Insulin enhances glucose-stimulated insulin secretion in healthy humans

Islet β-cells express both insulin receptors and insulin-signaling proteins. Recent evidence from rodents in vivo and from islets isolated from rodents or humans suggests that the insulin signaling pathway is physiologically important for glucose sensing. We evaluated whether insulin regulates β-cell function in healthy humans in vivo. Glucose-induced insulin secretion was assessed in healthy humans following 4-h saline (low insulin/sham clamp) or isoglycemic-hyperinsulinemic (high insulin) clamps using B28-Asp insulin that could be immunologically distinguished from endogenous insulin. Insulin and C-peptide clearance were evaluated to understand the impact of hyperinsulinemia on estimates of β-cell function. Preexposure to exogenous insulin increased the endogenous insulin secretory response to glucose by ≈40%. C-peptide response also increased, although not to the level predicted by insulin. Insulin clearance was not saturated at hyperinsulinemia, but metabolic clearance of C-peptide, assessed by infusion of stable isotope–labeled C-peptide, increased modestly during hyperinsulinemic clamp. These studies demonstrate that insulin potentiates glucose-stimulated insulin secretion in vivo in healthy humans. In addition, hyperinsulinemia increases C-peptide clearance, which may lead to modest underestimation of β-cell secretory response when using these methods during prolonged dynamic testing.     

Effect of obesity and insulin resistance on resting and glucose-induced thermogenesis in man. EGIR (European Group for the Study of Insulin Resistance)

OBJECTIVE: To assess the impact of obesity and insulin sensitivity on resting (REE) and glucose-induced thermogenesis (GIT). DESIGN: Data from 322 studies carried out in non-diabetic subjects of either gender, covering a wide range of age (18-80y) and body mass index (BMI, 18-50 kg/m2). MEASUREMENTS: Insulin sensitivity and thermogenesis were measured by combining the euglycaemic insulin clamp technique with indirect calorimetry. RESULTS: REE was inversely related to age (P = 0.001) and the respiratory quotient (P = 0.03), and positively related to BMI, lean body mass (LBM), fat mass, and percentage fat mass (all P<0.0001). In a multiple regression model, LBM-adjusted REE was estimated to decline by 9% between 18 and 80 y, independently of obesity and insulin sensitivity. In contrast, GIT was strongly associated with insulin sensitivity (P<0.0001) but not with gender, age or BMI. By multiple regression analysis, GIT was linearly related to insulin sensitivity after controlling for gender, age, BMI and steady-state plasma insulin levels. Furthermore, both of the main components of insulin-mediated glucose disposal (glucose oxidation and glycogen synthesis) correlated with GIT independently of one another. In the subset of subjects (n = 89) in whom waist-to-hip ratio (WHR) measurements were available, GIT was inversely associated with WHR (P<0.001 after adjustment by gender, age, BMI, insulin sensitivity and steady-state plasma insulin concentration). In this model, a significant interaction between WHR and gender indicated a stronger adverse effect on GIT of a high WHR in women than in men. CONCLUSIONS: In healthy humans, age, lean mass and respiratory quotient are the main independent determinants of resting thermogenesis. In contrast, insulin sensitivity and, to a lesser extent, abdominal obesity are the principal factors controlling glucose-induced thermogenesis.     

Failure of human insulin to influence endogenous basal insulin secretion in mild diabetics

In evaluating the possibility of self-regulation of insulin secretion in man, human insulin may be more appropriate as an inhibitor to be considered than insulins from other species because the differences in the structure of the hormones might play some part in this conflicting proposal. The present study was carried out to examine whether human insulin per se can exert a feedback effect on the insulin secretion of B-cell in mild diabetics under physiologic condition. Fifteen mild diabetics were given a two-hour infusion of human insulin at a constant rate of 40 mU/m/min after a priming dose of 160 mU/m/min for the first two minutes. The plasma glucose in nine of these patients were maintained at their basal level of 92.8 +/- 3.7 mg/dL (Group A) with a glucose clamp technique (the coefficient of variation = 5.0 +/- 0.8% during the clamp), while that in the remaining six patients were intentionally altered, within physiologic range, from 114.5 +/- 8.4 mg/dL to 83.8 +/- 4.9 mg/dL (Group B). During insulin infusion the plasma immunoreactive insulin (IRI) level were well-maintained at about 50 microU/mL level in both groups, whereas the C-peptide reactivity (CPR) in group B decreased from 1.28 +/- 0.15 ng/mL to 0.59 +/- 0.14 ng/mL in parallel to the change of plasma glucose, in contrast to the relatively stable CPR level of 0.92 +/- 0.08 ng/mL in group A.(ABSTRACT TRUNCATED AT 250 WORDS)     

Failure of insulin infusion during euglycemia to influence endogenous basal insulin secretion

Despite some evidence of self-regulation of insulin secretion, it is unclear whether endogenous insulin influences insulin secretion independently of blood glucose. The aim of the present study was to examine this question in humans. Seven healthy fasting men were given two-hour porcine insulin infusions (40 mU/min) with and without maintenance of euglycemia (glucose clamp). Intravenous glucose required to maintain basal blood glucose levels (4.2 ± 0.1 mmole/liter) during insulin infusion was 34.3 ± 3.0 gm with a mean rate of 273 ± 29 mg/min in the second hour of insulin infusion. During the glucose clamp, mean C-peptide levels were not significantly altered from fasting levels of 1.91 ± 0.24 ng/ml, but when blood glucose levels fell by approximately 1 mmole/liter, C-peptide fell to 0.37 ± 0.07 ng/ml. Plateau insulin levels were significantly higher during euglycemia than during mild hypoglycemia (53.2 ± 5.6 mU/liter versus 38.5 ± 3.6 mU/liter, P < 0.01). Plasma nonesterified fatty acids were suppressed equally in the two studies. However, a rise in plasma glucagon seen during mild hypoglycemia was absent when euglycemia was maintained. We conclude that insulin self-regulation (either direct or neurally mediated) is not physiologically important in the basal state in normal humans and that the blood glucose-insulin feedback loop dominates in the short-term control of basal insulin secretion.     

Effect of insulin and glucose on basal and cholecystokinin-stimulated exocrine pancreatic secretion in humans

Pancreaticobiliary secretion is reduced during acute hyperglycemia. In nondiabetics, this inhibitory effect also may result from hyperinsulinemia. Therefore we investigated the effects of acute hyperglycemia and euglycemic hyperinsulinemia on basal and cholecystokinin (CCK)-stimulated pancreaticobiliary secretion. Nine healthy volunteers (age, 22-52 years) were studied on three occasions in random order during (a) intravenous saline (control), (b) hyperglycemic hyperinsulinemic clamping (HG; plasma glucose at 15 mM), and (c) euglycemic hyperinsulinemic clamping (HI; plasma insulin at 150 mU/L, glucose at 4-5 mM). Duodenal outputs of bilirubin, amylase, trypsin, and bicarbonate were measured under basal conditions and during CCK infusion (0.25 and 0.5 IDU/kg/h). Basal pancreaticobiliary secretion was significantly (p < 0.05) reduced during both HG and HI. During low-dose CCK stimulation, HG significantly (p < 0.05) reduced bilirubin and trypsin output compared with control. In contrast, HI did not significantly reduce pancreatic enzyme and bilirubin output during low-dose CCK infusion. During high-dose CCK infusion, neither HI nor HG influenced pancreatic enzyme and bilirubin output. Pancreatic bicarbonate output was not influenced by CCK and remained significantly (p < 0.05) reduced during HI and HG compared with control. It is concluded that during both acute hyperglycemia and euglycemic hyperinsulinemia, basal pancreaticobiliary secretion is significantly reduced. CCK-stimulated pancreatic enzyme and bilirubin output is significantly reduced only during hyperglycemia. The inhibitory effect of hyperglycemia on pancreaticobiliary secretion in healthy volunteers may occur independent of insulin.     

Insulin sensitivity, insulin secretion, and glucose tolerance versus intima-media thickness in nondiabetic postmenopausal women

To study the association between insulin sensitivity and secretion vs. early manifestations of atherosclerosis, we performed a 5-yr prospective study in 84 nondiabetic, postmenopausal women, aged 58.7 +/- 0.4 yr (mean +/- SD). Insulin sensitivity was measured with the euglycemic, hyperinsulinemic clamp, and insulin secretion was measured as the acute response to iv arginine (5 g). Early atherosclerosis was studied by ultrasonography of the right carotid artery. Mean intima-media thickness (IMT), determined 1 cm proximal to the bifurcation, was 0.81 +/- 0.14 mm at baseline and increased by 0.012 +/- 0.014 mm/yr over the 5 yr (P < 0.001). The maximal IMT, determined in the carotid bifurcation, was 1.42 +/- 0.42 mm at baseline and increased by 0.035 +/- 0.049 mm/yr (P < 0.001). Neither basal IMT nor the increase in mean or maximal IMT correlated to insulin sensitivity or secretion. In contrast, both baseline IMT and the progression in IMT over the 5-yr follow-up (both mean common carotid artery IMT and maximal bifurcation IMT) correlated with systolic blood pressure and low-density lipoprotein cholesterol. We conclude that carotid intima-media thickness is not related to insulin sensitivity or secretion in nondiabetic, postmenopausal women. Instead, the strongest association is seen with systolic blood pressure and low-density lipoprotein cholesterol levels.     

Insulin resistance versus insulin secretion in the hypertension of obesity.

We measured the degree of association between obesity, blood pressure, insulin resistance, and insulin secretion in 72 male and female obese hypertensive, obese nonhypertensive, and normal weight control subjects. Baseline weight, body mass index, percent body fat, waist/hip ratio, and systolic and diastolic blood pressures were obtained. Insulin sensitivity was assessed according to Bergman's minimal model. Twelve-hour urinary c-peptide was measured after a standard liquid meal. Insulin action was inversely associated with blood pressure status, obesity status, and age. Meal-stimulated c-peptide excretion significantly correlated with systolic blood pressure and percent fat but not with body mass index or age. Multivariate regression analysis indicated that, of the measures of body composition, percent fat and waist/hip ratio had the strongest correlation with insulin action either alone or in combination with c-peptide excretion. Obese hypertensive patients had an index of insulin action (10(-4).min-1/[microunits/ml]) of 1.34 +/- 0.19, which was significantly (p less than 0.003) lower than in the obese nonhypertensive patients (index, 2.26 +/- 0.10) or the nonobese subjects (index, 5.41 +/- 0.26, p less than 0.001). Meal-stimulated c-peptide excretion (nmol/kg lean body mass) was increased only in the obese hypertensive group (0.32 +/- 0.01) and was significantly higher (p less than 0.001) than in the obese nonhypertensive (0.16 +/- 0.01) or the nonobese subjects (0.14 +/- 0.01). These results support the hypothesis that abnormalities in blood pressure regulation, insulin-stimulated glucose uptake, and insulin secretion coexist.     

Influence of galactose on insulin secretion in humans

Zusammenfassung Galaktose ruft sowohl nach oraler als auch nach intravenöser Gabe eine Insulinsekretion hervor. Die Insulinsekretion wird durch die im Galaktose-katabolismus entstandene Glucose ausgelöst.

---

Influence of galactose on insulin secretion in humans

Summary Following oral and intravenous galactose administration stimulation of pancreatic insulin secretion has been observed. It must be attributed to the conversion of galactose to glucose.

Mit Unterstützung der Deutschen Forschungsge-meinschaft Bad Godesberg.     

Feedback regulation of basal pancreatic secretion in humans

The effect of intrajejunal infusion of pancreatic juice on basal pancreatic secretion was studied in patients who had received pancreatoduodenectomy for pancreatic, biliary, or duodenal malignancy. Pure pancreatic juice was obtained through a drainage tube inserted into the main pancreatic duct. There was little fibrosis in the pancreatic remnant and daily pancreatic juice output was more than 200 ml. After intraluminal infusion of pancreatic juice, water, protein, bicarbonate, and enzyme outputs were decreased significantly by about 30%. Intraluminal trypsin also reduced pancreatic secretion. Trypsin inhibitor (aprotinin) suppressed the significant decrease caused by autopancreatic juice or trypsin solution. We conclude that basal pancreatic secretion in humans is under negative feedback control by intestinal pancreatic juice or tryptic activity.     

Insulin resistance and insulin secretion in non-diabetic acromegalic patients.

Insulin resistance (IR) can be induced by high amounts of growth hormone (GH). AIM: To set up, in acromegaly without diabetes mellitus, a correlation between the disease activity in GH-secreting adenoma (AA) - assessed by minimum GH serum level during an oral glucose tolerance test (OGTT) - and severity of insulin resistance (IR), assessed by HOMA-IR index. METHODS: 75 out of 88 consecutive patients with acromegaly hospitalized in our department were included in this study. 13 patients proved to have diabetes mellitus and were excluded. Serum glucose, GH and insulin levels were measured by immunoradiometricassay basal and at 30, 60 and 120 minutes after a 75 g OGTT in 88 patients with active or cured acromegaly. IR was assessed using HOMA-IR index (Homa-IR=basal serum glucose (mg/dl) x basal serum insulin (mU/L)/22.5 x 18). A value over 2.5 was considered indicating IR. RESULTS: Out of 75 patients without diabetes mellitus, 36 subjects (48%) were presenting with IR (34 with active disease, 2 cured). We found a significant positive correlation (r=0.56, p<0.001) between AA and HOMA-IR. The GH minimal level corresponding to the intersection of the exponential regression curve with the HOMA-IR level of 2.5 was 8.8 ng/mL, a cut-off point indicating IR with 82% specificity and 78% sensitivity. The odds ratio for developing IR becomes significant at a minimum GH level during OGTT of 2 ng/mL (odds ratio 7.6, 95% confidence interval 2-29). CONCLUSIONS: The severity of IR revealed by acromegaly correlates with GH production. A GH serum level higher than 2 ng/mL during OGTT indicates an increased risk for developing IR. This cut-off level of GH can be used as one of criteria of cured disease, regarding the lack of metabolic effects.     

Glucose control of basal insulin secretion in diabetes

Summary The plasma insulin response to both a small increase and decrease in the plasma glucose has been studied in normal and diabetic, non-obese subjects. In a second investigation the plasma insulin concentrations were measured during a gradual reduction of the raised fasting plasma glucose of diabetes to normal levels. In both studies, diabetic patients were found to have a markedly impaired response of the fasting plasma insulin to small changes in plasma glucose. These results do not support the suggestion that stimulated and not basal insulin secretion is impaired in diabetes. Both modes of secretion are probably via the same B-cell release mechanism, which is deficient in diabetes. There was a gradation of response between maturity onset and juvenile onset diabetics.     

Comparison of the inhibitory effect of insulin and hypoglycemia on insulin secretion in humans

Although both insulin and hypoglycemia are known to inhibit endogenous insulin secretion, their potency to suppress insulin secretion has not been directly compared thus far. The serum C-peptide concentration was measured during 28 euglycemic and 28 stepwise hypoglycemic (4.1,3.6, 3.1, and 2.6 mmol/L) clamp experiments using either a low-rate (1.5 mU x min(-1) x kg(-1)) or high-rate (15.0 mU x mU(-1) x kg(-1)) insulin infusion. The experiments lasted 6 hours and were performed in 28 lean healthy men. During both the euglycemic and hypoglycemic clamps, serum insulin was approximately 40-fold higher during the high-rates versus low-rate insulin infusion (euglycemia, 24,029 +/- 1,595 v 543 +/- 34 pmol/L; hypoglycemia, 23,624 +/- 1,587 v 622 +/- 32 pmol/L). Under euglycemic conditions, serum C-peptide decreased from 0.54 +/- 0.04 to 0.41 +/- 0.05 nmol/L during the low-rate insulin infusion (P < .05) and from 0.55 +/- 0.07 to 0.27 +/- 0.09 nmol/L during the high-rate insulin infusion (P < .001). Under hypoglycemic conditions, serum C-peptide decreased from 0.50 +/- 0.03 to 0.02 +/- 0.01 nmol/L during the low-rate insulin infusion (P< .001) and from 0.46 +/- 0.07 to 0.02 +/- 0.01 nmol/L during the high-rate insulin infusion (P< .001). In the euglycemic clamp condition, the high-rate insulin infusion reduced the C-peptide concentration more than the low-rate insulin infusion (P < .05). Independent of the rate of insulin infusion, the decrease in C-peptide was distinctly more pronounced during hypoglycemia versus euglycemia (P < .001). These data indicate that insulin inhibits insulin/C-peptide secretion in a dose-dependent manner. Hypoglycemia is a much stronger inhibitor of insulin secretion than insulin itself.     

Insulin sensitivity, glucose effectiveness, and insulin secretion in nondiabetic offspring of patients with non-insulin-dependent diabetes mellitus: a cross-sectional study.

To evaluate the factors that determine the worsening of intravenous glucose tolerance in subjects at high risk for developing non-insulin-dependent diabetes mellitus (NIDDM), 15 glucose-tolerant offspring of NIDDM patients and 21 control subjects were studied. Each subject underwent a frequently sampled intravenous glucose tolerance (FSIGT) test. The intravenous glucose tolerance index (K(G) index) was calculated between minutes 10 and 40 of a FSIGT test. Insulin sensitivity (S(I)), glucose effectiveness at zero insulin (GEZI), and first- and second-phase insulin responsiveness (phi1 and phi2) were estimated using glucose and insulin kinetic minimal models. The acute insulin response to glucose (AIRg) was calculated as the area under the insulin curve above the basal level between 0 and 10 minutes, and the suprabasal insulin effect was determined by the product of S(I) times AIRg. Offspring had a lower S(I) than control subjects (14.1 +/- 7.5 v 9.25 +/- 4.20 x 10(-5) x min(-1)(pmol x L(-1))(-1), P < .01), and their AIRg was similar (3,284 +/- 2,280 v 3,105 +/- 1,499 pmol x L(-1), NS). Sample division according to the median K(G) value showed that control subjects with low tolerance had a lower AIRg (4,417 +/- 2,531 v 2,043 +/- 1,068 pmol x L(-1), P < .05) and a lower suprabasal insulin effect (0.057 +/- 0.03 v 0.023 +/- 0.009 min(-1), P < .05) than control subjects with high tolerance. Offspring with low tolerance had a lower AIRg (2,574 +/- 1,197 v 3,707 +/- 1,656 pmol x L(-1), P < .05) and a lower GEZI (0.101 +/- 0.05 v 0.212 +/- 0.08 x 10(-1) x min(-1), P < .05) than offspring with high tolerance. Offspring with high and low tolerance showed lower phi1 (375 +/- 155 v 272 +/- 181 v 698 +/- 336 (pmol x L(-1))min(mmol x L(-1)), NS) than control subjects with high tolerance. In conclusion, our data suggest that decreases in GEZI and AIRg are the main factors responsible for the worsening of intravenous glucose tolerance in the offspring of NIDDM patients.     

Insulin secretion, insulin sensitivity and diabetes in black children.

Historically, type 2 diabetes has been considered rare in the pediatric population. However, over the last decade, there has been a disturbing upswing in the rate of non-type 1 diabetes in the pediatric age group, particularly adolescents, with a greater proportion of Black children being affected. In this review, the following questions will be addressed: (1) what are the clinical characteristics of youth-onset atypical diabetes, (2) how common is it, (3) what are the risk factors, and (4) how should it be treated?     

Insulin resistance and adiposity influence lipoprotein size and subclass concentrations. Results from the Insulin Resistance Atherosclerosis Study

BACKGROUND: Insulin resistance and obesity are associated with a dyslipidemia composed of high levels of triglycerides (TG), low levels of high-density lipoprotein cholesterol (HDL-C), and no change in level of low-density lipoprotein cholesterol (LDL-C). We examined the association of insulin resistance and adiposity with lipoprotein particle size, concentration, and subclass concentrations. METHODS: The Insulin Resistance Atherosclerosis Study is a multicenter cohort study of middle-aged men and women. Lipoprotein lipid concentrations were determined using standard methods. Lipoprotein size, particle concentration, and subclass concentrations were determined using nuclear magnetic resonance technology. Insulin resistance (SI) was determined based on the frequently sampled intravenous glucose tolerance test and the MINMOD program. A higher SI represents less insulin resistance. Fasting insulin, body mass index, waist circumference, and waist/hip ratio were assessed. RESULTS: Among the 1371 participants were 754 women and 617 men; 459 Hispanics, 383 African Americans, and 529 non-Hispanic whites; 437 with type 2 diabetes, 301 with impaired glucose tolerance, and 633 with normal glucose tolerance. The mean (SD) age was 55.5 (8.5) years, body mass index was 29.3 (5.8) kg/m2 , and SI was 1.6 (1.8) units. Adjusted for age, sex, and ethnicity, SI was not associated with LDL-C (r = 0.01); however, S I was associated with LDL size (r = 0.34, P < .001), LDL particle concentration (r = -0.28, P < .001), small LDL (r = -0.34, P < .001), intermediate LDL (r = -0.37, P < .001), and large LDL (r = 0.21, P < .001). In addition, S I was associated with TG (r = -0.36, P < .001), VLDL particles (r = -0.08, P < .01), large VLDL (r = -0.32, P < .001), VLDL size (r = -0.38, P < .001), HDL-C (r = 0.37, P < .001), HDL particles (r = 0.09, P < .001), large HDL (r = 0.31, P < .001), and HDL size (r = 0.33, P < .001). A factor analysis revealed a factor that accounted for 41.4% of the variance across the lipoprotein measures and that was correlated with SI (r = -0.33, P < .001). Similar results of opposing direction were observed for analyses of lipoprotein measures with fasting insulin and adiposity. CONCLUSIONS: The dyslipidemia associated with insulin resistance and obesity includes effects on lipoprotein metabolism that are missed when traditional lipoprotein cholesterol and total TG are examined. Lipoprotein size and subclasses should be examined in studies investigating the roles of insulin resistance and obesity in the pathogenesis and prevention of atherosclerosis.     

Effect of age on intravenous glucose tolerance and insulin secretion.

Twenty-seven normal volunteers whose ages ranged from 22 to 67 (mean, 37 years) were given intravenous glucose tolerance tests (IVGTT). Age was not correlated with the glucose disposal rate constant (KG), incremental insulin secretion (deltaI), or the "insulinogenic index (deltaI/DELATG). The volunteers were divided into three age groups: 1) mean age 25 years, 2) mean age 42 years, and 3) mean age 62 years. Groups 2 and 3 did not differ from Group 1 with respect to the mean level of fasting plasma glucose, KG, or deltaI/AG. Group 2 (but not Group 3) had a greater mean deltaI than Group 1, and Group 3 (but not Group 2) had a higher mean fasting insulin concentration than Group 1. No significant deterioration was detected in the intravenous glucose tolerance of healthy volunteers up to age 67. This may be attributable to the previously described feed-back system that calibrates the pancreatic beta-cell response according to the insulin sensitivity of the peripheral tissues.     

The insulin secretagogues glibenclamide and repaglinide do not influence growth hormone secretion in humans but stimulate glucagon secretion during profound insulin deficiency

In vitro data have recently suggested that sulfonylureas (SUs) enhance GH secretion by modulating the effects of GHRH and somatostatin in pituitary cells. The present study was undertaken to explore in more detail a possible influence of a single dose of SU (glibenclamide) and a non-SU (repaglinide) insulin secretagogue on circulating GH dynamics. Ten C-peptide-negative type 1 diabetic individuals were examined on three occasions in random order. Either glibenclamide (10.5 mg), repaglinide (8 mg), or placebo was administered after overnight normalization of plasma glucose by iv insulin infusion. Subsequently, GH concentrations were measured regularly after stimulation with GHRH (bolus 0.1 micro g/kg) alone and during concomitant infusion with somatostatin (7 ng.kg(-1).min(-1)). Insulin was replaced at baseline levels (0.25 mU.kg(-1).min(-1)) and plasma glucose clamped at 5-6 mmol/liter. Overall, there were no significant statistical differences in GH responses determined as either GH peak concentrations, integrated levels of GH, or secretory burst mass of GH during the experimental protocol. In contrast, plasma glucagon concentrations were significantly increased during glibenclamide and repaglinide exposure. The present experimental design does not support the hypothesis that acute administration of pharmacological doses of the oral antihyperglycemic agents glibenclamide and repaglinide per se enhance GH release in humans. Additionally, this study shows that these potassium channel inhibitors seem to stimulate glucagon secretion in people who have severe intraislet insulin deficiency (e.g. type 1 diabetes). However, extrapolation of our findings to type 2 diabetic individuals should be done with some caution.     

Acute suppressive effect of human growth hormone on basal insulin secretion in man.

The acute effects of human growth hormone (GH) on the basal levels of glucose and insulin in blood were investigated in 11 healthy men. GH doses of 5, 10, 20, and 40 mug/kg body weight were given iv as a constant-rate infusion over 30 min, and resulted in peak hormone levels (30 min) of 20.5 plus or minus 1.0, 48.5 plus or minus 2.2, 108.2 plus or minus 4.5, and 229.2 plus or minus 14.6 ng/ml, respectively. There was a small (max 9.8 plus or minus 2.6%) but significant decrease in the blood glucose level, observed already at 15 min after the beginning of the GH infusion and persisting up to 90 min. The highest dose of GH induced the most marked changes, but there was otherwise no clear correlation between dose and effect. The basal plasma insulin levels showed a more marked (max 16.0 plus or minus 4.7%) decrease which was not correlated, in time or in magnitude, with the changes in blood glucose. In some subjects, in whom no significant decrease in blood glucose was observed, plasma insulin still demonstrated a similar fall (max 20.2 plus or minus 7.6%). Neither were these changes in plasma insulin correlated to the dose of GH within the range used in this study. The findings suggested that the early, insulin-like effect of GH on blood glucose is distinct from its effect on the pancreas. The latter is a suppressive one, consistent with earlier findings on glucose-induced insulin release.