This was a retrospective observational cohort study including two cohorts of neonates with critical CHD (defined as biventricular physiology with or without aortic arch obstruction [BVP-AO or BVP, respectively], or single ventricle physiology [SVP]), who underwent neonatal open-heart surgery with the use of cardiopulmonary bypass (CPB), and were born between 2009 and 2017 at the Wilhelmina Children’s Hospital, University Medical Center Utrecht, Utrecht, The Netherlands. The surgical treatment of CHD did not change during this period. The first cohort was born between 2009 and 2012 (BVP-AO and SVP), and perioperative aEEG monitoring and MRI were performed for research purposes. This cohort consisted of 37 neonates; however, 2 did not have perioperative aEEG recordings and thus were excluded. The local Medical Ethical Committee approved this study, and parental informed consent was obtained. The second cohort was born between February 2016 and June 2017 (BVP, BVP-AO, and SVP), and aEEG monitoring and MRI scans were performed clinically in both the early postnatal and perioperative periods. The Medical Ethics Committee provided permission to use the routinely obtained aEEG and MRI data for research purposes. Neonates with a gestational age ≤۳۶ weeks, confirmed genetic disorders, or multiple congenital anomalies were excluded from this study. In both cohorts, aEEG monitoring was started 6 hours before surgery and continued during surgery until at least 48 hours after surgery. In the second cohort, aEEG monitoring was also started as soon as possible after birth and continued for at least 36 hours. For all neonates, an aEEG monitor (BrainZ; Natus, Seattle, Washington) with a sampling rate of 256 Hz was used. Four subcutaneous needle electrodes (F4-P4 and F3-P3) were applied with a central reference electrode (Fz) measuring the impedance. aEEG recordings were qualitatively evaluated for background pattern (BGP), sleep-wake-cycling (SWC), and (electroencephalographic) ictal discharges by 3 experts who were blinded to the neonatal clinical course and MRI results. BGP was classified as continuous normal voltage (CNV), discontinuous normal voltage (DNV), dense burst suppression (BS+), sparse burst suppression (BS-), continuous low voltage, or flat trace (FT).9 CNV and also DNV were considered normal BGPs, because even healthy term neonates show discontinuous activity during quiet sleep,10 and also because aEEG brain activity might be suppressed by drugs such as morphine and midazolam.11,12 Administration of these drugs is standard clinical care during and after cardiac surgery in neonates with CHD. BS+, BS-, continuous low voltage, and FT were considered abnormal BGPs. SWC was graded as absent, imminent (immature), or normal. Ictal discharges were classified as single, repetitive, or status epilepticus. Examples of BGP, SWC, and ictal discharges have been presented by Stolwijk et al.