Operative versus nonoperative management of pelvic ring fractures in older adults: A propensity-matched analysis of mortality and complications.

Abstract

Pelvic ring fractures are an increasingly common injury in the aging population, carrying mortality comparable to hip fractures and substantial functional decline. Most geriatric fractures are managed nonoperatively, with operative fixation reserved for unstable patterns, but high-quality evidence comparing both treatment strategies is lacking. This study compared mortality and complications between operatively and nonoperatively managed pelvic ring fractures in older adults across multiple time points. This retrospective cohort study used the TriNetX database to identify patients aged 55 years or older with a pelvic ring fracture. Patients were assigned to operative or nonoperative cohorts and matched 1:1. Co-primary outcomes were 30-day and 1-year all-cause mortality. Secondary outcomes were mortality at 90 days and 2 years, and VTE, pneumonia, and MI at each time window. A multivariable Cox model was constructed for 1-year mortality. Following matching, 5314 patients were retained in each cohort. Mortality did not differ at 30 days (p = 0.108) or 90 days, but was significantly lower in the operative cohort at 1 year (p = 0.034) and 2 years (p < .001). This benefit did not persist in the multivariable Cox model (HR 0.939; p = 0.296). VTE and pneumonia were significantly more common in the operative cohort at all time points, while MI did not differ. In the sensitivity analysis of isolated acetabular fractures (1772 matched patients per cohort), the 1-year mortality benefit attenuated to non-significance (p = 0.057), whereas the 2-year benefit was preserved (p < 0.001); the higher operative rates of VTE and pneumonia, the null MI finding, and the null adjusted Cox result (HR 0.888; p = 0.258) were all consistent with the primary analysis. Operative management was associated with lower 2-year mortality that was robust across analyses and a 1-year mortality signal that was not robust to case-mix, with neither benefit persisting after multivariable adjustment. Operative management was also consistently associated with higher rates of VTE and pneumonia. The decision to operate must balance improved mobility against perioperative risk, and prospective trials assessing fracture characteristics, frailty and functional measures are needed.