The effect of pediatric orthopaedic experience on interobserver and intraobserver reliability of the herring lateral pillar classification of Perthes disease.

An analysis was done of the effect of surgeons' pediatric orthopaedic experience on the classification of Perthes disease according to the lateral pillar classification described by Herring. Five observers with varied pediatric orthopaedic experience reviewed anteroposterior (AP) pelvis radiographs of 33 patients in the fragmentation phase of Perthes disease and classified each case on three separate occasions at least 24 hours apart. Frog-leg lateral-view radiographs taken at the same time were also classified using the same criteria based on the femoral head anterior column. Kappa statistics showed good agreement for intra- and interobserver reliability of classification for both AP and frog lateral radiographs. There were no statistically significant variations among the reviewers. Twenty-four percent of the cases had a classification one grade worse on the lateral radiograph compared to the AP view. The lateral pillar classification provides a reproducible radiographic technique for Perthes disease characterization independent of pediatric orthopaedic experience.