Abstract
<title>Abstract</title> <p> <bold>Background:</bold> This retrospective study aimed to evaluate the association of concomitant patellar denervation and peripatellar osteophyte excision (patelloplasty) with patellofemoral function and patient‑reported outcomes after fixed‑bearing unicompartmental knee arthroplasty (UKA). <bold>Methods:</bold> A total of 376 patients who underwent fixed‑bearing UKA for medial knee osteoarthritis between January 2022 and December 2024 were included. Treatment allocation (patellar denervation + osteophyte excision,n=218 vs. no patellar intervention,n=158) was based on the surgeon's intraoperative discretion,introducing a risk of confounding by indication. Outcomes were assessed preoperatively,at 6 months,and at final follow‑up (minimum 12 months,mean 26.9 ± 8.5 months). The primary outcome was the Feller patellar score; secondary outcomes included the Forgotten Joint Score‑12 (FJS‑12),performance in rising from a chair and stair climbing. Between‑group differences were compared using t‑tests with effect sizes (Cohen‘s d). Analysis of covariance (ANCOVA) was used to adjust for preoperative baseline scores. <bold>Results:</bold> At final follow‑up,the intervention group showed higher Feller scores (mean difference 1.12 points,95% CI: 0.53–1.71; Cohen’s d=0.40),higher FJS‑12 scores (mean difference 2.00 points,95% CI: 0.98–3.02; Cohen‘s d=0.41),and better performance in rising from a chair (mean difference 0.34 points; Cohen’s d=0.36) and stair climbing (mean difference 0.37 points; Cohen‘s d=0.59) compared with the control group. No significant differences were observed in HSS or WOMAC scores. After adjusting for preoperative scores,ANCOVA confirmed an independent association between the intervention and higher final Feller scores (F=11.12,p=0.001; partial η²=0.029). No patellar complications occurred. <bold>Conclusions:</bold> In this retrospective non‑randomized cohort,patellar denervation with osteophyte excision was associated with modest improvements in patellofemoral‑specific function and high‑demand activities. Effect sizes were small,and causality cannot be inferred. Prospective randomized trials are needed for validation. </p>