Long‐Term Efficacy of Arthroscopic Microfracture Combined With Autologous Collagen‐Induced Chondrogenesis for Knee Cartilage Defects: A 5‐Year Prospective Pilot Randomized Controlled Trial

OBJECTIVES: Cartilage defect of the knee joint is a common cause of knee pain and can result in significant functional disability due to its limited capability of spontaneous healing. Existing surgical options-such as microfracture, cartilage or chondrocyte transplantation, and joint replacement-remain limited by inconsistent restoration of durable hyaline cartilage. Autologous collagen-induced chondrogenesis (ACIC), which employs a collagen scaffold, has emerged as a promising single-stage alternative. Nevertheless, high-quality evidence evaluating its long-term efficacy relative to microfracture alone is still lacking. This study investigates the clinical effect of arthroscopic microfracture combined with autologous collagen-induced chondrogenesis for knee cartilage defects over a 5-year follow-up. METHODS: Twenty patients with knee cartilage defects were randomized to receive ACIC + microfracture (n = 10) or microfracture alone (MF, n = 10). Outcomes were assessed using Lysholm, VAS, and IKDC scores at baseline, 1 week, 3, 6, 12 months, and 5 years, alongside MRI-based MOCART scoring. Analyses employed linear mixed-effects models with multiplicity correction and effect size reporting. RESULTS: Both groups showed significant within-group improvements in Lysholm, VAS, and IKDC over time, but there were no between-group differences and no significant Group × Time interactions, indicating comparable functional recovery. In contrast, the MOCART score showed a significant long-term Group × Time interaction at 5 years (β = 53.1, 95% CI 29.0-77.2, p < 0.001), favoring ACIC + MF. At 5 years, ACIC + MF demonstrated a large structural advantage, although the unadjusted Mann-Whitney p = 0.0196 did not remain significant after multiplicity correction (adjusted p = 0.098). CONCLUSION: ACIC + MF resulted in superior long-term structural repair compared with MF, as reflected in MOCART scores, but this did not translate into superior patient-reported outcomes. These findings underscore the divergence between imaging-based repair and clinical function and highlight the need for prospective trials establishing anchor-based MCID and Patient Acceptable Symptom State (PASS) thresholds for MOCART to clarify its clinical significance. TRIAL REGISTRATION: Chinese National Medical Research Registration and Archival Information System: ChiCTR2400080094.