Prognostic evaluation of Legg-Calvé-Perthes disease by MRI. Part I: the role of physeal involvement.

Twenty-eight unselected hips affected by Legg-Calvé-Perthes disease (LCPD) in 24 patients from 1987 to 1994 had magnetic resonance imaging (MRI) performed. The patients were retrospectively reviewed at a mean follow-up of 5.4 years. An interobserver blind analysis was made to establish the reliability of four MRI parameters: extent of epiphyseal necrosis (EXT), lateral extrusion of the femoral head (LAT), physeal involvement (PHY), and metaphyseal changes (MET). The interobserver analysis resulted in a good reliability for all MRI parameters (concordance, >80%; K index, >0.45). A statistical correlation study (Spearman test) was then done between each MRI parameter and the condition of the hips at follow-up evaluated by Stulberg class and a personal scored system (total score) of clinical-radiographic condition. All MRI parameters appeared well correlated to the Stulberg class and to the total score (S > 0.66; p < 0.05). PHY resulted the strongest correlated parameter (S = 0.84 for Stulberg class; S = 0.91 for total score). Finally our statistical correlation study demonstrates physis involvement to have a high predictive value in LCPD; therefore it can be assumed as the main risk factor in formulating prognosis.