We used positron emission tomography (PET) to study the effects of mild hypoglycemia on cerebral glucose uptake and metabolism. Nine healthy men were studied under basal saline-infusion conditions, and during euglycemic and hypoglycemic clamp studies. Insulin was infused at the same rate (1 mU · kg−1 min−1) in both clamp studies. In euglycemic clamp studies, glucose was infused at a rate sufficient to maintain the basal plasma glucose concentration, whereas in hypoglycemic clamp studies, the glucose infusion rate was reduced to maintain the plasma glucose at 3.1 mM. Each study lasted 3 h and included a 30-min baseline period and a subsequent 150-min period in which insulin or glucose was administered. Blood samples for measurement of insulin, glucose, cortisol, growth hormone, and glucagon were obtained at 20- to 30-min intervals. A bolus injection of 5–10 mCi [18F]-2-deoxy-2-fluoro-D-glucose (2-DFG) was administered 120 min after initiation of the study, and plasma radioactivity and dynamic PET scans were obtained at frequent intervals for the remaining 40–60 min of the study. Cerebral regions of interest were defined, and concentrations of radioactivity were calculated and used in the three-compartment model of 2-DFG distribution described by Sokoloff. Glucose levels were similar during saline-infusion (4.9 ± 0.1 mM) and euglycemic clamp (4.8 ± 0.1 mM) studies, whereas the desired degree of mild hypoglycemia was achieved during the hypoglycemic clamp study (3.1 ± 0.1 mM, P < 0.05). The insulin level during saline infusion was 41 ± 7 pM. During the euglycemic and hypoglycemic clamp studies, insulin concentrations were 455 ± 21 and 421 ± 21 pM, respectively. The ratio of the forward- and reverse-transport rate constants for hexose transport across the blood-brain barrier (i.e., k1/k2 or the hexose distribution volume) was greater during the hypoglycemic clamp study (1.48 ± 0.47 ml/g) than during the saline-infusion (0.45 ± 0.05 ml/g) or euglycemic clamp (0.54 ± 0.11 ml/g) studies. The fractional utilization rate for 2-DFG [i.e., k1 · k3/(k2 + k3)] was similarly greater during hypoglycemia (0.06 ± 0.02 ml · g−1 · min−1) than during saline-infusion (0.03 ± 0.01 ml · g−1 · min−1) or euglycemic clamp (0.04 ± 0.01 ml · g−1 · min−1) studies. The increased hexose distribution volume and cerebral fractional utilization of glucose resulted in the maintenance of the cerebral glucose metabolic rate during hypoglycemia (46 ± 12 (μmol · 100 g−1 · min−1); thus, the measured rate in this condition was similar to those found during the saline-infusion (36 ± 3 μmol · 100 g−1 · min−1) and euglycemic clamp (42 ± 12 μmol · 100 g−1 · min−1). studies. In conclusion, in the presence of mild hypoglycemia, the brain extracts an increased fraction of 2-DFG from the blood. The constant rate of cerebral glucose metabolism suggests that this enhanced glucose uptake protects the brain from acute reductions in plasma glucose.

This content is only available via PDF.