211 - Relationship of postnatal medications with regional brain volume and cognition-neurobehavior outcomes in very preterm infants

Poster Board #: 211
Publication #: 2909-HP-QA.211

Background: Up to 30% of surviving very preterm infants (VPT) suffer from behavioral and cognitive impairments even in absence of severe intraventricular hemorrhage (IVH). Medications administered during intensive care can impact brain development.


Objective: To investigate the relationship between postnatal medications (steroids, opioids, benzodiazepines) and brain structure (3D volume at term equivalent age magnetic resonance imaging [TEA-MRI]) and function (18-month neurodevelopment in VPT infants.


Design/Methods: We conducted a prospective cohort study enrolling VPT infants born <32 weeks gestational age (GA) and admitted within the first 2 weeks of life in a level IV neonatal intensive care unit (NICU). Infants with known genetic syndromes, major anomalies and severe IVH (grade 3 or 4) were excluded. Infants underwent a TEA-MRI on a 3 Tesla GE scanner. 3D T2 CUBE images were segmented using ITK-SNAP to measure regional brain volumes. Infants were assessed at 15-18 months using Mullens Scales of Early Learning (cognitive impairment if < 2SD) and Modified Checklist for Autism in Toddlers (screen positive if ≥3), respectively. Cumulative medication doses (U/kg of daily med-calc weight) administered between NICU admission and TEA MRI were documented by prospective chart review and associations with brain volumes and cognitive-behavioral outcomes were investigated using bi- and multi-variate statistical tests.


Results: 57 VPT infants were included in this analysis (49 segmented TEA-MRI and 36 completed 15-18 mth assessment); born at 27.8±2.5 wks GA and underwent MRI at 39.4±2.6 wks post-menstrual age (Table 1). Cerebral white matter, hippocampus and amygdala volumes demonstrated negative associations with medications (Table 2), adjusted for covariates. Infants with cognitive-behavioral impairments had significantly lower white matter volumes and were administered higher cumulative doses of morphine, fentanyl and midazolam (Table 3). Systemic dexamethasone and hydrocortisone doses were associated with impaired cognitive outcome (Table 3). Adjusting for co-variates, GA was the dominant predictor of neuro-cognitive impairments.

Conclusion(s): We report lower cerebral white matter volumes, associated with higher doses of postnatal medications and accompanying cognitive and behavioral impairments in VPT infants. Future studies are needed to understand the role of potential mediators of impaired brain and child development, in absence of severe IVH.


Figure 1. Segmentation of 3D T2 CUBE images from term-MRI (coronal)


Table 1. Summary of clinical data and medication doses for all subjects in cohort


Table 2: Association of postnatal medication doses with regional brain volume


Table 3: Association of neurodevelopmental outcomes at 15-18 months age with regional brain volume and medication exposure from admission till term-corrected age MRI

---

Authors/Institutions: Sudeepta K. Basu, Children's National Hospital, Washington, District of Columbia, United States; Kushal J. Kapse, Children's National Hospital, Washington, District of Columbia, United States; Kristina M. Espinosa, Children's National Hospital, Washington, District of Columbia, United States; Anushree K. Kapse, Children's National Hospital, Washington, District of Columbia, United States; Stephanie Norman, Children's National Hospital, Washington, District of Columbia, United States; Devon Fischer, Children's National Hospital, Washington, District of Columbia, United States; Scott Barnett, Children's National Hospital, Washington, District of Columbia, United States; Nickie Andescavage, Children's National Hospital, Washington, District of Columbia, United States; Catherine Limperopoulos, Children's National Hospital, Washington, District of Columbia, United States