Effectiveness and Risks of Probiotics in Preterm Infants

OBJECTIVE: To evaluate the effectiveness and risks of probiotics among infants born before 34 weeks' gestation and with a birth weight less than 1000 g. METHODS: A population-based retrospective cohort study of infants born before 34 weeks' gestation and admitted to 33 Canadian Neonatal Network (CNN) units between January 1, 2016, and December 31, 2022. We excluded infants who were moribund on admission, died within the first 2 days, were admitted to CNN sites more than 2 days after birth, had major congenital anomalies, or never received enteral feeds. Logistic regression, propensity score-matched, and inverse probability of treatment weighting analyses were applied. RESULTS: Among 32 667 eligible infants born before 34 weeks' gestation, 18 793 (57.5%) (median IQR gestational age, 29 27-31 weeks) received probiotics, and 13 874 (42.5%) (median IQR gestational age, 31 29-33 weeks) did not receive probiotics. In these infants, probiotics were associated with decreased mortality rates (adjusted odds ratio aOR, 0.62; 98.3% CI, 0.53-0.73) but not decreased rates of necrotizing enterocolitis (NEC) (aOR, 0.92; 98.3% CI, 0.78-1.09) or late-onset sepsis (aOR, 0.90; 98.3% CI, 0.80-1.01). In 7401 infants with a birth weight less than 1000 g, probiotics were associated with decreased mortality rates (aOR, 0.58; 98.3% CI, 0.47-0.71) but not decreased NEC (aOR, 0.90; 98.3% CI. 0.71-1.13) or late-onset sepsis rates (aOR, 1.01; 98.3% CI, 0.86-1.18). Probiotic sepsis occurred in 27 (1.4/1000) infants born before 34 weeks' gestation and 20 (4/1000) infants with a birth weight less than 1000 g. Three infants with probiotic sepsis died, with probiotic sepsis deemed a possible cause in 2 cases. CONCLUSION: Probiotics used in Canadian neonatal units were associated with decreased mortality in infants born before 34 weeks' gestation and with a birth weight less than 1000 g with limited effects on NEC and late-onset sepsis. Probiotic sepsis was rare.