The kinetics of transport across the blood-brain barrier and metabolism in brain (hemisphere) of [14C]2-fluoro-2-deoxy-D-glucose (FDG) were compared to that of [3H]2-deoxy-D-glucose (DG) and D-glucose in the pentobarbital-anesthetized adult rat. Saturation kinetics of transport were measured with the brain uptake index (BUI) method. The BUI for FDG was 54.3 +/- 5.6. Nonlinear regression analysis gave a Km of 6.9 +/- 1.2 mM and a Vmax of 1.70 +/- 0.32 mumol/min/g. The Ki for glucose inhibition of FDG transport was 10.7+/-44 mM. The kinetic constants of influx (k1) and efflux (k2) for FDG were calculated from the Km2, Vmax, and glucose concentrations of the hemisphere and plasma (2.3 +/- 0.2 mumol/g and 9.9 +/- 0.4 mM, respectively). The transport coefficient (k1 FDG/k1 glucose)was 1.67 +/- 0.07 and the phosphorylation constant was 0.55 +/- 0.16. The predicted lumped constant for FDG was 0.89, whereas the measured hexose utilization index for FDG was 0.85 +/- 0.16. CONCLUSIONS The value for the lumped constant can be predicted on the basis of the known kinetic constants of FDG and glucose transport and metabolism, as well as brain and plasma glucose levels. Knowledge of the lumped constant is crucial in interpreting data obtained from 18FDG analysis of regional glucose utilization in human brain in pathological states. We propose that the lumped constant will rise to a maximum equal to the transport coefficient for FDG under conditions of transport limitation (hypoglycemia) or elevated glycolysis (ischemia, seizures), and will fall to a minimum equal to the phosphorylation coefficient during phosphorylation limitation (extreme hyperglycemia).