PC1099: Time to Achieve Full Oral Feeds as a Predictor for Neurological Outcomes in Neonatal Encephalopathy

Background: Neonatal encephalopathy as a result of perinatal hypoxia-ischemia (Hypoxic-ischemic encephalopathy, HIE) is a significant cause of neonatal mortality and morbidity worldwide. Therapeutic hypothermia (TH) is now standard of care in the developed world, and the outcomes of HIE need to be redefined in the setting of treatment with TH. We review the evolving epidemiology of HIE at a tertiary care NICU in Norfolk, VA with the goal of identifying clinical predictors of long-term neurological outcomes in infants with HIE.

Methods: A retrospective chart review was conducted of all patients diagnosed with HIE and admitted to the NICU at the Children’s Hospital of the King’s Daughters in Norfolk, VA from 2012-2017. Prenatal, perinatal, NICU course, MRI, short-term and long-term outcome data were collected. Neurodevelopmental scores assessed by the Capute scales [Cognitive Adaptive Test/Clinical Linguistic and Auditory Milestone Scale (CAT/CLAMS] were assessed multiple times between 6-36 months in the NICU follow-up clinic. A composite primary outcome of survival without neurological abnormality (SWNA) was created – this included infants with HIE who survived without complications such as cerebral palsy, tracheostomy/ventilator dependence, G-tube dependence, seizures/epilepsy, visual impairment and sensorineural hearing loss. Data were analyzed using JMP®, Version 16.2.0. SAS Institute Inc., Cary, NC.

Results: A total of 93 patients with a diagnosis of HIE were admitted between 2012 and 2017. Incidence of death for infants admitted with a diagnosis of HIE was 15%. Table 1 details risk factors and outcomes stratified by gender. 10 minute Apgar scores < 5, gestational age < 38 weeks, cord/Infant blood gas pH < 6.9, need for chest compressions in the delivery room, hypotension requiring vasopressors and evidence of end-organ injury (oliguria/renal failure, elevated liver enzymes) were the strongest predictors of death. Evidence of hypoxic/ischemic injury on brain MRI was negatively associated with SWNA. Mean time to full oral feeds was significantly shorter in the SWNA group (17.9±16 vs. 9.7±5.7). After controlling for gestational age, odds of SWNA were 13.6 times lower in babies requiring more than 10 days to achieve full oral feeds. Shorter time to full oral feeds was also associated with higher CAT/CLAMS developmental quotient scores at 6, 12 and 18 months of age.

Conclusion: In this preliminary analysis of HIE data from a tertiary care center, we find that time to achieve full oral feeds is a strong predictor of long-term neurodevelopmental outcomes. Data collection is ongoing to include patients from 2017-2020, and final analysis will include logistic regression modeling to elucidate the individual contribution of risk factors to long term neurodevelopmental outcomes in infants with HIE.