Factors associated with mortality in older sepsis patients: a nationwide retrospective cohort study in Germany

Résumé fourni par la source

BACKGROUND: Sepsis is a leading cause of mortality in older adults, but large unselected cohort data on factors associated with mortality remain limited. OBJECTIVE: To examine patient characteristics and acute illness features associated with mortality among older patients hospitalised with sepsis in Germany. METHODS: We conducted a retrospective, observational cohort study using statutory health insurance data (AOK) of patients aged ≥70 years and hospitalised with sepsis between 2016 and 2019. Outcomes were in-hospital and 90-day post-discharge mortality. Multivariate blockwise logistic regression models were applied to examine associations across conceptual domains. RESULTS: Of 227 110 older sepsis patients [median age, 81 years (IQR 76-85); 49.2% women], in-hospital mortality was 35.2%. Organ dysfunction, particularly hepatic and renal dysfunction, as well as receipt of organ support including mechanical ventilation and dialysis, showed strong associations with in-hospital mortality. Among 147 080 patients discharged alive, 19.5% died within 90 days. In this post-discharge phase, care dependency and metastatic cancer showed stronger associations with mortality, whereas associations with acute illness features were lower. Female sex was independently associated with lower mortality and conferred a survival advantage after discharge. CONCLUSION: The high mortality rate of older sepsis patients is associated with both baseline patient vulnerability and acute illness severity. Care dependency, used as a proxy for functional vulnerability overlapping with frailty, showed a strong association with mortality, particularly after discharge. These findings highlight the importance of integrating preexisting patient vulnerability and acute illness characteristics to guide individualised acute and post-discharge care for older sepsis patients.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.