Early administration to preterm infants of the probiotic bifidobacterium breve strain BBG to prevent infection and necrotising enterocolitis

Design: Placebo controlled double blind randomised trial Setting: Neonatal Units in Local Research Networks of the Medicines for Children Research Network and in central London. Search and review strategy: Prior to trial commencement trials of probiotics in preterm infants were identified through review of literature searches (Medline, Embase, Cochrane; databases are scanned automatically for more recent publications and these are reviewed regularly at investigator meetings. Target population: Infants born before 31 completed weeks gestational age. Health technologies to be assessed: Early enteral administration of a live bacterium (Bifido Bacterium breve strain BBG) whether or not milk feeding has been started. Measurements of costs and outcomes: The outcomes that have been selected are important determinants of neonatal survival and morbidity, that are easy to define unambiguously, easy to record clinically and whose inclusion is justified by our own pilot data or by published data. No formal economic evaluation is planned at this stage but data items will be included in the data collection that will enable a post-hoc economic evaluation if the intervention is found to be beneficial. Sample size: Primary analysis will be by Intention to Treat (ITT). Our aim is to recruit 1300 babies born before 31 completed weeks of gestation. We have used 90% power in calculations and based the incidence of the primary outcomes: episodes of sepsis, NEC and death, on our own pilot data analysed by ITT, on the incidence of NEC at Homerton over a 3 year period and the survival of babies born below 31 weeks gestational age in London from Thames Regional Perinatal Group data collection overseen by Kate Costeloe. This size of trial allows 20% relative risk reduction of babies with sepsis from 44% to 35% and 40% relative risk difference to detect a reduction from 15% to 9% for both NEC and death. Project timetables: The study began on 1st September 2009. There was a run-in period to set up the study in the participating centres, the first patient was recruited in July 2010. Recruitment is on target to complete during the summer of 2013. The average length of stay for study infants is be around 10 weeks therefore an additional 6 months is required to complete, validate and collate all clinical and microbiological data and a further 3 months to complete analysis and writing up of the results. Expertise in the team: Kate Costeloe, Michael Millar & Mark Wilks have been working co-operatively for several years and represent an experienced neonatologist, medical microbiologist and clinical scientist. Many hospitals in London, Greater London and the South East are participating in the study and are led by experienced consultant neonatologists/paediatricians. The study will be run through the National Perinatal Epidemiology Unit in Oxford, which has extensive experience of the successful delivery of large perinatal clinical trials. This study will generate evidence to support or refute the hypothesis that early probiotic feeding improves important clinical outcomes for preterm babies nursed in different units across networks. This will be used to inform patient management, assist in the development of clinical guidelines and lead to recommendations for further research.