Abstract
OBJECTIVE: Articles previously published by Sullivan et al. and Cristofalo et al. were reanalyzed using the proportion of cow milk-based nutrition received to determine whether that affected clinical outcomes during hospitalization for infants birth weight 500-1250 g. Abrams et al. showed in the same cohort incidences of necrotizing enterocolitis (NEC), NEC requiring surgery and sepsis increased proportionally to the amount of dietary cow milk. METHODS: The data from the two studies conducted under essentially the same protocol were combined yielding a cohort of 260 infants receiving a diet ranging from 0% to 100% cow milk. Data analysis utilized negative binomial regression which mitigates differences between subjects in terms of their time on study by incorporating that number into the statistical model. The percent of cow milk-based nutrition was the only predictor investigated. RESULTS: For all outcomes the larger the amount of cow's milk in the diet the greater the number of days of that intervention required. A trend toward statistical significance was seen for ventilator days; however, only parenteral nutrition (PN) days and days to full feeds achieved statistical significance. CONCLUSIONS: Incorporation of any cow milk-based nutrition into the diet of extremely premature infants correlates with more days on PN and a longer time to achieve full feeds. There was a nonstatistically significant trend toward increased ventilator days. These represent additional clinical consequences of the use of any cow milk-based protein in feeding EP infants.