Higher incidence of patellar incongruence after under correction of pre-arthritic coronal alignment following medial unicompartmental knee arthroplasty.

PURPOSE: This study aimed to compare patellofemoral joint alignment of knees with restored pre-arthritic coronal alignment versus knees with under- or overcorrection from their pre-arthritic coronal alignment following medial unicompartmental knee arthroplasty (UKA) and evaluate the effect of patellofemoral joint alignment on patient-reported outcomes. METHODS: A retrospective analysis of 517 knees following robotic-arm assisted medial UKA was conducted. Postoperative mechanical hip-knee-ankle angle (mHKA) was compared to estimated pre-arthritic coronal alignment, using the arithmetic hip-knee-ankle angle (aHKA). Knees were considered restored to the pre-arthritic coronal alignment if mHKA was within 2.0° of aHKA (Group 1). Non-restored knees were subdivided into > 2.0° over correction (Group 2), or > 2.0° under correction (Group 3) from the pre-arthritic coronal alignment. Patellar congruence and patellar tilt angles were compared between groups. The Kujala score and Knee Injury and Osteoarthritic Outcome Score for Joint Replacement (KOOS, JR) were analyzed by group. Continuous data was compared using analysis of variance or Kruskal-Wallis tests. Chi-squared tests were used to compare discrete variables. Logistic regressions were conducted to estimate the probability of postoperative patellofemoral malalignment. RESULTS: Group 1 included 357 knees (69.1%), Group 2 included 57 knees (11.0%), and Group 3 included 57 knees (11.0%). The prevalence of postoperative abnormal patellar congruence (i.e. ≥ 17°) was higher in Group 3 at 40.4% (p = 0.009), with a higher odds ratio of having an abnormal patellar congruence angle (2.3, p = 0.01) compared to Group 1. The prevalence of postoperative abnormal patellar tilt (i.e. ≥ 14°) was comparable between groups (n.s.). At mean follow up of 4.4 ± 1.6 years, Kujala was worse in Group 3 (76.6 ± 17.4 compared to 84.0 ± 14.9 and 85.0 ± 14.4 in Groups 1 and 2, respectively; p = 0.006). Mean KOOS, JR outcomes were comparable among groups. No significantly inferior Kujala or KOOS, JR outcomes were observed in patients across all groups with abnormal patellar congruence or tilt angles. CONCLUSION: Knees with under correction from their pre-arthritic coronal alignment following medial UKA were associated with a two-fold higher probability of having postoperative patellofemoral incongruence, as well as inferior mid-term Kujala scores compared to knees with restored pre-arthritic coronal alignment and knees with overcorrection from their pre-arthritic coronal alignment. Patellofemoral incongruence alone did not negatively impact functional outcome scores. LEVEL OF EVIDENCE: IV, case series.