MATERIAL AND METHODS

20 patients scheduled for total or partial knee arthroplasty were included after giving their informed consent. There were 7 men and 13 women with a median age of 65 years (range, 55 to 90). The median coronal deformation measured by X-rays was 8° of varus (range, 5° valgus to 22 ° varus). The same navigation system was used for both invasive and non-invasive measurements, but the basic algorithms were adapted for the non-invasive technique. For the non-invasive technique, metallic plates were strapped on the thigh and the calf to allow arrays fixation (fig. 1). Coronal femoro-tibial mechanical angle (CMFA) in maximal extension without stress was recorded by the non invasive system. This non-invasive analysis was immediately followed by surgery, and the same angle was measured intra-operatively with the invasive system. Comparisons between non-invasive and invasive measurements were performed using a Wilcoxon test, after checking that their distribution followed a normal distribution, and an equivalence testing with limits of ±3°. The correlation between the two sets of measurements was analyzed using a correlation test Spearman rank. The analysis of the concordance of the two sets of measurements was performed using Bland and Altman tests. The significance level p was set at 0.05.