A comparison of cancellous screws in a sliding compression configuration and angle-stable sliding compression implants for internal fixation of femoral neck fractures in the non-elderly predominantly below 65 years: a systematic review and meta-analysis.

BACKGROUND AND PURPOSE: Internal fixation is the preferred treatment in the non-elderly with femoral neck fractures, regardless of fracture displacement. High complication rates are reported, in particular for displaced fractures. We aimed to compare cancellous screws with angle-stable sliding compression implants for internal fixation of femoral neck fractures in the non-elderly.

METHODS

A systematic search was carried out in Medline, Embase, Scopus, and Cochrane. The search results were screened by 2 reviewers using Covidence and assessed for risk of bias. All comparative studies were included. The studies reported at least 1 of the following outcomes: avascular necrosis, fixation failure/cut-out, non-union, any complication, reoperation, femoral neck shortening, or Harris Hip Score (HHS). Dichotomous outcomes are reported as risk ratio (RR) and continuous outcomes as mean difference (MD). All effect measures use a random effects model.

RESULTS

The search yielded 23 studies eligible for inclusion: 4 randomized controlled trials (RCTs) and 19 retrospective cohort trials, including 1,844 fractures. Only 1 study had low risk of bias. The results demonstrated no difference in RCTs alone. Analysis of all studies showed superior outcomes in favor of angle-stable sliding compression implants for fixation failure/cut-out (RR 0.54, 95% confidence interval [CI] 0.31-0.94), any complication (RR 0.49, CI 0.28-0.87), shortening > 5 mm (RR 0.54, CI 0.37-0.80), and HHS 6-24 months (MD 3.1, CI 1.8-4.4).

CONCLUSION

RCTs alone showed no significant differences between implant types. When including retrospective studies, angle-stable sliding compression implants demonstrated some advantages. The strength of evidence is limited by the predominance of retrospective cohort studies and high risk of bias in the included studies.