[Variance of identification of femoral epicondyles in navigated total knee arthroplasty]

Abstract

Background: The aims of this study were the quantification of the accuracy of registration of the epicondylar axis (EA) in navigated total knee arthroplasty (TKA) and the identification of presumed factors influencing this accuracy.

Methods: A total of 32 navigated TKAs were performed and the surgical EA registered. Postoperatively, the difference from the surgical EA determined by computed tomography was calculated. Presumed factors influencing the accuracy were sex, preoperative malalignment, stability and range of motion, operated side, body mass index, and component size.

Results: The absolute error was calculated to be 1.4+/-1.3 degrees . Alignment according to the intraoperatively defined axes would have resulted in three outliers (>3 degrees malalignment). The operated side was the only factor showing a significant effect on the accuracy. The absolute error in left knee joints was calculated to be 0.9+/-0.7 degrees (max. 2.4 degrees ) and in right knee joints to be 2.0+/-1.5 degrees (max. 5 degrees, p=0.021).

Conclusions: The surgeon stood on the patient's right side in every case, so that right knee joints were operated from the lateral and left ones from the medial side. A medial position of the surgeon to the knee joint during registration of EA is recommended because it results in a higher accuracy than a lateral position.