We are underpowered to find a difference in the primary outcome.

This study tries to determine whether the oral administration of a specific probiotic (good bacteria) in premature infants hospitalized in a neonatal intensive care unit may prevent infections and the development of a severe inflammatory disease of the bowel called necrotizing enterocolitis. The investigators propose that premature infants not given probiotics will colonize their gut with bad bacterias and develop infection.

Neonatal infections currently cause about 1.6 million deaths annually in developing countries. Preterm infants are especially vulnerable to infections because of their immature immune responses and their exposure to the hospital milieu which promotes gastrointestinal colonization with Gram-negative pathogens. Multiple studies have shown that the colonization of the bowel with probiotics (nonpathogenic anaerobic bacteria) competitively inhibit the attachment of Gram-negative pathogens decreasing their likelihood for bacterial translocation and the development of life threatening infections.This potentially high-impact, low-cost intervention may significantly improve the survival and morbidity of preterm infants around the world.

RESEARCH OBJECTIVE

We seek to determine whether prophylactic administration of Lactobacillus reuteri decreases the incidence of death and nosocomial sepsis among susceptible preterm infants in a neonatal care setting in Colombia.

Specific aims:

1. To investigate the efficacy and safety of oral Lactobacillus reuteri for the primary endpoint prevention of death or nosocomial sepsis in preterm infants through a well powered multicenter, randomized, double-blinded, placebo-controlled clinical trial (RCT).
2. To assess the impact of Lactobacillus reuteri on four secondary endpoints: necrotizing enterocolitis, prevalence of Gram-negative pathogens, duration of hospitalization, and frequency of outpatient treatment and re-admission for infectious causes at 6 months post-discharge.

RESEARCH DESIGN AND METHODS

Study Design:

Overview

Multi-center, double-blind, randomized, placebo-controlled trial Primary outcome: Death or nosocomial sepsis Secondary outcomes: NEC, prevalence of Gram-negative pathogens, duration of hospitalization, and frequency of outpatient treatment and re-admission for infectious causes at 6 months post-discharge. Post-hoc analysis will evaluate differences in RCT outcomes in two key subgroups: breast-fed and formula-fed infants and vaginal delivery versus cesarean section, in both probiotic exposed and unexposed infants. In addition, we will look at death and nosocomial sepsis as separate outcomes to ensure there is no differential effect of treatment in either outcome. Location: 11 NICUs in Colombia.

Randomization and Stratification:

Stratification according to participating institutions will be done in order to control for center-related differences. Stratification by birth weight will be performed in 2 groups: < 1500 grams and 1501 to 2000 grams. When eligibility criteria are met, infants will be randomly assigned to treatment with probiotics or placebo by using a computerized stratified balanced block randomization design. Assignment to treatment will be accomplished using sealed, sequentially numbered, opaque envelopes, color-coded for strata, available in each NICU pharmacy.

Experimental Design:

Study patients will be randomized at birth to either placebo or probiotic administration (treatment group). The hospital pharmacy will be in charge of random allocation. Vials of probiotic and placebo will only be identified by the pharmacist and according to randomization will be sent to the NICU for administration to the patient. For those randomized to the treatment arm, L reuteri DSM 17938 will be administered at a dose of 10 to the eighth colony-forming units in 5 drops of a commercially available oil suspension once per day until discharge from the hospital. This oil suspension is stable for 21 months at 2o C to 8o C (as documented by the manufacturer, BioGaia AB, Stockholm, Sweden). For those randomized to the placebo arm, patients will receive an equal number of drops from an identical vial containing only the oil base, but without the probiotic. The above preparations will be administered even if the patient has not begun a feeding protocol if there are no contraindications for feeding. Patients will be stratified according to their exposure to breast milk or premature infant formula alone or a combination of both. Infants of mothers with insufficient production of breast milk will be offered premature formula.

Feeding and nutrition protocols will be standardized among participating institutions to minimize the confounding effects of diverse practices. A standard feeding protocol may be started at 24-48 hours of life at a volume of 10 ml/kg/day if the patient is hemodynamically stable, defined as having a stable blood pressure. The two milliliters of breast milk or premature infant formula will be subtracted from this total daily volume. Feedings will be advanced slowly at a volume between 10 and 20 ml/kg/day depending on tolerance.84 Umbilical lines may be removed once feeds are increased above 20 ml/kg/day. Parenteral nutrition will be started on the second day of life according to a standardized protocol. MCT oil (2 kcal/30ml) will be used as a dietary supplement to increases the energy content of breast milk or formula when needed. Human milk fortifiers may be started once full enteral feeds have been tolerated. Feedings will be discontinued and an abdominal radiograph taken if there are any signs of feeding intolerance (recurrent emesis, gastric aspirate > 50% of previous feeding with abdominal distension or the presence of macroscopic blood in stools). In situations where feeds have been temporarily discontinued, patients may re-enter the study at the initiation of feeds. Patients who develop NEC may continue in the study if they are able to resume feeds after treatment. Inotropic support, if being weaned, is not a contraindication to begin or continue the protocol. Administration of probiotic will end at discharge, but the infant's status will continue to be monitored until 6 months post-discharge. Patient's participation in the study will end at 6 months post-discharge, or upon the death of the patient.

Inclusion Criteria:

• Admission to the NICU
• Written parental consent
• Birth weight < 2000 grams
• Hemodynamically stable
• < 48 hours of age

Exclusion Criteria:

• Evidence or suspicion of congenital intestinal obstruction or perforation
• Prenatal or postnatal diagnosis of gastroschisis, large omphalocele, or congenital diaphragmatic hernia
• Major congenital heart defects
• Anticipated transfer to a NICU not involved in the study