We studied gut mucosal perfusion in 24 neonates requiring cardiopulmonary bypass (CPB). Group A patients (n = 12) had obstruction to their aorta such that gut perfusion before operation was dependent on flow through a ductus arteriosus (DA). Group B neonates were of similar age and size and required a similar duration of CPB, but did not have a DA. An orogastric tonometer allowed intermittent calculations of gastric intramucosal pH (pHi), and rectal mucosal perfusion ("flux") was monitored using laser Doppler flowmetry. Measurements of arterial base deficit, and lactate and pyruvate concentrations were made intermittently. Before CPB, mean femoral arterial pressure (MAP) and base deficit in group A were not significantly different from those in group B. However, mean flux before CPB was significantly lower and the lactate/pyruvate (L/P) ratio was significantly higher in group A compared with group B. Mean pHi was below normal (< 7.26) throughout the operative period in group A, although it remained normal (> 7.33) in group B. After corrective surgery, both during warm CPB and after CPB, we found no significant difference in MAP, L/P ratio or base deficit between the groups, but both flux and pHi were significantly lower in group A compared with group B. We conclude that neonates requiring aortic arch surgery may be at particular risk of gut mucosal hypoxia both before and after operation.