Non-invasive determination of frontal plane lower limb alignment using motion capture technique – An alternative for full-length radiographs in young patients treated by a temporary hemiepiphysiodesis?

•High correlation between X-ray- and non-invasive marker-based leg alignment.•Including BMI into the regression model significantly improves the prediction model.•Bland and Altman plot shows a small systematic bias of 1.8° between both methods.•Increased BMI overestimates the valgus malalignment in the marker-based method.•Bland and Altman plots show better agreement for patients with BMI below 25. Multiple full-length standing anteroposterior radiographs are common practice to quantify the mechanical axis angle (MAA) in young patients with lower limb malalignment in the frontal plane treated with a temporary hemiepiphysiodesis. Is it possible to predict the MAA measured with gold-standard radiographs from a non-invasive method using the marker-based motion capture technique in a standing position and has an increased body mass index (BMI) a negative effect on this prediction? Forty-six children and adolescents with valgus or varus malalignment of the knee were measured several times during the treatment period. In total 175 data sets were evaluated in this prospective study. BMI was included into the linear mixed effect regression to detect the influence of this variable on the prediction model. Bland and Altman plots were obtained to examine methods’ agreement. The X-ray-based MAA highly correlated (r = 0.808, p