Carriage of methicillin-resistant Staphylococcus aureus, ceftazidime-resistant Gram-negative bacilli, and vancomycin-resistant enterococci before and after intensive care unit admission.

Abstract

To measure patients' risk for acquiring antibiotic-resistant microorganisms associated with intensive care unit admission. Prospective, observational study. Ten public hospitals including one university medical center. Consecutive patients admitted to ten intensive care units. Serial patient surveillance cultures were screened for vancomycin-resistant enterococci, methicillin-resistant Staphylococcus aureus (MRSA), ceftazidime-resistant Gram-negative bacilli (CR-GNB), Acute Physiology and Chronic Health Evaluation II score, and antibiotic and medical device exposures. A total of 1,697 patient admissions in ten intensive care units were enrolled. The overall carriage rate of antibiotic-resistant bacteria at intensive care unit entry was 12.1% for MRSA, 14% for CR-GNB and 4.7% for both. At discharge from the intensive care unit, new carriage of MRSA, CR-GNB, and both was found in 11.1%, 14.2%, and 2.4% of the patients, respectively. The acquisition rates in the individual units correlated highly and positively with proportion of patients with carriage at intensive care unit entry for both MRSA (n = 10, Pearson's r =.89, p < 0.001) and CR-GNB (n = 10, Pearson's r =.92, p < 0.001). By logistic regression, severity of illness (odds ratio, 1.4), length of stay (odds ratio, 1.7), use of penicillins (odds ratio, 1.9), and number of antibiotics (odds ratio, 1.2) and medical devices (odds ratio, 1.2) were independently associated with intensive care unit acquisition of MRSA. In comparison, variables independently associated with intensive care unit acquisition of CR-GNB were Acute Physiology and Chronic Health Evaluation II score (odds ratio, 1.5), number of antibiotics (odds ratio, 1.1), and artificial airway (odds ratio, 1.5). These data suggest that hospitalization in the intensive care unit introduces significant risk to patients in terms of transmission of MRSA and/or CR-GNB. This risk seems to be influenced strongly by the proportion of patients with colonization at intensive care unit admission and is associated with severity of illness, length of stay, and exposures to antibiotics and medical devices.