Differential neuronal vulnerability varies according to specific cardiopulmonary bypass insult in a porcine survival model

Objective We investigated whether the degree of vulnerability of different areas in the developing brain varies according to the specific mechanism of the insults caused by cardiopulmonary bypass. Methods A meta-analysis of 2 experimental studies (n = 80) was conducted. The end points of the otherwise identical studies were tissue oxygen index in the first experiment, whereas cerebral microvessel vasoconstriction and inflammatory response of endothelial cells were directly visualized in the second study. We assigned ultra–low flow bypass at 25 °C for 60 minutes as control; circulatory arrest at 25 °C for 60 minutes as ischemic stress under circulatory arrest (ischemia–CA); and ultra–low flow bypass at 34 °C for 60 minutes as the stress under ultra–low flow bypass (ischemia–ULF). Histologic neuronal damage was the primary outcome. Secondary measures included neurologic recovery. Results Vasoconstriction after ischemia and inflammation after bypass were independent predictors of severe histologic damage. The caudate nucleus was significantly vulnerable to ischemia–CA and was significantly influenced by vasoconstriction. In contrast, the hippocampus was significantly vulnerable to ischemia–ULF. The different forms of ischemic insults did not influence Purkinje cells, whereas Purkinje damage significantly correlated with inflammation. Tissue oxygen index had the ability to differentiate accurately regional damage. Neurologic recovery under ischemia–CA was significantly worse compared with ischemia–ULF. Neurologic recovery correlated with neuronal damage in the caudate nucleus, but it did not correlate with damage in the hippocampus. Conclusions Neuronal vulnerability in different areas of the developing brain varies according to mechanisms of bypass-induced ischemic stress. Certain regional damage may not be apparent in assessing acute neurologic recovery.