Interobserver and Intraobserver Reliability of Three-Dimensional Preoperative Planning Software in Total Hip Arthroplasty

Abstract

BackgroundThe purpose of this study is to clarify interobserver and intraobserver reliabilities of the three-dimensional (3D) templating of total hip arthroplasty (THA). MethodsWe selected preoperative computed tomography from 60 hips in 46 patients (14 men and 32 women) who underwent primary THA. To evaluate interobserver and intraobserver reliability, 6 orthopedic surgeons performed 3D templating twice over a 4-week interval. We investigated intraclass correlation coefficients (ICCs) and percent agreement of component size and alignment, comparing morphological differences in the hip. Reproducibility was also compared between groups with osteoarthritis (OA) and those with osteonecrosis (ON). ResultsThe interobserver reliabilities for mean cup size and stem size were excellent, with ICC = 0.907 and 0.944, respectively. The value was significantly higher in the ON group than in the OA group. In the OA group, the reliability of cup size and alignment decreased in hips with severe subluxation. Percent agreement of stem size was significantly different between the shapes of femoral canal. For intraobserver reliability, the mean ICC of cup size was 0.965 overall, while the value in the ON group was significantly higher than in the OA group. The mean ICC of stem size was 0.972 overall. ConclusionComputed tomography–based 3D templating showed excellent reliability for component size and alignment in THA. Deformity of the affected joint influenced the reliability of preoperative planning.