The FNTA and FNAA are completely different anatomical measurements [7–9]. First, the former is defined as the angle between the long axial plane of the femoral neck cross-section and the coronal plane of the proximal femur, and the latter is defined as the angle between the 3D axis of the femoral neck and the coronal plane of the femur. Second, the sizes of the two angles differ from each other. Third, the results reported in the literature using the 3D CT measurement method show that the FNAA is approximately 10°, while the FNTA is approximately 30° [7, 8]. Unfortunately, current studies often confuse the two angles [1, 6, 14]. In other words, the expression of the angles (femoral torsion angle and femoral neck torsion angle) is not standardized and consistent at present. For example, the expression of the FNTA was mentioned by Yin, Hartel, and Zhao, but in fact, it was actually the FNAA, according to the measurement method and results reported in their articles [1, 6, 14].
Many methods have been established to define the femoral neck axis. In the early stage, the axis of the femoral neck was determined by the anteroposterior and lateral centerline of X-ray or two-dimensional CT, but both methods were affected by the femoral position during fluoroscopy, and the axis was ultimately two-dimensional [3, 4]. Nakanishi and Yin [5, 21] searched for the layers including both the femoral head and the femoral neck on coronal slices of 3D CT images, and they defined the connecting line between the femoral head center and the femoral neck isthmic center as the femoral neck axis. However, this method was also affected by the spatial position of the femur. Bonneau et al. [22] first proposed the concept of the 3D axis of the femoral neck. However, the reconstruction of the femoral neck medullary cavity is complicated because of the special distribution of bone trabeculae in the femoral neck (Fig. 4). In our study, the actual 3D axis of the femoral neck was generated using a 3D method. The shape of the femur is not a standard cylinder, the femoral trochanteric medullary cavity is irregular, and the femur length and curvature differ between men and women [17]. Therefore, the present study adopted the method introduced by Hartel et al. [1] to determine the axis of the proximal femur. Based on the traditional coronal plane of the femur, the coronal plane of the proximal femur was created using the method of establishing a plane perpendicular to a specified plane through two points (details are provided in Sect. 2.3 of the Methods).
The FNTA of the isthmus that we measured was very similar to the values reported by Kate (30°) and Zhu (31.34 ± 2.08°), but these authors did not report the specific position of the femoral neck cross-section [7, 8]. Kate measured 1000 femur specimens in India, but the specific measurement method was not described in detail. Zhu et al. rebuilt the proximal femurs of 30 healthy adult volunteers and fitted the ellipse with the “concentric circle” method, but did not clearly define the position of the coronal plane of the proximal femur. Unfortunately, the lack of a definition in both of these articles significantly reduced the repeatability of their research methods. For the first time, the size of the FNTA at different positions (FNI and FNB) of the femoral neck was measured using 3-Matic software in the present study. The torsion of the femoral neck is not presumed to increase completely at one time from the FNB to FNI but may be increased gradually. The FNTAs at the FNI and FNB of the male patients are significantly greater than the female patients, which is of guiding significance for the treatment and posttreatment evaluation of patients of different sexes with femoral neck related diseases, such as the choice of the model of the internal fixation device. However, the FNTAs at FNI and FNB between left and right side were not significantly different, indicating that the anatomical morphology of the healthy side can be used as a reference for the treatment of the affected side in patients with femoral neck related diseases.
Three cannulated screws in parallel are currently still the first choice for femoral neck fracture fixation [13, 23]. The presence of a torsion angle directly affects the nailing point and screw configuration on the lateral wall of the greater trochanter. Therefore, the spatial distribution of the three screws should match the morphology of the transverse plane (including the FNTA) of the femoral neck isthmus as much as possible to abut the screws to the femoral neck cortex without iatrogenic penetration and to obtain the maximum occupancy effect of the three screws [9, 10, 13, 24]. Similarly, the screw hole design of the proximal femoral plate should refer to the FNTA. The attachment of the plate should be satisfactory while reducing the penetration rate of the femoral neck screw [25, 26]. Due to the presence of the FNTA in basilar part, the long axis of the FNB cross-section was not located in the coronal plane of the proximal femur. Thus, forward deviation of the opening was likely to occur in the operation, resulting in difficult prosthesis placement, proximal femoral splitting, and periprosthetic fracture. Postoperative complications such as anterior femoral pain and early loosening of the prosthesis are common. Therefore, the optimal opening point of the femoral medullary cavity during hip replacement should be the posterior position of the top of the femoral neck cross-section [10–12].
In the present study, correlation analyses and stepwise linear regression analyses were performed to determine the correlation between patient attributes and the FNTA as well as the degree of their influence. Height exerted the greatest effect on the isthmic FNTA and the iFNTA, which may be related to local muscle strength, as more muscle strength may be needed to coordinate the posture of a taller individual [27].
This study has one limitation: the patients in this study were relatively old. Thus, the reference range of the measured morphological parameters does not represent the overall population. Studies examining an expanded age group or comparing the data with findings obtained from other research centers are necessary to circumvent this limitation.