621 - Feasibility and safety of early duodenal feeds in very low birth weight infants

Publication Number: 2606.621

McKenna Straus, Johns Hopkins All Children's Hospital, St Petersburg, FL, United States; Aaron M. Germain, Johns Hopkins All Children's Hospital, St Petersburg, FL, United States; John M. Morrison, Johns Hopkins All Children's Hospital, Saint Petersburg, FL, United States; Noura Nickel, Johns Hopkins All Children's Hospital, St Petersburg, FL, United States

Background: Premature infants have high rates of bronchopulmonary dysplasia (BPD) and are at increased risk for gastroesophageal reflux and aspiration which may exacerbate lung injury. Transpyloric feeds may decrease this risk. Duodenal feeds have been shown to be non-inferior to gastric feeds with regards to nutritional outcomes. However, there are no studies evaluating the feasibility and safety of early duodenal feeds in premature infants.

Objective: To determine the feasibility and preliminary safety of early duodenal feeds in very low birth weight (VLBW) infants.

Design/Methods: We conducted a prospective, interventional pilot study of infants < 1501 grams that were admitted to the NICU within 72 hours of life. Exclusion criteria included infants with perinatal depression, congenital anomalies, nil-per-os status or low level of respiratory support (high flow nasal canula (HFNC) ≤2L/min). Consented infants were transitioned from gastric to duodenal feeds once advanced to a feeding volume of 60 ml/kg/day. Duodenal feeds were continued until respiratory support was weaned to HFNC ≤2L/min, allowing for initiation of oral feeds. If a duodenal tube dislodged, it was replaced per protocol and included in the feasibility outcome. If parents did not consent to duodenal feeds, they were consented to the observational, non-interventional gastric feeds arm (i.e. control group). Feasibility was defined as >80% successful tube placement within 3 attempts confirmed by radiographic imaging. Safety was assessed as number of intestinal perforations related to tube placement. Secondary outcomes included mechanical ventilation days, BPD, sepsis, necrotizing enterocolitis, number of replaced enteral tubes and related radiographs obtained, length of stay and growth. Feasibility, safety, and secondary outcomes were summarized with counts and proportions.

Results: 24 patients were enrolled, 12 per study arm. Patient demographics were similar between study groups (Table 1). For the outcome of feasibility, 39/40 (98%) duodenal tubes were successfully placed/replaced. No intestinal perforations related to duodenal tube placement were observed. Secondary outcomes revealed a similar incidence of necrotizing enterocolitis, sepsis, BPD, and growth parameters at 36 weeks PMA (Tables 2, 3).

Conclusion(s): Early duodenal feeds in VLBW infants are both feasible and safe. Growth parameters and incidence of necrotizing enterocolitis, sepsis, and BPD were similar between duodenally and gastrically fed infants. Based on these studies, future prospective studies are required to determine the benefit of early duodenal feeds on respiratory outcomes.

Table 1. Demographic and clinical characteristics of cohort by study arm


Table 2. Outcomes at 36 weeks PMA by study arm


Table 3. Comorbidities by study arm
Incidence of comorbidities occurring during admission