Impact of diabetes on mortality among patients with community-acquired bacteremia

Objectives: To assess the impact of diabetes on 30-day mortality in patients with community-acquired bacteremia. Methods: We conducted a hospital-based observational study in patients with community-acquired bacteremia who were admitted from the emergency department (ED). Consecutive admitted patients with positive blood cultures obtained in the ED were interviewed and their charts were reviewed. We compared 30-day mortality in diabetic patients with nondiabetic patients. Cox proportional hazards regression was used to estimate the independent effect of diabetes on 30-day mortality. Results: Among 839 patients in this study, 242 (29%) had diabetes. The median age of these patients was 67 years (interquartile range: 53-77) and 48% were women. The probability of survival at 30 days was not different between the diabetic and the nondiabetic group (84% vs. 77%, respectively; P = 0.15 by Wilcoxon test). On multivariate analysis, diabetes was not associated with an increased risk of 30-day mortality (hazard ratio, 0.82; 95% confidence interval, 0.53-1.26). By contrast, age and higher comorbidity index increased the risk of 30-day mortality. Several markers of disease severity at ED presentation (tachycardia, hypotension, and bandemia) predict 30-day mortality. Conclusions: Diabetes appears not to have significant impact on 30-day mortality in this study population. This finding should be interpreted in the context that diabetes patients may have an increased incidence of bacteremia. (c) 2007 The British Infection Society. Published by Elsevier Ltd. All rights reserved.