RESUMO
INTRODUCTION: It is still debated whether to use cannulated cancellous screw (CCS) or sliding hip screw (SHS) implants for femoral neck fracture (FNF) patients but there are no studies large enough to evaluate on smaller differences. The aim were to compare the reoperation and mortality rates of a large cohort of FNF patients above 60 years of age with internal fixation using CCS and SHS implants, with stratification by fracture classification. METHODS: This register study employed prospectively collected data from 2014 to 2018 from the Danish Fracture Database, including 2-year follow-up and data on age, sex, surgery, comorbidity and fracture classification. Reoperation data was retrieved from the Danish National Patient Register. The primary outcome was major reoperation, defined as re-osteosynthesis, conversion to arthroplasty, non-union surgery, Girdlestone procedure or deep infection 1-year post-surgery. Relative risk (RR) for the CCS group compared with the SHS group was estimated using Cox proportional hazards modelling, adjusted for age, sex and comorbidity (including 95% confidence intervals). RESULTS: A total of 2,598 FNFs were included, 1,731 with CCS (67%) and 867 with SHS (33%). The overall 1-year major reoperation rate was 15% with CCS and 13% with SHS, yielding an adjusted RR of 1.10 (0.88; 1.37) for CCS compared to SHS. The adjusted RR was 1.15 (0.77; 1.69) for undisplaced FNF, 1.09 (0.68; 1.75) for displaced FNF and 1.35 (0.94; 1.94) for transcervical FNF, which increased to 1.42 (1.01; 2.00) after 2 years. The minor reoperation rate was higher for CCS (3%) than SHS (1%), yielding an adjusted RR of 2.28 (1.14; 4.51). The adjusted RR for CCS compared to SHS was 0.83 (0.62; 1.12) for 30-day mortality and 0.96 (0.82; 1.14) for 1-year mortality. CONCLUSIONS: We found no difference between CCS and SHS in major reoperation rates; however, for transcervical FNF, CCS was associated with higher reoperation rates. Furthermore, CCS was associated with a higher risk of minor reoperation than SHS. There were no differences in mortality rates.
