Department of Pediatrics, Division of Neonatology, University of Virginia, USA (t: +1 434 9245428, e: DAK4R@hscmail.mcc.virginia.edu)
The review of the literature by Drs Corvaglia and Aceti in this issue of Cure&Care: Journal of Neonatal Intensive Care has captured the complexity of the data regarding probiotics and necrotising enterocolitis (NEC) in preterm infants. Although the meta-analyses have shown impressive findings, in regard to the beneficial effects of probiotics on NEC and survival, they are based on heterogenous studies, so there is room for improvement, in terms of improving the quality of meta-analytical studies, in order to recommend better use of probiotics to advance patient care [1–3]. The review points out, importantly, that different probiotic strains (within a genus of, for example, Lactobacillus or Bifidobacterium) have different effects on the gastrointestinal tract. Even after considering the meta-analyses of studies with probiotics and NEC, it still remains unclear which probiotic strain or combination to select for clinical use, and at what dose and duration. To illustrate this wide variation, Table 1 shows the diversity of the probiotic strains, study populations and feeding categories that were included in studies examined in a recent meta-analysis.
There is a lack of certainty about which patient population may benefit from probiotics. The largest study to date (ProPrems) demonstrated a reduction in NEC in the study population of infants <1500 g who were given a probiotic combination, although a subanalysis showed that NEC rates were not reduced with the same probiotic combination in infants <1000 g or <28 weeks’ gestation . Furthermore, a recent Cochrane review concluded there was a lack of efficacy in reducing NEC or mortality in infants <1000 g . The benefit of probiotics may therefore only be evident in infants 1000–1500 g, and additional studies are needed to answer the question of whether probiotics are beneficial for infants below this weight.
There is a need to establish what the optimal daily dose should be for each probiotic strain. In general, the daily doses used in studies have ranged widely – from 50 million to 6 billion organisms. Are different doses needed for individual probiotic strains, or for infants <1000 g compared with larger infants, and does the dose affect the efficacy observed in individual studies? In addition to the type and strain of probiotic studied, it would be beneficial for meta-analyses to control for variables that are known to affect the risk for NEC, such as the use of human milk (i.e. mother’s milk, donor milk and the milk volume amount per day), antibiotics and feeding advance schedules (e.g.volume advanced daily such as 20 ml/kg per day). Many studies do not give a more detailed breakdown of their results or control for these variations in practice, so the meta-analyses are limited in their scope by what is reported in the individual studies. Interventions that have been shown to be associated with reduced NEC rates are listed in Table 2; it is important for clinicians to know if probiotics are as efficacious as these interventions or additive, if some are already used in their neonatal intensive care unit (NICU). The implication from the ProPrems study is that high breastfeeding rates lowered NEC rates in the control group . Additionally, since one way probiotics act is related to favourably changing the microbiome, the mode of delivery (vaginal delivery or cesarean section), as well as the type of antibiotic or feeding used, may also affect outcomes .
Consideration of the variability of probiotic studies could improve the quality of meta-analyses, as the inclusion of some studies may be problematical. For example, the recent Cochrane review included three studies that were not designed to evaluate the effect of probiotics on NEC and did not include NEC data in their publications, whilst another study was designed to evaluate the combination of a probiotic and lactoferrin .
If NEC prevention occurs because of the contribution of probiotics to a healthy microbiome, as one of the main mechanisms of action involved, there is potential for researchers to develop a method for analysing the microbiome, to discover the factors associated with lower NEC rates. Such an approach would be useful for the evaluation of individual probiotic strains or combinations of strains, both now and in the future. If such an analysis were available, large randomised controlled trials to evaluate the efficacy of probiotics would not always be required, if they are shown to be safe and to demonstrate specific beneficial changes on the preterm gastrointestional microbiome.
Download the full article