In the present study, we investigated the mechanical properties of PFBN, DHS + DS and PFNA for the treatment of vertical BFNF using finite element analysis. Our study found that the primary stability of BFNF fixed with PFBN was significantly improved compared with DHS + DS and PFNA. Besides, the peak stress and stress distribution of PFBN and proximal femur were lower than that of DHS + DS and PFNA. PFBN demonstrated better ability to resist shearing force of BFNF, which may be crucial to improve the clinical outcomes of BFNF.

Currently, the selection of the most suitable implant for internal fixation in the treatment of BFNF is a subject of ongoing debate. Osteosynthesis using DHS or CCS is the standard care [16,17,18]. However, the postoperative femoral neck shortening has raised growing concerns. This was due to weakness of abductor muscles and inferior hip function resulted from the subsequent decrease of abductor lever arm. In addition, femoral neck shortening after BFNF can increase the risk of femoral head collapse [19]. Given that BFNF treated by sliding implants is not as stable as previous believed, various devices or techniques for length-stability of femoral neck have been developed [20,21,22]. The use of intramedullary fixation system has been suggested by some authors [23, 24]. In a retrospective clinical study, Guo [25] reported that intramedullary nails had a trend to decrease the femoral neck shortening compared with CCS in treatment of unstable BFNF (5.0% vs 14.29%). Other researchers have investigated the biomechanical properties between cephalomedullary nails and DHS. In a comparative study with synthetic femora, Imren et al. [26] found the PFNA has higher failure loads and possesses biomechanical benefits for fixation of unstable basicervical fractures compared with DHS. Nevertheless, Seyhan and colleagues discovered a heightened likelihood of encountering the following conditions within the PFNA group: reverse displacement of the proximal screw, proximal femur shortening, and a reduction in the varus angle of the proximal femur [27]. When managing unstable hip fractures in geriatric patients, PFNA demonstrates a mechanical failure rate of 7.5% [28]. This encompasses a range of complications, including implant cut-out with an incidence between 5.4 and 13% [29, 30], coxa vara occurring at a rate of 2.5% [31], and a 1% incidence of internal fixation failure [32]. Due to the substantial risk of a reverse wedge effect associated with PFNA fixation for basicervical fractures, this may not effectively mitigate neck collapse or prevent mechanical cut-out failure, even with the enhanced rotational stability provided by the helical blade [33].

The occurrence of various mechanical failures may be attributed to the mismatch between these implants and the proximal femoral anatomical structure and mechanical transmission. In their investigation of complications associated with internal fixation, Zhang et al. have suggested that a triangular stabilization structure (Chinese patents: ZL200920254063.4, ZL200920254062.x, ZL201120370391.8) could potentially reduce the likelihood of internal fixation failure, thereby contributing to the development of proximal femoral bionic nail (PFBN). The innovation of PFBN lies in its double triangle structure, composed of the supporting screw, fixating screw, and main nail. This design closely replicates the cantilever beam structure typically found in a normal proximal femur, resulting in a significant improvement in the postoperative stability following BFNF. The first component, known as the mixed triangle, is formed by the combination of the supporting screw, fixating screw, and the cancellous bone of the femoral head. This configuration significantly enhances the rotational stability of the hardware within three-dimensional space, ensuring a stable transfer of body-weight load to the junction of the supporting screw and fixating screw. The second component, referred to as the metal triangle, is constructed from the fixating screw, supporting screw, and main nail. The combined triangle and the main nail together create a stable cantilever beam structure that aligns with the anatomical structure and mechanical characteristics of the femoral neck. Moreover, the fixating screw is supported by both the main nail and the supporting screw, resulting in a double-pivot fixation. Consequently, this shortens the force arm, ultimately reducing stress concentration and enhancing fracture stability. Moreover, the tension stress of fixating screw can be significantly shared by the supporting screw due to horizontal placement. In our study, the PFBN group reduced stress on the implanted femur by 39.6% and 22.6% compared to the DHS and PFNA groups, respectively.

As shown in displacement distribution, the fracture section stability and overall construct stability were higher in the PFBN model than that of DHS + DS and PFNA models, which meant that PFBN had good ability to resist compression and tension force. The peak stress of DHS + DS and PFNA was 1.3 and 1.1 times higher than that of PFBN. We believe the supporting screw plays an important role in reducing the stress concentration of fixation screw in PFBN, which could decrease the incidence of screw withdrawal, cut-out, and hip varus. In elderly patients with BFNF, who often have osteoporosis, PFBN could potentially offer improved stability and support for early postoperative rehabilitation exercises. However, it should be noted that our study did not specifically investigate the mechanical performance of PFBN in an osteoporotic model, and further research is needed to validate these assumptions.

In addition, the peak stress of proximal femur in the PFBN model had the least value among all models, which demonstrated that PFBN had a decreased dependence on the integrity of femoral medial cortex. However, the peak stress and stress concentration of proximal femur in DHS + DS and PFNA models were located at the medial cortex of femur, which was different from that of PFBN. We consider that the results could be partially explained by the poor construct stability of the DHS + DS and PFNA. In a study on fracture morphology of BFNF patients, Collinge et al. [34] found 96% of cases had femoral neck comminution, which was located at the inferior in most cases (94%). Therefore, the PFBN is a suitable internal fixation for treating the BFNF, especially associated with comminution of medial cortex. Taken together, our study indicated that PFBN can not only enhance the mechanical stability of BFNF model, but also make improvement in the stress distribution of implant and proximal femur.

To our knowledge, our study is the first study to test the mechanical properties of PFBN in BFNF. However, there are some limitations to this study. Firstly, despite superior biomechanical stability of PFBN, we did not verify whether the use of PFBN would result in better clinical outcomes. Further randomized comparative studies are needed to verify the clinical benefit. Additionally, it is important to consider that patients with osteoporosis may exhibit lower mechanical properties than healthy patients, potentially leading to higher displacements in such cases.