Continuous versus Bolus Gastric Tube Feeding in Very Low Birth Weight Infants Supported with Noninvasive Respiratory Support: A Randomized, Pilot Study

Abstract

Objectives This study aimed to compare time to full feeding (TFF) between continuous gastric feeding (CGF) and bolus feeding (BF) in very low birth weight (VLBW) infants supported with noninvasive ventilation (NIV) and to evaluate feasibility and identify methodological pitfalls for future large-scale studies. Study Design This study is a randomized controlled, prospective, pilot study. VLBW premature infants, supported with NIV, were randomized while still on trophic feeding <20 mL/kg/day to receive feeding over 2 hours of CGF or over 15- to 30-minute BF. The primary outcome was TFF. Analysis was done by intention to treat. Results Overall, 32 infants were included in this analysis, 17 in the CGF group and 15 in the BF group. Infants in the CGF group were significantly younger than the BF group (mean ± standard deviation [SD] gestational age [GA] 26.9 ± 1.2 vs. 28.9 ± 1.5 weeks, respectively). TFF was comparable with median (interquartile range [IQR]) for the two groups, 10.0 (10.0, 19.0) days in the BF group versus 12.0 (9.0, 13.0) days in the CGF group (p = 0.59). Feeding length was not found to significantly affect TFF in multivariate analysis correcting for GA. Groups were comparable in weight gain, gastrointestinal complications, length of NIV, bronchopulmonary dysplasia incidence, and age at discharge. Most infants from both groups (60% of BF and 70% of CGF) required changes in feeding length. Conclusion In this pilot study, among VLBW infants supported with NIV, TFF was comparable between the BF and CGF groups. These results should be interpreted with caution due to the small sample size and despite the multivariate analysis correcting for the different GA. Interestingly, most infants required changes in feeding length regardless of their allocation. This feasibility study emphasizes the need for careful attention to randomization and strict feeding protocols including criteria for switching allocation in future large-scale studies aimed at determining the preferred feeding length during NIV in VLBW infants. Key Points