Background Interpretation of published trials of probiotics in preterm infants is complicated by the use of multiple bacterial strains and exclusion from some trials of babies at high-risk of complications.
Objective To evaluate a single strain of Bifidobacterium breve in an unselected population of preterm infants.
Methods Multi-centre double blind Randomised Placebo Controlled Trial of Bifidobacterium breve BBG-001, 2.1 to 5.3 × 108 cfu daily, (B breve) in infants below 31w gestation randomised before 48h. Primary outcomes were necrotising enterocolitis (NEC) ≥Bell stage 2, late onset sepsis (LOS) and death. Results are presented by intention to treat adjusted for sex, gestation and randomisation within 24 h and allowing for clustering of multiples.
Results 1310 infants were randomised, median gestation 28.0w, median birthweight 1010g and median age starting the intervention 44 h. No adverse events related to the intervention were reported.
Conclusions This intervention was not associated with any advantage in this population of babies. This result highlights the need to assess the efficacy of different probiotic strains and challenges the validity of combining trials using different probiotic interventions in meta-analyses.