Article Text
Abstract
Introduction Owing to concerns regarding their efficacy and safety, the use of probiotics for the prevention of necrotising enterocolits (NEC) remains a hotly debated topic amongst neonatologists. Despite previous meta-analyses consistently showing that probiotics prevent NEC, randomised controlled trials (RCTs) continue to demonstrate conflicting results regarding the efficacy of probiotics. Here we present the largest and most comprehensive meta-analysis to date, and the first to use a random-effects model, to investigate the efficacy of probiotics in very-low-birth-weight (VLBW) infants. We also attempt to explain for the first time the apparent heterogeneity in results from RCTs by performing meta-regression and sub-group analysis to identify particular probiotic agents and other measured variables that may significantly influence the outcome.
Method PubMed/MEDLINE, EMBASE and the Cochrane Library databases were searched for studies published before 01/08/2016. Only RCTs involving VLBW infants with outcomes of NEC stage ≥2 and mortality were selected. All studies were independently scored to quantitatively assess the methodological quality. A minimum Jadad score of 3 had to be achieved for a study to be included in the analysis. 23 RCTs were identified. Data from these studies were pooled using a random-effects model for the meta-analysis and meta-regression analysis.
Results Probiotics significantly reduced the incidence of NEC stage ≥2 (RR=0.57 [95% CI 0.43–0.74, p<0.0001]) and all-cause mortality (RR=0.72 [95% CI 0.57–0.92, p=0.009]), with a trend towards reduction in NEC-related mortality (RR=0.64 [95% CI 0.38–1.07, p=0.09]). Genus-specific subgroup analysis revealed that only formulations containing a combination of Lactobacillus and Bifidobacterium species, led to significant reduction in both incidence of NEC stage ≥2 (RR=0.41 [95% CI 0.25–0.66, p=0.0003]) and mortality (RR=0.47 [95% CI 0.27–0.80, p=0.006]). Multivariate meta-regression analysis showed that probiotics were most effective in cohorts with higher levels of exclusive breast milk feeding and lower levels of prenatal steroid exposure.
Conclusion This analysis shows with considerable statistical confidence that probiotics prevent NEC and reduce mortality in VLBW infants. We have identified that differences in probiotic agents and the influence of prenatal steroids and feeding regimens may explain the differences in outcomes between studies. Our findings support further scientific investigation and ultimately a strain-specific meta-analysis to identify an effective probiotic protocol.
Disclosure of Interest None Declared
- Necrotising enterocolitis
- probiotics