Absence of MRSA Transmission in Adult Intensive Care Units Without Isolation of Colonized Patients

Background: Transmission of methicillin-resistant Staphylococcus aureus (MRSA) occurs when patients colonized or infected with MRSA infect others through cross-transmission.  Stringent infection control measures are designed to prevent nosocomial transmission from asymptomatic carriers (“silent transmission”) as well as from those with overt disease. Contact isolation, including a private room, has been recommended for all patients including those who are colonized but have no evidence of clinical infection. 

Objective: The objective of this study was to determine the incidence of silent MRSA transmission in 2 adult intensive care units (ICUs) in an academic tertiary care hospital practicing universal nares screening, universal precautions, and MRSA prevention bundles but not contact isolation for patients who are merely colonized.

Methods: A retrospective cohort study was performed using infection control and laboratory databases of all patients with MRSA infections in two adult medical-surgical ICUs over a five-month period in 2009.  Patients were included in the study if they developed a laboratory-confirmed MRSA infection during their ICU stay or if they were colonized with MRSA and were in the ICU concurrently with another patient who developed an infection. Pulsed field gel electrophoresis (PFGE) was performed as a measure of cross-transmission on all patients with available archived samples from all positive nares screens and infection cultures. Strain patterns were compared to determine if any patients shared the same strain.

Results: 46 patients with clinical and laboratory-documented MRSA infections were identified.  Of these, 8 patients were already colonized with MRSA upon admission and their nares screen showed the same PFGE pattern as the isolate obtained from a clinical culture.  All of these patients (8/8, 100%) had community-acquired MRSA strains.  8 patients had negative nares screens but developed MRSA infections while in the ICU.  Of these 5 of 8 (63%) had community-acquired MRSA strains, while the remaining 3 (37%) had hospital-acquired strains.  When their PFGE patterns were compared to the nares screen isolates of 10 other patients concurrently in the same ICUs who were colonized with MRSA but not infected, none of these isolates had a similar PFGE pattern.  The remaining patients had no samples available for testing. 

Conclusions: PFGE enables healthcare epidemiologists to investigate MRSA transmission in the ICU.  Silent transmission of MRSA was not observed during the study period, suggesting adequate infection control measures from the MRSA prevention bundles, even in the absence of patient isolation. The lack of documented transmission also suggests that nares-only screening policies may fail to detect some colonized patients who subsequently develop MRSA infections during their ICU stay.