Summary:We compared local cerebral glucose meta bolic rates (LCMR g 1u) that were determined with [ 18 FJfluorodeoxyglucose (FDG) and [ 14 C]glucose labeled in the 1, 2, 3-4, and 6 positions. Double label digital auto radiography was used with published kinetic models to determine LCMR gl u for FDG and glucose in the same an imals. Glucose showed metabolic rate dependent under estimation of LCMR gl u compared to FDG, which wors ened with increasing experimental times. The least un derestimation occurred with glucose labeled in the 6 Kinetic models have been proposed for the mea surement of local cerebral metabolic rate of glucose (LCMRg1u) with both radiolabeled glucose (Hawkins et aI., 1979; Lu et aI., 1983; Hawkins et aI., 1985) and glucose analogs such as deoxyglu cose (DG) (Sokoloff et aI., 1977) and fluorodeoxy glucose (FDG) (Reivich et aI., 1979; Phelps et aI., 1979;Miller and Kiney, 1981; Lear et aI., 1984; Olds et aI., 1985; Sako et aI., 1985; Lear and Ack ermann, 1988). DG and FDG cross the blood-brain barrier using the same facilitated transport system as glucose. In the brain, DG and FDG are phosphorylated by hexokinase to DG-6-phosphate (DGP) and FDG 6-phosphate (FDGP), as glucose is metabolized to Received December 22, 1987; accepted March 22, 1988. Address correspondence and reprint requests to Dr. J. L.Lear, at Nuclear Medicine (A034) , University of Colorado Health Sciences Center, 4200 E. 9th Ave., Denver, CO 80262 U.S.A.Abbreviations used: DO, deoxyglucose; DOP, deoxyglucose 6-phosphate; FDO, fluorodeoxyglucose; FDOP, fluorodeoxyglu cose 6-phosphate; LCMR gt u, local cerebral glucose metabolic rates.
575position at 6 min, reaching 10% in areas of high metabo lism. Labeling in the I position, the 2 position and the 3-4 position caused progressively worse underestimation at all times. In addition, some structures showed differ ences not directly related to metabolic rate, indicating re gional variations in relationships between individual ki netic constants of FDG and glucose. Key Words: Deoxy glucose-Fluorodeoxyglucose-Glucose-Autoradiog raphy -LCMR gl u-Metabolism. glucose 6-phosphate. Unlike glucose 6-phosphate, however, further metabolism of DGP and FDGP is extremely slow; therefore DGP and FDGP accu mulate within brain cells at a rate proportional to metabolism. Thus over time, the radio tracer in the brain will change from predominantly precusor (DG or FDG) to predominantly metabolic product (DGP or FDGP). DG and FDG generally are allowed to circulate for approximately 45 min, so that by the end of the experiment the majority of the radioac tivity in the brain is DGP or FDGP. This is impor tant because neither position emission tomography scanning nor autoradiography can differentiate be tween unmetabolized DG or FDG and the com pound of interest, DGP or FDGP. After estimation of the fraction of total brain activity which is DGP or FDGP, a "lumped constant" is then used to re late cerebral DGP or FDGP concentration to glu cose metabolism. This lumped constant represents the steady-st...