Automated Web Tool Measurement of Total Hip Arthroplasty Acetabular Component Inclination and Anteversion Angles

Periprosthetic hip dislocation after total hip arthroplasty is a devastating postoperative complication. It is often associated with suboptimal orientation of the acetabular component, characterized by the acetabular abduction and anteversion angles obtained from anteroposterior pelvic radiographs. We introduce a novel automated web tool to streamline the subjective and lengthy process of this manual measurement and compare it to manual human measurements.

One board-certified orthopaedic surgeon used the web tool to make automatic measurements of anteroposterior radiographs of 97 patients who underwent unilateral hip arthroplasty. Manual and web tool measurements included abduction angle and calculated anteversion angle by Liaw's method. Differences between manual and web tool measurements were compared with a paired t-test and Bland-Altman analysis.

There were no statistically significant differences between the average of manual measurements as compared to the web tool measurement in abduction angle (43.29 ± 7.05 vs 43.00 + 6.22, P = .85), anteversion angle (20.43 ± 7.62 vs 20.82 ± 7.37, P = .52), and ratio of the minor axis of the acetabular cup circumference in the AP radiograph to the total length of the acetabular head (0.42 ± 0.15 vs 0.44 ± 0.15, P = .18). The mean difference of average for abduction angle, anteversion angle, and ratio between the short axis of the transverse ellipse to the total length of the acetabular cup were -0.28, 0.39, and 0.02, respectively. Bland-Altman analysis for all 3 measurements displayed negligible systemic bias with random scattering.

Automated measurements obtained with a novel web tool are in strong agreement with the manually obtained ground truth measurements. The web tool helps to eliminate interobserver differences that arise with manual annotation. The web tool has the potential to streamline acetabular measurements with enhanced accuracy.