Hi, everyone. I'm Todd Lawrence from the Children's Hospital of Philadelphia, and this talk is on the basics of MPFL surgery in kids. I have no relevant disclosures related to this presentation. My goals for this talk are to review some of the relevant anatomical and biomechanical principles regarding the soft tissue restraints on the patella, and see how we apply those in our approach to MPFL reconstruction in immature patients. My hope is that you come away from this talk with the understanding that an MPFL reconstruction is a good option for providing a check grain to lateral dislocation in most immature patients, but that it can also work as a tether in some cases if your hand is forced, and addressing containment of the patella or overall alignment is not an option. Since the femoral attachment site has the greatest influence on restraint provided by the MPFL complex, techniques where the tension and tracking can be assessed dynamically before finalizing fixation make the most sense to me, and I'll explain how I do that. Finally, in immature patients, the femoral insertion of the MPFL is just distal to the physis, and so in patients with more than two years of growth remaining, if you're going to fix the graft in a tunnel in the femur, you should probably stay all epiphyseal. As you know, the patella is a sesamoid bone in the quad tendon. It doesn't usually begin to ossify until about age five, and like many joints, the patella and the trochlea appear to develop in conjunction with one another. The overall lower extremity alignment produces a slight angle between the pole of the quad and the patellar tendon, and this is referred to as the Q angle. When force is applied through the quad, this places a laterally directed force on the patella, and anything that increases the Q angle increases the lateral force on the patella. The patella tracks in the trochlea, but the patella does not usually fully engage the trochlea until about 20 to 30 degrees of knee flexion. Factors controlling patellar stability change throughout range of motion such that the overall net lateral force is balanced mainly by soft tissue restraints in early flexion and bony constraints later in flexion. Of note, patellar height and trochlear morphology determine the degree of knee flexion needed for this transition. This arrangement is kind of like if you're pulling a barge along a canal from the land on one side. Eventually the boat's going to run into the land unless the lateral force vector on the tow rope is balanced by an equal force pulling towards the other side. One way to balance the system is to pull from the opposite side with enough force to keep the boat in the middle, and in the knee we call this the MPFL. Another way to keep the boat in the middle would be to set up a trough in the canal that would constantly push the boat towards the middle, and in the knee we call this trough, the trochlea. The standard patella femoral joint is actually arranged something like this, with the boat sitting just at the edge of the area where the flat water turns into a trough. In this scenario the pull from the opposite shore doesn't have to be hard at all to keep things in line, it really just has to be there in case something else perturbs the system. This is the concept of a check rein. However, patella femoral joints come in lots of different shapes and sizes, some even with rocks in the middle of the canal trying to push the boat off course. These joints make the soft tissue restraints do lots of work to keep things straight. When addressing one of these joints biomechanically it always makes sense to me to pull the boat from within the canal and take out the rocks rather than just pulling harder from the opposite shore. However, sometimes you can't do that because of growth considerations and you have to revert to using your MPFL as a tether instead of just a check rein. With that concept in mind, it is no surprise that the MPFL is the most important restraint to lateral translation, accounting for about 50% of the restraint to lateral translation of the patella. The MPFL is a thin, flat ligament in the second layer of the medial knee. It attaches to the femur between the adductor tubercle and the medial epicondyle, and then spreads out as it approaches the patella, inserting onto the proximal two-thirds of the patella. More recent descriptions have noted a proximal extension of the ligament attaching onto the distal quad, and some call this portion of the medial quadriceps tendon femoral ligament, or MQTFL. However, anatomically, the MPFL and MQTFL are one continuous fan of tissue. Since the MPFL complex is always torn when you dislocate, and the MPFL is the primary restraint to lateral translation, most now agree that addressing it should be part of any surgical treatment plan. Multiple randomized studies have been published comparing early MPFL repair to conservative treatment, and this 2019 meta-analysis noted that there was very little advantage to early repair over conservative treatment, especially if you're looking specifically in adolescents. So, if repair doesn't work, what about MPFL reconstruction? Well, multiple techniques have been described for MPFL reconstruction, and no papers have really demonstrated the superiority of one technique over another, although there is some data to suggest that two limb reconstructions may have a slightly lower retair rate compared to single limb reconstructions. Biomechanical studies have shown that quad-based graphs more closely resemble the elasticity of the native MPFL, but fixation of the patella can provide more rigid lateral restraints. When it comes to graph selection, multiple studies have noted no difference in autograft versus allograft, and from a fixation standpoint, many authors are moving away from techniques that involve drill holes in the patella because of concerns over risks of patella fracture. These risks may actually be greater in younger patients with smaller patellas. The ideal femoral attachment is usually described as being a shotless point, which is a radiographic landmark taken from a perfect lateral image. In skeletally immature patients, this point appears to project just proximal to the physis on that lateral X-ray. However, because of the undulation of the physis, the actual insertion is usually described as being just distal to the physis. Biomechanically, the femoral attachment has the greatest effect on graft isometry. This is because of the cam effect of the distal femur. The native MPFL actually loosens slightly as the knee is flexed past about 20 to 30 degrees, placing thus, if you put a graft too proximal, it makes the graft too tight in flexion. On the other hand, placing a graft too distal means it's too loose in flexion. However, if you had to pick one or the other, because atrochlea provides much of the stability in flexion, you should probably pick a little bit too distal. This notion of going just a little bit distal was recently confirmed with a simulated epiphyseal and adductor-based tendon reconstruction assessment, where epiphyseal-based reconstructions provided a more isometric graft in early to mid-knee flexion compared to the other physio-respecting techniques. So that's probably the best one to use. Thus, in the case of skeletally mature patients, it's usually recommended that the femoral fixation point be within the epiphysis, just distal to the physis. Not only does this optimize biomechanics, it also prevents the attachment point from moving proximally as the patient grows. And in a patient with significant growth remaining, consideration should be given to staying completely epiphyseal with the femoral fixation. And when it comes to tensioning in MPFL, just a little bit is all that it needs, since tensioning over about 2 newtons increases contact pressures of the patella-femoral joint. Outcomes of isolated MPFL reconstruction, even in skeletally immature patients, are pretty good. With a recent systematic review of adolescent patients, noting a redislocation rate of only 2% to 4%, with good improvement in patient-reported outcomes. So if you're going to do an MPFL on an immature patient, how do you do it? Well, I start all patella-femoral reconstructions with an arthroscopy to assess the status of the articular cartilage and address any intraarticular pathology. This setup also allows me to assess the lateral retinocular tightness. If the patella will not centralize within the trochlea, a lateral retinocular lengthening is needed. This is a case where the patella initially tracked very laterally in the trochlea, almost right on the ridge. However, with just a little medially directed force on the patella, you can see it easily centralizes into the trochlea. I also assess for excessive lateral laxity and make sure that with significant medial force on the patella that it will not subluxate medially. Here you can see the patella getting just up to the medial trochlea ridge, but not dropping over it. So this would be a case that would not need any sort of lateral release or lengthening. The scope is then removed and a midline incision created. A sharp dissection is carried out down to the extensor mechanism, and small flaps are raised to either side of the patella. A medial parapatellar incision is then carefully made with a blade through layers one and two, leaving the capsule intact. I think this layer is easiest to find at the proximal, about third of the patella. But you know you're in the right layer when just gentle finger dissection deep to that layer that you've released opens up the interval between the layers easily. You've obviously gone too far if you get joint fluid. And then once you find that layer, finger dissection can be performed all the way down to the medial epicondyle. So the finger in the knee, down at the medial epicondyle, the skin incision for the femoral attachment can then be localized and made. If I'm making a standard femoral tunnel, angling proximal and anterior, I'll tend to make that incision slightly distal and posterior to the tip of my finger. However, for an epiphyseal tunnel like the one here, I like to stay slightly proximal. That helps with the angle of the tunnel. For this incision, I dissect bluntly actually down to level the fascia and then split that fascia in line with the surgical incision with a bovie. Basically, that is posteriors I can get it right onto my finger. Since it doesn't seem to matter, I usually give the patients the option of an autograft or an allograft. However, in ligamentously lax patients, I usually lean slightly towards the allograft. I prefer a two-limbed graft because I feel it recreates the anatomy a little bit better, but other single limb techniques are also well proven and widely employed at our institution. The quadriceps turn down technique is one of the most popular single limb techniques. To do this technique, you harvest a one centimeter wide strip of the superficial quad only, getting essentially as much length as you can before detaching it proximally. Then as the tendon comes down onto the patella, it has to be detached subperiosteally about halfway down the patella, stopping at an oblique angle so that it can be folded over medially. Reinforcing stitches are placed at the corners and the medial border of the patella, and then the end of the graft is whip stitched. My preferred technique, however, is the basket weave. For this procedure, a slit is made in the medial patella and the fibers of the MPFL. Full thickness slits are then made about every centimeter across the mid-aspect of the patella, and then use a periosteal elevator or sharp freer to elevate a full thickness subperiosteal sleeve of tissue in the alternate intervals between the slits. In the distal quad, I make two partial thickness slits just proximal to the superior pole of patella, and then make tunnels medially and laterally from there. You then take a semi-T allograft, usually about 28 centimeters in total length, and whip stitch the end three centimeters on each side with high-strength suture. The whip stitch ends are then woven through the tunnels and stitched in place at every weave point with high-strength suture. The whip stitch ends are then tied together at the lateral patella. This basically leaves a loop of graft at the medial patella, recreating both the MPFL and MQTFL components of the MPFL complex. For the femoral insertion point in an immature patient, I usually localize by feel first, trying to get the guide pin in the trough between the adductor tubercle and medial pacondyle. I start with an AP shot, making sure I'm distal enough to the physis so that a tunnel directed in this location would not go into the physis. I then leave the pin there and swing to the perfect lateral and fine-tuned localization of the starting point. Sometimes it does take a few back-and-forth rotations to get the pin in a good kind of compromised position. Once you have your starting point, you have to angle towards this magic triangle in the notch of the knee, like this 2019 paper notes. If you angle distal at about the slope of the lateral physio curve in the AP view, and then shoot for the center of the angle formed by Blumenstance lines and the physis on lateral, you're normally pretty good. I aim to be in the center of that triangular space on the lateral view by the time the pin is halfway across on the AP view. Once the pin is in, I then route the graft down to the medial knee deep to layer two. I wrap it around the pin and check patellar tracking, both clinically and with the scope. The amount of graft past the pin is then the minimum depth for the femoral tunnel. But I usually try to drill at least five to eight millimeters more if I can to give me room to fine-tune the graft tension and patellar tracking. Once I know I like the pin, I drill it out the lateral side of the knee. In an immature knee, often this would mean drilling out the articular cartilage, though. So if it looks like this might happen, I'll actually stop short with the guide pin, drill the tunnel, and then remove the pin and redirect the pin in the tunnel. And with this wiggle room, you can usually get the pin across the knee safely without damaging any cartilage. The pin can then be used to shuttle tensioning sutures for the graft across the knee. Tensioning sutures on the graft are used to pull the graft into the femoral tunnel, and the tensioning sutures are held out on the lateral aspect of the knee. My usual approach is to apply a good bit of tension to the graft at about 30 degrees of flexion, and then release all tension on those tensioning sutures, and cycle any multiple times, allowing the knee to take out any excess tension on the MPFL graft. I check it clinically, looking for one to two quadrants of lateral translation in full extension, but good tracking into the trochlea with no movement by 30 to 45 degrees. I'll also check with the scope at this point if there's any question. The beauty of this system is that I can back off the tension a bit if it seems too tight, or increase it a bit if there's still a bit too much play in the system. Once I'm happy with the tension and the tracking, I'll insert an interference screw alongside the graft in the femur to finalize the fixation. If you've done it right, you should be able to stick the scope back in, and as we've been watching here, see your graft indenting the medial capsule, and then watch the patella track centrally within the trochlea. Closure starts at the medial incision, with closure of the fascia at this location, and then take out redundant tissue at the medial patella with a double-row pants-over-vest closure. We mobilize these patients in a brace locked in extension, and limit weight-bearing post-operatively. We initiate knee range of motion and muscle activation early, but limit impact loading and running until about four months post-op, and you have to have appropriate strength to progress to that as well. Return to activities is usually achievable at about six to nine months. So with that, hopefully I've convinced you that NPFL reconstruction is a good option for providing a check range lateral dislocation in immature patients. Since the femoral attachment site has the greatest influence on the restraint provided by the NPFL complex, I would encourage you to consider a technique where the tension and tracking can be assessed dynamically before finalizing your fixation. And in immature patients with more than two years of growth remaining, stay all epiphyseal with ephemeral fixation, whatever technique you choose. Thank you very much for your attention.

MPFL surgery

soft tissue restraints

MPFL reconstruction

femoral attachment

patient-reported outcomes