Analysis of Computed Tomography–Based Technique for Measuring Femoral Anteversion: Implications for Measuring Rotation After Femoral Intramedullary Nail Insertion

Objectives To assess the reliability of the current computed tomography (CT) based technique for determining femoral anteversion, and quantify the prevalence and magnitude of side-to-side differences. Design Cross-sectional cohort study. Setting Academic trauma center. Patients We reviewed CT scans from 120 patients with bilateral full-length axial cuts of both femurs. Two hundred forty femurs with no fractures or other identifying features in their femora were included. Ten unique data sets were created to measure anteversion of the left and right sides. Main outcome measurements Intraobserver and interobserver reliability were calculated using intraclass correlation coefficients (ICC) and pooled absolute differences. The mean absolute difference between the sides was determined using a fixed-effects model. Results Interobserver reliability was high (ICC: 0.85, 95% confidence interval [CI]: 0.83-0.88). The pooled mean absolute magnitude of variation between reviewers was small at 1.6 degrees (95% CI: 1.4-1.8 degrees) per scan. The intraobserver reproducibility was high (ICC: 0.91, 95% CI: 0.88-0.93) with a mean error of 2.7 degrees (95% CI: 2.2-3.1 degrees) per repeat viewing of the same scan by the same person. The magnitude of side-to-side variation was 2.0 degrees (95% CI: 1.5-2.6 degrees). Twenty-one subjects (18%, 95% CI: 12-25%) had a mean side-to-side calculated femoral anteversion difference of ≥10 degrees, while 6 (5%, 95% CI: 2-10) subjects had a calculated mean side-to-side difference ≥15 degrees. Conclusions CT based femoral anteversion measurement techniques demonstrate good precision. Only 1 in 20 patients had side-to-side differences of 15 degrees or more. Level of evidence Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.