Validation of computerized tomography antetorsion angle measurement of the femur

Objective: Several CT-procedures exist to determine rotational variances of the femur. However, solely on the basis of few CT-scans and without a three-dimensional reconstruction of the femur, it is impossible to define the exact course of the femoral neck axis. Looking for determination of the exact course of the femoral neck axis, auxiliary constructions need to be called upon. Aim of the study was to validate several existing CT-procedures and analyse their respective error rates.

Methods: Fourty-five femora were CT-scanned and subsequently reconstructed three-dimensionally in this study. Femoral anteversion was measured using three different CT-methods in each of these femora and in addition, the respective anatomical anteversion was determined in each femur. In order to test the reliability of such methods, flexion of the femora around the center of the femoral head as well as a varus deformity of the longitudinal axis of the femora were simulated and anteversion measurements were repeated in such simulated positions. Results were statistically analysed using SPSS.

Results: All tested CT-methods, when compared to the anatomical anteversion of a femur, showed differing anteversion-angles with statistical significance (p < 0.001). Equally, the results of the individual methods tested differed with significance. Due to a high correlation with the anatomical anteversion, all methods examined expressed trend of anteversion. Under simulated hip flexion, the "Ulmer Method" produced different values with statistical significance which were, however, without correlation to anteversion values in joint extension. With statistical significance, the "Essener Method" produced different anteversion values when the longitudinal axis of the femur was modified; however, these values highly correlated to anteversion values of femora with unchanged longitudinal axis.

Conclusions: Our results lead us to conclude that the "Essener Method" being an interindividual method and independent from positioning, is the most suitable procedure as it allows for the correction of errors with respect to anatomical anteversion.