Carriage of vancomycin-resistant Enterococcus faecium in infants following an outbreak in the neonatal intensive care unit: time to clearance of carriage and use of molecular methods to detect colonization

**Background:** Vancomycin-resistant Enterococcus faecium (VRE-fm) is classified by the WHO as a high-priority pathogen. Fecal colonization often precedes infection, and carriage in adults can persist for months to years. Data on the duration of VRE carriage in infants after NICU discharge are lacking. In 2017, a monoclonal VRE-fm outbreak (vanA gene) occurred in the NICU at Tel Aviv Sourasky Medical Center, Israel, involving 49 colonized infants without clinical infection.

**Methods:** This retrospective cohort study included all 49 newborns identified as colonized with VRE-fm between February 20, 2017, and April 11, 2017. After discharge, infants underwent sequential stool testing for VRE clearance. Testing began no sooner than 3 months after the initial positive test. A negative test was confirmed by repeat testing 2 weeks later; a positive test was followed by repeat testing 2 months later. VRE-fm presence was determined by two methods: standard culture on selective/chromogenic media and PCR detection of vanA and the housekeeping ddl-fm genes. A positive test was defined as positive culture and/or positive PCR for both vanA and ddl. Clearance was defined as two consecutive negative tests at least 2 weeks apart. The date of the first positive test was the start point; the date of the first of the two negative tests was the endpoint. Median time to clearance was estimated using nonparametric maximum likelihood estimation (NPMLE). A paired t-test compared time to clearance by culture alone vs. culture plus PCR. Bivariate and multivariable logistic regression identified risk factors for prolonged carriage (>6 months). Follow-up ended September 1, 2018.

**Key Results:** Among 49 colonized infants, 34 were preterm. Median length of NICU stay was 24 days (IQR 12–45 days). 80% received antibiotics. Of 46 infants who underwent post-discharge stool testing (range 1–8 tests), 6 were lost to follow-up before clearance. Thus, 40 patients cleared VRE as defined by culture and PCR after a median of 4 tests. Time from initial positive test to clearance ranged from 91 to 361 days. The estimated time until 50% of patients achieved clearance was 217 days. By culture alone, 41 patients achieved clearance between 91 and 294 days, with an estimated median of 113 days. The difference between methods was significant (P < .0001). After one negative test, the next test was also negative in 32/40 (80%). A negative culture followed by a positive culture occurred in only 1 case. No patient developed a clinical VRE infection during 16–18 months of follow-up. In bivariate analysis, mechanical ventilation, central line, oral iron supplement, and second-line antibiotics appeared protective against prolonged carriage (P < .10), but none remained significant in multivariable analysis.

**Clinical Implications:** This is the first study to describe the natural history of VRE carriage in NICU infants after discharge. All infants who completed follow-up cleared carriage within 1 year, and no clinical infections occurred. These data can reassure parents and guide infection control teams. The authors recommend beginning clearance testing at 3 months after the initial positive test. PCR-based detection yields a significantly longer estimated carriage duration than culture alone (median 217 vs. 113 days), which should be considered when interpreting test results. Limitations include non-uniform testing intervals, potential false-positive PCR results (mitigated by requiring both vanA and ddl positivity), and lack of data on post-discharge risk factors.