To evaluate anatomic factors of donor and recipient trochleae associated with osteochondral allograft matching for superolateral and central defects in a computer-simulated model.
Ten cadaveric trochlear specimens were utilized for this study. Three-dimensional (3D) point cloud models of the trochleae were created using a segmentation software program from CT scans (Mimics, Materialise Inc., Leuven, Belgium). Circular defect models were created virtually at both superolateral (18mm and 22.5mm) and central (18mm, 22.5mm, 30mm) locations. The mean articular step-off was calculated along the periphery of the graft. Demographic and anatomic factors measured include age, sex, laterality, lateral facet width, medial facet width, lateral condyle radius of curvature, trochlear groove radius of curvature, tibial width, trochlear length, and trochlear width. Sulcus angle and sagittal angle at the center of the defect and grafts for the central trochlear location were calculated by a custom-designed program.
On multivariate analysis, sulcus angle difference (β=0.019, p=0.002), sagittal angle difference (β=0.018, p=0.001), and lateral condyle radius of curvature (β=0.015, p=0.003) were significant predictors of articular step-off for 18mm central defects (R2=0.45). For 22.5mm central defects, sulcus angle difference (β=0.029, p<0.001), sagittal angle difference (β=0.025, p<.001), medial facet width (β=0.026, p=0.021), and lateral radius of curvature (β=0.013, p=0.025) were significant. Similarly, for 30mm central defects, sulcus angle difference (β=0.03, p=0.001), lateral radius of curvature (β=.032, p=0.004), and trochlear length (β=0.041, p<0.001), were significant predictors. For 18mm superolateral defects, no significant predictors were identified on multivariate analysis. For 22.5mm superolateral trochlear defects, medial facet width (β=.006, p<0.001) and trochlear length (β=.002, p=0.003) were significantly associated.
Minimizing mismatch in anatomic factors such as sulcus angle, sagittal angle, lateral condyle radius of curvature, and medial facet width may contribute to more optimal graft matching.