The Effect of Patellar Surface Morphology On Subchondral Bone Alignment When Matching Patellar Osteochondral Allografts (206)

Objectives: Recent research has shown that implanting a patellar osteochondral allograft with a non-matched surface morphology (i.e., Wiberg classification) does not create increased chondral surface deviation or circumferential step-off in the donor plug compared to the native patella. While much of the research on patellar osteochondral allografts has been focused on chondral surface matching, little has been done to determine if the subchondral bone alignment at the donor:native interface plays a role in graft healing, local force distribution, and long term success of the allograft transplant. Previous work in our lab demonstrated that even when the patellar cartilage surface was well matched, notable differences in subchondral bone alignment were observed. The purpose of this study was therefore to use surface contour mapping of subchondral bone to determine if differences in Wiberg classification play a role in the ability of donor patellar osteochondral allograft subchondral bone to align with the native patellar subchondral bone when treating osteochondral defects of the patellar apex. The hypothesis was that patellar surface morphology would have an effect on subchondral bone surface height deviation and circumferential step-off when performing osteochondral allograft transplants of the patellar apex. Methods: Sixty fresh frozen human patellae were acquired from a national donor procurement company. Twenty (10 Wiberg I and 10 Wiberg II/III) patellae were designated as the recipient and then nano-CT scanned. Each recipient was size-matched (within ±2mm tibial width) to both a Wiberg I and a Wiberg II/III patellar donor. A 16mm circular osteochondral “defect” centered on the central ridge of the patella was then created in the recipient patella. A randomly-ordered donor Wiberg I or Wiberg II/III plug was harvested from a homologous location and transplanted into the recipient. The recipient was then nano-CT scanner, digitally reconstructed, and superimposed on the initial nano-CT scan of the native recipient patella. After careful atraumatic removal of the first donor plug, the process was repeated using the other allograft plug. MATLAB was used to determine the root mean square (RMS) surface height deviation between the native and donor subchondral bone surfaces. Dragonfly 3D imaging software was used to measure the RMS subchondral bone step-off height at 3° increments around the circumference of the graft. Surface height deviation and circumferential step-off height were analyzed for the whole surface and by quadrant to determine if there were local differences. ANOVA was used to compare surface deviation and step-off heights between matched and unmatched grafts. Sidak’s multiple comparison test was used to complete sub-analysis between patellar graft quadrants. Comparisons were made between matched and unmatched grafts in terms of the RMS surface height deviation and step-off, as well as in the percentage of measurements that were more than 0.5mm, 1mm, and 2mm proud or sunken relative to the native surface. Results: There were no significant differences in RMS subchondral bone surface height deviation between matched and unmatched Wiberg plugs as a whole or by quadrant (RMS range = 0.69 to 0.97mm, p = 0.45 – 1.0). There was a significant difference in RMS circumferential step-off height between matched (1.14 ± 0.52mm) and unmatched (1.38 ± 0.49mm) Wiberg plugs ( p=0.015). The majority of these increased step-off measurements occurred in the lateral quadrant with lateral quadrant RMS step-off of 0.89 ± 0.43mm in matched grafts and 1.60 ± 0.78mm in unmatched grafts ( p=0.007). There was also a significant difference in the percent of step-off measurements greater than 2mm sunken in the lateral quadrant between matched and unmatched grafts (5.17 ± 20.87% matched, 24.5 ± 36.39% unmatched, p=0.028). There were no significant differences between matched and unmatched grafts for any other comparison using 0.5, 1, or 2mm cut-offs for circumferential step-off or surface height deviation. Combining all allografts, the respective proportion of surface deviation and circumferential step-off height measurements that were above the stated thresholds were as follows: 31% and 34% for a 0.5mm threshold, 15% and 21% for a 1mm threshold, and 2% and 8% for a 2mm threshold. Conclusions: While unmatched Wiberg patella osteochondral allograft implantation did not result in significantly different subchondral bone surface height deviations, there were significant differences in circumferential subchondral bone step-off heights. The majority of step-off height differences between Wiberg matched and unmatched osteochondral allografts occurred in the lateral quadrant. In comparison to previous data evaluating differences in the cartilage surface match in these patellar OCA transplants, the deviations and step-off heights in the subchondral bone identified in the current study were approximately 0.5mm greater than the differences in the cartilage surface. These findings therefore suggest there is greater variability in the alignment of the subchondral bone in these patellar osteochondral allografts than there is in the cartilage surface. Further investigation using finite element analysis modeling will help determine the implications of subchondral bone surface deviation and circumferential step-off on local cartilage:bone compression and shear force distribution. These studies may shed light on the mechanisms of failure in patellar osteochondral transplants and may help to better understand the contribution of subchondral bone alignment in OCA healing and long-term outcome.