We used proton magnetic resonance spectroscopy to study 11 children (age < 8 years) with congenital heart disease undergoing cardiopulmonary bypass to determine whether low (10 +/- 4; n = 6) vs high (20 +/- 4; n = 5) perfusate hematocrits during bypass resulted in changes in brain metabolites which correlate with neurologic injury. Long and short echo time single voxel magnetic resonance spectroscopy in occipital gray matter and neurologic assessment were performed preoperatively and 2 and 5 days postoperatively. We also determined whether prolonged periods at low flow rates during bypass affected spectroscopy variables. We found no significant differences in metabolite ratios between the low vs high hematocrit groups or the lower vs higher flow rate groups (repeated measures analysis of variance of observation ranks converted to normal scores). However, our study was limited by statistical power due to the small sample size, therefore no conclusions could be made. Additional studies involving a greater number of patients are necessary. In all 11 children, magnetic resonance spectroscopy detected a significant decrease in brain N-acetyl-aspartate, and increases in myoinositol and glutamate/glutamine after surgery (Quade test) demonstrating that magnetic resonance spectroscopy is sensitive in detecting subtle postoperative changes in brain metabolites.