Horm Metab Res 1984; 16: 164-166
DOI: 10.1055/s-2007-1014924
© Georg Thieme Verlag, Stuttgart · New York

Can Mixed Venous Blood Be Used to Measure Insulin Action during the Hyperinsulinemic Clamp?

J. Andrews, I. Klimes, Barbara Vasquez, M. Nagulesparan, G. M. Reaven
  • Human Diabetes Study Center, University of Texas; Phoenix Clinical Research Section, National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases, National Institutes of Health, Phoenix, Arizona; and the Department of Medicine, Stanford University School of Medicine, Palo Alto, California, U.S.A.
Further Information

Publication History

1983

1983

Publication Date:
14 March 2008 (online)

Summary

Mandatory use of arterialized venous blood for glucose measurement during insulin clamp studies can preclude its use in obese subjects. In order to assess the distortion produced by performing clamp studies with mixed venous blood, we have carried out the present study. Sixteen subjects (BMI range 21-53 kg/m2) whose glucose tolerance varied from normal to diabetic, had hyperinsulinemic clamps performed at plasma insulin concentrations approximately 250 μU/ml above basal, using arterialized venous (a.v.) blood for plasma glucose and insulin measurements. The test was repeated on the same subjects, but mixed venous (m.v.) blood was substituted for a.v. blood to determine plasma glucose and insulin concentrations. During the two studies, the mean (± SEM) steady state plasma glucose (a.v. = 110.6 ± 8.8mg/dl;m.v. = 111.3 ± 9.8 mg/dl) and the mean incremental increases in insulin concentrations (a.v. = 263 ± 20 μU/ml; m.v. = 273 ± 21 μU/ml) were essentially identical. Furthermore, there was almost a perfect correlation between the individual values obtained for both SSPG (r = 0.97, p < 0.001) and incremental insulin levels (r = 0.98, p < 0.001) with the two methods. In addition, the amounts of glucose metabolized calculated from the two methods were very similar (a.v. = 4.96 ± 0.67 mg/kg/min; m.v. = 5.25 ± 0.76 mg/kg/min) and the two determinations of M were highly correlated (r = 0.97, p < 0.001). Finally, the relative rank order of the M values obtained for the 16 subjects during the two studies were extremely comparable. These results suggest that the use of m.v. blood to perform insulin clamp studies may provide a reasonable alternative to the use of a.v. blood, and that either method should result in similar conclusions as to differences in in vivo insulin action between individuals.