Background: Diabetes has been associated with bacteraemia due to haemolytic streptococci but epidemiological evidence is limited.
Methods: We conducted a 15-year population-based case-control study of all adults with first-time bacteraemia with groups A, B, and G haemolytic streptococci and matched population controls. The study setting was Northern Denmark between 1992 and 2006. We computed odds ratios (ORs) for streptococcal bacteraemia according to diabetes and glycaemic control, using regression analysis for confounder adjustment.
Results: We identified 397 adult patients with bacteraemia due to haemolytic streptococci (median age 67 years, 51% women), of which 63 (17%) had diabetes. Persons with diabetes had a 2.1-fold increased risk of streptococcal bacteraemia compared with population controls (adjusted odds ratio (OR) 2.1; 95% confidence interval (CI), 1.5-2.9). For persons with type 1 diabetes, the adjusted OR was 14.8 (2.4-91.2). Longer diabetes duration and poor glycaemic control conferred higher risk estimates: adjusted OR 1.5 (0.8-3.0) for HbA(1c) level <7%, and OR 3.6 (1.6-8.1) for HbA(1c) level ≥9%. The association between diabetes and HS bacteraemia was independent of the underlying foci of infection and was strongest for group B streptococcal bacteraemia (OR 3.5; 1.8-7.0) and for group G streptococcal bacteraemia (OR 2.6; 1.6-4.4). There was no clear increase in risk for group A streptococcal bacteraemia (OR 1.2; 0.7-2.2).
Conclusions: Diabetes is a strong risk factor for group B and group G, but not group A, haemolytic streptococcal bacteraemia. The risk increase is particularly high for type 1 diabetes, long diabetes duration, and poor long-term glycaemic control.
Copyright © 2011 The British Infection Association. Published by Elsevier Ltd. All rights reserved.