Neonatal cardiac surgery: Preventing collateral damage

Abstract

Neonates with complex congenital heart disease are at risk for cerebral injury. Subsequently they may develop adverse neurodevelopmental sequelae at a later age. Studies using magnetic resonance imaging (MRI) have shown that both before, and after cardiac surgery, new cerebral lesions can be detected. Especially the cases needing surgery for an aortic arch obstruction, such as a coarctation or interruption of the aorta, or the more severe hypoplastic left heart syndrome, are at risk for peri-operative brain injury. This is considered to be due to the fact that continuous use of cardiopulmonary bypass (CPB) is not feasible during these procedures, as the arterial CPB cannula cannot remain in situ during the aortic arch reconstruction. Therefore, deep hypothermic circulatory arrest (DHCA) was developed, which is performed by cooling to approximately 18°C while on CPB, and removing the CPB cannulae from the surgical field to facilitate the aortic arch repair itself. Later, CPB is re-instituted and the patient rewarmed. A newer technique is antegrade cerebral perfusion (ACP), where after cooling, the CPB cannula is advanced into the innominate artery and in this way the brain is selectively perfused. It was not yet known which technique was superior in terms of cerebral outcome. Studies comparing neurocognitive outcome after the two types of surgery had failed to find a difference. We hypothesized that there may be a difference in cerebral injury when assessed soon after surgery, using MRI. In our randomized controlled trial, we included neonates undergoing aortic arch surgery and performed pre- and postoperative MRI scans. We observed that in 50% of infants there was cerebral injury already before surgery, and in 72% of the DHCA group, and 78% of the ACP group, new injury was seen after surgery - with no difference in incidence of new injury after the two techniques. Most injury was white matter injury, but specifically after ACP, there were cases with new central focal infarctions on the right. We hypothesize that this is due to particles coming from the CPB machine which enter the right carotid artery. We also assessed somatic outcome and general postoperative recovery, after DHCA and ACP. We found that longer duration of DHCA was associated with a longer length of stay on the intensive care unit, as well as longer mechanical ventilation, and higher markers of hemodynamic, renal and gastro-intestinal dysfunction. In contrast, ACP did not have any effect on these somatic outcome markers. We conclude that neonates with aortic arch obstructions remain at high risk for cerebral injury both before and after surgery. There was no difference in the total incidence of new injury between DHCA and ACP, but ACP does specifically carry a higher risk of cerebral infarctions. On the other hand, ACP can provide somatic protection during surgery. In future, efforts should be made to improve the ACP technique to eliminate the risk for infarctions. Furthermore, the many factors involved in peri-operative care need a thorough investigation to assess which are important to reduce the risk of cerebral injury.