476 - Comparison of echocardiographic measurements and cardiac biomarkers between neonates with neonatal encephalopathy who developed brain injury or not during therapeutic hypothermia

Poster Number: 476
Publication Number: 476.343

Anie Lapointe, CHU Ste-Justine, Montréal, PQ, Canada; Emmanouil Rampakakis, McGill University Faculty of Medicine and Health Sciences, Montreal, PQ, Canada; Jessica simoneau, McGill University Faculty of Medicine and Health Sciences, Montreal, PQ, Canada; Pia Wintermark, McGill University Faculty of Medicine and Health Sciences, Montreal, PQ, Canada; Gabriel Altit, McGill University Faculty of Medicine and Health Sciences, Montreal, PQ, Canada

Background: Despite therapeutic hypothermia (TH), many newborns with neonatal encephalopathy (NE) develop brain injury. An in-depth understanding of their hemodynamic alterations may identify cardiovascular-related targets to further improve their management and thus their outcomes.

Objective: Compare the echocardiographic (ECHO) parameters and cardiac biomarkers on day of life (DOL) 2 between neonates with NE undergoing TH developing or not brain injury.

Design/Methods: Prospective cohort study in a single tertiary unit enrolling neonates (born ≥36weeks and ≥1800 grams) undergoing TH for NE, between 2016 and 2020. All neonates presented on admission with moderate to severe NE by amplitude-integrated electroencephalogram (aEEG) profile and underwent a brain magnetic resonance (MRI) on DOL 2. Extent of injury was defined as per validated MRI scoring system. ECHO was performed on the MRI day and data were extracted offline by an expert blinded to the outcomes according to standards by the American Society of Echocardiography. Biomarkers were collected (highest troponins and creatine kinase during neonatal stay). Multiple logistic regression analyses were constructed using StataSE to determine which biomarkers best predicted brain injury or not.

Results: 34 neonates were recruited, of which 19 (56%) had brain injury. Those with brain injury had a higher gestational age (GA) at birth (39.6±1.6 vs 38.2±1.5, p=0.01) and more severe pattern on aEEG (p=0.007) (Table 1). Mortality (p < 0.001) was also more frequent. Left (LV) and right (RV) ventricular performance were similar between groups by standard ECHO measures, strain (Speckle-Tracking) and tissue Doppler imaging (Table 2). Although tricuspid annular plane systolic excursion was increased in those with brain injury (p=0.04), the Z-Score accounting for GA was similar. In the overall cohort, markers of significant RV and LV dysfunction were assessed: a TAPSE Z-score -16% in 6 (18%) and a RV-pLS >-14% in 12 (35%). Initial aEEG pattern was the best predictor of brain injury, when accounting for GA and ECHO parameters/biomarkers (Table 3).Conclusion(s): Although signs of biventricular dysfunction were frequent in this small cohort of neonates with NE, initial aEEG pattern and not hemodynamic profile was most predictive of brain injury on MRI. Future studies, with a larger sample size, may detect which hemodynamic measure better predicts brain injury, as well as later neurodevelopmental outcomes.
Table 1. Neonatal characteristicsResults are presented as count (percentage), mean ± standard deviation or median [Interquartile Range: 25th – 75th].
*p-value for comparison by X2 test for categorical data and Wilcoxon-Mann-Whitney test for continuous data.
Abbreviations: aEEG: amplitude-integrated electroencephalogram; DOL: day of life; NICU (neonatal intensive care unit); MRI: magnetic resonance imaging; PaO2 (partial pressure of oxygen in arterial sample), SNAPPE-2: Score for Neonatal Acute Physiology with Perinatal Extension-II; VIS: Vasoactive Index Score
Table 2: Comparison of echocardiographic measurements by brain injury statusResults are presented as frequency (%), mean ± standard deviation or median [25th – 75th].
* X2 test for categorical data and Wilcoxon-Mann-Whitney test for continuous data.
** Z-score as per Koestenberger M, et al. Systolic right ventricular function in preterm and term neonates: reference values of the tricuspid annular plane systolic excursion (TAPSE) in 258 patients and calculation of Z-score values. Neonatology. 2011;100(1):85-92.
Abbreviations: A (late atrial velocity of ventricular inflow by pulse wave Doppler); a’ (late myocardial diastolic velocity by tissue Doppler imaging); aEEG: amplitude-integrated electroencephalogram; E (early velocity of ventricular inflow by pulse wave Doppler); e’ (early myocardial diastolic velocity by tissue Doppler imaging); FAC (fractional area change); LV (left ventricle), MPI (myocardial performance index or Tei index calculated using the time intervals obtained by tissue Doppler imaging); PAAT (pulmonary artery acceleration time); RVET (right ventricular ejection time); RV (right ventricle); s’ (peak systolic velocity by tissue Doppler imaging), sPAP (estimated systolic pulmonary arterial pressure); sBP (systolic systemic blood pressure); TAPSE (tricuspid annular plane systolic excursion).