Abstract
OBJECTIVES: To evaluate the incidence of necrotizing enterocolitis before probiotic introduction, during probiotic administration, and after its cessation following the 2023 US Food and Drug Administration advisory discouraging probiotic use in preterm infants. STUDY DESIGN: Retrospective non-concurrent cohort study of very low birthweight (VLBW) infants (birthweight <1500 g), admitted to a level IV neonatal intensive care unit (NICU) from 2014 through 2024. Infants were grouped into three epochs based on Bifidobacterium longum ssp. infantis (B. infantis) EVC001 use: pre-EVC001 (no probiotic), EVC001 (routine use), and post-EVC001 (post-discontinuation). The primary outcome was NEC incidence. Secondary outcomes included NEC-associated mortality, all-cause mortality, NEC severity and NEC incidence among extremely low birthweight (ELBW; <1000 g) infants. Multivariable log-binomial and ordinal logistic regression models were used. RESULTS: Among 733 VLBW infants, NEC incidence was 12% pre-EVC001, 2.6% EVC001, and 16% post-EVC001 (P < .001). NEC risk was higher both pre-EVC001 (adjusted relative risk [aRR] 4.4, 95% confidence interval [CI] 2.2-9.0) and post-EVC001 (aRR 4.5, 95% CI 2.0-9.9; both P < .001) compared to during EVC001 administration. ELBW infants showed similar trends. EVC001 use was associated with reduced odds of severe NEC compared to epochs without EVC001 use (VLBW odds ratio 5.3, 95% CI 2.5-11.0; ELBW odds ratio 5.0, 95% CI 2.2-11.7; both P < .001). NEC-related mortality was lowest during EVC001exposed infants (0.9%) compared to unexposed (2.8, P = .05). CONCLUSIONS: NEC rates were higher before, lower during and rose after discontinuation of EVC001 in this single center retrospective study. There is a need for multicenter trials evaluating B. infantis for NEC prevention.