In isotopic experiments, the labeling pattern of glutamate opens a window on hepatic metabolism, particularly the citric acid cycle, gluconeogenesis and fatty acid oxidation. This is because glutamate is in isotopic equilibrium with alpha-ketoglutarate, whose labeling pattern is influenced by the following: 1) the contributions of glucose and fatty acids to acetyl-CoA, 2) the relative contributions of pyruvate carboxylase and pyruvate dehydrogenase to the entry of pyruvate carbon into the citric acid cycle, and 3) the rate of gluconeogenesis in relation to citric acid cycle activity. In humans and primates, hepatic glutamate can be sampled noninvasively via urinary phenylacetylglutamine, which is formed in liver from phenylacetate (a side product of phenylalanine catabolism) and glutamine (which equilibrates with liver glutamate and alpha-ketoglutarate). The (14)C- or (13)C-labeling pattern of the glutamate moiety of phenylacetylglutamine can be measured by sequential degradations to (14)CO(2), gas chromatography-mass spectrometry or nuclear magnetic resonance (NMR). When phenylacetylglutamine is labeled from singly labeled [(14)C]- or [(13)C]substrates, relative metabolic rates can be computed from the labeling pattern using Landau's model. In diabetic patients infused with [3-(13)C]pyruvate, the noninvasive sampling of hepatic glutamate via phenylacetylglutamine allows one to test the degree of liver insulinization via the (pyruvate carboxylase)/(pyruvate dehydrogenase) activity ratio. This ratio regulates gluconeogenesis in part. Its measurement may allow the identification of patients who might benefit from the intraperitoneal administration of insulin, or from recently developed antidiabetic drugs.