Power spectral analysis of heart rate variability during hyperinsulinemia in nondiabetic offspring of type 2 diabetic patients: evidence for possible early autonomic dysfunction in insulin-resistant subjects.

Sympathetic activation has been considered as a link between insulin resistance, hyperinsulinemia, and hypertension. However, little is known about the association between insulin sensitivity and autonomic regulation or about the effect of acute hyperinsulinemia on cardiac sympathovagal balance. The aim of this study was to investigate heart rate variability (HRV) during the euglycemic-hyperinsulinemic clamp in nondiabetic offspring of patients with type 2 diabetes. We studied 35 nondiabetic offspring of patients with type 2 diabetes and 19 control subjects. Probands were chosen from a 10-year follow-up study of patients with well-characterized type 2 diabetes according to their fasting C-peptide level (selected from both ends of the distribution) and from control subjects to form three groups: 1) a group including subjects who were offspring of type 2 diabetic patients with low C-peptide levels (deficient insulin secretion group [IS group], n = 17), 2) a group including subjects who were offspring of type 2 diabetic patients with high C-peptide levels (insulin-resistant group [IR group], n = 18), and 3) a control group without a history of type 2 diabetes in first-degree relatives (n = 19). HRV was assessed at baseline and at the steady state during the euglycemic-hyperinsulinemic clamp. Rates of whole-body glucose uptake (M value) were lower in the IR group than in the IS group and the control group (41+/-3 vs. 54+/-2 vs. 60+/-4 micromol x kg(-1) x min(-1), P < 0.01 and P < 0.01, respectively). In all groups, heart rate increased significantly during hyperinsulinemia. In the IR group, insulin infusion increased total power of HRV [from 7.70+/-0.15 to 8.05+/-0.15 ln(ms2), P < 0.01] and the low frequency-to-high frequency ratio (from 0.62+/-0.14 to 1.14+/-0.18, P < 0.01) and decreased power of the high frequency spectral component (from 5.73+/-0.17 to 5.43+/-0.16 ln(ms2), P < 0.05), whereas in other groups, changes in HRV were not significant. We conclude that the HRV response to acute hyperinsulinemia in the offspring of type 2 diabetic probands was likely to be modulated by the type 2 diabetic phenotype of the parent. In insulin-resistant subjects, autonomic dysfunction may be an earlier defect than hitherto acknowledged.