Osteochondral lesion of the talus: is there a critical defect size for poor outcome?

BACKGROUND Identifying factors associated with favorable or unfavorable outcomes would provide patients with accurate expectations of the arthroscopic marrow stimulation techniques. PURPOSE To investigate the prognostic significance and optimal measures of defect size in osteochondral lesion of the talus as treated with arthroscopy. HYPOTHESIS A critical, or threshold, defect size may exist at which clinical outcomes become poor in the treatment of osteochondral lesion of the talus. STUDY DESIGN Cohort study; Level of evidence, 3. METHODS In sum, 120 ankles underwent arthroscopic marrow stimulation treatment for osteochondral lesion of the talus and were evaluated for prognostic factors. Clinical failure was defined as patients' having osteochondral transplantation or an American Orthopaedic Foot and Ankle Society (AOFAS) Ankle-Hindfoot Scale score less than 80. Linear regression analysis and the Kaplan-Meier method were used to identify optimal cutoff values of defect size. RESULTS Eight ankles (6.7%) required osteochondral transplantation, and 22 ankles (18.4%) were considered failures because of AOFAS scores less than 80, which indicated fair or poor results. Linear regression analysis showed a high prognostic significance of defect area and suggested a cutoff defect size of 150 mm(2) for the optimum identification of poor clinical outcomes (P < .001). Only 10 of 95 ankles (10.5%) with a defect area <150 mm(2) showed clinical failure, whereas in patients with an area >or=150 mm(2), the clinical failure rate was significantly higher (80%, 20/25). There was no association between outcome and the patient's age, duration of symptoms, trauma, associated lesions, and location of lesions (P > .05). CONCLUSION Initial defect size is an important and easily obtainable prognostic factor in osteochondral lesions of the talus and so may serve as a basis for preoperative surgical decisions. A cutoff point exists regarding the risk of clinical failure at a defect area of approximately 150 mm(2) as calculated from magnetic resonance imaging.