Effect of isotope infusion and sampling sites on glucose kinetics during a euglycemic clamp

The importance of the location of isotope infusion and blood sampling on calculating glucose kinetics was studied in five mongrel dogs in the basal state and when glucose turnover was increased during a hyperinsulinemic euglycemic clamp. [U-14C]glucose was infused into the pulmonary artery, unlabeled glucose was infused into a femoral vein to maintain euglycemia, and blood was sampled from the right atrium (AV mode) and the femoral artery (VA mode). In the basal state there was no difference between the AV or VA mode in plateau specific activity; hence, the calculated rate of appearance of glucose was the same with either mode. During the euglycemic clamp procedure, plateau specific activity of the AV mode was significantly lower than that of the VA mode (P less than 0.05). The rate of appearance of glucose calculated from the VA mode was almost identical to the rate of infusion of unlabeled glucose (13.0 +/- 1.4 vs. 12.6 +/- 1.4 mg.kg-1.min-1, respectively), but the rate of appearance of glucose calculated from the AV mode was 12% greater. This study demonstrates that the calculation of glucose kinetics is sensitive to differences in sampling site when the turnover rate is high relative to the mass flow rate (cardiac output times substrate concentration).