Contribution of acquired meticillin-resistant Staphylococcus aureus bacteraemia to overall mortality in a general intensive care unit.

During a period of 11 years, 77 patients had meticillin-resistant Staphylococcus aureus (MRSA) bacteraemia > or =5 days after admission to the intensive care unit (ICU). Ten had no prior growth of MRSA, 13 had positive screens on admission and 54 initially tested negative for MRSA in ICU before positive blood culture. These 54 constituted 20.2% [95% confidence interval (CI): 15.6-25.0] of 267 who acquired MRSA > or =5 days after admission. Mortality among 77 patients with MRSA bacteraemia was 57.1% (46.0-68.2). Nineteen of these 77 patients with MRSA bacteraemia had growth of a second pathogen from blood. Those with only MRSA bacteraemia were each matched with five controls by diagnosis and initial Acute Physiological and Chronic Health Evaluation II score. Mortality was greater in bacteraemic patients [53.6% (40.5-66.7)] than in controls [31.8% (26.3-37.3)] [relative risk (RR) 1.69 (1.25-2.26), P < 0.01], implying an additional absolute mortality of 21.8% (8.0-40.1). Application of this estimate to all 77 patients suggests that ICU-acquired MRSA bacteraemia caused 17 (6-31) deaths, adding 0.3% (0.1-0.6) to the 30.1% hospital mortality of all admissions. Incidence of MRSA bacteraemia increased with length of stay, contributing an estimated 3.1% (1.1-5.7) towards 37.9% mortality of the 198 patients remaining > or =25 days. These data emphasise the importance of preventing initial MRSA colonisation/infection of long-stay patients.