The dawn phenomenon does not occur in normal elderly subjects.

To determine whether the dawn phenomenon occurs in normal elderly subjects and thus contributes to the progressive mild fasting hyperglycemia of aging, we examined the effect of physiological insulin levels on glucose disposal and hepatic glucose production (HGO) between 0530 and 0800 h, and 0930 and 1200 h. Paired euglycemic insulin clamp studies (8 mU/m2 X min) were performed on healthy old subjects (n = 5), employing [3H]glucose methodology to measure glucose production and disposal rates. Basal plasma insulin, GH, glucagon, and cortisol levels, and HGO and glucose disposal rates were similar before each study. Steady state plasma insulin values were slightly, but not significantly, lower during the dawn study [dawn: 20.3 +/- 1.1 (SE); control: 23.5 +/- 2.1 microU/ml, P = 0.08]. Insulin clearance rates were higher during the dawn study (dawn: 523 +/- 16; control: 430 +/- 19 ml/m2 X min, P less than 0.01). Maximum glucose disposal rates (dawn: 3.10 +/- 0.24; control: 3.03 +/- 0.23 mg/kg X min) and minimum HGO levels (dawn: 0.83 +/- 0.09; control: 0.62 +/- 0.03 mg/kg X min) were not significantly different in each part of the study. There was a significant decrease in plasma GH during the dawn (P less than 0.01, analysis of variance) but not the control studies. There was no difference in cortisol levels during the euglycemic clamp between the dawn and control studies. The mean decrement in glucagon during the insulin infusion was similar in each part of the study. We conclude that the dawn phenomenon does not occur in healthy elderly subjects despite an increase in insulin clearance during the dawn period.