Modeling adverse outcomes in very low birth weight infants based on an infant diet of mother’s breast milk and donor breast milk

Abstract

Modeling adverse outcomes in very low birth weight infants based on an infant diet of mother’s breast milk and donor breast milk Objective: This study evaluated the dose and threshold effects of mother’s breast milk (MBM), donor breast milk (DBM), and preterm formula (PTF) fed to very low birth weight (VLBW) infants on specific adverse health outcomes measured from birth to discharge from the hospital. This study contributes to the understanding of how much MBM is necessary during a premature infant’ hospitalization to improve health outcomes. It also provides information on the effects of DBM and PTF upon premature infant health outcomes. Methods: The current study is a retrospective analysis of VLBW infants who received MBM and/or DBM from birth to 1,800 grams. When MBM was not available, infants were fed PTF after 1,800 grams until discharge from the hospital. Any VLBW infant born at the University of Washington Medical Center Seattle, WA, after the initiation of the center’s DBM program was included in this study. Infants were excluded if any major congenital anomalies existed. The DBM program included infants born at 32 weeks or less gestational age or weighing less than or equal to 1,800 grams. VLBW infants were eligible for DBM until 33 weeks gestational age and weight was greater than 1,800 grams. Once an infant was no longer eligible for DBM, the infant was transitioned to preterm formula (PTF) if there was not enough MBM. Descriptive statistics were used to summarize quantitative variables. Dosage and threshold effects of MBM, DBM, and PTF for VLBW infants were evaluated using receiver operating characteristic curve (ROC) analysis. Bronchopulmonary dysplasia, necrotizing enterocolitis, late-onset sepsis, and mortality were analyzed using logistic regression models. Predictor variables included the percentage of MBM, DBM, and PTF in the infants’ total diet of all milk types from birth to discharge. Length of stay, weight gain, and total days on ventilator were regressed upon the percentage of MBM, DBM, and PTF in the infant’s total diet. Covariates included birthweight, gestational age, gender, Apgar scores at both 1 and 5 minutes, exposure to prenatal and postnatal steroids. Outcomes: The outcomes for this study included the incidence of bronchopulmonary dysplasia, necrotizing enterocolitis, late-onset sepsis, mortality, length of hospital stay, weight gain, and total days on ventilator for VLBW infants who were fed MBM, DMB, and PTF. Results: A total of 302 low birth weight infants were included in this study. The mean gestational age was 28.65 ± 3.08 weeks and mean birthweight is 1.02 kg ± 0.30 kg. The sample is 53.3% male and 46.7% female infants. The mean length of stay is 67.37 ± 37.01 days with the mean discharge weight of 2.92 kg ± 1.09 kg. Among the participants, 55.6% had bronchopulmonary dysplasia, 27.2% experienced medical necrotizing enterocolitis, 27.2% had late-onset sepsis, and 18.5% died. Logistic regression analyses showed that the percentage of MBM significantly predicted a decrease in BPD but not medical necrotizing enterocolitis, late-onset sepsis, or mortality. The increased percentage of MBM contributed to a reduction in length of stay and total days ventilated. The percentage of DBM yielded comparable results for these outcomes. ROC analysis showed that a diet composed of greater than 65.5% of MBM (odds ratio = 0.56, p = 0.050) was associated with survival, whereas a diet with greater than 31.5% of PTF (odds ratio = 1.94, p = 0.029) was associated with mortality.