Growth and Microbiome Development in Very Low Birth Weight Infants Fed Primarily Mother's Own Milk vs. Donor Human Milk

Infant, Very Low Birth Weight

human milk, enteral nutrition, microbiome, microbiota

A study to compare growth, development of the intestinal bacterial environment, and other short term outcomes in groups of babies fed primarily their own mother's milk compared to those who receive primarily donor human milk. The investigators hypothesize that infants who receive primarily their own mother's milk will have better growth, a more diverse intestinal bacterial environment, and possibly some improved short term outcomes such as better feeding tolerance and lower rates of infection.

Background: Human milk feeding provides numerous benefits to preterm infants due to improvements in gastrointestinal maturation, host defense, infection rates, and improved long-term outcomes in neurodevelopment as well as cardiovascular and metabolic disease. There is accumulating evidence that an exclusive human milk-based diet decreases the rates of necrotizing enterocolitis (NEC) and death, and is associated with better feeding tolerance in very low birth weight (VLBW) infants than a diet of bovine milk-based products. In order to provide VLBW infants the benefits afforded by human milk feeding, the use of donor milk (DM) in neonatal intensive care units (NICU) has increased as many mothers are unable to provide sufficient milk needed for their premature infants. While there have been numerous studies that have favorably compared feeding of mother's own milk (MOM) to formula as well as studies that compare DM to formula, there are relatively few that compare maternal milk to donor milk. In regard to feeding tolerance and infection prevention, it has been proposed that DM may be less beneficial than MOM due to reduction in biologically active components during pasteurization, including human milk oligosaccharides (HMOs) and other immunological factors, growth factors, and hormones. Finally, alterations in the intestinal microbiota of preterm infants are suspected to contribute to disease states such as NEC, specifically within infants who have decreased microbial diversity. To the investigators knowledge, no studies comparing the intestinal microbiota among infants fed primarily MOM versus those fed primarily DM have been published. Purpose: To compare growth velocities, time to reach full enteral feeding volume, intestinal microbiota, and short term outcomes (NEC, late-onset sepsis, white matter injury) between infants fed primarily mother's own milk versus pasteurized donor human milk. Hypothesis: Infants fed primarily (50% or greater) mother's own milk will have increased intestinal microbiome diversity compared to infants fed primarily pasteurized donor human milk. Design: This prospective cohort study will be conducted in the Level III NICU at Texas Children's Hospital - Pavilion for Women and the Level II NICU at Texas Children's Hospital -- West Tower. Infants less than 1500 g birth weight will be fed exclusively human milk (mother's milk and or donor breast milk) fortified with donor human milk-derived fortifier per the investigators hospital guidelines. Once enteral feeding is established, infants will be categorized into cohorts based on percentage of feeding volume consisting of mother's own milk, including broad categorization of greater than 50% maternal milk versus less than 50%, and possibly tiered analysis of infants who receive less 25% maternal milk, 25-75% maternal milk, and greater than 75% maternal milk. An enrollment goal of greater than 125 infants including twins and multiples will be targeted for adequate sample size. Procedure: Infants will be enrolled within 72 hours of birth and started on parenteral nutrition and enteral human milk feedings per standardized feeding protocols and discretion of the attending neonatologist on service. Decisions to decrease or discontinue enteral feedings due to medical instability will be made by the attending neonatologist. Infants will be preferentially fed their own mothers' milk when available. For mothers who are unable to express adequate milk volume for their infants, pooled, pasteurized donor human milk will be offered per established NICU protocol. Feeds will be supplemented with human-milk based fortifier per protocol. At time of initial consent for the study, mothers will also be asked for consent to obtain a small sample (0.2-0.5 mL) of colostrum or expressed milk produced in the first week of life, as well as weekly milk samples thereafter, in order to analyze bacterial content of milk as it compares to the developing infant microbiome. However, consent for milk collection is not required for the infant's participation in the study. For infants who receive primarily donor milk, weekly samples of the milk they receive may similarly be analyzed for bacterial content. Infant stool samples will be collected during the first week of life and then at weekly intervals for six weeks for research purposes. The samples will be analyzed via 16S rRNA sequencing to determine diversity of intestinal microbiota. Additional analysis for metabolomics will be considered if lab availability and cost allows. Once weekly, a research nurse or physician will document growth measurements, including weight, length, and head circumference. Outcome data from the infants' medical records will be recorded, including time to regain birth weight, feeding tolerance as indicated by time required to reach full enteral feeding volumes of 100 ml/kg/day (for hydration) and 130-160 ml/kg/day (final goal volume for nutrition), and rates of NEC, spontaneous intestinal perforation (SIP), late-onset sepsis, and bronchopulmonary dysplasia (BPD). As all preterm infants less than 1000 g birth weight born at TCH-PFW have routinely performed brain MRIs at term gestation or hospital discharge, enrolled infants who fall into this subgroup will have their MRI results reviewed for outcome analysis. Additional medical record data for collection will include mechanism of delivery (cesarean vs. vaginal), antibiotics received by mother during 2nd and 3rd trimester as well as at time of delivery and while nursing/expressing milk, antibiotics received by baby during hospitalization, length of hospitalization, postmenstrual age at discharge, days NPO, days of parenteral nutrition, percentage of patients with patent ductus arteriosus, and rates of intraventricular hemorrhage, retinopathy of prematurity, and death. Although there is no long-term follow up currently designed for this study, at time of initial enrollment there will be an optional consent to allow study personnel to access medical records for patients who go on to have neurodevelopmental follow up visits at clinics within the investigators institution. There will also be optional consent for families to be contacted when the infant is approximately 6, 12, 18, and 24 months for developmental follow up. This will allow for possible comparison of neurodevelopmental outcomes at approximately 18-24 months in the subgroup of patients whose mothers consent to these aspects of the study. No labs will be requested for research purposes. No interventions are part of this protocol.

United States

Participants by arm