Bovine Colostrum as a Fortifier Added to Human Milk for Preterm Infants

Overview

Very preterm infants (\<32 weeks gestation) with very low birth weight (VLBW, \<1500 g) show immaturity of organs and have high nutrient requirements forgrowth and development. In the first weeks, they have difficulties tolerating enteral nutrition (EN) and are often given supplemental parenteral nutrition (PN). A fast transition to full EN is important to improve gut maturation and reduce the high risk of late-onset sepsis (LOS), related to their immature immunity in gut and blood. Conversely, too fast increase of EN predisposes to feeding intolerance and necrotizing enterocolitis (NEC). Further, human milk feeding is not sufficient to support nutrient requirements for growth of VLBW infants. Thus, it remains a difficult task to optimize EN transition, achieve adequate nutrient intake and growth, and minimize NEC and LOS in the postnatal period of VLBW infants. Mother´s own milk (MM) is considered the best source of EN for VLBW infants and pasteurized human donor milk (DM) is the second choice, if MM is absent or not sufficient. The recommended protein intake is 4-4.5 g/kg/d for VLBW infants, when the target is a postnatal growth similar to intrauterine growth rates. This amount of protein cannot be met by feeding only MM or DM. Thus, it is common practice to enrich human milk with human milk fortifiers (HMFs, based on ingredients used in infant formulas) to increase growth, bone mineralization and neurodevelopment, starting from 7-14 d after birth and 80-160 ml/kg feeding volume per day. Bovine colostrum (BC) is the first milk from cows after parturition and is rich in protein (80-150 g/L) and bioactive components. These components may improve gut maturation, NEC protection and nutrient assimilation, even across species. Studies in preterm pigs show that feeding BC alone, or DM fortified with BC, improves growth, gut maturation and NEC resistance during the first 1-2 weeks, relative to DM, or DM fortified with conventional HMFs.On this background, we hypothesize that BC, used as a fortifier for MM or DM, can induce similar growth and better NEC and LOS resistance, than conventional fortifiers. A pilot trial is required 1) to test the feasibility and initial safety of BC as a fortifier (e.g. similar growth rates and clinical variables as conventional fortification), 2) to calculate the sample size for a later, larger RCT with NEC +LOS as the primary outcome, and 3) record paraclinical outcomes associated with type of fortifier.

The main objectives of this multicentre, non-blinded, pilot, RCT are: 1. To investigate the safety, tolerability and the preliminary effects of BC, used as an HMF for MM and DM in very preterm infants. 2. To facilitate the determination of sample size for a later, larger RCT with NEC- and LOS-free survival as the primary outcome. 3. To assess the feasibility of study procedures, incl. recruitment rates, parental consent, adherence, sample collection, and clinical routines. Participants Parents to eligible very preterm infants admitted to the Neonatal Intensive Care Units (NICU) at Nanshan People's Hospital (NAN) and Baoan Maternal and Children's Hospital in Shenzen, China will be asked for participation. Since this is a pilot trial, a conventional sample size calculation, using only one primary outcome, is not required. The aim is to include 200 infants (100 per group), which is expected to give sufficient strength to demonstrate effects on the chosen feasibility outcomes and secondary blood and stool variables (see protocol). Statistical analyses will be performed blindly on both intention-to-treat and per protocol basis. Continuous outcomes will be summarized as mean and standard deviation (e.g., body weight) or median and interquartile range (e.g. time to reach full enteral feeding). Binary outcomes (e.g. incidence of NEC) will be presented as counts and percentages. To test the preliminary effects of BC, clinical and para-clinical outcomes will be compared between the two groups. The estimates will be presented as relative risk and absolute risk difference, difference between means, or hazard ratio, depending on the type of outcome. The estimates will be presented with a 95% confidence interval. Further statistical analyses are described in the protocol.