Severe bloodstream infections: A population-based assessment

Objective: Although bloodstream infection commonly results in critical illness, population-based studies of the epidemiology of severe bloodstream infection are lacking. We sought to define the incidence and microbiology of severe bloodstream infection (bloodstream infection associated with intensive care unit admission within 48 hrs) and assess risk factors for acquisition and death. Design: Population-based surveillance cohort. Setting: Multidisciplinary and cardiovascular surgical intensive care units. Patients: All adults with severe bloodstream infection in the Calgary Health Region (population ≈1 million) during 2000-2002. Interventions: None. Measurements and Main Results: Three hundred forty patients had 342 episodes of severe bloodstream infection (15.7 per 100,000 population/year). Several demographic and chronic conditions were significant risk factors for acquiring severe bloodstream infection (relative risk, 95% confidence interval) including age ≥65 yrs (7.0, 5.6-8.7), male gender (1.3, 1.1-1.6), urban residence (2.4, 1.2-5.6), hemodialysis (208.7, 142.9-296.3), diabetes mellitus (5.9, 4.4-7.8), alcoholism (5.6, 3.8-8.0), cancer (7.5, 5.3-10.3), and lung disease (3.8, 2.6-5.4). The most common etiologies were Staphylococcus aureus, Escherichia coli, and Streptococcus pneumoniae (3.0, 3.0, and 1.9 per 100,000/year, respectively). The case-fatality rate was 142 of 340 (42%) for an annual mortality rate of 6.5 per 100,000. Increased Acute Physiology and Chronic Health Evaluation II score (odds ratio, 1.1 per point; 95% confidence interval, 1.1-1.2) and presence of a co-morbidity (odds ratio, 2.5; 95% confidence interval, 1.4-4.3) were significant independent predictors of death. Conclusions: Bloodstream infections are commonly severe enough to require management in an intensive care unit and are associated with a high mortality rate. Identification of risk factors for severe bloodstream infection may allow targeting of preventive efforts to individuals at greatest potential benefit.