Despite the favorable success rates of surgical procedures for various types of hip fractures, treatment of femoral neck fracture (FNF) can be a burden to surgeons, as the fracture itself is associated with many complications such as osteonecrosis of the femoral head, loss of reduction, malunion, and nonunion.1,2,3) Moreover, the treatment remains challenging when fixation failure occurs after primary internal fixation of FNF. In young patients with FNF, internal fixation with cannulated compression screw (CCS) or dynamic hip screw (DHS) showed favorable results and could be a treatment of choice.4) However, in the elderly patients, internal fixation showed unsatisfactory results in some studies and other studies reported that arthroplasty is favored for patients over 60 years of age with displaced FNF.5,6,7,8) Although patient’s age, initial degree of displacement, timing of fixation, reduction quality, and reduction methods are established risk factors for failure in internal fixation of hip fractures, the risk of failure of internal fixation, particularly after FNF, has yet to be widely reported.7,8,9,10) Hence, in this study, we tried to identify the risk factors for internal fixation failure after FNF.

This study was approved by the Institutional Research Ethics Committee at Dankook University Hospital (approval No. 2023-04-014) and the informed consent was waived by the Institutional Research Ethics Committee. Eighty-three patients who were diagnosed with FNF and underwent internal fixation between July 2002 and March 2017 at our center were analyzed retrospectively. Among the patients who were diagnosed as FNF and took the internal fixation, anyone who diagnosed as pathologic fractures, previous fracture history of lower extremity and had less than one year of follow-up period were excluded. Among FNF patients, Garden type 1 and 2 fracture patients were operated with CCS and Garden type 3 and 4 fracture patients were operated with DHS.5)

Furthermore, in this study, we arbitrarily classified failure after internal fixation of the FNF as follow our radiographic criteria: (1) loss of reduction, (2) nonunion. Loss of reduction was defined when reduced fracture fragment or fixation devices was displaced on radiographic examination. Nonunion was defined when three of four cortices of the fracture site had not shown continuity over six months on plain radiographs.

A total of 83 patients were selected on the basis of the inclusion and exclusion criteria of this study and the mean follow-up duration was 17.0±2.2 months. No significant differences were observed between the two groups (internal fixation failure group vs. internal fixation nonfailure group) in terms of the demographic data including age, follow-up duration, sex, location of the fracture in the hip, height, weight, body mass index, mechanism of injury, fixation method, and smoking status (Table 1). Among the 83 patients, 17 patients (20.5%) were defined as failure of internal fixation. During the follow-up, six patients (7.2%) showed loss of reduction and they were gradually progressed to varus hip resulting in post-traumatic arthritis. Eventually, they underwent total hip arthroplasty (THA) one year after surgery. Three patients (3.6%) were identified as having nonunion and underwent THA one year after surgery. Eight patients (9.6%) who showed loss of reduction were additionally diagnosed with AVN. Among them, four patients underwent THA, while the rest were decided to conservative treatment as patients’ request. When comparing the two groups according to the radiographic data, Pauwels’ angle and reduction quality showed significant differences, respectively (p<0.001; p<0.001) (Fig. 1). However, fracture level did not show the significant difference (Table 2). Furthermore, unlike the surgical time and estimated blood loss, significant difference was noted in the time interval from injury to surgery, especially whether surgery was performed within 12 hours after injury (p<0.001; Table 3) (Fig. 2).

Fig. 1
A 62-year-old female demonstrated pain in the left hip and diagnosed as femur neck fracture. (A) Preoperative pelvis anteroposterior (AP) plain radiographs showing 38 degrees of initial Pauwels’ angle. (B) Postoperative hip AP plain radiographs using cannulated screws showing good anatomical reduction quality. (C) Final follow up plain radiographs demonstrating well maintenance of reduction quality.

• Download Figure

Fig. 2
A 65-year-old female demonstrated pain in the left hip and was diagnosed with a femoral neck fracture. (A) Plain radiographs of preoperative pelvis in anteroposterior (AP) view showing displaced fracture of the femoral neck. Due to the general condition of the patient, surgery was performed after 15 hours of the initial trauma. (B) Plain radiographs of postoperative hip in AP view showing osteosynthesis with cannulated screws. (C) Plain radiographs of hip in AP view at follow-up demonstrating varus hip due to the avascular necrosis of the femoral head and neck. (D) Plain radiographs at final follow-up demonstrating a state of total hip arthroplasty due to the internal fixation failure of the femur neck fracture.

• Download Figure

Table 1
Demographics of the Fixation Failure and the Non-Failure Groups Suffering Femoral Neck Fractures

• Download Table

Table 2
Comparison of Radiographic Variables between the Fixation Failure and the Non-Failure Groups in Femoral Neck Fractures

• Download Table

Table 3
Comparison of Surgical Variables between the Failure and the Non-Failure Groups with Femoral Neck Fractures

• Download Table

Internal fixation is one of the conventional treatment of choice for undisplaced FNF in young and active patients.14,15,16) On the contrary, in elderly patients with FNF, internal fixation has been reported higher rates of revisional surgery, complication, non-union, delayed union, and poorer functional outcomes.17) Hence, some studies recommend the arthroplasty rather than internal fixation in elderly patients with FNF.17) However, it is judged that internal fixation is still has several advantages in elderly patients. It enables shorter operation time, less blood loss and less incidence of deep wound infection.17) Moreover, other studies have reported that age is not associated with mortality, rate of fixation failure, or the ultimate functional results.18)

Some studies reported the factors which influence the outcomes after internal fixation in FNF.18,19) Gregersen et al.20) found that poor reduction quality of FNF which is defined as greater than 5 mm of fracture displacement, outrange the Garden angle AP plain radiograph which the interval of 160°-175°, or greater than 20° of AP angulation is associated with high risk of revision surgery. This finding was consistent with our study that poor reduction led to a higher risk of treatment failure following internal fixation of FNF. Chang et al.21) reported the patient and surgical factors for outcomes of internal fixation of FNF. Though Chang et al.21) mentioned that preoperative Pauwels’ angle is not associated with the postoperative outcomes, in this study, lower preoperative Pauwels’ angle group showed favorable outcomes compared to the higher one.

Previous studies reported that the failure rate as 28.0%-53.5% and the AVN rate as 10.0%-46.2% after internal fixation of displaced FNF, respectively.7,14,16) Although in our study, the percentage of Garden type 1 and 2 were higher than that of Garden type 3 and 4, the overall failure rate was noted as 20.3% and the AVN rate as 9.6%, respectively. Jain et al.22) reported that early reduction and fracture fixation within 12 hours of injury after a displaced FNF could reduce the incidence of AVN. On the contrary, Papakostidis et al.23) argued that there were no significant differences in the AVN rate of the delayed internal fixation group. In the current study, the group in which internal fixation was performed within 12 hours after injury showed a failure rate of 5.4% and an AVN rate of 2.7%. In comparison, the group in which internal fixation was not performed within 12 hours showed a higher failure rate of 32.6% and an AVN rate of 15.2%. Hence, it is assumed that early fracture reduction reduces the rate of postoperative AVN, resulting in decreases the rate of fixation failure. Thus, performing surgery within 12 hours after injury could be a key factor for achieving successful results.

This study had several limitations. First, owing to the retrospective nature of the study, the factor of time was not precisely established. It would have been better to collect data from the onset of clinical symptoms to the time of surgery. Second, the number of cases was relatively small. A small sample size often leads to a type II error, but the adequacy of the present study was proven on the basis of the post hoc analysis results with a power of 80.7%. Third, in this study, we did not use a logistic regression analysis to find the risk factors for internal fixation failure in FNF. If a logistic regression analysis had been done, we could have been more accurately known the association between risk factors and internal fixation failure in FNF. Finally, we could not involve the bone mineral density data of patients due to the lack of osteoporosis evaluation. While Barrios et al.24) argued that osteoporosis has no statistically significant difference in internal fixation failure between failure and non-failure group, Spangler et al.25) stated that presence of osteoporosis could be the predictive factor of revision surgery in femur fracture patient. As many studies have argued different theory of osteoporosis as the factor associated with internal fixation failure, further data collection and assessment will be required in our next study.

Pauwels’ angle, reduction quality and time to surgery are major factors that can predict the possibility of internal fixation failure of FNF. Hence, early and accurate anatomical reduction is needed to decrease complications after internal fixation of FNF.