ABSTRACT
PURPOSE: To determine whether bipolar lesions (BL) are associated with inferior clinical outcome following articular cartilage regeneration (CR) compared to unipolar lesions (UL). METHODS: A registry-based study, including patients undergoing isolated CR for focal knee cartilage lesions was performed. Lesions were considered UL or BL depending on the opposing cartilage. Propensity score matching was applied to eliminate potential confounders. Two groups comprising 119 patients with similar baseline characteristics were matched. The Knee Injury and Osteoarthritis Outcome Score (KOOS) at baseline, 6, 12, 24, and 36 months following CR served as primary outcome measure. KOOS improvement, reaching the minimal clinically important difference (MCID), KOOS subcomponents, and failure rates were calculated. RESULTS: Autologous chondrocyte implantation (ACI) was the most frequently performed procedure in both groups with 63.0% (BL) and 46.6% (UL). There was a significant difference regarding KOOS at 24 months between UL (76.39 ± 14.96) and BL (69.83 ± 18.83; P = .028), which did not exceed the threshold of MCID. No significant difference was detected at any other follow-up. KOOS improvement from baseline was lower in the BL group at all follow-ups and peaked at 36 months in both groups (UL [26.00 ± 16.12] vs. BL [16.63 ± 17.29]; P = .024). The failure rate in the BL group was higher at 8.2% (9/110) compared to the UL group at 3.9% (4/98) (P = .256). CONCLUSIONS: BL were associated with worse clinical outcome 2 years following CR compared to UL. However, both groups showed an ongoing clinical improvement up to 3 years postoperatively and a low failure rate. While inferior clinical improvement and a lower clinical response rate may be expected in BL patients, the observed differences do not justify excluding these patients from CR. LEVEL OF EVIDENCE: Level III, retrospective comparative prognostic trial.