Accuracy of planar anteversion measurements using anteroposterior radiographs

Abstract

Abstract Background Several methods using simple anteroposterior (AP) radiographs have been suggested for the measurement of anteversion of the cup component after total hip arthroplasty. Herein, we compared six widely used anteversion measurement methods using two different types of AP radiograph, the conventional pelvis AP and hip-centered AP radiographs, to identify the measurement method and the type of radiograph that would provide the highest accuracy and reliability. Methods We developed two custom-made bi-planar anteversion measurement models for the validation test. The models were designed for pelvis AP and hip-centered AP radiographs, respectively. The radiographs were acquired using the inclination angles of both models, changing from 10° to 70° at 10° increments. For each inclination angle, anteversion was changed from 0° to 30° at 5° increments. The measurements were obtained independently by two orthopedic surgeons blinded from each other’s measurements, using the methods of 1) Pradhan et al., 2) Lewinnek et al., 3) Widmer et al., 4) Liaw et al., 5) Hassan et al., and 6) Ackland et al. The measurements were repeated after 2 months. The accuracy, compared with that of the reference angle, and intra-observer and inter-observer reliabilities of each method were calculated. Results The highest accuracy was found when the method of Liaw et al. was used with hip-centered AP radiographs, which showed a difference of 1.37° ± 1.73 from the reference angle. Moreover, regardless of the type of radiograph, the methods by Pradhan et al., Lewinnek et al., and Liaw et al. showed excellent correlations with the reference anteversion. However, substantial differences were found when the methods by Widmer et al., Hassan et al., and Ackland et al. were used, regardless of the type of radiograph used. When anteversion was measured in an inclination between 30° and 50°, the method of Pradhan et al., when used with pelvis AP radiographs, showed the highest accuracy (1.23° ± 0.92°). We also found no significant difference in anteversions between the measurements made on pelvic and hip-centered AP radiographs. Both interobserver and intraobserver reliabilities were high for all the measurements tested. Conclusions The methods by Pradhan et al., Liaw et al., and Lewinnek et al. may provide relatively accurate anteversion measurements with high reliability, regardless of the type of radiograph.