Lack of Association Between Mortality and Timing of Surgical Fixation in Elderly Patients With Hip Fracture

Authors:
  • LBL A. Beaupre
    L A. Beaupre
  • DJD W. C. Johnston
    D W. C. Johnston
  • DDD A. Dick
    D A. Dick
  • JCJ G. Cinats
    J G. Cinats
  • HJH X. Jiang
    H X. Jiang

Abstract
Conventional wisdom suggests high-quality care for most patients with hip fractures is surgical fixation within 24 hours to reduce mortality and complications, although there is little evidence to support this standard. We sought to determine the relationship between timing of hip fracture surgery and early mortality. This was a retrospective population-based cohort study of 3981 patients with hip fractures>60 years of age that were admitted to hospitals in one Canadian health region from 1994-2000. We collected sociodemographic, prefracture comorbidity, and postoperative complication data. Timing of surgery was classified as within 24 hours ("early surgery," the referent group for all analyses), 24-48 hours, and beyond 48 hours. Main outcome was in-hospital mortality. We used multivariable logistic regression methods, including adjustments with propensity scores and a validated hip fracture-specific mortality index, to determine the independent association between early versus later surgery and mortality. Median age of patients was 82 years, 71% were women, and 26% had >4 prefracture comorbidities. Unadjusted in-hospital mortality was 6%; it was 5% for those who had surgery within 24 hours or from 24 to 48 hours, 10% for surgery beyond 48 hours, and 21% for patients that did not have surgery. Compared with those who had surgery within 24 hours, there was no independent association between timing of surgery and in-hospital mortality (24-48 hours, adjusted odds ratio 0.89, 95% confidence interval 0.62-1.30, P=0.55; beyond 48 hours 1.30, 95% confidence interval 0.86-2.00], P=0.21). The timing of surgical fixation of hip fracture was not associated with early mortality in carefully adjusted analyses, and the use of "surgery within 24 hours" as a measure of high quality care may be inappropriate.

This publication has 34 references indexed in Scilit: