Improved accuracy of CT based measurements for anterior prominence of acetabular prosthesis using a novel protocol based on anatomical landmarks

Abstract Background Anterior hip pain after total hip replacement (THR) can be caused by iliopsoas impingement (IPI). Validation of CT measured cup version is well published; however quantification of the anterior cup prominence has not been validated. Therefore, we did an in vitro study aiming to: (1) assess the accuracy of the current method of measuring the prominence of the acetabular component using a dry bone model; (2) develop a CT protocol based on bony anatomical landmarks that improves the accuracy of these measurements. Hypothesis We hypothesise that utilising reproducible anatomical landmarks on the acetabulum will improve the accuracy of measurement of anterior prominence at the psoas valley. Methods A dry bone pelvic model was implanted with bilateral acetabular cups. The cup was manipulated into ten different combinations of inclination and version, and anterior prominence directly measured in relation to the acetabular rim. A CT scan was performed for each position. Anterior prominence was then measured by two radiologists, first using standard methods (axial slice at level of centre of femoral head) and then again using a novel protocol (based on the position of the psoas valley in relation to the inferior medial acetabular notch). Results Mean errors between direct measurement of anterior prominence and radiologist reported measurements using standard protocol were 6.94 mm ± 5.24 (SD) (95%Confidence Interval: 3.25–4.27) and 5.14 mm ± 3.07 (95%CI: 1.9–2.5) with a tendency towards overestimation of prominence. Using an anatomical landmark based protocol; mean error was reduced to 3.0 mm ± 1.9 (95%CI: 1.16–1.53) and 4.3 mm ± 4.3 (95%CI: 2.67–351) on the first attempt, and to 1.7 mm ± 1.9 (95%CI: 1.15-1.52) and 2.1 mm ± 1.4 (95%CI: 0.87–1.15) on the second attempt. Concordance correlation improved from 0.43 (95% CI: 0.19–0.68) and 0.59 (95% CI: 0.44–0.79) using the standard protocol, to 0.88 (95%CI: 0.61–0.91) and 0.9 (95%CI: 0.56–0.92) using the novel protocol. Conclusions There is a difference in mean error and reliability between radiologically measured values for anterior prominence and values measured directly with present methods. The use of our novel protocol based on bony acetabular landmarks, significantly improves the accuracy of measurement. This protocol may improve reproducibility and the accuracy of this measurement and aid in the decision between tenotomy or revision of the acetabular component. Level of evidence III, in vitro study comparing diagnostic tools.