Standardized AP radiographs do not provide reliable diagnostic measures for the assessment of acetabular retroversion

Abstract

Diagnosis of acetabular retroversion is essential in femoroacetabular impingement (FAI), but its assessment from radiographs is complicated by pelvic tilt and the two-dimensional nature of plain films. We performed a study to validate the diagnostic accuracy of the cross-over sign (COS) and the posterior wall sign (PWS) in identifying acetabular retroversion. COS and PWS were evaluated from radiographs and computed tomography (CT) scans as the standard of reference in 50 hips of subjects with symptoms of FAI. A CT-based method using three-dimensional (3D) models was developed to measure the COS, PWS, true acetabular version, and pelvic tilt relative to the anterior pelvic plane. The new CT-based method aimed to eliminate errors resulting from variations in the position and orientation of the pelvis during imaging. A low level of agreement for COS and PWS was found between radiographs and CT scans. A positive COS strongly correlated with pelvic tilt. These results suggest that COS and PWS determined from anteroposterior radiographs are considerably limited by pelvic tilt and inherent limitations of radiographs. Their use as the sole basis for deciding whether or not surgical intervention is indicated seems questionable.