Introduction: Insulin exerts several opposing hemodynamic vascular actions, including stimulation of both nitric oxide-mediated vasodilation and endothelin-1-mediated vasoconstriction. Notably, hyperglycemia switches the vascular response to insulin from dilation to constriction in pre-clinical studies. To our knowledge, hyperglycemia’s effect on insulin’s action on vascular stiffness in humans is unknown. The American Heart Association recently recommended that vascular stiffness be determined noninvasively by carotid-femoral pulse wave velocity (cfPWV) and/or using wave separation analyses (WSA) instead of aortic augmentation index (AIx).
Methods: Thirteen lean, healthy subjects twice received a 1 mU/kg/min 2-hour insulin clamp. In one study, subjects were euglycemic and in the second, hyperglycemic (∼200 mg/dL) for 2 hours before and throughout the clamp. In each study, octreotide was infused to block endogenous insulin secretion. We assessed vascular stiffness with cfPWV, AIx, and WSA at baseline and after the 2-hour insulin clamp. Linear mixed model was used for statistical analysis, with Bonferroni correction for between-study comparisons. Spearman correlation was used to evaluate the relationship between PWV and WSA components.
Results: Compared to baseline, insulin increased cfPWV (p=0.03) during hyperglycemia without affecting AIx or components of WSA. No changes were observed during euglycemia, and insulin levels did not differ between groups. Neither AIx nor WSA components (i.e., backward and forward wave amplitude and reflection magnitude) correlated with cfPWV.
Discussion: We provide the first evidence that 4 hours of moderate hyperglycemia unmasks an action of physiologic hyperinsulinemia to increase central aortic stiffness (as measured by cfPWV) in healthy humans. Second, our results suggest that neither AIx nor WSA components correlate with cfPWV as measures of vascular stiffness.
W.B. Horton: None. L. Jahn: None. L. Hartline: None. J.T. Patrie: None. E. Barrett: None.