This study tested whether glucose labeled at the C-6 position generates metabolites that leave brain so rapidly that C-6-labeled glucose cannot be used to measure brain glucose phosphorylation (CMRGlc). In pentobarbital-anesthetized rats, the parietal cortex uptake of [14C]glucose labeled in the C-6 position was followed for times ranging from 10 s to 60 min. We subtracted the observed radioactivity from the radioactivity expected with no loss of labeled metabolites from brain by extrapolation of glucose uptake in an initial period when loss was negligible. The observed radioactivity was a monoexponentially declining function of the total radioactivity expected in the absence of metabolite loss. The constant of decline was 0.0077.min-1 for parietal cortex. Metabolites were lost from the beginning of the experiment. However, with correction for the loss of labeled metabolites, it was possible to determine an average CMRGlc between 4 and 60 min of circulation of 64 +/- 4 (SE; n = 49) mumol.hg-1.min-1.