Surgical treatment of FNFs mainly comprises closed or open reduction and internal fixation and primary arthroplasty. The internal fixation implants are CCSs, SHSs, DHSs and Hansson pin systems, while primary arthroplasty includes total hip arthroplasty (THA) and hemiarthroplasty[18]. Many factors including displacement of the femoral neck, presence of hip osteoarthritis, age, reduction quality, and stable internal fixation affect the surgeon's decision on the surgery method[15].
The ideal implant is considered as a conduct with the characteristics of strong fixation of fractures, prevention of femoral neck shortening, and avoidance of tilting and rotation of the femoral head. The FNS is a newly developed femoral neck internal fixation device in recent years. It contains a bolt, an anti-rotation screw and a femoral lateral plate. This plate has 1 hole or 2 holes for the standard 5.0 mm locking screw. After assembly, the FNS forms a stable structure with an angle of 130° in the femoral neck and femoral shaft. This stable structure combines the advantages of angular stability and minimally invasive surgical techniques, and allows the bolt and anti-rotation screw to slide together in the plate barrel to dynamically compress the fractured end, similar to a DHS. However, the surgical incision for an FNS is smaller than that for a DHS, thereby reducing soft tissue damage and protecting the blood supply. Therefore, FNS is considered the next generation of internal fixation devices for the treatment of FNFs[19]. It combines many advantages, including providing sufficient angular stability, reducing blood supply damage, dynamic compression and anti-rotation.
A few advantages of the FNS are due to its biomechanical characteristics. Fan et al. indicated that the internal fixation stress of FNS was higher than that of CCS in finite element analysis, which is approximately 1.6-3.0 times that of CCS in Pauwels III fractures at 50°, 60°, and 70°[20]. Especially at 70°, the displacement of the double-hole FNS was the smallest in the various groups. A biomechanical loading test conducted by Stoffel et al. evaluated the performance of FNS in comparison with DHS and CCS[21]. The experiment increased at a rate of 0.1 N/cycle until the termination criteria were achieved. The study found that cycles until 15 mm leg shortening and 15 mm femoral neck shortening in FNS were significantly higher than those in CCS. Similarly, Schopper et al. evaluated the biomechanical performance between FNS and Hansson pin systems in models of Pauwels II FNFs[22]. They indicated that the angular stability of the FNS provided superior resistance against varus deformation and performed in a less sensitive way to variations in implant placement.
Several comparative studies have reported the clinical outcomes between FNS and various internal fixations, indicating that the short-term efficacy of FNS is satisfactory[9-11, 15, 19]. However, FNS related complications, such as SFNS, ANFH, nonunion or delayed healing, and screw-out, have also been reported in these literature.
A total of 6 patients developed SFNS in this study. One of the characteristics of the FNS is dynamic compression. The precollapsed insertion allows the anti-rotation screw and bolt to slide in the maximum 20 mm packaging to meet femoral neck shortening during fracture healing. Similar to a DHS, femoral neck shortening after FNS placement is also a common phenomenon. The principle of fracture site compression utilized by surgical constructs may promote healing. However, SFNS is associated with worse patient-reported outcomes and objective functional measures. Most studies defined SFNS as a shortening of 10mm or longer in length. Both the retrospective FAITH trial[23] and the prospective SHOC trial[24] showed that SFNS after internal fixation was associated with inferior functional outcomes. Similarly, Zlowodzki et al. found differences in scores related to the degree of shortening, indicating worse functional outcomes with a greater degree of shortening[25]. Therefore, in this study, half of SFNS patients chose to receive THA, and their function was partially restored after surgery.
In this study, all EFFNSs existed in young and middle-aged patients (under 65 years old).The general treatment strategy for FNFs is widely considered to be that internal fixation is more suitable for young and middle-aged patients, while THA is more suitable for elderly patients with poor physical condition and bone quality. However, this strategy did not form a consensus. A meta-analysis by Xu et al. reviewed 2065 patients with FNFs from 17 case-control studies and found no association between age and osteonecrosis of the femoral head[26]. However, another meta-analysis by Slobogean et al. reviewed 1558 cases of FNFs from 41 studies, indicating that the high total incidence of ANFH in patients under 60 years old was 14.3%, and nonunion was 9.3%[7]. In addition, for elderly patients, the best functional results could be achieved in patients with a well-healed femoral neck without ANFH after urgent reduction and internal fixation of displaced FNFs[27]. Therefore, patient selection and surgical skill were important factors influencing clinical outcomes. We suggest that future studies emphasize the importance of surgical indications in the young and middle-aged patient populations. Furthermore, the promotion of surgical skills is key for avoiding postoperative complications and EFFNS.
Garden classification was an important parameter when considering surgery in FNF patients. There were a few studies supporting the Garden classification as one of the risk factors for ANFH after internal fixation of FNFs[12, 26, 28]. In this study, displaced fractures (Garden III and IV) in the failure group accounted for 90%, while in the heeling group, they accounted for only 75%. Although there was no statistical significance, EFFNS may be associated with the high proportion of displaced fractures. However, after regression analysis, we did not find support for Garden classification as a risk factor for EFFNS. The reasons may be the short follow-up time (6 months) of this study and the imaging diagnostic criteria of ANFH (cortical collapse of the femoral head) using X-ray films.All cases during follow-up did not undergo routine magnetic resonance imaging (MRI) examinations but only X-rays. Therefore, the manifestation of cortical collapse of the femoral head can only be seen on X-rays of Ficat III and IV patients[29]. However, MRI is unanimously considered the gold standard technique in the early stages, and is capable of detecting bone marrow changes such as oedema and sclerosis. Therefore, ANFH patients in Ficat I-II may be ignored during outpatient follow-up. Therefore, we suggest that regular review of MRI could be conducted to detect ANFH earlier for treatment. Moreover, a longer follow-up allows observation of the long-term clinical outcomes after FNS fixation.
In addition, we could not find the relationship between EFFNS and sex, BMI, injured side, injury mechanism, reduction method, Pauwels angle, femoral neck-shaft angle, or Pauwels classification.
This study had some limitations. First, it was a single-centre study. Therefore, there must have been some selective bias. Second, this study did not investigate other possible risk factors, such as weight-bearing time, anatomical classification, length of the screws, and preoperative bone quality, that may have significantly affected the prognosis of the patients. Hence, future studies need to be conducted to clarify these issues.