Cerebral vascular effects of aortovenous cannulations for pediatric cardiopulmonary bypass

The effects of aortovenous cannulations for pediatric cardiopulmonary bypass on cerebral blood flow velocity (CBFV) and electroencephalography (EEG) were evaluated. CBFV and EEG were continuously recorded before (baseline), during, and after cannulations until initiation of cooling (mean +/- 95% confidence interval). Vasopressors and/or volume replacement were administered if mean arterial pressure (MAP) decreased below 35 mm Hg. Cannulation-related EEG slowing was used as a criterion for electrocortical alteration. We studied 124 children (3 days to 17 years of age). Aortic and venous cannulations decreased mean CBFV by 10+/-3% and 13+/-4%, respectively, from baseline (p < 0.001). MAP diminished (p < 0.01) by 8+/-3% and 12+/-4%, respectively, from precannulation values (53+/-2 mm Hg). Right atrial cannulation, which was often chosen because the patient was hemodynamically unstable, was more frequently associated with pharmacologic intervention when compared with superior vena cava (SVC) cannulation (p < 0.01). Transient EEG alterations (n = 20) were associated with persistently low MAP (< 30 mm Hg), low CBFV (< 69%), and aortic (n = 4) or SVC (n = 7) cannula malposition. Infants with right atrial cannulation and intervention had more frequent EEG alterations (p = 0.04). Patients requiring intervention were younger (p < 0.01) and had longer hospital stay (p < 0.01) than those without intervention. Cerebral effects of cannulations are greater in young infants. This was found to be associated with low MAP during heart manipulation or consequence of cannula malpositions.