Minimal Effect of Implant Position on Failure Rates in Femoral Neck Fractures: Commentary on an article by Anne Marie Nyholm, MD, et al.: "Osteosynthesis with Parallel Implants in the Treatment of Femoral Neck Fractures. Minimal Effect of Implant Position on Risk of Reoperation".

Abstract

Commentary Nyholm and colleagues are to be commended for their study on reoperation rates among patients undergoing osteosynthesis of a femoral neck fracture with parallel fixation implants. The authors extracted their data from a national database and included 1,206 cases. An overall reoperation rate of 13% and a 1-year mortality rate of 18.9% were recorded. A thorough analysis of the risk factors for reoperation identified several patient-related factors, including age, sex, higher American Society of Anesthesiologists (ASA) score, and fracture displacement. The only surgery-related risk factors included insufficient fracture reduction, placement of the implants at an angle to the shaft of ≤125°, and perforation of the femoral head by the implant. Interestingly, many other surgery-related factors, such as the screw distance to the posterior cortex, screw distance to the calcar, screw tip-caput distance, number of screws, and parallel screw placement did not influence the risk of reoperation. This study allows for the conclusion that many outcomes in hip fracture surgery are determined by various patient-related factors, which the treating surgeons cannot control. The study also emphasizes that orthopaedic surgeons need to pay close attention to achieve the best possible reduction in order to minimize the risk of reoperation. One must also emphasize the somewhat unexpected negative findings; it appears that many variables regarding implant placement may be less important than widely assumed in the orthopaedic community. The study had strengths and limitations. The methodology was thorough, and the authors identified a large number of patients eligible for this study. Moreover, the data were extracted from a well-maintained national database that includes data from all Danish patients. Within this system, surgical reimbursements for hospitals are tied to data submission into this registry, which inherently increases the accuracy of the database. With respect to limitations, the results of this study are certainly subject to selection bias. During the study period, a total of 5,774 procedures for a femoral neck fracture were recorded in the database; the most common treatment was hemiarthroplasty (n = 3,110), as one would expect given current evidence and recommendations from the literature1. The question remains: which specific patients were treated with internal fixation during the study period and which specific patients were treated with arthroplasty? Similarly, there may be potential selection bias regarding the number of screws. The authors did not find any significant association between reoperation risk and the use of 2 versus 3 screws. However, it would be reasonable to assume that fixation constructs with 3 screws were applied more frequently for fracture patterns that were thought to be more unstable by the treating surgeon. An additional limitation was the wide age range of patients enrolled in the study (21 to 102 years); 2 very different patient populations were therefore encountered, including young patients with high-energy femoral neck fractures and elderly patients after ground-level falls. Finally, the authors report 1-year outcome data; although that is very appropriate when studying the reoperation rate for patients with femoral neck fractures, we must be aware that the actual reoperation rate was most certainly slightly higher than recorded. This becomes obvious when carefully looking at Figure 3 in the article, which suggests that the incidence of reoperation had not reached a plateau at the 1-year time point. Despite these shortcomings, the study is probably as good as any information currently in the literature regarding this topic. One main take-home point is that the outcomes for patients with femoral neck fractures are influenced by patient-related factors, on which surgeons have little influence. In addition, we can confirm that appropriate fracture reduction was significantly associated with a lower reoperation risk and should remain a major treatment focus for these injuries. Implant placement seems to be less important than most surgeons would probably expect. The authors convincingly demonstrated that a screw angle to the shaft of ≤125° may result in unfavorable reoperation rates. However, other technical details, such as screw distance to the posterior cortex, screw distance to the calcar, screw tip-caput distance, number of screws, and parallel screw placement, that we frequently pay close attention to during surgery and in resident education seem to play a less-important role in predicting reoperation rates. Finally, this study leaves room for additional questions that remain unanswered. Future studies may further investigate the advantages and disadvantages of 2 versus 3 screws, which currently remains subject to geographic differences, surgeon preference, and biomechanical considerations2.