Cerebral oxygen saturation does not normalize until after stage 2 single ventricle palliation

Abstract

Background: In an effort to optimize neurologic outcome, cerebral oxygen saturation (SCO2) is often measured intraoperatively and postoperatively. We hypothesized that SCO2 would be related to stage of palliation in children with single ventricle congenital heart disease.

Methods: Cerebral oxygen saturation was continuously recorded intraoperatively in 34 infants and children undergoing palliative surgery on cardiopulmonary bypass for single ventricle congenital heart defects and in a control group of 12 neonates with ductus-dependent circulation undergoing complete repair. Saturations were correlated with the patient's stage and outcome.

Results: Baseline SCO2 was 61% in single ventricle neonates (group P1, n = 10), 55% in neonates undergoing repair (group R), 42% in infants undergoing stage 2 palliation (group P2, n = 6), and 70% in children undergoing Fontan (group P3, n = 14). Baseline was lowest (41%) in infants undergoing interstage operations (group I, n = 4). After bypass, there was a significant improvement in SCO2 to 53% in group P2 infants (p = 0.04); there were no significant changes in the other groups. By the end of the operation, there was a significant decrease in SCO2 to 48% in group P1 (p = 0.001), with other groups unchanged from baseline. There were five perioperative deaths. Cerebral oxygen saturation at the conclusion of surgery was lower in children who died (38% versus 61%, p = 0.01).

Conclusions: In children with single ventricle physiology, SCO2 decreases after initial palliation, remains low before second-stage palliation, but is normal before and after the Fontan. This has implications for perioperative mortality, neurologic injury, and potentially for interim mortality. Low postoperative SCO2 predicts perioperative mortality.