Arthroscopic treatment of osteochondral lesions of the talus with correlation of outcome scoring systems.

PURPOSE: The goal of this study was to perform a retrospective review of arthroscopically treated osteochondral lesions of the talus (OCLTs) to determine their outcome and to analyze the correlation of 3 subjective outcome scoring systems for the ankle.
METHODS: Between 1985 and 1999, 37 arthroscopic debridement or subchondral drilling procedures were performed on 33 ankles to treat OCLTs. The charts of these patients were reviewed for general demographic, preoperative, surgical, and postoperative information. Patients were sent a questionnaire that included 3 subjective outcome scoring systems: the Martin Score questionnaire; the Berndt and Harty scale; and the Single Assessment Numeric Evaluation (SANE) question. Twenty-eight people responded to the questionnaire. Twenty-four patients reported a history of trauma. There were 17 lesions on the medial aspect of the talus, 10 laterally, and 1 centrally. The lesions were classified according to their arthroscopic appearance. There were 7 stage I, 2 stage II, 10 stage III, and 9 stage IV lesions. After all data were analyzed, the Pearson product-moment correlation coefficient (r) and correlation of determination (r2) were performed among the 3 outcome scales.
RESULTS: The mean follow-up time was 66 months (range, 6-169 months). Outcome results varied according to the scoring system. Using the Berndt and Harty scale, 13 (46%) patients had good, 13 (46%) fair, and 2 (8%) poor results. According to the Martin Score, 1 patient (4%) had excellent, 10 (35%) good, 9 (32%) fair, and 8 (29%) had poor results. According to the SANE score, 5 patients (18%) had excellent, 8 (29%) good, 9 (32%) fair, and 6 (21%) had poor results. Seventeen patients reported some form of pain at follow-up examination. Analysis of variables, including age at operation, stage of the lesion, length of follow-up, and previous surgery revealed no statistical significance. All 3 scales showed positive correlations. Based on these outcome data, the Berndt and Harty scale showed good correlation with both the SANE and the Martin outcome systems (r =.81 and r =.69, respectively). The Martin Score showed moderate correlation with the SANE outcome system (r =.57).
CONCLUSIONS: The outcome results for our patient population were not as successful as has been previously reported. Our analysis showed positive correlation among the 3 subjective outcome systems. However, it remains difficult to compare these data with previous results secondary to the variety of outcome measures employed by previous reports in the literature.