Early Enteral Feeding versus Intravenous Fluids and Necrotizing Enterocolitis among Preterm Infants at Kakamega County General Hospital: A Cohort Study

Abstract

Context: Necrotizing enterocolitis (NEC) is a leading cause of morbidity and mortality among preterm infants, with limited data on how initial feeding strategy shapes NEC-related outcomes in resource-limited African settings. Aim: This study aimed to evaluate early enteral feeding versus intravenous fluids and necrotizing enterocolitis outcomes, specifically NEC incidence and NEC-related mortality, among preterm infants at Kakamega County General Hospital, comparing those initiated on each feeding regimen within the first 24 hours of life. Methods: In this prospective observational cohort study, the study followed 209 preterm neonates born at 28-32 weeks gestation (birth weight 1000-1500g) for two weeks from admission at Kakamega County General Hospital, newborn unit using a structured data collection form. Neonates were categorized by initial feeding regimen received within the first 24 hours of life: early enteral feeding (n=59) or intravenous fluids (n=150). The study compared outcomes (NEC incidence, discharge status) using Fisher's exact test and multivariable logistic regression. Results: Overall NEC incidence was 15.8% (33/209). Neonates on early enteral feeding had lower NEC incidence (8.5%) than those on intravenous fluids (18.7%) (p= 0.091); absolute risk reduction 10.2%; number needed to treat ≈10). Although the early feeding group had a lower incidence, this difference was not statistically significant (p=0.091) and should be interpreted cautiously. Mortality was three times higher with intravenous fluids (10.0%) than early feeding (3.4%; OR=3.17, 95% CI: 0.70-14.30, p=0.161). NEC itself was strongly associated with mortality: case fatality was 27.3% among NEC-positive neonates versus 4.6% among unaffected neonates (OR=7.88, 95% CI: 2.76-22.48, p=0.0002). In the adjusted model, birth weight was the sole significant independent predictor of NEC (aOR=1.69 per 100g decrease, 95% CI: 1.09-2.60, p=0.018). Conclusion: The observed differences in NEC incidence and mortality by feeding type remain inconclusive and should be interpreted cautiously, warranting further adequately powered studies to confirm these observations.