyseal part was not fixed at all, and came out with the metaphyseal part. There was no damage to the proximal femur, and we were able to insert another ESOP of the same size or one size up.
The ESOP device may be used in all conditions, and at any age. The only limiting factor in patient selection is the quality of the metaphyseal bone stock. If, after the stability tests, there is the slightest doubt about the eventual primary stability of the implant, a different system will need to be used; however, ESOP users have found this to be a very rare occurrence. ESOP should not be used for revisions, unless the surgeons involved are very experienced, and the patient meets stringent requirements (Type 1, with a sound metaphysis).
Full weight-bearing is permitted immediately after surgery; the patient should use two aids for four weeks, as for cemented implants: This is intended to protect the soft tissues: The integrity of implant fixation is not predicated upon the use of walking aids.
The patients were followed up at 2 months, 4 months, 1 year, 2 years, and 5 years. The radiographs were analyzed using a software that gives the ARA score. A computer programme for patient follow-up has been developed and used within our Working Group. The full results will be published in the near future.
In this study, the clinical results were seen to be excellent in the first weeks, as with cemented implants; there was no instance of thigh pain.
(Figs. 2a, 2bc - 3a1, a2, b1, b2, c1, c2)
 |
 |
 |
Figure 2a: Bilateral ESOP HA followed up for 5 years; radiograph at 4 and 3 years, respectively. 42-year-old, 80-kg manual worker. Perfect result to date, despite an unduly oblique medial cut. Figure 2b and c: Follow-up radiograph at 3 years. 55-year-old, 125-kg patient; surgery for bilateral avascular necrosis. Excellent result at 4 years. On the left, customized implant at 6 years. On the right, ESOP HA: no subsidence or migration; patient has "forgotten" he had a hip replacement. |
 a1 - a2: Postoperative radiographs |
 |
 b1 - b2: At 1 year |
 |
 c1 - c2: At 5 years |
 |
| Figure 3a 1.2 - b 1.2 - c 1.2 Example of radiographic follow-up of an ESOP HA in a 57-year-old 100-kg manual worker with a cemented implant in the contralateral hip. Postoperative radiographs and reviews at 1 year and 5 years. A reinforcement device had been inserted to manage acetabular insufficiency - one of the three screws is seen to be broken at 5 years. |
|
The radiological follow-up showed
(1) No secondary subsidence or tilt (these phenomena were routinely checked for with measurements);
(2) No radiolucencies in the metaphysis;
(3) Normal or slightly increased metaphyseal cancellous bone density in Zones 2 and 6, without any stress shielding at the longest follow-up;
(4) No thickening of the shaft cortex, and no pedestal formation. The CL8 index remained unchanged over time; there was no change in the axes, and the ARA score was between 5 and 6, i.e. excellent.
Compared with the cementless prostheses used in the past, these implants with metaphyseal fixation have superior reliability; their clinical results are similar to those of cemented prostheses, with the benefit of absence of thigh pain. The question arises whether these arguments provide sufficient justification for the use of these implants. Or, to put it differently, can these implants do something that the "classical" prostheses could not do?
There are three advantages that must be taken into account:
(1) A mechanical advantage: Stress transmission occurs in the metaphysis.
The radiographs taken in the medium term show condensation of the metaphyseal cancellous bone, which provides evidence of the transmission of stresses at this level. All implants with diaphyseal fixation - be they cemented or cementless - are associated with thinning of the metaphyseal and proximal diaphyseal cortices; while this thinning does not normally cause loosening, it contributes to the decline in the quality of the bone stock around the implant, which will go on to major osteolysis. So, preventing this factor of bone stock compromise is actually very useful.
(2) A biological advantage: Polyethylene microparticles are prevented from tracking around the implant
The bioactive coating produces an intimate contact between the implant and the host bone, without any interposed fibrous tissue of the kind routinely observed at cement/bone interfaces. This fibrous membrane provides a thoroughfare along which wear debris can travel around the implant, especially in the femoral shaft. This is what Schmalzried and Harris have called the effective joint space. The metaphyseal portion of the device has an all-round bioactive HA coating that prevents debris from tracking, which is why femoral shaft lysis has never been seen.
(3) Removal is easy and tissue-sparing
Since the implant is coated only over the proximal few centimetres, it can be extracted without difficulty. There are none of the removal problems that bedevil the cementless implants with diaphyseal fixation, nor those of the cemented ones, where the removal of the cement leaves the bone fragile and devitalized.