Discrepancy between morphological findings in juvenile osteochondritis dissecans (OCD): a comparison of magnetic resonance imaging (MRI) and arthroscopy.

ABSTRACT

PURPOSE: The aim of this study was to assess the reliability of preoperative MRI for the staging of osteochondritis dissecans (OCD) lesions of the knee and the talus in juvenile patients, using arthroscopy as the gold standard of diagnosis. METHODS: Sixty-three juvenile patients (range 8-16 years) with an OCD of the knee or the talus underwent arthroscopy after MRI. In 54/9 out of 63 cases, 1.5/3 T MR scanners were used. The OCD stage was classified according the staging criteria of Dipaola et al. Arthroscopic findings were compared with MRI reports in each patient. RESULTS: From the 63 juvenile patients, MRI/arthroscopy revealed a stage I OCD in 4/19 patients, stage II in 31/22 patients, stage III in 22/9 patients and stage IV in 6/6 patients. No osteochondral pathology was evident in arthroscopy in seven out of 63 patients. The overall accuracy of preoperative MRI in staging an OCD lesion of the knee or the talus was 41.3%. In 33 out of 63 patients (52.4%), arthroscopy revealed a lower OCD stage than in the preoperative MRI grading, and in four out of 63 cases (6.4%), the intraoperative arthroscopic grading was worse than in preoperative MRI prior to surgery. The utilization of the 3 T MRI provided a correct diagnosis with 44.4%. CONCLUSIONS: Even with today's modern MRI scanners, it is not possible to predict an accurate OCD stage in children. The children's orthopaedist should not solely rely on the MRI when it comes to the decision to further conservative or surgical treatment of a juvenile OCD, but rather should take surgical therapy in consideration within persisting symptoms despite a low OCD stage provided by MRI. LEVEL OF EVIDENCE III.