The Natural History of Non-operatively Managed Legg–Calvé–Perthes’ Disease

Abstract

Abstract Background The purpose of this study was to examine the evolution of Legg–Calvé–Perthes’ disease (LCPD) among children from British Columbia (BC), Canada who were treated non-operatively and to compare the results to a previously conducted study in India. Methods This was a retrospective review of patients treated non-operatively for LCPD in BC between 1990 and 2006 compared with a cohort from India. Demographic and treatment information were collected from medical records. Radiographs were assigned modified Waldenstrom, Catterall, Salter–Thompson and Herring classifications and intra- and interobserver reliability were assessed. We evaluated epiphyseal extrusion (EE) and metaphyseal width (MW), and assessed radiographs using the Mose and modified Stulberg classifications. Results 102 hips (90 patients) had radiographs available for evaluation. 95% of the BC cohort presented as Waldenstrom stages I and II, whereas, 90% of the Indian cohort presented as IIIa. Final EE was similar for both groups (BC 26.8%, India 27.3%) and final MW was 119% in both groups. Modified Waldenstrom and Herring classifications had substantial intra- and interobserver reliability, while Salter–Thompson and Catterall classifications had moderate agreement at best. Most hips were Catterall IV (80%) and Herring C (89%) for the BC cohort compared to only 44% and 43% of Indian hips, respectively. Most hips were irregular according to the Mose classification (BC 43%, India 52%) and aspherical according to the Stulberg classification (BC 78%). Conclusions We found similar radiographic progression and final radiographic appearances of LCPD in India and BC though differences in the distribution of the classification systems warrant further study.