Prebiotics to prevent necrotising enterocolitis in very preterm or very low birth weight infants

**Background:** Necrotising enterocolitis (NEC) affects about 1 in 20 very preterm or VLBW infants and carries a mortality rate of approximately 20%. Dietary supplementation with prebiotic oligosaccharides to modulate the intestinal microbiome has been proposed as a strategy to reduce NEC risk. This Cochrane review assessed the benefits and harms of enteral prebiotic supplementation versus placebo or no treatment for preventing NEC and associated morbidity and mortality in very preterm or VLBW infants.

**Methods:** The authors searched CENTRAL, MEDLINE, Embase, Maternity and Infant Care database, and CINAHL from earliest records to July 2022, along with clinical trial registries and conference proceedings. They included RCTs and quasi-RCTs comparing prebiotics with placebo or no prebiotics in very preterm (< 32 weeks' gestation) or VLBW (< 1500 g) infants. Two review authors independently assessed risk of bias, extracted data, and synthesised effect estimates using risk ratio (RR), risk difference (RD), and mean difference (MD) with 95% confidence intervals (CIs). The GRADE approach was used to assess certainty of evidence.

**Key Results:** Seven trials with a total of 705 infants were included (mean sample size 100). Three trials had potential sources of bias including lack of clarity on allocation concealment and masking. The studied prebiotics included fructo- and galacto-oligosaccharides, inulin, lactulose, and human milk oligosaccharides (2′-fucosyllactose and lacto-N-neotetraose). Meta-analyses of data from 7 trials (686 infants) showed: NEC occurred in 83 per 1000 with prebiotics versus 86 per 1000 with control (RR 0.97, 95% CI 0.60 to 1.56; RD none fewer per 1000, 95% CI 50 fewer to 40 more; low-certainty evidence). All-cause mortality occurred in 10 per 1000 with prebiotics versus 25 per 1000 with control (RR 0.43, 95% CI 0.20 to 0.92; RD 40 per 1000 fewer, 95% CI 70 fewer to none fewer; low-certainty evidence). Late-onset invasive infection occurred in 175 per 1000 with prebiotics versus 237 per 1000 with control (RR 0.79, 95% CI 0.60 to 1.06; RD 50 per 1000 fewer, 95% CI 100 fewer to 10 more; low-certainty evidence). Only one trial (van den Berg 2010) reported neurodevelopmental outcomes at 2 years corrected age (76 infants): Bayley MDI < 85 (RR 0.84, 95% CI 0.25 to 2.90), PDI < 85 (RR 0.24, 95% CI 0.03 to 2.00), cerebral palsy (RR 0.35, 95% CI 0.01 to 8.35); all very low-certainty evidence. Sensitivity analyses of 3 trials (467 infants) at low risk of bias showed consistent results: NEC (RR 0.98, 95% CI 0.58 to 1.65), mortality (RR 0.38, 95% CI 0.17 to 0.89), infection (RR 0.82, 95% CI 0.59 to 1.14).

**Clinical Implications:** The available trial data provide low- to very low-certainty evidence about the effects of prebiotics on NEC, mortality, infection, and neurodevelopmental impairment in very preterm or VLBW infants. The certainty of evidence is limited by methodological weaknesses in some trials (risk of bias) and imprecision of effect estimates, with 95% CIs including both large benefit and small or no benefit. The authors conclude that the true effects may be substantially different from the estimates, and that large, high-quality trials are needed to inform policy and practice decisions. These findings contrast with Cochrane reviews of probiotics and synbiotics, which suggested reductions in NEC and mortality but were also limited by concerns about trial quality, heterogeneity, and publication bias.