No significant side-to-side differences in retropatellar load distribution using more than 7.5 years after isolated MPFL reconstruction: A CT-osteoabsorptiometry pilot study in nine patients.

Abstract

To evaluate long-term retropatellar load distribution after isolated medial patellofemoral ligament (MPFL) reconstruction using computed tomography-osteoabsorptiometry (CT-OAM). A retrospective single-centre pilot study (<i>n</i> = 9) was conducted, including patients with isolated MPFL reconstruction with a minimum follow-up of 7.5 years and without any contralateral knee injury or surgery. Bilateral patellae were analysed using CT-OAM. Maximum-intensity projections were subdivided into four patellar regions: medial facet (MF), central ridge (CR), medial portion of the lateral facet (MLF) and lateral portion of the lateral facet (LLF). High-density area (HDA) fractions were defined as >1000 Hounsfield units (HU). Operated and contralateral patellae were compared using paired Wilcoxon signed-rank tests with Holm correction and effect sizes were reported as rank-biserial correlation (<i>r</i>). Interregional differences were analysed using Friedman tests followed by pairwise Wilcoxon post hoc comparisons with Bonferroni correction. Sensitivity analyses were repeated using HU thresholds of 900 and 1100. No significant differences in HDA fractions were detected in any patellar region between the operated and contralateral knee (all adjusted <i>p</i> > 0.05; rank-biserial <i>r</i> = 0.02-0.08). Interregional variation was significant (<i>p</i> < 0.001), with highest HDA fractions in MLF, followed by LLF, CR and MF. Varying HU thresholds (900 and 1100) yielded similar regional rankings and unchanged inferential outcomes. At a minimum of 7.5 years after anatomically executed, isolated MPFL reconstruction, CT-OAM showed no statistically significant side-to-side differences in retropatellar subchondral mineralization patterns and the regional distribution remained consistent with physiological loading. While these preliminary findings provide no CT-OAM-based evidence of altered long-term mechanical adaptation or pathological medial overload, the small sample size (<i>n</i> = 9) precludes definitive conclusions about equivalence. CT-OAM demonstrated feasibility of characterizing load-bearing adaptations after patellofemoral stabilization surgery. Level IV, retrospective case series.