Article Text

Download PDFPDF
Randomised controlled trial
Probiotics reduce the risk of necrotising enterocolitis (NEC) in preterm infants
  1. Nicholas Embleton,
  2. Janet E Berrington
  1. Newcastle Neonatal Service, Newcastle Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK
  1. Correspondence to Dr Nicholas D Embleton
    Newcastle Neonatal Service, Newcastle Hospitals NHS Foundation Trust, Richardson Road, Newcastle upon Tyne NE1 4LP, UK; nicholas.embleton{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Commentary on: OpenUrlAbstract/FREE Full Text


Survival rates for preterm infants have increased dramatically over the last two decades, but necrotising enterocolitis (NEC) remains a major problem. Around 30–50% of NEC-affected infants require surgery, approximately 30% may die, and survivors have increased risks of adverse neurodevelopmental sequelae. NEC has a multifactorial aetiology, and may represent the end stage of a variety of pathological processes, but in many cases appears strongly associated with patterns of gut microbial colonisation.1 Probiotics are the live bacteria that confer a health benefit; most are from the genus Lactobacilli or Bifidobacterium. Several recent meta-analyses have suggested probiotics reduce the risk of NEC and NEC-related death, but there remain methodological concerns.2 ,3


Fernandez-Carrocera and colleagues conducted a blinded randomised controlled trial (RCT) in preterm infants (birthweight <1500 g) over 3 years in a single neonatal unit in Mexico using a multispecies probiotic combination. Allocation concealment is implied. Effectiveness of randomisation is not reported. Trial products were given when enteral feeds started, preparation was administered by milk bank staff and caregivers were blinded. The background rate of NEC was high (15–20%), although ∼80% were small for gestational age, a factor known to increase the risk of NEC. Some high-risk infants (low Apgar score or ‘asphyxia’) were excluded from the trial, but it is not clear how many there were. Most infants received a combination of breast and formula milk.


The study (n=75) was powered to detect a reduction in the rate of NEC from 20% to 5%, but failed to show a significant effect (relative risk (RR)=0.54, 95% CI 0.21 to 1.39) although a trend to reduced NEC frequency was noted (probiotic 8% vs control 16% p=0.13). When the combined risk of NEC or death was calculated in a post hoc analysis, there was a significant reduction (RR=0.39, 95% CI 0.17 to 0.87). In the intervention group there was one death and this was attributed to sepsis. In the control group, there were seven deaths: four attributed to sepsis and three to causes unlikely to be affected by any putative probiotic mechanisms (cardiac anomalies, bronchoaspiration and hydrocephalus). There were no NEC-related deaths. The groups were well-matched, there was no significant difference in relevant secondary outcomes and no safety concerns with the probiotics. The infants’ age averaged 4–5 days at the start, although some were over 3 weeks. Although ∼70% had received antenatal steroids, and a similar number received surfactant, a high proportion (65–72%) also received indomethacin, presumably as treatment for a patent arterial duct.


While this blinded RCT contributes to the evidence base, several unresolved areas remain, and many clinicians may feel that this study does little to resolve their uncertainty. This study was conducted in a unit with a much higher prevalence of NEC than most units in North America, Australia and Europe, especially in a relatively mature population (median 31-week gestation). However, probiotics appear safe, cheap and widely available, and may also be of benefit in resource-poor settings where NEC has a higher case death rate. In keeping with existing RCTs, the probiotic combination appeared safe, and no reports of sepsis with probiotic organisms were detected. More than 20 different probiotics or combinations have now been tested, and while this may strengthen arguments for probiotic prophylaxis as a class of agents, it remains impossible to determine the optimal dose, combination and duration of treatment that might maximise benefit, while minimising any harm. Uncertainties remain because in many trials, including this one, infants at the highest risk may not have been enrolled, or are relatively under-represented. Overall, probiotics in preterm infants appear safe, although concerns about antibiotic resistance and the uncertainties of altering early gut flora persist. A large Australian trial group (ProPrems ACTRN12607000144415) have presented preliminary data suggesting a reduction in NEC, but peer-reviewed publication is awaited. The largest trial to date will conclude this year.4 These two large RCTs will add substantially to the evidence base. While the precise efficacy of probiotics remains uncertain, this study and the meta-analysis are difficult to ignore. Most countries (including the UK) do not have regulatory approval for probiotic products to be used under a medicinal licence. However, given their apparent safety, and possible efficacy, many will feel it is difficult to withhold probiotics from parents and infants.


View Abstract


  • Competing interests None.