The molecular epidemiology of incident methicillin-resistant Staphylococcus aureus cases among hospitalized patients in Alberta, Canada: a retrospective cohort study

Kathryn Bush, Jenine Leal, Sumana Fathima, Vincent Li, David Vickers, Linda Chui, Marie Louie, Geoffrey Taylor, Elizabeth Henderson

Infection Prevention and Control (IPC) surveillance for incident methicillin-resistant Staphylococcus aureus (MRSA) in hospitalized patients is performed in a complete provincial surveillance network of all acute care facilities in Alberta, Canada. IPC surveillance is centralized using a web-based data entry platform so that each patient is counted only once. All diagnostic laboratories submit the first clinical MRSA isolate associated with a patient without previous MRSA positive clinical cultures in the preceding year to the Provincial Laboratory for Public Health (ProvLab) for molecular typing. This study will investigate the relationship between the IPC epidemiological classification based on time of detection following admission to hospital (Hospital Acquired and Community Associated) and the matched laboratory MRSA surveillance data using a retrospective cohort study design. Incident IPC MRSA cases were classified according to IPC epidemiologic definitions. DNA sequencing of the Staphylococcus protein A (spa) gene and pulsed-field gel electrophoresis (PFGE) typing was performed. IPC MRSA surveillance data were matched to the ProvLab molecular surveillance data. Univariate comparisons of proportions were performed for categorical variables and the Student's t test for continuous variables. MRSA molecular typing data were available for matching for 46.7 % (2248/4818) of incident IPC cases. There was agreement in definitions for traditional nosocomial clones (USA100/CMRSA2) with Hospital Acquired (HA)-MRSA (65.1 % of all IPC HA-MRSA) and traditional community clones (USA400/CMRSA7 and USA300/CMRSA10) with Community Acquired (CA)-MRSA (62.4 % of CA-MRSA). However, we observed discordance for both traditional nosocomial/CA-MRSA (30.4 % of CA-MRSA) and for traditional community/HA-MRSA (26.9 % of HA-MRSA). We note agreement between traditional nosocomial clones and HA-MRSA, and traditional community clones and CA-MRSA. However, approximately one-quarter of HA-MRSA are those of traditional community clones while approximately one-third of CA-MRSA are those of traditional nosocomial clones. Collaborative provincial MRSA surveillance is important as the distinction between IPC case attribution in acute care settings and the historical definitions of MRSA clones as community- or healthcare-associated have blurred.