Abstract
Background. It has been well established that human milk feeding contributes to limiting lung diseases in vulnerable neonates. The primary aim of this study was to compare the need for mechanical ventilation between human milk-fed neonates with sepsis and formula-fed neonates with sepsis. Methods. All late preterm and full-term infants from a single center with sepsis findings from 2002 to 2017 were identified. Data on infant feeding during hospital admission were also recorded. Multivariate logistic regression analyses were performed to assess the impact of feeding type on ventilation support and main neonatal morbidities. Results. The total number of participants was 322 (human milk group = 260; exclusive formula group = 62). In the bivariate analysis, 72% of human milk-fed neonates did not require oxygen therapy or respiratory support versus 55% of their formula-fed counterparts (p < 0.0001). Accordingly, invasive mechanical ventilation was required in 9.2% of any human milk-fed infants versus 32% of their exclusively formula-fed counterparts (p = 0.0085). These results held true in multivariate analysis; indeed, any human milk-fed neonates were more likely to require less respiratory support (OR = 0.44; 95% CI:0.22, 0.89) than those who were exclusively formula-fed. Conclusion. Human milk feeding may minimize exposure to mechanical ventilation.