Juvenile osteochondritis dissecans of the knee: predictors of lesion stability.

BACKGROUND: Recent data suggest magnetic resonance imaging (MRI) is the best method to analyze the status of the cartilage and subchondral bone in patients with juvenile osteochondritis dissecans (JOCD). METHODS: MRI analysis of 122 knees and 132 JOCD lesions in 109 patients who underwent arthroscopic treatment for osteochondritis dissecans lesions of the knee between March 2003 and January 2011. RESULTS: Agreement between MRI and arthroscopic grading was 62.1%. MRI sensitivity was 92% and specificity was 55%. Positive predictive value of MRI was 33% and negative predictive value of MRI was 97%. In a multivariable logistic regression model, the odds of a unstable lesion on the lateral femoral condyle nonweight-bearing location were 15.7 times greater than the odds of an unstable lesion on the medial femoral condyle weight-bearing area (95% confidence interval: 2.6-95.7, P=0.003.) The odds of the lateral femoral condyle weight-bearing lesion having an unstable grade were also greater than for a medial femoral condyle weight-bearing lesion, but the results were not statistically significant (odds ratio, 1.70, P=0.349). CONCLUSIONS: A high T2 signal retrograde to the lesion may commonly appear with an early, stable arthroscopic grade lesion. MRI continues to be reliably sensitive to JOCD lesions and a good predictor of low-grade, stable lesions. However, MRI predictability of high-grade, unstable JOCD lesions is less reliable. Lesions in atypical locations, such as the nonweight-bearing surface of the lateral femoral condyle, more commonly present as higher, arthroscopic grade lesions. LEVEL OF EVIDENCE: Level IV, retrospective case series.