The Effect of Growth Hormone on Glucose Metabolism and Insulin Secretion in Man*

Abstract

Pharmacological doses of GH are known to impair glucose tolerance. In the present study we have employed the euglycemic insulin and hyperglycemic clamp techniques to examine the effect of physiological elevations in plasma GH concentrations (27 ± 2 ng/ml) on the tissue responses to insulin and on glucose-stimulated insulin secretion. Three types of studies (low dose insulin clamp, high dose insulin clamp, and hyperglycemic clamp) were performed in young healthy volunteers before and after the infusion of GH (2 μg/kg-h) for 2 and 12 h. In the low dose insulin clamp studies, the plasma insulin concentration was acutely raised and was maintained at 59 ±4 μU/ml, while plasma glucose was maintained at basal levels. After 2 h of GH infusion, insulin-mediated glucose metabolism was slightly, although not significantly, decreased (4.48 ± 0.56 vs. 5.04 ± 0.39 mg/kg-min) in the control study. After 12 h of GH infusion, however, insulin-mediated glucose uptake decreased by 32 ± 9% (3.36 ± 0.40 mg/kg-min; P < 0.01). Basal endogenous glucose production (2.02 ± 0.07 mg/kg-min) was suppressed by 89 ± 4% in the control study. After short term GH infusion, the degree of suppression (90 ± 5%) was similar to the control value. After 12 h of GH infusion, however, suppression of endogenous glucose production (78 ± 5%) was slightly less than that in controls (P < 0.05). In the high dose insulin clamp studies, GH was infused for either 2 or 12 h, after which plasma insulin levels were increased to 2292 ± 96 μU/ml while maintaining euglycemia (high dose insulin clamp study). Insulin-mediated glucose uptake was 11.22 ± 0.53 mg/kg-min in the control study, slightly lower during short term GH infusion (9.56 ± 1.00 mg/kg-min) and significantly diminished after long term GH infusion (7.16 ± 1.12 mg/kg-min; P <0.02). In the hyperglycemic clamp studies, plasma glucose was raised and maintained at 125 mg/dl above the basal level. Glucose metabolism in controls (8.68 ± 0.53 mg/ kg-min) was decreased by 31 ± 7% after 2 h and by 44 ± 6% after 12 h of GH infusion. The mean increment in plasma insulin in response to hyperglycemia (control, 56 ± 10 μU/ml) was unaltered after 2 and 12 h of GH administration. Total binding of [125I]insulin to monocytes was 7.0 ± 0.5% before GH infusion. After 2 h of GH administration, no change in total specific insulin binding (7.4 ± 0.4%) occurred. After 12 h of exposure to GH, however, total binding was decreased (5.9 ± 0.4% P < 0.02) due to a decrease in receptor affinity. We conclude that physiological elevations in GH 1) induce state a of insulin resistance within 2–12 h, 2) only slightly impair insulin's suppressive effect on endogenous glucose production, indicating that the primary site of insulin resistance resides in peripheral tissues, 3) do not alter the plasma insulin response to hyperglycemia, and 4) cause a decrease in insulin binding that results from a decrease in receptor affinity. The inability to overcome the defect in glucose metabolism at high plasma insulin concentrations suggests that a significant postreceptor defect (in addition to the receptor defect) contributes to the observed insulin resistance. (J Clin Endocrinol Metab55: 973, 1982)