Reduction in Methicillin-Resistant Staphylococcus aureus (MRSA) Surveillance in a Low-Prevalence Neonatal Intensive Care Unit Does Not Lead to Increase in Vancomycin Utilization

Abstract

Background: Methicillin-resistant Staphylococcus aureus (MRSA) infection in neonates is associated with significant morbidity, mortality, and hospital cost. Multiple studies have shown that these infections are often preceded by colonization, but no consensus has been established for MRSA surveillance. The impact of changing the surveillance strategy on vancomycin utilization has not been evaluated previously. Methods: Retrospective chart review of infants who underwent MRSA screening in a level IV NICU with all outborn neonates. A weekly surveillance PCR was obtained from the nares between July 2016 and June 2017 (phase 1) and only on admission and discharge between July 2017 and June 2018 (phase 2). Patients with a positive PCR were placed on contact precautions without decolonization. The χ² test was performed to compare the 2 phases of screening, and the Student t test and the Fisher exact test were used to compare the characteristics of MRSA colonized infants. Vancomycin utilization was measured in days of therapy (DOT) per 1,000 NICU patient days. Results: In total, 689 infants underwent MRSA screening during the study period; 324 infants had weekly MRSA surveillance and 365 infants had screening at admission and discharge. There was no statistically significant difference in MRSA colonization rates (4.3% vs 3.0%) or MRSA colonization acquisition (negative to positive, 1.8% vs 1.0%) between the phases. Among MRSA-colonized patients, nearly 60% were colonized on admission. Nearly 40% of the infants became colonized with MRSA during their hospitalization, none of whom developed MRSA infections prior to discharge. Mean vancomycin utilization decreased from 38.55 to 30.16 DOT per 1,000 NICU patient days between the 2 study periods. Conclusions: In a level IV NICU with relatively low MRSA prevalence, the change in MRSA screening practice from weekly surveillance to surveillance upon admission and discharge demonstrated no difference in MRSA acquisition or infection. Overall vancomycin utilization also decreased during this period, suggesting a culture shift around antibiotic utilization. Further study is needed to evaluate the utility of MRSA screening, decolonization, and isolation practices in low-prevalence NICUs and to identify additional drivers of vancomycin utilization.

Funding: None

Disclosures: None