Probiotics, prebiotics infant formula use in preterm or low birth weight infants: a systematic review

Systematic review of randomized trials in hospitalized preterm or low birth weight infants who were exclusively formula-fed. Included infants were born before 37 weeks' gestation or weighed less than 2.5 kg at birth, with individual trials enrolling very preterm or very low birth weight populations.

Infant formula was supplemented with either probiotics or prebiotics and compared with placebo or unsupplemented formula. Probiotic regimens included Saccharomyces boulardii, Lactobacillus reuteri ATCC 55730, Lactobacillus acidophilus, and Bifidobacterium lactis at doses ranging from 9 x 10^6 cfu/mL formula to 2 x 10^7 cfu/g milk powder, generally started in the first days of life and continued for 14 to 30 days. Prebiotic regimens included GOS, FOS, scGOS/lcFOS, or GOS/FOS mixtures at doses from 0.4 g/100 mL to 1 g/dL, usually for 14 to 28 days.

Overall, evidence was insufficient to support routine probiotic or prebiotic supplementation of formula in exclusively formula-fed preterm infants. For probiotics, weight gain was not significantly improved (MD 1.96 g/day, 95% CI -2.64 to 6.56; 2 studies, n=34), and major clinical outcomes were also not significantly affected, including necrotizing enterocolitis (RR 0.42, 95% CI 0.15 to 1.16; 2 studies, n=162), sepsis (RR 0.40, 95% CI 0.11 to 1.45), and mortality (RR 0.33, 95% CI 0.04 to 2.85). Stool frequency increased with probiotics (MD 1.60, 95% CI 1.20 to 2.00). For prebiotics, pooled weight gain was also not significant overall (MD 0.04 g/day, 95% CI -2.65 to 2.73; 2 studies, n=50), although some subgroup analyses suggested modest effects on growth and increased bifidobacteria counts. The review concluded that larger, longer randomized trials are needed before any clinical recommendation can be made.

The evidence base was small, with few trials, small sample sizes, short intervention durations, and limited methodological quality. Many outcomes had few events or were reported inconsistently, several analyses were underpowered, and key clinical endpoints such as long-term growth, necrotizing enterocolitis, sepsis, and mortality were not clearly improved. Heterogeneity in strains, doses, and outcome measures further limits confidence and generalizability.