As Ian says, it's my duty to talk through how to reconstruct the atrocious membrane. And the slides will move on. So we need to make the point that when you see this radial head fracture, you should not be thinking about the radial head fracture. You should be looking at that X-ray and that CT scan and thinking, what is the underlying injury? With every radial head fracture, that should be the prime question in your mind. And here, the radial neck is resting against the capitellum. So we know that there is longitudinal instability of the forearm. That's the only way this situation could be created, assuming the ulnar humeral joint is well reduced. If we look at the distal end, we've seen this picture several times in these talks. You get prominence of the distal ulnar because of the proximal migration of the radius. And this is a significant problem. And we know because of the talks we've been through that the primary pathology is not the radial head fracture. It is the injury to the interosseous membrane and primarily to the central condensation of the interosseous membrane within the forearm. We shouldn't ignore the other elements of the interosseous membrane, but the central condensation is key to the longitudinal stability of the forearm. As Sumedh has told us, it's responsible for transfer of load, and it's said to prevent bowing of the radius and the ulnar. As Sumedh has already pointed out, and this is work from our own unit in Wrightington, as we move the forearm from pronation to supination, the strain within the interosseous membrane increases so that in neutral forearm rotation we achieve a peak of strain within the central band of the interosseous membrane. What are the physical properties of the interosseous membrane? Well, it's extremely strong and extremely stiff, as you would expect for any ligament. But what's really interesting is that when you test it to failure, it doesn't avulse from the bone on the radial side. It doesn't avulse from the bone on the ulnar side. It typically tears in the mid-substance of the central condensation. And so this means that attempting a repair, a direct repair of this structure, is extremely difficult. It's not simply a matter of putting anchors into bone and stitching it back down. In most cases, you're trying to sew bits of a shaving brush to each other, and that doesn't work well. And as Ian has alluded to, it's very likely that these interosseous membrane injuries, these central condensation tears, do not heal with time. We've already made the analogy of the forearm anatomy to that of the knee, and we know perfectly well that those intrasubstance tears of the ACL do not heal with time. And I think the injury is really very much the same. So yes, you can cross-pin the distal radial joint. Yes, you can put in a radial head. But experience is showing us that with time, these forearms continue to migrate proximally. Not only our experience, I'm sure many of you have had this experience yourselves, and it's well reported in the literature, that you follow a very classic sequence. The radial head goes in, they get more elbow pain, they get wear of the capitellum, the radial head comes out. They then get wrist pain because of proximal migration of the radius, and so the ulna gets shortened. And it's a sequence of events that we've all seen numerous times in a number of patients. And these radial heads, they fail because the loads through them are totally abnormal. They're impacting onto the capitellum, they erode the capitellum, or they dislocate posteriorly from the proximal radial ulnar joint. And in extreme cases, they just burr their way into the capitellum because of the massive loads that are being applied through the radial capitellar joint. As Ian has alluded, we've attempted to address this problem by replacing both sides of the joint, but inevitably the forces are so great that you get early failure of the radial side of the joint. So what are the other treatment options for these injuries? Well, we've looked at tightrope reconstructions, or other units have looked at tightrope reconstructions, using buttons with a suture in between to reconstruct the central condensation. And what they found, this is the intact interosseous membrane, and this is excursion in millimeters of the y-axis, is that when you take the radial head out, you get increased excursion. Cut the interosseous membrane, you get further excursion. But by using the suture button technique, you can reduce the excursion of the radius relative to the ulna, but nowhere near down to the same levels as the intact specimen. Another way to look at this is to look at the load through the distal ulna. Obviously, if the radius moves proximally, then you're going to get increased loading through the distal ulna when you load the wrist. And so another group have looked at this. This is the intact specimen here, and this is force in newtons on the distal ulna. If you cut the interosseous membrane, then you get a massive rise in the pressures through the distal ulna. Use the tightrope reconstruction, you can actually reduce these forces through the distal ulna, certainly close to the intact situation. But then put in a radial head, and suddenly you've got a much more dramatic drop. And why is that? It's because you're increasing the load at the radial capitella joint. And this is shown in this study from Los Angeles, where they move the arm into varus and valgus alignment. And with a radial head implant in there, in valgus, where you're increasing the load through the radial capitella joint, there's really no difference with the IOM section or the IOM intact. And that's because all the force is being taken up through the radial capitella joint. So what we're actually much more interested in is not the forces on the distal ulna, it's the forces being transmitted to the proximal radius. So typically in the normal situation, you load the distal radius, 94% of the force through the wrist transmitted to the distal radius, 75% will end up at the proximal radius. The rest is dissipated through the interosseous membrane with resultant forces at the proximal and distal radial ulna joints. If we then cut the central condensation of the interosseous membrane and load the wrist, we get less force applied through the distal radius, more of it is transmitted to the ulna, but all of the force that's transmitted to the distal radius is transmitted directly to the proximal radius. Because the IOM can't take up and transmit those forces to the ulna. So this group attempted a reconstruction using an FCR graft. And what they found with a single bundle, so 94% was transmitted through the distal radius and 80% reached the proximal radius compared to 75% in the intact. It was only by using a double bundle graft of FCR that they were able to find a reproduction of the normal anatomical situation. But they also reported that while in the intact state you have normal alignment of the distal radial ulna joint, with the FCR graft in place, even with a properly tensioned graft, you still get a few millimeters of proximal migration of the radius. And why is that? That's because with cyclical loading of an FCR graft, tendons are not the same as ligaments. They have more plasticity in them. And with cyclical loading you get creep and you get plastic deformation. So Lee Osterman has proposed that rather than using a tendon graft, we use a ligament graft. And he's used patellar tendon bone grafts and reconstructed those with screwing the bone fragments to the radius and ulna on either side. Laboratory studies looking at the strength of each of these reconstructions does show that indeed patellar tendon grafts are stiffer than FCR, than flex carporadialis grafts and tendon grafts. But they're nowhere near as stiff as the intact introsus ligament. They also tend to fail fairly catastrophically when they do fail. And they fail by cutting out of the screws from the bone, where the bone fragments are screwed into the radius and the ulna. And those other grafts, the tendon grafts, which pass through bone tunnels, tend to fail in a much more gradual manner. But Lee Osterman's own series of 16 patients that he reported here to this society is actually pretty impressive. He's looked at 16 patients with an average follow-up of 78 months. And in 15 out of 16 patients, the pain had improved. The grip strength had improved to up to 84% of the opposite side. However, their complication rate was quite high, at around 50%. Four had knee pain, obviously from the donor site. Two had radial capillary osteoarthritis. One had an ulnar nonunion, and one a delayed ulnar union. And one had extensive tendon adhesions, because he puts these grafts in from the dorsal side. I've published my own experience of using a different technique, which I'd like to take you through this afternoon. And this is using a synthetic graft called a Lars ligament, which is a polyester rope, which is extremely strong, but also extremely stiff, with very little residual strain. And it's also very robust to cyclical loading, with no damage seen after 5 million cycles in the laboratory. So briefly, to run through the technique that I use, I approach the radius through a Wohler-Henry's approach, and plan my screw hole into the radius, depending on the position of my ulnar screw hole. Now, in my experience, 33% from the ulnar styloid, in most adult patients, is approximately 6 centimeters. If you then, on your preoperative x-rays, plot out an angle of 21 degrees from that point of 6 centimeters to the radius, in almost every case, it ends up within a few millimeters of 12 centimeters from the radial styloid. So that is where I place my radial pilot hole. And I drill with a 3.5 millimeter drill, with an exit hole of 2.5 millimeters to reduce the weakening of the bone. I then approach the ulnar through a subcutaneous approach on the ulnar side of the wrist, and drill a 3.5, 3.5 millimeter drill hole through the ulnar. The graft is then brought in through the Wohler-Henry's approach and secured over an ender button to the radius. The graft is cut to length, and two whip stitches of number 2 orthocord suture, which is a strong synthetic braided suture, are passed into the graft and brought out through the ulnar. With one fellow holding the elbow at 90 degrees, the arm is reduced, the radial length is reduced by longitudinal traction, with the forearm in neutral rotation, because that's where the strain of the forearm is at its greatest. And the graft is then secured to a second ender button on the subcutaneous border of the ulnar. And this is the appearance of the reconstruction that we should end up with. This is a radiographic appearance immediately postoperatively, showing its acceptable reduction of the length of the radius. Additional procedures are often required, such as ulnar-shortening osteotomy, if you can't completely reduce the forearm, revision radial head replacement, and I'll come to talk a little bit more about that in just a moment, and capitellum excision. I'm not too worried about loading the radiocapitellar joint. What I want to achieve is loading of the proximal radioulnar joint. Again, the analogy of the knee is extremely useful here. If you performed an ACL reconstruction, but the patient was missing the medial femoral condyle, the knee would fall very rapidly into varus when unloading. And exactly the same thing is true with the forearm. So this is the first patient I did a reconstruction on, and this is the MOPIC radial head stem, which is an extremely difficult implant to remove when it's been placed in for a while. And I foolishly didn't put in a radial head back into this patient initially. And what you see is, as she makes a clenched fist, the proximal radius swings in, and she gets impingement at the proximal end. So in that situation, what you have to do is go back in and put a radial head back in this individual so you restore both condyles of the forearm in order to restore stability. And this has been shown in cadaveric studies, too, that with radial head excision and reconstruction of the interosseous membrane, you get proximal convergence of the radius and the ulna with loading. Our experience to date is with seven procedures and chronic injuries with a minimum follow-up of 18 months. One patient within that group had persistent axial instability. What actually happened was her radius and ulna were scissoring on the interosseous membrane reconstruction. And that taught us that really we need to be doing further reconstructive procedures at the distal radial ulnar joint in order to maintain stability there. In our series, we've seen no change in length on loading and, importantly, no change in length of the graft over time. So we've seen no radiographic recurrences of ulna plus at the wrist in any of our patients. So no evidence of radius migration, no evidence of fractures, and no cortical erosions. Over time, the bone tunnels are disappearing, which would appear to suggest integration, but we obviously have no histological evidence of that. In terms of the patient-reported outcomes, the median preoperative DASH score in our series is reduced from 77 to 41. So these patients, they've had a lot of procedures in the past. We're certainly not giving them a normal forearm, but we are able to reduce their pain. And in 6 out of 7 patients, they're satisfied with the outcome. So future considerations for interosseous membrane reconstructions are to stabilize the distal radial ulnar joint, and it may be that we need to look at doing something like an Adams procedure at the same time. Stabilization of the proximal radial ulnar joint may also be necessary, and possibly we should be considering using double bundle grafts. Thank you very much.

interosseous membrane

radial head fractures

longitudinal instability

Lars ligament

surgical procedure