Incidence of probiotic sepsis and morbidity risk in premature infants: a meta-analysis.

Background: For preterm infants, supplementation with probiotics improves rates of necrotizing enterocolitis (NEC) and other morbidities. Case reports of probiotic sepsis have prompted warnings from the American Academy of Pediatrics and the Federal Drug Administration. However, incidence rates of probiotic sepsis are lacking, making it challenging to evaluate risk-benefit tradeoffs. We performed a meta-analysis and review of probiotic sepsis events in preterm infants to evaluate tradeoffs against NEC, mortality, and clinical sepsis outcomes.

Methods: Dual-reviewers screened 160 articles, selecting 77 for review. Pooled estimates of incidence were computed using random-effect models. Case reports captured infant demographics, hospital course, and outcome.

Results: For 20,323 exposed infants across 63 studies, 8 probiotic sepsis cases were identified [estimate: 0% (95% CI: 0-10%)]. Risk-benefit calculations note an additional 62 cases of NEC, 42 deaths, and 92 clinical sepsis events in the unexposed cohort per case of probiotic sepsis. Case reports identified 27 probiotic sepsis events, mostly in extremely-low-birthweight infants (median GA/BW: 28 weeks, 970.0 g) and those at risk for bacterial translocation.

Conclusion: Probiotic sepsis is extremely rare in preterm infants, with the greatest risk in an identifiable sub-population. Estimates highlighted increased morbidities in unexposed cohorts compared to probiotic sepsis incidence, suggesting consideration of risk-benefit may be warranted.

Impact: This study quantifies the risk of probiotic sepsis in preterm infants utilizing a meta-analysis. In over 20,000 exposed infants across 40 randomized trials and 23 observational studies, 8 cases of probiotic sepsis were identified (<0.04%). Assessing this risk against improvements in morbidities with probiotic use, we can expect 62 more cases of NEC, 42 more deaths, and 92 more cases of clinical sepsis per case of probiotic sepsis (1:2500) avoided in the unexposed group. While the use of probiotics carries risk, rates for probiotic sepsis presented by this analysis highlight a favorable benefit/risk ratio in preterm infants.