Autologous chondrocyte implantation in the knee: mid-term to long-term results.

BACKGROUND: From 1998 to 2008, 1000 skeletally mature patients underwent autologous chondrocyte implantation for an osteochondral defect of the knee. We evaluated the functional outcomes in 827 of 869 patients who had undergone autologous chondrocyte implantation with Chondron or periosteum (ACI-C/ACI-P) or matrix-assisted chondrocyte implantation (MACI) and attempted to identify factors that influenced outcome.
METHODS: The age of the patient, the size and site of the osteochondral lesion, previous surgery, and the presence of early osteoarthritis were assessed for their influence on outcomes. Each factor was evaluated in a separate Cox proportional hazards model with use of hazard ratios (HRs), with 95% confidence intervals (CIs), describing the likelihood of failure for that particular factor. Outcomes were assessed with use of the modified Cincinnati score, visual analog scale pain score, and Stanmore functional score.
RESULTS: The mean duration of follow-up was 6.2 years (range, two to twelve years). The mean age was thirty-four years (range, fourteen to fifty-six years), with 493 males and 334 females. The average size of the defect was 409 mm2 (range, 64 to 2075 mm2). Four hundred and twenty-one procedures (51%) were performed on the medial femoral condyle; 109 (13%), on the lateral femoral condyle; 200 (24%), on the patella; and fifty (6%), on the trochlea. Kaplan-Meier survival analysis revealed that the unadjusted graft survival rate was 78.2% at five years and 50.7% and ten years for the entire cohort. No difference was found between the survival rates of the ACI-C/ACI-P and MACI techniques (HR = 0.948, 95% CI = 0.738 to 1.219, p = 0.678). There was a significant postoperative improvement in the function and pain scores of all three outcome measures (p < 0.002). Survivorship in the group with a previous cartilage regenerative procedure was inferior to that in patients with a previously untreated lesion, with failure five times more likely in the former group (HR = 4.718, standard error [SE] = 0.742, 95% CI = 3.466 to 6.420, p < 0.001). Degenerative change in any compartment had a significant detrimental effect on survivorship, with survivorship worsening as the osteoarthritis grade increased (Grade 1: HR = 2.077, 95% CI = 1.299 to 3.322, p = 0.002; Grade 2: HR = 3.450, 95% CI = 2.646 to 4.498, p < 0.001; and Grade 3: HR = 3.820, 95% CI = 2.185 to 6.677, p < 0.001).
CONCLUSIONS: Our study demonstrated an overall graft survival of 78% at five years and 51% beyond ten years following both autologous chondrocyte implantation techniques. Despite study limitations, our results demonstrate that autologous chondrocyte implantation for the treatment of osteochondral defects of the knee can achieve good results. LEVEL OF EVIDENCE: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.