Discussion
The results of our study demonstrate that ballistic femoral neck fractures had very high complication rates and poor outcomes. With a minimum follow-up of 1 year, 5 (71%) out of 7 patients with ballistic femoral neck fractures failed, with all 5 of those failures being non-union. Of those five failures, one patient also had avascular necrosis and was converted to a THA. Although, the outcomes were not significantly different between cohorts, likely due to small sample size, this study provides insight into the high complication rates of young adults ballistic femoral neck fractures relative to those with closed blunt injuries. Notably, the 50% failure rate of our blunt-injury cohort is similar to the 45% failure rate of young femoral neck fractures receiving ORIF as reported by Collinge et al.13 Similar to this study, Collinge et al’s definition of failure included non-union, osteonecrosis, and any secondary major reconstructive operation.13 However, they also included malunion, which was defined as vertical or femoral neck shortening of ≥10 mm, which was not assessed in this present study.13
The non-union rate of our ballistic femoral neck fractures is greater than the non-union rates reported in previous studies examining these fractures. Zhang et al reported an overall non-union rate of 9% in 69 patients with ballistic hip fractures.7 However, their cohort also included peritrochanteric fractures, and thus the failure rate of their ballistic femoral neck fractures independent of other hip fractures is unclear. Interestingly, they reported that 65% of their ballistic fracture patients were active tobacco users, which is similar to our cohort. These findings suggest that the active tobacco use in this patient population may partially contribute to the high rate of non-union, as tobacco use has been identified as a risk factor for non-union in multiple studies.16 17 Similarly, in an abstract presented at the Orthopaedic Trauma Association 2020 meeting, Jo et al reported a non-union rate of 15% in 20 patients with ballistic femoral neck fractures, which is far lower than the non-union rate in our cohort of 83%.18 The discrepancy between Jo et al’s non-union rate and the non-union rate in this study may be due to the fact that we had only included patients in our outcomes analysis if they had a minimum 1-year follow-up or if they failed earlier than 1 year. Nevertheless, ballistic femoral neck fractures generally seem to have far greater rates of non-union than that of young blunt-injury femoral neck fractures, which have a non-union rate of 8% to 9%.11 19 Moreover, our displaced fracture failure rate of 71%, which were primarily due to non-union, is similar to the failure rate of Maqungo et al’s cohort.8 When examining only their patients with greater than 6 months of follow-up, 100% (4 of 4) of their ballistic femoral neck fractures required revision operation.8
We hypothesize that the high rate of non-union in ballistic fractures may be due to severe fracture comminution, which can preclude successful ORIF, and the high rates of active tobacco use in this patient population. Recently, Collinge et al demonstrated that young patients with femoral neck fractures undergoing ORIF have a technical error rate of 50%, which was associated with an increased risk of treatment failure, especially in displaced fractures.14 As ballistic femoral neck fractures can have far more comminution than blunt-injury fractures, one would expect that there may be a greater rate of technical errors in these patients secondary to an increased difficulty of achieving an appropriate reduction. Interestingly, in our cohort, the fracture reduction quality for patients with displaced ballistic femoral neck fractures with a minimum 1-year follow-up were graded as either excellent (50%) or good (33%) (online supplemental table 2), suggesting that technical errors regarding reduction may not have played a major role in the poor outcomes in our patients.
Given the high rates of failure in these injuries, there may be a role for evaluation of the femoral neck blood supply after these fractures to determine the feasibility of ORIF.20 21 Furthermore, the high rates of failure in these young patients suggest that primary THA may be a potential treatment option in these patients. Recently, Bell et al reported performing primary THA for a ballistic femoral neck fracture with severe comminution of the neck.10 Interestingly, the injury radiographs in their patient are similar to the injury radiographs of our patient shown in figure 1, which demonstrated severe comminution of the femoral neck that was irreducible. They demonstrate that at 2-year follow-up, the patient had excellent outcomes with full return to activity.10 Nevertheless, we were hesitant to perform primary THA in these ballistic fractures due to their young age and the need for future revision, as studies have demonstrated 10-year THA revision rates to be about 10% in young patients.22 23 Additionally, younger patients will likely need to undergo multiple revision procedures, with each revision having shorter survival periods.24
Limitations
A major limitation of this study is sample size, as there were only 14 ballistic injuries over a 5-year period between three institutions. This small sample size also precluded us from performing multivariable logistic regression to adjust for potential confounders regarding failure and non-union. For instance, although we had observed that a large proportion of patients with ballistic fractures had active tobacco use (58%), we were unable to draw conclusions regarding its effect on the risk of failure or non-union. Additionally, the retrospective nature of this study, which included the grading of fracture reduction by the treating surgeon, precludes us from excluding assessment bias, as surgeons may tend to underestimate the number and severity of their own complications in chart documentation. Furthermore, the limited 1-year follow-up of about 50% in both cohorts limits our ability to form conclusions regarding avascular necrosis, post-traumatic hip osteoarthritis, and conversion to THA. Nevertheless, the high rate of non-union and failure in our cohort, despite the relatively limited follow-up, suggests that ballistic femoral neck fractures are devastating injuries with poor outcomes after ORIF.