Does using the hip joint distance (x-ray) as an input change the kinematic, kinetic output and is this clinically relevant?

There are a number of methods for determining the centre of the hip joint (HJ). The most common are regression equations or functional methods. In individual cases, however, we do not know how well the HJ centre is actually determined. Several papers present Harrington's regression formula as the best choice (Harrington et al., 2007; Kainz et al., 2015; Peters et al., 2012). If an image of the pelvis is available, the HJCD can be determined from it, and this can be used to optimise the determination of the joint centre in the regression formula. Does using the hip joint distance (x-ray) as an input change the joint parameters? A retrospective analysis of the gait laboratory database identified patients who had a calibrated radiograph and a 3D gait analysis. The calculated HJCD from the gait data was compared with that from the radiograph. In addition, the ASIS distance was calculated using the hip joint distance from the radiograph, and again the HJ position was determined using the newly obtained ASIS distance in the Harrington formula. The gait data were statistically compared using SPM analysis and the maximum distance between the two methods was determined over all curves. This was compared with the minimal detectable changes (MDC) (Wilken et al., 2012). Data from 349 patients (legs n=698, age: 4-22 years) with anterior knee malalignment without neuromuscular disease were analysed. HJCD correlations between radiographs and 3DGA values were 0.662 (p<0.001) using the Harrington method. The Bland-Altman plots for HJCD showed minimal differences using the Harrington regression formula. However, there were differences of up to 40 mm between the two methods of determining the HJCD. A comparison of the gait results with the two calculated equations shows significant differences (SPM). In most cases the differences between the two methods were negligible, but in some patients (legs) they were above the MDC value.Download : Download high-res image (85KB)Download : Download full-size image On average, the HJ distance from the radiograph and the gait analysis data were in good agreement, but not in every patient (up to 40 mm). The gait curves show significantly different results according to SPM analysis. In most cases the differences are below the MDC, but in individual patients there may well be clinically relevant differences in the results. Therefore, if pelvic imaging is available, we recommend using it to calculate the HJ centre.