Evaluation of the Optimal Femoral Fixation Site for Medial Patellofemoral Ligament Reconstruction in the Skeletally Immature Patient

Abstract

The aim of this study was to assess which femoral fixation site, distal to the physis, most closely emulates the native medial patellofemoral ligament (nMPFL) length relationship in the pediatric knee. Lateral knee radiographs were taken incrementally, from 0° to 120° of flexion. The femoral origin of the nMPFL and 3 other MPFL femoral fixation sites distal to the physis were identified (P1, 1 cm distal to physis along the posterior femoral cortex; P2, 2 cm distal to physis along the posterior femoral cortex; P3, 1 cm distal and 1 cm anterior). Virtual MPFL lengths were measured from all sites. Measurement reliability was assessed using intraclass correlation coefficients. Primary comparison between simulated MPFL length at native and femoral fixation sites was made using a repeated measures analysis of variance at 30° flexion. Bivariate Pearson correlation was used to assess the relationship between fixation sites and nMPFL through the range of motion (alpha=0.05). The intra-class correlation coefficient for intraobserver reliability of MPFL length measurement was excellent. Analysis of variance showed virtual MPFL lengths from proposed femoral fixation sites at 30° were significantly different from nMPFL length ( P <.01). The P1 MPFL had the least difference in means compared with nMPFL (−2.6±1.8 mm). Correlational analysis showed that P1 MPFL had the closest relationship with nMPFL compared with other femoral fixation sites ( R =0.91, P <.01). The P1 MPFL had the length change most comparable to that of nMPFL (23.9±3.1 mm vs 13.3±2.85 mm) compared with the other simulated MPFLs. Femoral graft fixation site 1 cm distal to the physis in the lateral view along the posterior femoral cortex closely represents the length relationship of the nMPFL. [ Orthopedics . 2023;46(2):108–113.]