Epidemiology and long-term outcome of sepsis in elderly patients

Objectives: A study was made of the epidemiological data of sepsis requiring admission to the ICU in patients over 65 years of age, with an evaluation of independent predictors of survival at 2 years. Design: A prospective cohort study was made. Patients: Patients meeting criteria for sepsis upon admission to the ICU. Setting: A 40-bed ICU in a tertiary hospital. Results: The study group included 237 elderly patients (>= 65 years of age) and 281 controls (< 65 years of age) (n = 518). At least one chronic comorbid condition was present in 70% of the elderly patients as compared to only 56.1% of patients under age 65 (P < .01). There were several epidemiological differences between the groups: the prevalence of chronic diseases (diabetes, chronic obstructive pulmonary disease, and chronic heart failure), presentation as septic shock (52.3% vs 42%; P < .05), and the abdomen as the source of sepsis (52% vs 36%; P < .01) were all more frequent in elderly patients. Nine percent of the global patients discharged from hospital died in the 2-year follow-up period, but this rate reached 20% among the elderly. Independent predictors of 2-year mortality in the elderly were: chronic heart failure (adjusted hazard ratio [aHR] 2.24, 95% confidence interval [CI 95%] 1.28-3.94; P < .01), acute renal failure (aHR 3.64, 95%Cl 2.10-6.23; P < .01), acute respiratory failure (aHR 3.67, 95%Cl 2.31-5.86; P < .01), and inappropriate empirical antimicrobial therapy (aHR 2.19, 95%Cl 1.32-3.62; P < .01). Conclusions: Sepsis showed different demographic characteristics and clinical presentations in the elderly. In the aging cohort, after adjusting for potential confounders, inadequate empirical antimicrobial therapy was associated to a 2-fold decrease in survival at two years. (C) 2012 Elsevier Espana, S.L. and SEMICYUC. All rights reserved.\