Brief Summary:

Anesthesia-related neurotoxicity in the developing brain is still a concern although evidence in humans is debatable. Moreover, it is unclear whether repeated and/or prolonged exposures are harmless and whether their effects are more pronounced in newborns and infants with brains more vulnerable to injury. One such specific group of patients is children with congenital heart disease (CHD). Nearly, half of the school-age survivors with CHD exhibit neurodevelopmental symptoms. It is thus important to elucidate whether any plausible neurotoxicity of the commonly used anesthetic agents can be observed in this population, and whether specific neuroprotective strategies can be demonstrated within the frame of a randomized controlled trial (RCT). Animal data have shown that dexmedetomidine (DEX) induces neuroprotective effects only at well-adjusted doses. One major issue with trials of anesthetic neurotoxicity is the latency between the conduct of these studies and the assessment of neurodevelopmental outcome. In contrast, the use of biomarkers of neuronal injury could be extremely valuable. Serum Neurofilament Light (NfL) has been shown to be a sensitive and specific marker of neuronal injury and is associated with neurologic outcome of children with various pathologies. The investigators hypothesize that in congenital heart surgery, use of DEX as main anesthetic agent in conjunction with low dose sevoflurane results in less release of serum NfL and is thus potentially less neurotoxic compared to the current standard of care. The hypothesis is tested with a RCT including patients between 0 - 3y undergoing surgery with cardiopulmonary bypass. To avoid any neurotoxicity due to anesthetic overdose, intraoperative burst suppression will be avoided. In addition to postoperative comparison of serum NfL, postoperative electroencephalogram and neurodevelopmental outcome of both groups will be compared taking into consideration the genetic background.