Brazilian Journal of Implantology and Health Sciences · Nov 13, 2026

Predictors of mortality in older adults with proximal femoral fracture: an integrative review

By Jorge GCF, Fontana GDSEC, da Costa LF et al.

INTRODUCTION: Proximal femoral fracture in older adults is associated with high mortality, and various factors, scores, and care practices have been proposed as predictors or modulators of this outcome. OBJECTIVE: To integrate evidence from studies with different designs on predictors of mortality in older adults with proximal femoral fracture. METHODS: An integrative review of ten studies published between 2017 and 2026 (four systematic reviews, five cohorts, and one randomized clinical trial), selected according to predefined eligibility criteria. Data were extracted in a standardized manner and synthesized into thematic categories, without meta-analysis. RESULTS: Four categories emerged: baseline clinical factors, inflammatory and nutritional markers, time to surgery, and prognostic models. Age, male sex, higher ASA score, dementia, and institutional residence were associated with higher mortality at 30 days (systematic review with 33 studies in the meta-analysis) and at 1 year (cohort with 685 patients; ASA, adjusted OR of 2.14 per class; 95% CI: 1.54–2.99). The systemic inflammation response index was associated with mortality in one cohort (HR 1.447 in the upper tertile; 95% CI: 1.093–1.917). Observational cohorts indicated higher mortality at 30 days with a wait longer than 12 hours (OR 1.43) or 24 hours, but the only randomized trial, which compared accelerated and standard surgery (medians of 6 and 24 hours), showed no difference in mortality at 90 days (9% and 10%; p = 0.40). The prognostic models had AUC values between 0.710 and 0.964, with a high risk of bias in most. CONCLUSIONS: There is convergence regarding age, male sex, ASA, dementia, and institutionalization; the effect of time to surgery remains uncertain when observational studies and the randomized trial are compared, and the models require external validation. No included study is Brazilian.

doi.org/10.36557/2674-8169.2026v8n11p1206-1218Open accessPopulations

AgingResearch.News