Quantitative measures for evaluating the radiographic outcome of Legg-Calvé-Perthes disease

Background: Large sample sizes are needed if discrete outcome measures are used to perform outcome studies, whereas more practicable sample sizes are sufficient if continuous outcome measures are used. Recognizing the need to devise reliable continuous outcome measures for assessing the outcomes of treatment of Legg-Calvé-Perthes disease, we evaluated a method of quantifying the shape and size of the femoral head and the femoral-acetabular relationship from computer images of anteroposterior and lateral radiographs with the help of image-analysis software. Methods: Radiographs of 121 skeletally mature individuals with healed Legg-Calvé-Perthes disease were analyzed, and the sphericity deviation score, femoral head enlargement, femoral neck growth inhibition, and composite femoral congruency arc were calculated. Results: Each of these measurements was reproducible, with distinctly different values for hips in each of the Stulberg classes. Measures of sphericity and congruence (the sphericity deviation score and the composite femoral congruency arc) could very clearly discriminate between spherical and aspherical hips, and there was a strong correlation between these two measurements. Conclusions: The study shows that it is possible to have reliable quantitative measures of the outcome of Legg-Calvé-Perthes disease at skeletal maturity.