Purpose
Femoral neck anteversion (FNA) is an anatomical index to evaluate the pathophysiology of hip and patellofemoral joint diseases. In the past, it has been reported that the measurement angle of FNA differs depending on the measurement method of FNA. However, the limb position during the image capture were not explicitly described in previous studies, and the effect of the limb position on the measurement of FNA is not clear. The purpose of this study was to investigate the effect of the limb position on the measurements of FNA.
Methods
The DICOM date of 20 femurs of 10 patients (5 males and 5 females, with patellar dislocation, ACL injury, or knee osteoarthritis, average age 32.3 years old) who underwent plain computed tomography (CT) images were used. The Data were imported into software (Mimics 23.0 and 3-matic 14.0). The angle between the line passing through the center of the femoral head and the center of the femoral neck and the tangential line to the posterior femoral condyles on axial slices was measured as FNA angle (FNAA). First FNAA was measured using axial slices and the angle was defined as the original FNAA. Second, to accurately examine the effects of the limb position on the FNAA measurement, the axial cutting plane was tilted to arbitrary directions to mimic limb position change. The cutting direction of the axial plane was changed to virtually change hip flexion/extension and abduction/adduction angles from -20° to 20° at 5° intervals respectively. Further, the cutting direction was changed in combined directions by changing each direction from -10° to 10° at 5° intervals. FNAA was measured on each slice and the amount of the angle change from the original FNAA was calculated. The relationship between the hip flexion/extension or adduction/abduction angle change and the change in FNAA was examined using Spearman’s rank correlation coefficient.
Results
The average original FNAA was 17.6° (-13.5° to 40.7°). The hip flexion/extension change showed a strong negative correlation with the FNAA angle change (Correlation coefficient: -0.96, p < 0.001. Regression line: y = 0.73x-0.4 ). The hip adduction/abduction change was weakly correlated with the FNAA change (Correlation coefficient: 0.35, p < 0.001), while the average maximum difference was 9.7°± 3.3. The average possible maximum difference in the FNAA measurement by the combination of flexion/extension and abduction/adduction was 21.0° ± 4.9.
Discussion
The FNAA changed in associated with the change in cutting direction of the axial slice, suggesting that the measured value can differ depending on the limb position. This trend was particularly pronounced in the flexion-extension direction, with a 1° change in hip flexion angle resulting in a 0.73° change in FNA. Careful attention needs to be paid to limb position and slice conditions in order to correctly and consistently evaluate FNAA.