Abstract
BACKGROUND: Total hip arthroplasty (THA) has theoretical advantages and disadvantages over hemiarthroplasty for femoral neck fractures. Numerous studies have suggested equivalent reoperation rates between the procedures. The purpose of this study was to use the reverse fragility index (RFI) to evaluate the statistical robustness of randomized controlled trials (RCTs) reporting nonsignificant differences in reoperation rates between hip hemiarthroplasty and THA for femoral neck fractures. METHODS: Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, all RCTs that compared reoperation rates between THA and hemiarthroplasty for femoral neck fractures were identified. Studies that reported nonsignificant differences in reoperation rates were included. The RFI and patients lost to follow-up were calculated for each study. RESULTS: We identified 9 RCTs comparing THA and hemiarthroplasty for femoral neck fractures with nonsignificant reoperation rates. The overall median reoperation rate was 6.7%, 7.9% in the hemiarthroplasty group and 5.0% in the THA group. The median RFI was 3, indicating that had the outcome of 3 patients in 1 treatment arm been reversed, the studies' nonsignificant result would change to statistically significant (P < .05). Six of the 9 included studies had a loss to follow-up greater than or equal to the studies' RFI. CONCLUSIONS: The results of clinical trials reporting nonsignificant differences in reoperation rates between hemiarthroplasty and THA for femoral neck fractures would become significant if the outcomes of a few patients were reversed. The number of patients needed to reverse these studies' results was frequently less than those lost to follow-up; thus, the neutrality of these studies is fragile. LEVEL OF EVIDENCE: 1.