I'm Peter Fabrikant from Hospital for Special Surgery, and I want to thank Andreas and Aaron for inviting me to speak on management of cartilage-only fragments after patellofemoral dislocation, specifically with a focus on when and how to fix. These are my disclosures, and none of them are financial or related to this talk. I want to acknowledge Ben Hayworth from Boston Children's Hospital, who served as senior author on this paper that we wrote from which a lot of the concepts in this talk are taken. I've broken down this presentation into two parts. In part one, we're going to define what is a cartilage-only fragment, and we're going to discuss why we might consider fixing cartilage-only fragments to bone with a focus on outcomes. And then in part two, we're going to discuss assessment of these pieces intraoperatively, deciding when and how to fix, and reinforcing how to have a backup plan intraoperatively. So first, what is a cartilage-only fragment? Well, it's been defined in our paper and by others as a fragment where there's an inability to visualize the fragment on injury radiograph, so there's a lack of radiographic bone, and then an inability to discern bone on the fragment intraoperatively through gross inspection, so there's a lack of visualized bone on the piece. So just as a point of clarity, the biology between a traumatic chondral shear injury, which is what we're talking about today and can be seen on the left, is much different than the biology of an OCD loose body, and you can see the parent bone on the right side is under stress and has chronic remodeling, and so biologically this is much different than an acute traumatic chondral shear injury, which is what we're talking about today. So why even consider fixing a cartilage-only fragment? So we were inspired to study this in a larger setting based on a case report from Nakamura et al. from Japan, which was published in JBJS in 2004, where they reported a case report of an 11-year-old male who sustained a twisting injury to the knee and developed a chondral-only fragment, which they fixed. At a second look arthroscopy at six months for biopsy, both grossly and histologically, the cartilage had a normal appearance with a normal cartilage architecture, indicating that these can heal. So we set out to study this in a multicenter fashion by compiling patients from Boston Children's Hospital and Children's Hospital of Philadelphia, and we published these results in OJSM in 2018. We looked at 15 patients who had chondral-only loose bodies of the knee that were generated traumatically and fixed at an average of 11.2 days. The median age of the patients was 12.7 years old, and there were two reoperations, only one of which was for a dislodged fragment, which happened from a fall onto a flexed knee one week post-operatively. Of the 15 patients, nine of them had an MRI at an average of 12 months post-operatively, and eight of those nine showed normal or nearly normal cartilage and osteochondral junctions. This was pleasantly surprisingly similar to the results from a French study looking at 14 patients who had osteochondral fractures that were fixed at an average of 5.2 days, similar age of patients, and although they had three operations for patellofemoral instability, there were no reoperations for the fragment or for fragment non-healing. We were pleasantly surprised by these results, so we dug into the literature and found a basic science paper published in 1996 that was using a bovine osteochondral model and loaded the cartilage with a traumatic shear force, and they noted that in the juvenile bovine cartilage, there were these fingers of cartilage tissue that were penetrating more deeply into the cartilage and thus created a less well-demarcated osteochondral transition zone. So when these fragments were sheared off, there was likely some microscopic bone on the back of the piece. In the mature bovine osteochondral junction, it was a much sharper transition zone, and therefore the fragments that were cartilage-only were more likely truly cartilage-only and lacking even microscopic osseous tissue on the back of the fragment. These findings were replicated and published by a multicenter group, including Aaron Kreitsch in AJSM in 2019, where they looked at 10 male patients with a mean age of 14.6 years, and eight of those 10 patients were skeletally immature. They similarly had chondral-only fragments that were fixed at an average of 17.6 days, and they noticed no clinical failures at one year. So even though there's limited data from multicenter case series due to the rarity of these injuries, and those studies are heterogeneous with regard to the fragment, the fixation, and the follow-up, it is worth considering fixing these cartilage-only fragments as they can heal. And even if there's only partial healing, a second surgery for cartilage restoration may be more manageable in that there may be a smaller area that needs cartilage reconstruction. When assessing these lesions intraoperatively, I kind of look at three main factors. One is the location. So is it from the patella, the lateral trochlea, or the far lateral femoral condyle? Because this helps us understand if the piece is weight-bearing or less weight-bearing. What is the size and shape of the loose body? Is it deformed, or damaged, or swollen, or does it not fit into the site that it originated from? And what's the chronicity? A lot of authors have indicated that two weeks is somewhat of a cutoff, but from the studies presented previously here, certainly these can heal even after two weeks. I tend to favor fixing loose fragments that come from the patella and the lateral trochlea as these are more weight-bearing than the far lateral femoral condyle. The larger the fragment is, the more likely we are to fix it, especially as they get bigger than 1 or 1.5 centimeters. And it's more favorable to fix pieces that are acute and haven't had time to swell and still fit into the donor site. As far as implants go, it's dealer's choice. Absorbable implants, metal implants, and suture have all been described. I tend to favor implants that have a head, which can provide compression in the thin chondral fragment. Headless compression screws tend to be more challenging to use because there's less material in the fragment for it to grab and compress. And it's always important to have a backup plan. So depending on the size of the donor site, you could consider doing an Oates autograft or implanting a cellular cartilage matrix that's been rehydrated with bone marrow aspirate, or consider taking an autologous chondrocyte biopsy for later reimplantation. I just want to highlight these concepts with a few cases. Here in the first case, it's a 12-and-a-half-year-old male with a patellofemoral instability event that generated a loose body. The loose body was cartilage only. You can see here the donor site still has the intact subchondral bone tide mark with a sharp demarcation between the normal cartilage and the defect. Interoperatively, you can see the fragment here in the lateral gutter, which appeared healthy but had no bone on the back. You can see the donor site on the right side, both arthroscopically and through an open lateral parapatellar arthrotomy, which showed cartilage that still remained in the donor site, which is another sign of a cartilage-only fragment. As part of the surgical management of this injury, I prepared the bed and then pinned the piece back in place and fixed it with four headed absorbable implants, as you can see here. At four months postoperative, you can see an MRI showing restoration of the articular cartilage contour and healing of the fragment. Here's another case of a 13-year-old male who had a patellofemoral instability event and a loose body that was formed. This presented after a few weeks, and you can see the piece arthroscopically was essentially being chewed up underneath the lateral femoral condyle. On the right side, you can see the donor site, which was at the far lateral aspect of the lateral femoral condyle, and as outlined by the blue bracket there, is essentially completely contained and weight-bearing on the meniscus rather than the tibial cartilage. This was treated with simple excision, and you can see in a follow-up MRI the fibrocartilaginous infill appears healthy and that it largely bears weight on the meniscus rather than the tibial cartilage. This third case is that of a 15-year-old male with a patellofemoral instability event and a loose body. You can see it comes from the underside of the patella at the distal pole, a really junky-looking donor site that was cleaned up, and the piece was fixed back with two head and absorbable implants. You can see here at a follow-up MRI, complete restoration of the articular surface, and although there's a little bit of fissuring medially, as you can see on the axial slice, the cartilage has healed with an intact osteocondyle junction. So in conclusion, I think it's important to consider fixing cartilage-only fragments in the setting of an acute patellofemoral instability event, which generates a traumatic piece through a shear force, and again, understanding that these are biologically much different than OCD lesions, which are more chronic in nature and have abnormal bone underneath. Healing rates in kids are really comparable to osteocondyle fracture fixation because there's likely a microscopic amount of bone on the back of these cartilage-only fragments, In my hands, I like to protect these repairs by stabilizing the patellofemoral joint using both bony and or soft tissue procedures as we typically would, and I think that getting these to heal can certainly avoid the need for a future cartilage restoration, and even if it only partially heals, any future cartilage restoration has the potential for being a smaller and more straightforward procedure. Thank you again for the invitation to present.

cartilage-only fragments

management

patellofemoral dislocation

intraoperative assessment

fixation