Measurements of acetabular cup position and pelvic spatial orientation after total hip arthroplasty using computed tomography/radiography matching

Abstract

This study presents a clinical validation of postoperative measurements of acetabular cup alignment following total hip arthroplasty (THA). The methodology was based on concurrent anatomic three-dimensional (3D) measurements of both the acetabular cup alignment and pelvic orientation, using an original CT/X-ray matching algorithm named Xalign. The subjects were 19 patients who had undergone bilateral THA using CT-based surgical navigation. All patients had postoperative pelvic CT scans and multiple antero-posterior (AP) pelvic X-rays. Using a proprietary software algorithm, the X-rays included in the study were matched with the corresponding postoperative CT scans. The goal of this method was to allow 3D anatomic pelvic and acetabular measurements on two-dimensional AP X-rays. The postoperative cup abduction, version and pelvic flexion angles were determined in three different ways: using CT images directly, applying the Xalign method, and finally by performing conventional (abduction only) measurements on AP pelvic X-rays. The cup orientation measured on CT images was taken as the ground truth. The Xalign measurement errors were defined as the difference between the CT cup values and those obtained by applying the matching method. The mean cup abduction error was 0.85° ± 1.3° (± standard deviation) and the mean version error was 0.01° ± 1.99°. Conventionally measured cup abduction ranged from 44° to 62° and correlated significantly (p = 0.001, r = −0.5) with pelvic flexion angle, proving the linear negative correlation between pelvic flexion and the error in conventional radiographic cup measurements. The Xalign method offered reasonable accuracy for cup orientation, and allowed cup and pelvic 3D anatomic measurements at different times.