Eur J Intern Med. 2026 Jun 6:106986. doi: 10.1016/j.ejim.2026.106986. Online ahead of print.
ABSTRACT
BACKGROUND: Evidence on sex-related mortality differences in critically ill older adults is inconsistent, and frailty is rarely incorporated despite its major prognostic relevance in geriatric intensive care.
METHODS: We performed a pooled analysis of acute ICU admissions from three prospective multinational cohorts within the Very Old Intensive Care Patients project: VIP1 and VIP2, which enrolled patients aged 80 years or older, and COVIP, which enrolled patients aged 70 years or older with COVID-19. Frailty was assessed before the acute illness using the Clinical Frailty Scale. The primary endpoint was 30-day mortality. Associations between sex and mortality were examined using robust Poisson regression adjusted for age, frailty, Sequential Organ Failure Assessment score, organ support, and treatment limitations, with Bayesian models used to estimate posterior probabilities of clinically relevant effect sizes.
RESULTS: Among 10,363 acute ICU admissions, 43% were women. Women were older and more often frail, whereas men more frequently received invasive mechanical ventilation, vasopressors, and renal replacement therapy. Crude 30-day mortality was higher in men than in women (44%vs. 40%). Frailty was a strong independent predictor of mortality, with each 1-point increase in the Clinical Frailty Scale associated with an 8% increase in adjusted mortality risk (incidence-rate ratio 1.08, 95% CI 1.06-1.10). After multivariable adjustment for frailty and clinical covariates, male sex remained associated with a small residual increase in 30-day mortality (incidence-rate ratio 1.08; 95% CI 1.01-1.15). This association was attenuated with greater illness severity and more intensive organ support. Bayesian analyses yielded adjusted posterior median incidence-rate ratios of 1.04 to 1.06, and the probability of at least 10% excess mortality in men was low after adjustment (2% to 13%).
CONCLUSIONS: In older critically ill adults, pre-admission frailty is a major driver of mortality. After explicit adjustment for frailty and illness severity, male sex was associated with only a small residual increase in 30-day mortality. Bayesian analyses suggest that large clinically meaningful sex-based mortality differences are unlikely after accounting for case mix and treatment intensity.
PMID:42250989 | DOI:10.1016/j.ejim.2026.106986