Pre- and postoperative magnetic resonance imaging in neonatal arterial switch operation using warm perfusion.

Neurological morbidity is a major concern in pediatric cardiac surgery. Cardiopulmonary bypass is one of the few modifiable factors affecting neurodevelopmental outcome. This study aimed to measure the incidence of abnormalities apparent by magnetic resonance imaging (MRI) after neonatal arterial switch operation using warm surgery. Neonates admitted for transposition of the great arteries underwent pre- and postoperative brain MRI. They were operated on using a warm perfusion method. The data collected included antenatal diagnosis, place of birth, gestational age, total maturation score as described by Childs, weight, cyanosis as assessed by minimal SpO(2) sustained for at least 10 min, balloon atrial septostomy, prostaglandin E1 infusion, need for neonatal intensive care, bypass time, time to extubation, and length of stay in intensive care. All of the MRI results were interpreted by the same senior specialist in pediatric neuroimaging, and lesions were classified as white matter injury, infarct, or hemorrhage. On preoperative exam, nine patients (42%) had one or more lesions, with infarct in four patients, white matter injury in four patients, and hemorrhage in five. We were unable to find any correlation between the data collected and brain injury. On postoperative exam, there was one new infarct, two new cases of white matter injury, and three cases of hemorrhage but no worsening of the preoperative lesions. Based on this initial experience with brain imaging, there is no deleterious effect of warm perfusion and no rationale to postpone surgery in neonates with "subclinical" brain injury.