morning. We have actually 95 people scheduled for this ICL, but it's 645 Saturday following the alumni dinner, so I'm sure it's going to be a little bit of time for people to wake up and trickle in. I'm very excited about this instructional course lecture. The title of it is Shoulder and Elbow Fractures, When to Fix Versus Replace. We have an exceptional faculty. Dr. Chai Mudgal from the MGH here locally in Boston will be starting us off discussing radial head fractures. Dr. Harry Hoang from Cleveland will be discussing distal humerus fractures, and then Dr. Chris Schmidt from Pittsburgh will be discussing proximal humerus fractures. We're going to go through types of fractures, indication techniques, and obviously most importantly, what are indications and pros and pitfalls for fixing these fractures versus replacing these fractures. So why don't we get started, and Chai, do you want to set it up? I'm trying to find a pointer here. I'm trying to find a pointer here, and I'm trying to find a pointer here, and negative. Okay, good morning. For the next 15 minutes, we're going to have some fun. I know it's early. So particularly, I have no disclosures. I'm part of AO faculty, and that's the only thing I have going. Most of these illustrations you'll see over the next 15 minutes are mine, but I have to be grateful to the archives of the AO North America, and particularly to John Capo and Tom Warecha, and my patients who make beautiful jewelry from distal radius plates. So the next 15 minutes, we're going to try and understand the anatomy of the radial head as it pertains to treatment, its role in elbow stability, indications for internal fixation versus arthroplasty, and pearls and pitfalls for each technique. But the real bottom line for me is to try and demystify this injury so that at the end of this thing, in your own mind, you can come up with a consistently reproducible plan or algorithm to approach this injury. Now, you'll notice that we are not talking about radial head excision, either partial or total, and that's because we've decided that we are going to keep the radial head. When we say, do we fix or do we replace, we've decided that it is to be kept. So it's like the Clash song, should I stay or should I go? There's no should I go, I stay. So we're going to try and figure out why, when, and how. Dr. Jupiter told us many years ago that if you look at the elbow as a ring concept, then when you hurt one side, such as the radial head, it is inevitable that at least some part of the ring on the other side should be injured. And if you look at the data from Roger Van Riet, in type 1 radial head fractures, you can have up to 5% injury to the structures on the medial side. And in complex fracture dislocations, it can reach almost 80%. So where does the radial head fit into all this? Well, when you put a compressive force across the elbow, up to two-thirds of it goes through the radial capitellar joint. And it is also the secondary restraint to valgus after the AMCL. That's okay, but if you take out the radial head and you injure the coronoid and the AMCL, then the elbow tends to fall apart. Because along with the coronoid, the radial head is responsible for preventing posterior translation. Therefore, in the absence of the AMCL, it follows that radial capitellar contact should be, and some would say, must be restored. Now, the anatomy of the radial head also has implications on treatment. Firstly, the radial head is not perfectly spherical. It is elliptical, which means that it has a smaller diameter and a larger diameter, which need to be taken into consideration while fixing as well as replacing. More importantly, the radial head makes a 15-degree angle with the shaft. Now, this is really critical because if you put a plate on the radial head and shaft and it's not perfectly contoured, you change the inclination. And when that happens, you affect the rotation adversely. More importantly, when you're using a stem device for replacement, the stem should not ideally go past the bicipital tuberosity because you change this angulation, unless it's a bipolar device. Now, if you look at the right radial head end-on, as seen in the middle picture, Havertstock et al looked at 18 cadavers and they did CTs. And what they found at the green quadrant, which is the anterolateral quadrant, which is most commonly injured, has much less bone if from the left to the right, you're going from proximal to distal. And you'll notice that the blue on the posterior medial quadrant, which is the red one, has much more blue in it, which indicates dense bone. And the anterolateral quadrant, which is the green one, has much less bone in it with all the black there, which therefore means that the amount of bone fixation available for fixation is much less in the anterolateral quadrant. This has significant implications for treatment and screw purchase. How about vascularity and how does it impact what we do? Well, think about it like a femoral head. We are very careful about preserving vascularity to the femoral head, but yet we tend to approach the radial head with relative disdain. Now, we have interosseous vessels. You have an arcade of vessels entering through the capsule and the soft tissues. But if you break the neck, you lost the interosseous connections. Furthermore, if you dissect down the neck, you also lose the external vasculature. And yet, we go around putting in plates there with impunity and wondering why radial neck fractures don't heal. So once we know this anatomy, now we want to understand the pathology of what a fracture does. We get x-rays. X-rays are singularly unhelpful in understanding the complexity and the degree of comminution. They consistently undercall it. To me, a CT is invaluable for any fracture that I'm going to consider fixing. Not only does it show me the fracture geometry, it also identifies other fractures like this capillary fracture which is interposed in the radial head, as you can see there. However, it's been shown in multiple studies that if you were to have interobserver reliability for either plain x-rays or CTs, it is singularly poor, which therefore means that it makes our job difficult when you're trying to understand classifications. All the classifications we have currently are fraught with poor interobserver reliability. Therefore, I tend to follow Hotchkiss's modification of Mason's classification, which is based on treatment. To me, two things matter, Hotchkiss 2, Hotchkiss 3. Either you can fix the fracture or you can't fix the fracture. So which fractures need surgery? Well, either 2 or 3 with block to forearm rotation. That's all that boils down to. And then, of course, all your fracture dislocations. Okay, we decide to fix it. What are we trying to achieve? From what we just heard, you need to restore radial capillary contact and restore that anterior buttress to provide stability against valgus and posterior translation. If it is one mantra I tell my residents about, it is a stable, congruent, and concentric joint, even if somewhat stiff, is still vastly superior to a fully mobile but unstable joint. You can always mobilize an elbow, but making it stable or stiffer is a Herculean task. This is what I keep in the OR, and one of the best-kept secrets in orthopedic surgery is the use of threaded K-wires. They are invaluable, and along with headless scantilated screws are great for radial head fractures. How do I decide what to do? It's very simple. If it's less than 3 pieces and the neck is okay, I'll fix them. If it's more than 3 pieces, I'll replace them. Usually the head screws are about 20 to 24 millimeters, so if you measure 36, that's a problem. I use general anesthetic and a block usually, supine on a hand table, as we'll show you. Always use a sterile tourniquet and put a large bump under the elbow. The approaches I use are predicated on what injury I'm dealing with. If it's an isolated fracture, I'll use the lateral approach using the Coker interval, and if it's a fracture dislocation, I'll use a posterior approach, again using the Coker interval. More often than not, I start off thinking I'm going to use the Coker, but the injury has already made my interval for me. You keep the forearm pronated at all times, and with the elbow at 90 and the forearm at neutral, if you draw a line extending from the lateral epicondyle to the crista supinatoris, that line indicates the location of the lateral and locolateral complex. So making a capsular incision anterior to that ensures that you won't injure it, and always proceed from deep to superficial. So there's the positioning with the bump under the elbow, a sterile tourniquet, and marking the incision. Now, how do you identify the interval between the ECU and the anchoneus? You almost never can identify it proximally. The best place to find it is distally, where you can identify a fat stripe, as shown in the top two pictures by the arrows. If you open that interval and then open the capsule based on the line that we just showed you, you identify the capitellum first. Knowing about the posterior interosseous nerve is critical, so the first three to four centimeters are really dangerous territory, beyond which the radial nerve is, posterior interosseous nerve is very close to the radial neck. Then we start off looking at what are the things we need to remember. Remember the osteopenia of the anterolateral quadrant. Therefore, the horizontal screws may not have purchase, and if by any chance you are long, you're going to impinge on the superior radial joint. Therefore, direct your screws slightly obliquely distally. Threaded key wires work like really, really small headless screws, if you cut them perfectly. Let's look at some cases. This is what I call the kissing lesion, where you have contiguous lesions of the capitellum and the radial head, and you can see that the plane radiographs do not identify the posterolateral blowout of the distal humerus. So I put in a headless screw, and as you can see, I didn't go obliquely enough. Not surprisingly, I didn't get enough purchase and was unhappy. So I put in the key wires that you see on the top left, and those are going obliquely and distally. Cut them flush, and then you get much better purchase. Fix the capitellum with screws from back to front, and the posterolateral blowout with a simple tension band, and you can get that kind of function. How about plating? Well, you have to follow the safe zone, which Garth Smith told us about. It's a 110 degree arc, which is based off the mid-lateral line with the forearm at neutral, 55 degrees in the front and 55 in the back, but you still have to be cautious about placing a plate there and going distally with a soft tissue dissection because of the posterior interosseous nerve. Try and get as many screws into the radial head, and locking constructs allow us that ability even better. If possible, and if necessary, you can get a screw from the plate into the head angled obliquely, which gives you much better construct. The bottom line is no matter what you do, if you fix a comminuted radial head, it better be stable enough for you to move it. Otherwise, it's a wasted exercise. Now, plates are not a panacea. They have their limitations. Distally, you've got to worry about the posterior interosseous nerve, and time and again, I was told as a resident, and I've heard people tell their residents that you can take the radial head up to the back table, reconstruct it, and put it back inside. No. We just talked about the vascularity. How can we be so full of hubris that we think we can take this dead piece of bone, stick it back in there, and expect the body to heal it? Not surprised it may work as a spacer, but really, is it going to heal? Let's look at this young lady here. Bad radial head, radial neck also. So we place the plate, and you can see in supination and pronation, following the safe zone principles, and it can give you a reasonable looking x-ray. Needless to say, she did lose some forearm rotation, despite using all the safe zone principles, and I find that it's almost inevitable that happens. When you put a plate, you almost always lose some rotation. How about radial head arthroplasty? I use this device. I have no relationship to them, but I just like it because of its simplicity and its modularity. You want to match the radial head size, the width, to that of the capitellum, remembering that it's elliptical. Go with the smaller diameter, so if in doubt, go smaller, and avoid overstuffing. How do you do that? Try to match the top of the radial head to the lateral edge of the coronoid. That usually works very well. I don't recommend this. If you need to put six screws in the radial head, that means there were many pieces. That radial head needs replacing. This is not my case. I inherited this recently from someone else. On the other hand, if you get something like this, where you get three distinct pieces, in addition to a radial neck fracture, even if the patient's a younger patient, you've got to replace it. You reconsider the radial head on the back table, find out which is the smaller diameter, put in a trial, get your radial head implant ready on the back table, and then stick it in and repair your lateral structures with Mitex suture anchors. Now, when I say lateral structures, I oftentimes will erase the muscles that come off the lateral ridge and repair them with Mitex. You'll get that appearance. You'll notice that the radial head matches the lateral edge of the coronoid. The width of the radial head matches that of the capitellum, and you can get that kind of function. How about fracture dislocations? This young man came to me nearly two weeks out. His dislocation was reduced, but he had no forearm rotation. So this was relatively straightforward. Now, this is the same person I inherited with the six screws in the radial head. These were the incisions. He had bilateral injuries, and these were the incisions. I have never found the need to make two incisions. I much rather do it through one. So I use a posterior approach, which allows me to raise large flaps, and then when you come to the cochlear interval on the side, in fracture dislocations, especially terrible triads, you'll find that the lateral epicondyle is completely bald. In up to 66% of these patients, there is some avulsion of the common extensor. The radial head sags because the lateral collateral complex is also avulsed. So you fix the radial head, again, with headless screws, and if you notice, these screws I managed to direct slightly distally to get better purchase. How about someone like this who comes to you with terrible triad three weeks out from a fall, dislocated? Well, we do the same thing, the posterior approach, and that circle on the right is showing you the rent that the injury made for me. So I just followed that. You take out the radial head. The bottom line is, you put back in how much you take out, and when you section the radial head, you do it right below the flare of the radial head. This is from Mike McKee's paper, and you can see that it allows you unlimited access to the coronoid. You take a whip stitch in the coronoid, and you pass it through drill holes, which come out dorsally, as shown by this wire, which is emerging on the fractured surface. I use just a simple suture passer to pass the stitches, and then replace the radial head. This is five years down the line, and you can see that he's already developed arthritic changes. The arrow shows the holes which are made for the pull-through suture for the evolution of the coronoid, but no pain and reasonable function. How about someone with a posterior montesia? This is a big guy. He's about 230 pounds, and he has the classic description by Jupiter of the anterior quadrilateral plate fragment, including the coronoid, so we fixed that and replaced his radial head. Again, taking great care to make sure that the radial head height matched that of the coronoid, and the width matched that of the capitellum, with that kind of an outcome. Interestingly, on the other side, he has a terrible triad, which was fixed by my partner, David Ring. So, if you had to look at the data, it reminds me of muddy waters, because the waters are muddy. There is no attempt at putting everything together. Patient populations are heterogeneous, but what I can conclude is that trying to maintain your native radial head is the best, because it's modulus of elasticity allows maintenance of capitellar cartilage, and prevention of capitellar erosion long-term. If you do indeed have a displaced fracture with a forearm block, repair can give you up to 100 percent, that's doubtful, good to excellent results, or replacement can give you 76 to 94 percent good to excellent results. So, restoring radial capitellar contact appears to be better than excision. However, and especially in fracture dislocations, however, you can excise a radial head, but we're not going to talk about it. If you do excise a radial head, expect to see arthritis in up to 81 percent, and after arthroplasty in up to 74 percent. Interestingly, this is 8 to 12 year follow-up for both of these numbers, but the number of patients who complain of pain, despite this radiographic appearance, is extremely small. So, I'm going to stop there. I think we managed to achieve the goals that we set out initially. I'm just going to make a shameless plug and invite you all to a conference that we'll have in Houston in November, promises to be an excellent and exciting place for learning. So, please come, we'd love to have you with us. Thank you.

Shoulder and Elbow Fractures

Fix Versus Replace

Anatomy of Radial Head

Fracture Treatment Techniques

Radial Head Arthroplasty