Background: To evaluate the efficacy and safety of gastric lavage (GL) in neonates with coffee-ground hematemesis due to swallowing maternal blood by postnatal day 1.
Methods: This multicenter, randomized controlled trial included term neonates presenting coffee-ground hematemesis by postnatal day 1. Patients were randomly assigned to GL or non-GL groups. The primary outcome was the rate of feeding intolerance (FI) within 24 h. Secondary outcomes included onset of FI, residual gastric volume, vomiting episodes, weight changes, and bilirubin levels. Adverse events were monitored.
Results: Eighty neonates were analyzed (40 per group). FI occurred in 15.0 % of the GL group and 27.5 % of the non-GL group (p = 0.17). However, in the subgroup of neonates exclusively fed with formula, FI was significantly lower in the GL group than in the non-GL group (16.0 % vs. 47.4 %, p = 0.02). GL significantly reduced residual gastric volume but showed no significant differences in other secondary outcomes. No adverse events related to GL were observed. Post hoc analysis indicated that a larger sample size would be needed to detect statistical significance; and subgroup findings suggested potential benefit in selected populations.
Conclusions: GL did not significantly reduce the incidence of FI in neonates with coffee-ground hematemesis and thus it cannot be considered superior based on current results. However, it was well tolerated, and the study may have been underpowered. Exploratory analyses suggested potential benefits, particularly among formula-fed neonates, warranting further investigation.

