The overall failure rate of FNS in our study, with a mean follow-up of 11.8 months, was 9.3% (4/43 patients), and the results were similar or superior to those reported in previous studies (8.8–21%) 3,14,15. Even in patients with Pauwels type 3 fracture, which is a relative contraindication to MCS fixation, complete bone union was achieved without major complications such as nonunion or angular deformation in most patients (Table 1). With regard to the favorable clinical outcome of FNS, Stoffel et al. demonstrated the biomechanical strength of FNS compared to the existing MCS and DHS through a biomechanical study conducted in 2017 12. According to their study, it was confirmed that the FNS had a higher axial stiffness and better resistance to varus tilting and construct failure compared to MCS and DHS. A lower incidence of femoral neck and leg shortening was observed with FNS than that observed with MCS. Despite the minimally invasive mechanical properties, the biomechanical stiffness was similar to that of DHS. Recently, Moon et al. reported that FNS can provide stronger structural stability than DHS in displaced basicervical neck fractures, which are considered rotationally unstable 13. Since our study included data on patients with all types of FNFs within a certain period for which osteosynthesis was planned, our data would be insufficient to statistically prove the clinical superiority of FNS in certain criteria or specific fracture patterns. However, based on the favorable clinical results, although limited, we believe that FNS can be used easily and safely for various types of FNFs.
The major postoperative complications included fracture nonunion, osteonecrosis, and screw cut-out. The nonunion rate reported in previous studies varied from 6–33%, and the rate in our study was much lower (4.7%, 2/44 patients) 16,17,18. Similarly, post-traumatic osteonecrosis was reported in only one case (2.3%) during the follow-up period, and the result was also significantly lower than that reported in previous studies (4.5–11%) 3,19. Most cases of fracture nonunion are diagnosed within 10–12 months, and given our mean follow-up period was 11.8 months, the favorable outcome might be attributable to the superior biomechanical stability of the FNS. In contrast, osteonecrosis may be detected even 2 years or more after trauma. Since we did not routinely use more invasive imaging modalities (such as magnetic resonance imaging and computed tomography) in addition to simple radiographs, the actual incidence of osteonecrosis might have been overlooked.
Although the clinical significance of surgical timing is controversial, multiple factors such as patients’ sex, BMI, age, fracture type, time elapsed from injury to surgery, and the quality of reduction are known to affect the outcomes and risk of reoperation 3,15,17,20. Stassen et al. reported that the patient's age and presence of chronic lung disease were closely related to reoperation 15. Meanwhile, Davidson et al. revealed that the patient’s age, surgeons’ seniority, and proper placement of FNS were closely associated with reoperation 14. Our results (Table 3) found that prior stroke history, TAD, and bolt sliding length were related to reoperation. It was confirmed that both case identification numbers 13 and 14, presented in Table 2, had a prior history of stroke. Despite the history of stroke, the pre-injury ambulatory function of the patients was not severely limited (Koval’s grade 1 and 3), and the CCI score, which indicates the degree of patients’ morbidity, showed no difference compared to the group of patients that did not undergo revision surgery. Therefore, clinical significance of patients’ stroke history should be validated through further detailed analysis.
Radiologically measured TAD and bolt sliding length were also the major risk factors for reoperations. First, the TAD exceeded 25 mm in three out of four patients who underwent reoperation, and the TAD in the other case was 24.9 mm, which was also near the 25 mm value. Only a few studies have studied the relationship between FNS and TAD, and the results have been inconsistent. Jung et al. reported that the length and position of the bolts play an important role in obtaining biomechanical stability in Pauwels type 3 fractures 21. However, Stassen et al. and Davidson et al. did not observe a statistically significant association between TAD and revision surgery 14,15. Nevertheless, both authors emphasized that the precise central location of the bolt is of substantial importance in surgical prognosis. The bolt length is provided in 5 mm increments; therefore, some difficulties in finely adjusting the TAD exist. However, it is advisable to keep the TAD of the bolt within 25 mm and insert it in a position as central as possible. Meanwhile, bolt sliding is a process that occurs acutely during intraoperative inter-device compression and gradually during the natural healing process. The bolt allowed for sliding on the angular plate up to a maximum length of 20 mm. In our study, in the patients that underwent revision surgery, however, the degree of sliding was longer than the mean value and occurred very rapidly, especially in patients with bolt cut-out and nonunion due to reduction failure. Therefore, verifying the causal relationship may be limited despite a statistically significant correlation.
Finally, locking screw stripping occurred in 60% of the cases (3/5 cases) during implant removal after the complete bone union. This has not been reported in previous studies and various mechanisms have been considered to explain this phenomenon.
The footprint of the plate in contact with the lateral femoral cortex, which secures angular stability, is small, and only one or two 5.0 titanium locking screws support the load. The excessive stress applied to the thread between the plate and screw head may result in mechanical locking or jamming of the threaded head of the screw, which may cause shear failure of the screw head. Another hypothesis is that the location of the locking screw insertion may also be related to this phenomenon. The distal locking screw secured the plate by obtaining a bicortical fixation at the level of the lesser trochanter of the proximal femur. In this area, the cortical bone is very thick, and the calcar femorale is located at the posteromedial junction of the femoral neck and diaphysis, making the actual working length of the thread much longer. Therefore, the torque used for screw removal exceeded the strength of the thread, and the thread pattern was destroyed during the removal process, making it impossible to release the thread.
Since implant removal is primarily considered in active young patients, the good bone quality and thick cortical bone in these patients may act as obstacles to implant removal. To avoid such unexpected difficulties during surgery, it is necessary to prepare any available devices, such as a reverse-threaded screw extractor set, diamond burr, or trephine drill. In some cases, a plate cutting tool may be helpful (Fig. 3). Meticulous attention must be paid not to spread metal debris that is inevitably generated during this undesired procedure. Moreover, notifying the patients before the surgery is critically important.
Based on our clinical experience with a mean follow-up of 11.8 months and other recent studies, we believe that the FNS can be safely and easily used for various types of FNFs. However, future large-scale randomized controlled studies are required to validate mid- to long-term clinical outcomes between DHS and/or MCS with FNS, as well as post-market surveillance regarding implant failure and screw stripping.
Our study is limited by its non-randomized design, lack of patient-reported outcome measures, and a relatively small number of participants. However, this study is significant because the main purpose of this study was to discuss the initial experience of using a new fixation device (FNS) and evaluate the radiological complications, such as nonunion, osteonecrosis, and screw cut-out. Furthermore, despite the small number of study participants, we identified several statistically significant factors related to surgical failure.