Usefulness of epiphyseal quotient measurement on magnetic resonance images for outcome prediction in patients with early-stage Legg-Calve-Perthes disease.

Abstract

To investigate the usefulness of epiphyseal quotient (EQ) measurement for outcome prediction in patients with Legg-Calve-Perthes disease (LCPD), magnetic resonance (MR) imaging was performed in 26 patients with unilateral LCPD within 6 months of the disease onset and subsequently once every 3 months. All the patients were treated conservatively in our institute between 1990 and 1999. The mean follow-up period was 80 months. Clinical outcome was evaluated based on the plain radiograms of the femoral head at the final examination according to Stulberg's classification: the patients whose outcome was class I were regarded as 'excellent' (excellent group), class II as 'good' (good group), and classes III, IV and V as 'poor' (poor group). The excellent group consisted of 10 patients, the good group of 12, and the poor group of four. The EQ measured on MR images (MR-EQ) decreased remarkably in the poor group chronologically, whereas there was no decrease in the excellent group and a slow decrease in the good group. The mean MR-EQ at or before the sixth month (range, 2-6 months; mean, 4.9 months) was 86.2 (range, 76.8-94.8) in the excellent group, 78.3 (57.0-93.4) in the good group, and 67.4 (57.8-74.5) in the poor group (P<0.05 among the three groups). According to our results, because all patients having an MR-EQ at or before the sixth month of 75 or higher were classified in the excellent or good groups, it is thought that the ongoing method of treatment can be continued. On the other hand, since half of the patients having an MR-EQ at or before the sixth month of lower than 75 were classified in the poor group and had a high potential for a poor outcome, it is thought that these patients require strict containment therapy, and altering the surgical procedure could be considered depending on the particular patient. It was thought that the MR-EQ is a useful predictive factor of LCPD prognosis.