Acquisition of methicillin-resistant Staphylococcus aureus in a large intensive care unit

OBJECTIVES: To determine the prevalence of MRSA colonization on admission to the ICU and the incidence of MRSA colonization in the ICU. DESIGN: Prospective cohort study. SETTING: University hospital. PARTICIPANTS: Patients admitted to the ICU in 2000-2001. METHODS: Patients were screened for MRSA with nose, throat, groin, and axilla swabs on admission and discharge. MRSA acquisition was defined as a negative admission screen and a positive discharge screen. Risk factors analyzed included previous wards/current unit, gender, age, and length of stay prior to and in the ICU. Univariate and multivariate analyses were performed using logistic regression. RESULTS: Of screened patients, 6.8% were MRSA colonized on admission to the ICU. Some patients (11.4%) became newly colonized during their stay in the ICU. Factors that remained significant in the multivariate analysis of MRSA colonization on admission were previous admission to various wards and length of stay prior to ICU admission of more than 3 days. In the multivariate analysis of MRSA acquisition in the ICU, being a trauma patient and length of stay in the ICU greater than 2 days remained significant. Thirty-six percent of patients had both admission and discharge swabs taken. This percentage increased in the presence of a supervisory nurse. CONCLUSION: Significant acquisition of MRSA occurs in the ICU of our hospital, with trauma patients at increased risk. Patients who had been on the cardiothoracic ward prior to the ICU had a lower risk of MRSA colonization on admission. Presence of a supervisory nurse improved compliance with screening.