60 - Implementation of an enteral feeding algorithm to improve outcomes in infants with moderate and severe hypoxic-ischemic encephalopathy undergoing therapeutic hypothermia

Publication Number: 1053.60

Ahreen Allana, UT Southwestern Medical Center, Dallas, TX, United States; Rachel Clark, Ut southwestern, Dallas, TX, United States; Emily Bordier, University of Texas Southwestern Medical School, Dallas, TX, United States; Jennifer Joukhdar, University of Texas Southwestern Medical School, Dallas, TX, United States; Emma L. Ramon, University of Texas Southwestern Medical School, Wylie, TX, United States; Valerie Simcik, University of Texas Southwestern Medical School, Dallas, TX, United States; Tetyana H. Nesterenko, University of Texas Southwestern, Dallas, TX, United States; Lina Chalak, Childrens Health Dallas UT Southwestern, DALLAS, TX, United States; Rachel L.. Leon, University of Texas Southwestern Medical School, Dallas, TX, United States

Background: Infants with moderate and severe hypoxic-ischemic encephalopathy (HIE) commonly undergo therapeutic hypothermia (TH) for neuroprotection. Recent evidence indicates that early enteral feeding during TH is safe and may improve outcomes, however, practices remain variable. Our unit historically deferred feeds until rewarming. Implementing a standardized feeding algorithm for clinically stable infants with HIE undergoing TH aimed to optimize nutritional outcomes and practice consistency.

Objective: To reduce the time to achieve full enteral feeds by 20% over 18 months among neonates with HIE undergoing TH in a level III NICU.

Design/Methods: A multidisciplinary team developed a feeding algorithm outlining eligibility criteria and guidelines for early enteral nutrition (Figure 1). Eligible infants received trophic feeds of mother's own or donor milk at 12-24 hours of life, initiated at 10-20 mL/kg/day via naso- or orogastric tube, maintained at this volume during TH and rewarming, and advanced thereafter. The primary outcome was time to full enteral feeds (day when parenteral nutrition was discontinued). The process measure was adherence to the algorithm and balancing measures included incidence of feeding intolerance and necrotizing enterocolitis (NEC). Key drivers focused on standardization, education, staff engagement, and patient safety (Figure 2). The algorithm was implemented in PDSA Cycle 1 (September 2024). PDSA Cycle 2 (March 2025) reinforced education and reminders. Baseline data (January 2023-February 2024) informed mean time to full feeds. An X-bar S chart monitored performance over time and special cause variations.

Results: Thirty-two infants received early enteral feeds after algorithm implementation. The baseline mean time to full feeds was 7.8 days. Following PDSA cycles 1 and 2, a downward trend was observed, with most infants achieving full feeds earlier than 7.8 days; however, special cause variation was not detected (Figure 3). Adherence to the algorithm was 59%, improving modestly during PDSA 2. No cases of feeding intolerance or NEC occurred post-implementation.

Conclusion(s): A standardized feeding algorithm for infants with HIE undergoing TH was safe and associated with a trend toward earlier achievement of full enteral nutrition, though not yet statistically significant. Low adherence and a small number of patients likely decreased the impact of new practice. Future PDSA cycles focused on provider engagement, real-time data feedback, and sustained adherence are warranted to achieve measurable improvement.

Figure 1
Feeding algorithm for infants undergoing therapeutic hypothermia (TH)

Figure 2
Key driver diagram summarizing interventions driving changes aimed at achieving the specific aim of decreasing the time to full enteral feeds by >20% over an 18-month period.

Figure 3
Xbar-S chart for the mean time to full enteral feeds and variability from January 2023 though July 2025, encompassing two PDSA cycles.
Xbar chart (top panel) demonstrates overall stable process. Following PDSA 1 and PDSA 2, some points fall below the baseline mean, suggestiong improvement tendencies, however not sustained enough to achieve a special cause variention.
The S chart (bottom panel) indicates consistent and controlled variability throughout the study period. Variability remained within control limits, indicating a stable process with a possible trend toward improvement but no confirmed process shift during the observed period.