Dementia and Long-Term Postoperative Outcomes After Ankle Fracture Fixation: A National Propensity-Matched Analysis.

Wang, Brendon; Kumar, Arun; Wang, Irene; Burton, Alex; Kraus, Jonathan · J Foot Ankle Surg · 2026

retrospective_cohort · Level III

Where this comes from

Abstract

Dementia affects more than one in nine adults aged 65 and older (approximately 10-12%) and has been associated with adverse surgical outcomes, yet its impact following ankle fracture fixation remains poorly characterized. This study examined whether pre-existing dementia independently increases the risk of infectious, systemic, and healthcare utilization outcomes compared with matched controls following ankle fracture fixation. Retrospective cohort study with propensity score matching. Adult patients undergoing ankle fracture fixation (2006-2026) were identified from the TriNetX® US Collaborative Network. Patients with dementia (N=3,791) were matched 1:1 to controls, balancing age, race, sex, diabetes, hypertension, and other comorbidities. Outcomes were assessed at 90 days, 2 years, and 5 years and reported as risk ratios (RRs) with 95% confidence intervals. At 5 years, patients with dementia had a significantly higher risk of osteomyelitis (RR 1.613, 95% CI 1.351-1.926) and sepsis (RR 1.956, 95% CI 1.730-2.211) compared with matched controls (both p < 0.001). Pneumonia (RR 1.631, 95% CI 1.466-1.815) and pressure injuries (RR 2.097, 95% CI 1.795-2.450) were also significantly elevated at 5 years (both p < 0.001). Hardware-related infection was significantly more frequent in the dementia cohort at all time points, reaching RR 1.698 (95% CI 1.168-2.468, p = 0.005) at 5 years. Delirium was markedly more common among patients with dementia, with risk ratios ranging from 3.946 (95% CI 2.759-5.644, p < 0.001) at 90 days to 5.124 (95% CI 4.101-6.401, p < 0.001) at 5 years. Hospital readmission was significantly higher at every time point, reaching 71.0% vs. 56.0% at 5 years (RR 1.267, 95% CI 1.223-1.312, p < 0.001). Mortality was significantly elevated at 2 years (RR 1.444, 95% CI 1.266-1.646) and 5 years (RR 1.537, 95% CI 1.397-1.690, both p < 0.001). No significant differences were observed in hardware removal, reoperation, reoperation for nonunion or malunion, or amputation. Dementia independently increases infectious complications, including hardware-related infection, readmission, delirium, and long-term mortality following ankle fracture fixation without elevating most procedure-specific complication rates, suggesting excess morbidity is driven by patient-level vulnerability rather than surgical failure. Enhanced postoperative surveillance and care coordination are warranted for cognitively impaired patients undergoing ankle fracture fixation.