Mortality in patients with trochanteric hip fractures (AO/OTA 31-A) treated with sliding hip screw versus intramedullary nail: A retrospective national registry study of 9547 patients from the Danish Fracture Database

Abstract

BackgroundWhether trochanteric hip fractures (AO/OTA 31-A) should be treated with an intramedullary nail (IMN) or sliding hip screw (SHS) is debated. Recent studies suggest an association between IMN and excess mortality rates compared to SHS, but higher quality studies fail to show this association. Furthermore, there is an increased usage of IMN with sparse evidence supporting this rise. Our aim was to compare mortality rates between IMN and SHS in patients with AO/OTA 31-A fractures. Secondarily, to investigate choice of implant in relation to fracture subtype. MethodsThis national registry study is based on data from the Danish Fracture Database (DFDB). Data were retrieved on patients aged ≥65 years treated for a non-pathological AO/OTA type 31-A fracture with IMN or SHS from January 1, 2012 to December 31, 2018. Data from DFDB were merged with data from the Danish Civil Registration System (CRS) for time of death. Mortality rates were recorded at 30 days, 90 days, and 1 year and presented as crude mortality and adjusted for age, gender, ASA-class, and AO/OTA-subtype. ResultsA total of 9,547 patients were included. The mean age was 83 years, 69% were female, and 55% were ASA-class 3–5. We found higher mortality rates for IMN-patients at 30 days, 90 days, and 1 year. The crude relative mortality risk for IMN was also significantly higher at both 30 days (1.20) and 90 days (1.11). Adjusted relative mortality risk for IMN was 1.12 [0.96; 1.31] at 30-days, 1.03 [0.91; 1.17] at 90-days, and 1.01 [0.92; 1.11] at 1 year. Most patients suffered a 31-A2 fracture (56%) and, overall, 74% of patients were treated with IMN. ConclusionWe found significantly increased crude relative mortality risk at 30 days and 90 days in patients treated with IMN. However, when adjusting for confounders the two groups had similar mortality risks. In total, 74% of all patients in this cohort were treated with IMN. Level of evidenceThis study has level of evidence: III.