|
 |
|||||||||||||||||||
|
|
|||||||||||||||||||
Bernhard Georg Weber has profoundly influenced Swiss orthopaedics; his many papers have gained him international renown. After a long and busy career, his metal-on-metal prosthesis has put him once again in the forefront of orthopaedic innovation. We went to interview him - and found a grand old man full of youthful enthusiasm.
M.O. How did you become Maurice Müller's successor at St. Gall?
B.G.W. After training under Prof. Francillon at Zurich University Hospital, I found a post in Prof. Stinchfield's department, in New York. While I was waiting for my United States visa, I decided to tour the main centres of orthopaedic surgery in Europe. When I was in England, I got a telegram from Maurice Müller asking me to come and work with him. I thought, well, perhaps it would be better anyway to work in Switzerland. So I accepted, and spent six and a half years as Maurice Müller's Senior Registrar. Then, when he got the Chair in Berne, I became his successor at the Canton Hospital in St. Gall.
B.G.W. He had met me when I was a house officer and he was Senior Registrar at Balgrist. Then, when he was in private practice, he started sounding out his colleagues in Switzerland to see who had what it took to be his senior registrar at the centre he was going to - St. Gall. By then, I had a reputation as a hard worker and a good sport, someone who could win or lose. That was exactly what Maurice Müller was looking for, to set up "his" orthopaedic centre that was to be a completely new departure in Switzerland. The combination of orthopaedics and traumatology had not been thought of before. The following years showed that this sort of centre is ideal for multiply injured patients and those with single injuries alike; it also proved an ideal pattern for training in conventional orthopaedics. That is because in traumatology, the junior surgeon gets to know the pathophysiology of trauma, and because every orthopaedic operation is in itself a trauma. Also, traumatology offers a much larger surgical material than would pure orthopaedics. In 1960, when the St. Gall hospital was founded, there were only three conventional orthopaedic centres. Today, 35 years later, there are some fifty centres - and it was St. Gall that had provided the impetus for this development. Many orthopaedic surgeons who now head departments - not only in Switzerland but also abroad - did their training in St. Gall.
B.G.W. Maurice Müller developed new concepts in all branches of the orthopaedic discipline, which have been influential world-wide. I was incredibly lucky to be able to work in his department as a young registrar. Not only did I benefit from his knowledge, he was also so busy writing up his ideas that I was able to run the centre more or less as I saw fit. I was virtually in control, and able to do my own thing. Maurice Müller checked what we did, but gave us a plenty of scope to do things the way we wanted, provided that his principles for the management of cases were adhered to. Also, whenever we made suggestions that were along his lines, we found him very willing to listen. This provided us with a vast field of activity, and accounts for the large number of papers that came out of that centre. Our scientific output was remarkable not only in terms of quantity, but also because of the high quality of our publications. That sort of thing was unique in the history of Swiss orthopaedics. And it was Maurice Müller who had provided the framework within which all these things could go ahead.
B.G.W. He had ambitions, and he wanted a Chair. Getting a Chair was very, very important to him. Later on, he was also made President of SICOT, and by the end of his career he had reached a level few other orthopaedic surgeons could claim to have risen to. I myself have never been tempted by the honours of academia. There are differences between the two of us that play a role even nowadays. Not a negative role - but what I would call mutual monitoring. He certainly keeps an eye on what I am doing in hip surgery. Fractures did not present a problem, but in hip surgery we were rivals to some extent. However, when he left St. Gall, he wanted his work to go on, and by leaving me in charge he knew that it would be continued by one of his disciples and his friends.
B.G.W. When we started at St. Gall, in 1960, Maurice Müller did not know total hip replacement, and his favourite subjects were osteotomy and fusion. I, on the other hand, had been to England, where I had picked up Sir John Charnley's ideas. I told Maurice Müller what I had learnt across the Channel, and he immediately saw how important this surgery was. Since he is a great "doer", he had made and implanted his first total hip just a few weeks after I had described Charnley's work to him. He went on to take a profound interest in THR, and became friends with John Charnley. He was also interested in the McKee prosthesis, and, in fact, everything and anything that was going on in joint replacement in the Sixties. I was a Senior Registrar, so I did not have the time to immerse myself in this subject; however, when I became independent, in 1967, I was able to put my own ideas into practice. I was thinking of a device with a rotating trunnion at the neck (Fig. 1). When it came on the market, it was the first modular prosthesis, and it was metal-on-metal on the neck. In other words, it was the first device with movement in two different planes. At the time, many felt that, given the poor performance of the McKee, with its metal-on-metal system, my device was doomed. In actual fact, things turned out quite differently, and my implant did very well.
B.G.W. I had developed this device completely on my own. I had been thinking about it and designing it since Easter 1961. However, the designs stayed in a drawer until the time I became a consultant. In 1968, I had the first model, and by 1971 the first proper trunnion-bearing device (Fig. 2).
Fig. 1 Drawing (23 April, 1961) of a trunnion-bearing THR. Movement occurs in two different planes. (Original text: "Die Hüftbeweglichkeit ist im Gelenk immer kombinierte Bewegung in zwei verschiedenen Gleitflächen.")
Fig. 2a, b Development of the THR since 1971
a - Trunnion-bearing prosthesis with different neck length heads; 1971 model
b - The same design, with an alumina ceramic head; 1973 model
B.G.W. And how! He was a super Chief, but he also saw himself as the lord of the manor.
B.G.W. First of all, ankle fractures, a subject on which I wrote a textbook that came out in 1968. After that, I became interested in non-union, and wrote about that subject with O. Cech, a colleague in Prague who had spent two years working with us. It was a great textbook on non-union, and is still highly relevant 22 years later. Non-union was classified in terms of the bioactivity of the tissue involved. We knew that the healing chances of non-union depended on two factors: Firstly, union is a function of the viability - i.e. the blood supply - of the fragments. That is something that Robert Judet had also found; and we were able to confirm this fundamental concept. Secondly, union depends on the stabilization of the fragments, so that the ends can knit. This means that there are two problems that need to be controlled - a biological one, and a mechanical one. At the time, surgeons were not sufficiently aware of this dual requirement. We were able to make this point, and thus to contribute to the success of our textbook.
B.G.W. Indeed we did - in fact, we described in our textbook all the possibilities that were of proven value at the time. What we suggested was based upon our experience in the management of 800 cases of non-union, at the Canton Hospital and in Prague. We knew the nature and the natural history of most of the patterns of non-union in our patient material, and were thus able to establish a detailed prognosis for every case. This prognosis was based on the dual assessment of the biological condition at the site of non-union and of the mechanical stabilization that could be provided. When all is said and done, non-union is quite straightforward to treat: It all boils down to knowing the bioactivity at the site, and providing stable fixation.
B.G.W. Our work on non-union, and especially on infected cases of non-union, gave us vast experience with the technique of stabilization using an external fixator. So I wrote a textbook on that as well. I think it's the best of all my books. My co-author was F. Magerl, who described the use of the external fixator in spinal conditions. This part written by Magerl was so original that many surgeons felt that they were on to a good thing. As indeed they were, when they took the idea one step further and developed the internal spinal fixator.
B.G.W. Absolutely. He cut off the pins and got himself an internal fixator. After that, I took an interest in recurrent shoulder dislocations. Reviewing patients who had undergone a Bankart or a Putti-Platt procedure clearly showed that while dislocation had been controlled, there was a 20 to 40 degree loss of external rotation. I thought that was terrible. An athlete, a mountaineer for example, must surely have a full range of rotation. So I said to myself, why don't we devise a simple anterior Magnuson-type procedure for tensioning the subscapularis, combined with a transverse 30* rotational osteotomy of the humerus? I did my first case in 1963, when Maurice Müller was away on holiday. I discreetly followed up this athlete until I was my own master; and then I started using this procedure on a large scale. Within a few years, we had clocked up over 300 cases. The results were sensational. This procedure was very popular with surgeons who were used to doing internal fracture fixation; those who did not have this experience were worried that the osteotomy might not heal (Fig. 3).
Fig. 3a, b Subcapital osteotomy in the treatment of habitual shoulder dislocation
a - Bankart, Putti-Platt, or Magnuson-Stack procedure.
Compensation of the external rotation deficit by a rotational osteotomy.
b - Stabilization of the osteotomy by a compression angle plate
B.G.W. Of course it does, but only if the surgeon is not familiar with the technique.
B.G.W. Felix Hardegger, one of my senior registrars, has reviewed 250 procedures, and found only four cases of non-union. These were re-operated on, and then healed. Recurrence of dislocation was very, very rare. The surgical technique is straightforward and logical; however, it cannot be picked up from a book - the learner needs hands-on experience. Once mastered, the technique is virtually foolproof.
B.G.W. This was the procedure of choice for athletes, e.g. gymnasts, mountain climbers, or ice hockey players. The disadvantages were the risk of non-union as a result of poor surgical technique, and - admittedly - the size of the scar.
B.G.W. This procedure is, in fact, a Bankart-type repair, with a complementary rotational osteotomy.
B.G.W. The hardware did not cause any problems. However, these were young, athletic patients, and they were advised to have the hardware removed after one year. I have always wondered whether we could not have done a supracondylar humeral osteotomy. I never did cut the humerus at that site, but I think it would have worked just as well.
B.G.W. The procedure has gone out of fashion, and in the States they are now advocating what is known as an anterior and inferior capsular shift. This reminds me of what Eugene Ionescu once said - "If you run with the times, you will always be lagging behind." The fact remains that other types of surgery are now "in." However, I do think that the procedure is ideal for those with athletic or professional demands. For instance, we had five or six parachutists who needed full external rotation to control their free-fall jumps. A conventional operation would have left them with an external rotation deficit that would have made them come spiralling down. We have also developed several other original techniques, such as internal fixation using a dished metal plate.
B.G.W. These plates were used in the treatment of femoral non-union that was difficult to manage otherwise. We had many such cases, either after attempts at conservative management or after intramedullary nailing or plating, in particular with segmental fractures. Even where first-class internal fixation with a plate or a nail had been performed, the fragments would sometimes not unite because of the poor biological condition of the bone ends. In these cases, where the hardware would break after a few months, I thought that the plate should be made to bridge the site of non-union, with a graft placed under the bulge of the plate to lie between the plate and the lesion (Fig. 4). The practical application of this principle was somewhat empirical, but the success of this procedure was so convincing that we began to manage recent fractures in critical zones with this technique. We then applied the method to limb lengthening, to speed up healing. We applied a dished plate and a graft, and the whole thing healed very quickly. Much faster than with an Ilizarov. With an Ilizarov, lengthening takes a year, whereas with a Wagner followed by a plate-and-graft, it only takes four to five months.
Fig. 4a, b The bulge plate
a - The bridge has been weakened by the saw cuts made by the two pranksters. However, it does not collapse in the ensuing pursuit because it is stabilized by the bracing on the underside.
b - Analogous situation in the femur: The "brace" consists of a bulge plate that is kept away from the lateral cortex by a massive cortico-cancellous graft. Medial loading is shifted to the lateral cortex.
B.G.W. That's because I did not make a huge effort to make it known. I did not want to create a whole range of implants to cater for every conceivable pattern. What I wanted was to obey the laws of biomechanics with a minimum of hardware, but with a hardware that the surgeon in the theatre could adapt to the individual patient's needs. In other words, the surgeon should be able to twist and bend the plates to make them fit the different, and very complex, patient patterns. This policy has been called "Minimax", since it achieves a maximum of effect with a minimum of outlay.
B.G.W. Yes. As I was saying, at St. Gall we were a group of young surgeons who were forever discussing and bouncing ideas off each other. Not all of our new ideas were popular with the Chief. Very early on - I think in 1962 - Alexander Boitzy refused to do anatomical reductions and stable internal fixation of the tibia in certain cases. The fractures concerned were complex multi-fragment patterns, and he disobeyed the house rule which said that such cases should be managed with anatomical reduction and rigid fixation (Fig. 5). Boitzy thought it would be enough to link the main distal fragment and the proximal fragment using a subcutaneous plate that did not interfere with the injury zone. That's how biological fixation was born, and Boitzy actually coined the term "biological internal fixation." Thirty years later, this term is still very modern. But when we, the young ones, were talking about "elastic" biological fixation, we were almost branded traitors to the AO/ASIF cause. It is a good example of how things evolve. In surgery, a bad idea does not take long to be recognized as such. After a couple of years, everyone knows it's rubbish. But good ideas take a long time to establish themselves - twenty years is not unheard of.
Fig. 5a, b "Biological" internal fixation
a - Recent complex, multi-fragment fracture
b - The plate bridges the complex fracture site and is fixed only to the two main fragments. The small peroneal plate has allowed the fractured tibia to be aligned. Excellent healing, thanks to preservation of the blood supply to the fracture site.
B.G.W. To me, nailing always has had very limited indications: fractures of the middle third of the femur or the tibia; either transverse or very short oblique fractures. All other cases would be managed either conservatively, with a Sarmiento plaster, or plated, or given an external fixator. The interlocked nail that now exists is really an unstable intramedullary plate! One might just as well use an external fixator, which would not interfere with the intramedullary blood supply.
B.G.W. No, it would work better. And it can be done using a unilateral fixator with three upper and three lower pins (Fig. 6). I have written a paper on complex femoral fractures managed with an external fixator. The results were remarkably good, and I cannot understand why this technique has encountered so little interest. Everybody wants to use interlocking nails, because they are the "in" ironmongery.
Fig. 6a, b, c, d
a - Complex fracture aligned by means of an external fixator
b - The fracture has united three months after the accident.
c - External fixator in situ, shortly before its removal
d - Excellent healing, one year after the accident
B.G.W. In 1967, I was contacted by a medical engineering team which invited me to join in the setting up of a company to develop a hip prosthesis. I had my project, which had been sitting in a drawer since 1961, with designs for a completely novel THR and designs of a bipolar prosthesis. These devices were novel, since they provided for axial movement at the neck. The idea was to allow for movement in different planes, and it was assumed that movement would occur wherever friction was least at any given moment. The two engineers who were working with us thought that this might have a beneficial effect on friction and on wear. So, in 1968, the first trunnion-bearing prosthesis was made; however, one of the materials used - polyester - did not perform well. The first heads were made of PE, while the sockets were metal.
B.G.W. Yes. But a couple of years later it was found that that was causing loosening and extensive granulomas. The retrieved implants were sent to Hans-Georg Willert, who, I think in 1973, published his first paper on macrophage tissue reactions and tissue proliferation in the presence of polymer debris. This finding came before the observations later made by Willert and by many others, of PE wear debris. As soon as these clinical findings were out, we stopped using PE heads. In 1971, we changed over to metal heads with a metal-on-metal combination at the neck and a metal head running in a PE socket.
B.G.W. No - because the metal synovitis problem had occurred between the McKee type head and socket, not in the cylinder. The engineers felt that the cylinder would behave much better. Also, there was a little channel through the head of the prosthesis, to allow the flow of liquid.
B.G.W. Not only that system, but also the cylindrical joint around the implant neck. We thought that, during walking, small pistoning movements in the neck could act as a pump and allow joint fluid to circulate (Fig. 7). Then, at a time when only Boutin had a ceramic device, we became very interested in ceramic materials. In fact, wetting is better with ceramics than with metal; so, in 1972, we went over to 32 mm ceramic heads.
B.G.W. It was - firstly, because there was a fin to enhance rotational stability; and secondly, because the stem cross-section was trapezoidal. I wanted the lateral part of the implant, which is subject to tensile stresses, to be made thicker than the medial part, so as to be better able to resist these stresses. The medial part could be left thinner, but obviously not to the point of producing a cutting edge. That decision was taken back in 1967 - and the cross-section has remained the same ever since (Fig. 8).
Fig. 7a, b, c, d THR lubrication
a - Trunnion-bearing prosthesis: Joint fluid circulates through the trunnion-bearing which communicates with the joint space through an opening in the head.
b - In fixed-head prostheses, this circulation cannot take place.
c - Diagrammatic view of the lubrication pattern in a trunnion-bearing prosthesis
d - No lubrication in non-trunnion-bearing prosthesis after slight PE wear - dry frictionFig. 8a, b Stem cross-sections of the Weber prosthesis
a - More metal on the lateral than on the medial side. Trapezoidal cross-section. Note flattening of medial surface.
b - Medial aspect of implant. Flattened medial surface, implant flares from medial to lateral.
B.G.W. At the time, the advantages of this prosthesis were not all that obvious. Surgeons would have needed to study and analyze the design in considerable depth. But that's not how the market works. Naturally, surgeons are more inclined to use something that is obvious and does not require complicated reasoning to understand its merits. I carried out a ten-year review of 108 of my implants with ceramic heads; I had myself inserted all of them. A colleague at the Canton hospital reviewed a similar series of 138 cases who had received metal heads. We published the study. I think that the results in the patients with ceramic head implants were spectacular: there were only 5 cases of loosening, while there were 12 cases of loosening in the metal head group. This study allows the conclusion that ceramic heads are superior to metal ones. We did publish these results, but they are subject to criticism since the series was not very large and since the follow-up was undertaken by the surgeon/sponsor himself. However, over 100,000 trunnion-bearing prostheses have been implanted and used by a number of surgeons that has remained steady over the years.
B.G.W. That was mainly for financial reasons; and also because I wanted to compare the fixed-head devices with the trunnion-bearing ones. With the follow-up I now have, I do feel that the ceramic head on a trunnion was the best design on the market. This rotation counteracts PE wear and also stem loosening. Compared with fixed ceramic heads, ceramic-headed trunnion-bearing prostheses have shown less PE wear and a lower rate of loosening.
B.G.W. They did indeed - but only the trunnion-type ones (Fig. 7).
B.G.W. I was at a large hospital, with many surgeons; and I could see the difference between good and bad technique. We also had a large number of revisions building up in a short time. I often wondered why a particular joint replacement had failed. It rapidly became clear to me that the cementing technique was crucial, and that all the details of that technique mattered. This was on the lines of what I had initially been taught by John Charnley - lessons that had always been extremely useful to me. I think that cement as such is neither a good thing nor a bad thing - it's all a question of how the surgeon handles this material. Now that we no longer have to use polyethylene, we have only our surgical technique to blame if the implant loosens.
B.G.W. Because of the long-term results, at 15 years and beyond. There is firstly the problem of wear debris and the macrophage reactions produced by that debris; and then there is the mechanical problem brought about by wear. When the head has penetrated 2 or 3 mm into the PE, it begins to get stuck. So in the end there is an adverse biological and an adverse mechanical pattern. We had to look for a better material combination, and we chose metal-upon-metal for a number of reasons. Firstly, some patients who had been given metal-on-metal Müller-McKee implants with 42 mm heads, back in 1967-68, had no signs of loosening after 25-30 years. Also, it was realized that the McKee had failed because of manufacturing defects. The heads did not completely match the sockets. In the cases where McKee prostheses had been retrieved after having been in situ for a very long time, it was found that the heads were perfectly spherical and of a diameter just slightly below that of the socket. However, that had happened by chance rather than by design. And finally, in the revisions of my trunnion-bearing prosthesis, wear at the cylindrical metal-on-metal neck joint was found to be negligible compared with the PE wear of the socket. This is why, in 1983, the Sulzer engineers at Winterthur and myself set out to tackle the problem at the root. It took us five years to find the ideal dimensions between the metal socket and the metal head.
B.G.W. The problem was how to make a prosthesis with perfect dimensions. The manufacturing process is complicated. It took several years to build machines that can manufacture spherical heads with the required tolerances.
B.G.W. For the acetabular component, we chose a rigid socket with a minimum thickness of 3 mm. For the head, the ideal diameter is between 28 and 32 mm.
B.G.W. Seven years - since 1988. And to date there have not been any problems.
B.G.W. So far, I have not seen any problems.
B.G.W. Dislocation does not seem to be a problem. The engineers tell me that any scratches will level off eventually.
B.G.W. Metal allergy.
B.G.W. After a lifetime trying to master my craft, I find that there is one field that deserves particular attention. I am talking about postgraduate education and training for surgeons, in particular the acquisition of skills. Obviously, one can read textbooks and listen to papers at scientific meetings, and acquire a lot of theoretical knowledge in this way. However, I think that, in postgraduate education, too much emphasis is placed on theory, and too little importance is given to the acquisition of practical skills.
B.G.W. There should be practical courses, provided on a regular basis.
B.G.W. I am sure they could. That's where training starts. It's like playing the piano. PGE should not be all theory; there must also be practical workshops.
B.G.W. The surgeons who attend them. If I wanted to learn to play an instrument, I would need to pay my music teacher. Equally, if I wanted to improve my tennis or my skiing, I would need to take paid lessons. Surgeons should be encouraged to achieve greater mastery over the technical aspects of their craft. The courses could be organized by the surgeons themselves.
B.G.W. They would be a beginning, a means of learning a new surgical technique - "Learning by doing." The next stage would be a visit from a colleague who is well versed in that technique; and the final stage would be finding someone that puts the instruments into the learner's hands and tells him to get on with it.
