Association Between Donor Human Milk Initiation Timing and Necrotizing Enterocolitis, Mortality, and Feeding Outcomes in Very-Low-Birth-Weight Infants: A Multicenter Retrospective Cohort Study Using Two Japanese Registries.

Background/

Objectives

Early enteral feeding in preterm infants remains controversial because, despite promoting intestinal maturation, concerns about necrotizing enterocolitis (NEC) persist. Donor human milk (DHM) is recommended when the mother's milk is unavailable, but the optimal initiation timing is unclear. This study evaluated the association between DHM initiation timing and outcomes in very-low-birth-weight (VLBW) infants in Japan.

Methods

This multicenter retrospective cohort study used data from a national human milk bank registry (2018-2023) and the Neonatal Research Network of Japan (NRNJ; 2022). Infants with birth weight <1500 g were categorized by DHM initiation timing (≤24, 25-48, or >48 h), with infants in the NRNJ cohort serving as the comparison group. The primary outcome observed was NEC incidence, including all stages; secondary outcomes included in-hospital mortality and time to full enteral feeding. Bayesian regression models were applied. Detailed feeding data were unavailable in the NRNJ cohort.

Results

Among 2962 infants, NEC incidence was low across groups. NEC occurred in 1.31% of the ≤24 h DHM group and 1.83% of the NRNJ group, with mortality rates of 4.53% and 6.23%, respectively. Although NEC incidence was numerically lower in the ≤24 h DHM group, estimates were imprecise because of limited events. Early DHM initiation was associated with lower in-hospital mortality and earlier full enteral feeding, particularly in infants with <1000 g birth weight.

Conclusions

Early DHM initiation was associated with lower mortality and earlier achievement of full enteral feeding in VLBW infants without elevated NEC risk. However, because detailed feeding information was unavailable in the comparison cohort, these associations should be interpreted cautiously. Given the observational design and heterogeneous nutritional exposures, further prospective studies are warranted.