Hip resurfacing revision rates: radiological audit of risk factors

Abstract

We read the article by Ramisetty et al with interest. Nevertheless, we have some concerns with the methodology. The study was retrospective and so there was no standardisation of the radiographs. The authors’ method of measuring the neck angle has been demonstrated to be inaccurate1 in that as little as 7o of external rotation of the hip can result in more than 10o change in the apparent neck shaft angle (NSA) and up to 10o can be elicited by internal rotation. Wilson et al reported that when using a picture archiving and communication system to measure the NSA of the proximal femur, the reliability of the measurement is only ±6o for different observers even under the best conditions.2 In the study by Ramisetty et al the valgus angle of the hip resurfacings that were revised only varied from 4o to -6o. With an error of 6o in measurement alone, it is impossible to predict failure from this. Murray described that the cup inclination angle (CIA) was better measured with computed tomography as opposed to a plain anteroposterior (AP) radiograph.3 This is in part due to the variation of angles from which AP radiographs are taken. The correct calculation is the angle between the major axis of the ellipse and the transverse body axis. If the cup is anteverted, then the apparent inclination is an overestimate of the radiographic inclination: Tan(RI) = Tan(CI) / Cos2(OA). The greater the anteversion, the greater the overestimation.4 The topic of this paper is of interest but, unfortunately, due to the inherent errors highlighted above, the results do not translate into the ability to clinically predict failure.