Assessing the Concordance of MRSA Carriage Screening With MRSA Infections

Abstract

OBJECTIVES: This study focused on children with confirmed methicillin-resistant Staphylococcus aureus (MRSA) infections to determine MRSA screening utility in guiding empirical anti-MRSA treatment of children without history of MRSA infection. We examined the concordance of screens to assess differences by infection type and used statistical analysis to determine significant contributors to concordance. METHODS: Pediatric hospital patients admitted from 2002 through 2022 were included. Subjects had MRSA infections subsequent to MRSA surveillance screens performed the preceding year. Statistical analysis identified associations between MRSA screens and infections. Number needed to treat analysis calculated the utility of rescreening. RESULTS: Among 246 subjects, 39.0% had concordant screens; 151 (61.4%) screens were obtained in the 2 weeks preceding infection. Sensitivity for bacteremia was 50.0% (n = 42), for endotracheal/respiratory 44.4% (n = 81), and 29.4% (n = 102) for skin and soft-tissue infection. For children aged younger than 6 months, sensitivity was 35.9% (n = 78). Multivariable analysis significantly associated days since screening with decreasing likelihood of concordance. Regression modeled the probability of concordance to drop below 50.0% for all infections after 4 days, after 6 days for bacteremia specifically, and 12 days for endotracheal/respiratory infections. CONCLUSIONS: The concordance of screens was far lower than negative predictive values found previously; earlier studies were possibly impacted by low prevalence and exclusion of children at high risk to inform high negative predictive values. We suggest that negative MRSA screens should not invalidate reasonable suspicion for MRSA infection in patients with high pretest probabilities.