Abstract
BACKGROUND: The rate of failure of internal fixation for femoral neck fractures has remained largely unchanged over the past 30 years. The current study attempted to identify the controllable variables influencing the failure of internal fixation of femoral neck fractures. METHODS: The study included 190 patients aged from 20 to 65 with femoral neck fracture caused by low energy violent injuries (fall from standing height), who were treated with multiple cannulated screws over the period 2005-2019 at a single centre. Kaplan-Meier (KM) survival analysis was firstly utilized to evaluate the potential interaction between each variable and cumulative rates of reoperation. If P < 0.1 in KM survival analysis, the variables would be included in subsequent Cox survival analysis to explore the influencing need for reoperation of a femoral neck fracture. Next, all of the 190 patients were divided into perfect reduction group (Garden Alignment Index I) and imperfect reduction group (Garden Alignment Index II, III, IV). Propensity score matching (PSM) analysis resulted in 39 pairs. After the baseline variables were balanced between the two groups, cox survival analysis was utilized again to explore the variables influencing the need of reoperation of a femoral neck fracture. Finally, KM survival analysis was utilized to compare the cumulative rate of reoperation between perfect reduction (Group PR) and imperfect reduction (Group IR) as a subgroup analysis. RESULTS: Before PSM analysis, the mean age was 49.96 ± 12.02 years and the total reoperation rate was 17.40%. Cox survival analysis showed that only reduction quality was interrelated with the need for reoperation before PSM analysis and after PSM analysis. Kaplan-Meier cumulative reoperation rate was higher in Group IR than in Group PR after PSM analysis. CONCLUSION: To prolong the service life of the original femoral head, it is essential to achieve a completely anatomical reduction and maintain the reduction quality until the patient fully recovers.