Contribution of defects in glucose uptake to carbohydrate intolerance in liver cirrhosis: assessment during physiological glucose and insulin concentrations

It is well established that subjects with liver cirrhosis are insulin resistant, but the contribution of defects in insulin secretion and/or action to glucose intolerance remains unresolved. Healthy individuals and subjects with liver cirrhosis were studied on two occasions: 1) an oral glucose tolerance test was performed, and 2) insulin secretion was inhibited and glucose was infused in a pattern and amount mimicking the systemic delivery rate of glucose after a carbohydrate meal. Insulin was concurrently infused to mimic a healthy postprandial insulin profile. Postabsorptive glucose concentrations were equal (5.36 +/- 0.12 vs. 5.40 +/- 0.25 mmol/l, P = 0.89), despite higher insulin (P < 0.01), C-peptide (P < 0.01), and free fatty acid (P = 0.05) concentrations in cirrhotic than in control subjects. Endogenous glucose release (EGR; 11.50 +/- 0.50 vs. 11.73 +/- 1.00 mumol.kg(-1).min(-1), P = 0.84) and the contribution of gluconeogenesis to EGR (6.60 +/- 0.47 vs. 6.28 +/- 0.64 mumol.kg(-1).min(-1), P = 0.70) were unaltered by cirrhosis. A minimal model recently developed for the oral glucose tolerance test demonstrated an impaired insulin sensitivity index (P < 0.05), whereas the beta-cell response to glucose was unaltered (P = 0.72). During prandial glucose and insulin infusions, the integrated glycemic response was greater in cirrhotic than in control subjects (P < 0.05). EGR decreased promptly and comparably in both groups, but glucose disappearance was insufficient at the prevailing glucose concentration (P < 0.05). Moreover, identical rates of [3-(3)H]glucose infusion produced higher tracer concentrations in cirrhotic than in control subjects (P < 0.05), implying a defect in glucose uptake. In conclusion, carbohydrate intolerance in liver cirrhosis is determined by insulin resistance and the ability of glucose to stimulate insulin secretion. During prandial glucose and insulin concentrations, EGR suppression was unaltered, but glucose uptake was impaired, which demonstrates that intolerance can be ascribed to a defect in glucose uptake, rather than abnormalities in glucose production or beta-cell function. Although insulin secretion ameliorates glucose intolerance, impaired glucose uptake during physiological glucose and insulin concentrations produces marked and sustained hyperglycemia, despite concurrent abnormalities in glucose production or insulin secretion.     

Glucose turnover in compensated hepatic cirrhosis

Glucose turnover and recycling from glucose derived 3-carbon intermediates were examined in overnight fasted patients with compensated hepatic cirrhosis and in age- and weight-matched normal control subjects. Fasting blood concentrations of glucose, lactate and glycerol were similar in both groups but blood pyruvate (60 +/- 10 vs. 80 +/- mumol/l, P less than 0.05), blood alanine (0.23 +/- 0.02 vs 0.34 +/- 0.02 mmol/l, P less than 0.01) were decreased and serum insulin increased (19 [13-24]v 7 [4-11] mU/l, P less than 0.01) in cirrhotic subjects. Absolute glucose turnover, assessed by analysis of decay of [3H]-3-glucose specific activity was decreased in cirrhotic patients (8.1 +/- 0.6 v 12.1 +/- 0.7 mol/kg-1 min-1). Glucose "recycling", assessed by the difference between absolute glucose turnover and that given by [14C]-1-glucose data, was normal in cirrhotic patients suggesting that Cori cycle (glucose-lactate-glucose) activity was normal. These data support previous findings of decreased peripheral glucose utilisation and insulin resistance in cirrhotic patients.     

The influence of insulin on circulating ghrelin

Ghrelin is a novel peptide that acts on the growth hormone (GH) secretagogue receptor in the pituitary and hypothalamus. It may function as a third physiological regulator of GH secretion, along with GH-releasing hormone and somatostatin. In addition to the action of ghrelin on the GH axis, it appears to have a role in the determination of energy homeostasis. Although feeding suppresses ghrelin production and fasting stimulates ghrelin release, the underlying mechanisms controlling this process remain unclear. The purpose of this study was to test the hypotheses, by use of a stepped hyperinsulinemic eu- hypo- hyperglycemic glucose clamp, that either hyperinsulinemia or hypoglycemia may influence ghrelin production. Having been stable in the period before the clamp, ghrelin levels rapidly fell in response to insulin infusion during euglycemia (baseline ghrelin 207 +/- 12 vs. 169 +/- 10 fmol/ml at t = 30 min, P < 0.001). Ghrelin remained suppressed during subsequent periods of hypoglycemia (mean glucose 53 +/- 2 mg/dl) and hyperglycemia (mean glucose 163 +/- 6 mg/dl). Despite suppression of ghrelin, GH showed a significant rise during hypoglycemia (baseline 4.1 +/- 1.3 vs. 28.2 +/- 3.9 microg/l at t = 120 min, P < 0.001). Our data suggest that insulin may suppress circulating ghrelin independently of glucose, although glucose may have an additional effect. We conclude that the GH response seen during hypoglycemia is not regulated by circulating ghrelin.     

Alterations in liver, muscle, and adipose tissue insulin sensitivity in men with HIV infection and dyslipidemia

Dyslipidemia is common in patients with HIV infection. In this study, a two-stage euglycemic hyperinsulinemic clamp, with infusion of stable isotopically labeled tracers, was used to evaluate insulin action in skeletal muscle, liver, and adipose tissue in HIV-infected men with dyslipidemia (HIV-DL; plasma triglyceride >250 mg/dl and HDL <45 mg/dl; n=12), HIV-infected men without dyslipidemia (HIV w/o DL; n=12), and healthy men (n=6). Basal rates of glucose production (glucose R(a)), glucose disposal (glucose R(d)), and lipolysis (palmitate R(a)) were similar between groups. The relative suppression of glucose R(a) (63+/- 4, 77+/- 2, and 78+/- 3%, P=0.008) and palmitate R(a) (49+/-4, 63+/-3, and 68+/-3%, P=0.005) during ow-dose insulin infusion (plasma insulin approximately 30 microU/ml), and the relative stimulation of glucose R(d) (214+/-21, 390+/-25, and 393+/-46%, P=0.001) during high-dose insulin infusion (plasma insulin approximately 75 microU/ml) were lower in HIV-DL than in HIV w/o DL and healthy volunteers, respectively. Suppression of basal glucose R(a) correlated with plasma adiponectin (r=0.44, P=0.02) and inversely with plasma IL-6 (r=-0.49, P<0.001). Stimulation of glucose R(d) correlated directly with adiponectin (r=0.48, P<0.01) and inversely with IL-6 (r=-0.49, P=0.02). We conclude that dyslipidemia in HIV-infected men is indicative of multiorgan insulin resistance, and circulating adipokines may be important in the pathogenesis of impaired insulin action.     

The effect of growth hormone administration on lipids and lipoproteins in growth hormone-deficient patients

Plasma lipids, lipoproteins, and lipoprotein cholesterol levels were studied in a group (n = 8) of prepubertal growth hormone-deficient patients before and after growth hormone (GH) administration. Determination of plasma lipoproteins by a sensitive agarose gel electrophoretic technique demonstrated: (a) in the patients with two prebeta bands an intensification of the fast prebeta lipoprotein fraction after growth hormone administration; and (b) in the patients with one prebeta band the appearance of a second prebeta band after growth hormone administration. The mean (+/- SD) plasma triglyceride level before GH was 86 +/- 60 mg/dl and 158 +/- 95 mg/dl after GH (P less than 0.01). Mean (+/- SD) plasma cholesterol level before GH was 196 +/- 25 mg/dl and 174 +/- 28 mg/dl after GH (P less than 0.05). High-density lipoprotein cholesterol concentrations decreased significantly (P less than 0.001) from mean (+/- SD) 55 +/- 12 mg/dl before GH to 37 +/- 10 mg/dl after GH. Very-low-density lipoprotein cholesterol concentrations increased significantly (P less than 0.05) from mean (+/- SD) 13 +/- 12 mg/dl before GH to 23 +/- 15 mg/dl after GH. Low-density lipoprotein cholesterol concentrations decreased (N.S.) from mean (+/- SD) 123 +/- 15 mg/dl before GH to 114 +/- 15 mg/dl after GH. These lipid and lipoprotein changes could be mediated through the insulin antagonism, hyperinsulinemia, and a decrease in lipoprotein lipase activity caused by growth hormone.     

Portal glucose infusion increases hepatic glycogen deposition in conscious unrestrained rats.

It has been demonstrated in the conscious dog that portal glucose infusion creates a signal that increases net hepatic glucose uptake and hepatic glycogen deposition. Experiments leading to an understanding of the mechanism by which this change occurs will be facilitated if this finding can be reproduced in the rat. Rats weighing 275-300 g were implanted with four indwelling catheters (one in the portal vein, one in the left carotid artery, and two in the right jugular vein) that were externalized between the scapulae. The rats were studied in a conscious, unrestrained condition 7 days after surgery, following a 24-h fast. Each experiment consisted of a 30- to 60-min equilibration, a 30-min baseline, and a 120-min test period. In the test period, a pancreatic clamp was performed by using somatostatin, insulin, and glucagon. Glucose was given simultaneously either through the jugular vein to clamp the arterial blood level at 220 mg/dl (Pe low group) or at 250 mg/dl (Pe high group), or via the hepatic portal vein (Po group; 6 mg. kg(-1). min(-1)) and the jugular vein to clamp the arterial blood glucose level to 220 mg/dl. In the test period, the arterial plasma glucagon and insulin levels were not significantly different in the three groups (36 +/- 2, 33 +/- 2, and 30 +/- 2 pg/ml and 1.34 +/- 0.08, 1. 37 +/- 0.18, and 1.66 +/- 0.11 ng/ml in Po, Pe low, and Pe high groups, respectively). The arterial blood glucose levels during the test period were 224 +/- 4 mg/dl for Po, 220 +/- 3 for Pe low, and 255 +/- 2 for Pe high group. The liver glycogen content (micromol glucose/g liver) in the two Pe groups was not statistically different (51 +/- 7 and 65 +/- 8, respectively), whereas the glycogen level in the Po group was significantly greater (93 +/- 9, P < 0.05). Because portal glucose delivery also augments hepatic glycogen deposition in the rat, as it does in the dogs, mechanistic studies relating to its function can now be undertaken in this species.     

The serum concentration of tumor necrosis factor alpha is not an index of growth-hormone- or obesity-induced insulin resistance

BACKGROUND: The tumor necrosis factor alpha (TNF-alpha) might play a central role in insulin resistance, a frequent correlate of obesity likely contributing to some obesity-associated complications. Adult growth hormone (GH) deficiency syndrome (GHDA) shares with obesity excessive fat mass, hyperlipidemia, increased cardiovascular risk, and insulin resistance. On the other hand, GH has been shown to induce transient deterioration of glucose metabolism and insulin resistance when administered in normal humans and in GHDA patients. No information is presently available on the relationship between serum TNF-alpha levels and insulin sensitivity in GHDA. METHODS: We compared the serum TNF-alpha levels found in 10 GHDA patients before and after a 6-month recombinant human GH therapy (Genotropin), in an insulin resistance prone population of 16 obese (OB) patients and in 38 normal-weight healthy blood donors (controls). The insulin sensitivity was assessed by a euglycemic-hyperinsulinemic glucose clamp in all the GHDA patients and in 10 OB and in 6 control subjects. RESULTS: The serum TNF-alpha levels were not significantly different in OB patients (42.2 +/- 12.81 pg/ml), in GHDA patients at baseline (71.3 +/- 23.97 pg/ml), and in controls (55.3 +/- 14.28 pg/ml). A slight decrease of TNF-alpha values was noted in GHDA patients after 6 months of recombinant human GH treatment (44.5 +/- 20.19 pg/ml; NS vs. baseline). The insulin sensitivity (M) was significantly reduced in OB patients (2.4 +/- 0.30 mg/kg/min) as compared with control subjects (7.5 +/- 0.39 mg/kg/min) and in GHDA patients both at baseline (6.6 +/- 0.6 mg/kg/min) and after recombinant human GH therapy (5.6 +/- 0.7 mg/kg/min). The insulin sensitivity in the GHDA patients, similar to that of controls at baseline, worsened after recombinant human GH treatment (p < 0.05 vs. baseline; p = 0.05 vs. controls). Linear regression analysis showed no correlation between TNF-alpha and M values (see text) in all patient groups. CONCLUSIONS: These data indicate that circulating concentrations of TNF-alpha do not reflect the degree of insulin resistance in obesity and GHDA. They, however, do not exclude that TNF-alpha may induce insulin resistance at tissue level.     

Metabolic regulation of growth hormone by free fatty acids, somatostatin, and ghrelin in HIV-lipodystrophy

Human immunodeficiency virus (HIV)-lipodystrophy is a syndrome characterized by changes in fat distribution and insulin resistance. Prior studies suggest markedly reduced growth hormone (GH) levels in association with excess visceral adiposity among patients with HIV-lipodystrophy. We investigated mechanisms of altered GH secretion in a population of 13 male HIV-infected patients with evidence of fat redistribution, compared with 10 HIV-nonlipodystrophic patients and 11 male healthy controls similar in age and body mass index (BMI). Although similar in BMI, the lipodystrophic group was characterized by increased visceral adiposity, free fatty acids (FFA), and insulin and reduced extremity fat. We investigated ghrelin and the effects of acute lowering of FFA by acipimox on GH responses to growth hormone-releasing hormone (GHRH). We also investigated somatostatin tone, comparing GH response to combined GHRH and arginine vs. GHRH alone with a subtraction algorithm. Our data demonstrate an equivalent number of GH pulses (4.1 +/- 0.6, 4.7 +/- 0.8, and 4.5 +/- 0.3 pulses/12 h in the HIV-lipodystrophic, HIV-nonlipodystrophic, and healthy control groups, respectively, P > 0.05) but markedly reduced GH secretion pulse area (1.14 +/- 0.27 vs. 4.67 +/- 1.24 ng.ml(-1).min, P < 0.05, HIV-lipodystrophic vs. HIV-nonlipodystrophic; 1.14 +/- 0.27 vs. 3.18 +/- 0.92 ng.ml(-1).min, P < 0.05 HIV-lipodystrophic vs. control), GH pulse area, and GH pulse width in the HIV-lipodystrophy patients compared with the control groups. Reduced ghrelin (418 +/- 46 vs. 514 +/- 37 pg/ml, P < 0.05, HIV-lipodystrophic vs. HIV-nonlipodystrophic; 418 +/- 46 vs. 546 +/- 45 pg/ml, P < 0.05, HIV-lipodystrophic vs. control), impaired GH response to GHRH by excess FFA, and increased somatostatin tone contribute to reduced GH secretion in patients with HIV-lipodystrophy. These data provide novel insight into the metabolic regulation of GH secretion in subjects with HIV-lipodystrophy.     

Basal insulin, glucagon, and growth hormone replacement

Conclusions drawn from the pancreatic (or islet) clamp technique (suppression of endogenous insulin, glucagon, and growth hormone secretion with somatostatin and replacement of basal hormone levels by intravenous infusion) are critically dependent on the biological appropriateness of the selected doses of the replaced hormones. To assess the appropriateness of representative doses we infused saline alone, insulin (initially 0.20 mU.kg(-1).min(-1)) alone, glucagon (1.0 ng.kg(-1).min(-1)) alone, and growth hormone (3.0 ng.kg(-1).min(-1)) alone intravenously for 4 h in 13 healthy individuals. That dose of insulin raised plasma insulin concentrations approximately threefold, suppressed glucose production, and drove plasma glucose concentrations down to subphysiological levels (65 +/- 3 mg/dl, P < 0.0001 vs. saline), resulting in nearly complete suppression of insulin secretion (P < 0.0001) and stimulation of glucagon (P = 0.0059) and epinephrine (P = 0.0009) secretion. An insulin dose of 0.15 mU.kg(-1).min(-1) caused similar effects, but a dose of 0.10 mU.kg(-1).min(-1) did not. The glucagon and growth hormone infusions did not alter plasma glucose levels or those of glucoregulatory factors. Thus, insulin "replacement" doses of 0.20 and even 0.15 mU.kg(-1).min(-1) are excessive, and conclusions drawn from the pancreatic clamp technique using such doses may need to be reassessed.     

Hyperglycemia compensates for diet-induced insulin resistance in liver and skeletal muscle of rats

High-fat and high-sucrose diets increase the contribution of gluconeogenesis to glucose appearance (glc R(a)) under basal conditions. They also reduce insulin suppression of glc R(a) and insulin-stimulated muscle glycogen synthesis under euglycemic, hyperinsulinemic conditions. The purpose of the present study was to determine whether these impairments influence liver and muscle glycogen synthesis under hyperglycemic, hyperinsulinemic conditions. Male rats were fed a high-sucrose, high-fat, or low-fat, starch control diet for either 1 (n = 5-7/group) or 5 wk (n = 5-6/group). Studies involved two 90-min periods. During the first, a basal period (BP), [6-3H]glucose was infused. In the second, a hyperglycemic period (HP), [6-3H]glucose, [6-14C]glucose, and unlabeled glucose were infused. Plasma glucose (BP: 111.2 +/- 1.5 mg/dl; HP: 172.3 +/- 1.5 mg/dl), insulin (BP: 2.5 +/- 0.2 ng/ml; HP: 4.9 +/- 0.3 ng/ml), and glucagon (BP: 81.8 +/- 1.6 ng/l; HP: 74.0 +/- 1.3 ng/l) concentrations were not significantly different among diet groups or with respect to time on diet. There were no significant differences among groups in the glucose infusion rate (mg x kg(-1) x min(-1)) necessary to maintain arterial glucose concentrations at approximately 170 mg/dl (pooled average: 6.4 +/- 0.8 at 1 wk; 6.4 +/- 0.7 at 5 wk), percent suppression of glc R(a) (44.4 +/- 7.8% at 1 wk; 63.2 +/- 4.3% at 5 wk), tracer-estimated net liver glycogen synthesis (7.8 +/- 1.3 microg x g liver(-1) x min(-1) at 1 wk; 10.5 +/- 2.2 microg x g liver(-1) x min(-1) at 5 wk), indirect pathway glycogen synthesis (3.7 +/- 0.9 microg x g liver(-1) x min(-1) at 1 wk; 3.4 +/- 0.9 microg x g liver(-1) x min(-1) at 5 wk), or tracer-estimated net muscle glycogenesis (1.0 +/- 0.3 microg x g muscle(-1) x min(-1) at 1 wk; 1.6 +/- 0.3 microg x g muscle(-1) x min(-1) at 5 wk). These data suggest that hyperglycemia compensates for diet-induced insulin resistance in both liver and skeletal muscle.     

Growth hormone secretion, serum, and cerebral spinal fluid insulin and insulin-like growth factor-I concentrations in pigs with streptozotocin-induced diabetes mellitus.

Diabetes mellitus was induced using streptozotocin in five gilts between 8 and 12 weeks of age. Gilts were maintained with exogenous insulin (INS) except during experimental periods. Four litter-mate gilts served as controls. At 9 months of age, all gilts were ovariectomized, and 30 days after ovariectomy, Experiment (Exp) 1 was conducted. Jugular vein catheters were inserted and blood samples were collected every 10 min for 8 hr. Experiment 2 was conducted when gilts were 11 months of age. Venous blood and cerebrospinal fluid (CSF) samples were collected in the absence (Phase I) or presence (Phase II) of INS therapy. In Experiment 1, plasma glucose concentrations were greater (P < 0.05) in diabetic (465 +/- 17 mg/100 ml) than in control (82 mg +/- 17 mg/100 ml) gilts, whereas serum INS was lower (P < 0.0001) in diabetic gilts (0.3 +/- 0.02 vs 0.9 +/- 0.05 ng/ml) and insulin-like growth factor-I was similar in diabetic and control gilts (32 +/- 3 vs 43 +/- 4 ng/ml, respectively). Mean serum GH concentration was 2-fold greater (P < 0.02) in diabetics (2.8 +/- 0.4 ng/ml) than in control gilts (1.2 +/- 0.2 ng/ml). Diabetic gilts exhibited a greater (P < 0.05) number of GH pulses than control gilts (3.2 +/- 0.4 vs 1.5 +/- 0.3/8 hr, respectively). In addition, GH pulse magnitude was markedly elevated (P < 0.02) in diabetic (5.8 +/- 0.4 ng/ml) compared with control gilts (3.3 +/- 0.6 ng/ml). Mean basal serum GH concentrations were greater (P < 0.07) in diabetic (2.2 +/- 0.5 ng/ml) compared with control gilts (1.0 +/- .1 ng/ml). In Experiment 2, CSF concentrations of insulin-like growth factor-I, INS, GH, and protein were similar for diabetic and control gilts in both phases. Serum GH levels were similar for diabetics and controls in Phase I, but were greater (P < 0.05) in diabetics than in controls in Phase II. CSF glucose levels were greater in diabetic than in control gilts in both the presence (P < 0.003) and absence (P < 0.0002) of INS therapy, whereas plasma glucose was greater (P < 0.003) in diabetic than in control gilts in the absence of INS, but returned to control concentrations in the presence of INS. However, serum GH levels were unchanged after INS therapy in the diabetic gilts. In conclusion, altered GH secretion in the diabetic gilt may, in part, be due to elevated CSF glucose concentrations, which may alter GH-releasing hormone and/or somatostatin secretion from the hypothalamus.     

Enteral infusion of glucose at rates approximating EGP enhances glucose disposal but does not cause hypoglycemia

Portal infusion of glucose at rates approximating endogenous glucose production (EGP) causes paradoxical hypoglycemia in wild-type but not GLUT2 null mice, implying activation of a specific portal glucose sensor. To determine whether this occurs in humans, glucose containing [3-3H]glucose was infused intraduodenally at rates of 3.1 mg. kg-1. min-1 (n = 5), 1.55 mg. kg-1. min-1 (n = 9), or 0/0.1 mg. kg-1. min-1 (n = 9) for 7 h in healthy nondiabetic subjects. [6,6-2H2]glucose was infused intravenously to enable simultaneous measurement of EGP, glucose disappearance, and the rate of appearance of the intraduodenally infused glucose. Plasma glucose concentrations fell (P < 0.01) from 90 +/- 1 to 84 +/- 2 mg/dl during the 0/0.1 mg. kg-1. min-1 id infusions but increased (P < 0.001) to 104 +/- 5 and 107 +/- 3 mg/dl, respectively, during the 1.55 and 3.1 mg. kg-1. min-1 id infusions. In contrast, insulin increased (P < 0.05) during the 1.55 and 3.0 mg. kg-1. min-1 infusions, reaching a peak of 10 +/- 2 and 18 +/- 5 micro U/ml, respectively, by 2 h. Insulin concentrations then fell back to concentrations that no longer differed by study end (7 +/- 1 vs. 8 +/- 1 micro U/ml). This resulted in comparable suppression of EGP by study end (0.84 +/- 0.2 and 0.63 +/- 0.1 mg. kg-1. min-1). Glucose disappearance was higher (P < 0.01) during the final hour of the 3.1 than 1.55 mg. kg-1. min-1 id infusion (4.47 +/- 0.2 vs. 2.6 +/- 0.1 mg. kg-1. min-1), likely because of the slightly, but not significantly, higher glucose and insulin concentrations. We conclude that, in contrast to mice, selective portal glucose delivery at rates approximating EGP does not cause hypoglycemia in humans.     

Homocysteine levels in acromegaly patients

Acromegaly is associated with a two to three-fold increase in mortality related predominantly to cardiovascular disease. The excess mortality is associated most closely with higher levels of growth hormone (GH). Survival in acromegaly may be normalized to a control age-matched rate by controlling GH levels; in particular, GH levels less than 2.5 ng/mL are associated with survival rates equal to those of the general population. Hyperhomocysteinemia has also been recognized as a risk factor for cardiovascular disease, yet there are limited data on the prevalence of hyperhomocysteinemia in patients with acromegaly. Eighteen acromegaly patients (7 male, 11 female, mean age 42.8 +/- 11.0 years) in our endocrine clinic consented to having the following tests performed: complete blood count (CBC), thyroid hormones, folic acid, vitamin B12, plasma homocysteine levels, uric acid, fibrinogen, CRP, fasting glucose, insulin, C-peptide, total serum cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, GH, insulin-like growth factor-1 (IGF-1) and GH levels after an oral glucose tolerance test (OGTT). By history, fourteen had macroadenomas and four had microadenomas; eight had hypertension; two had glucose intolerance, and four had diabetes. Fifteen had had transsphenoidal or transfrontal surgery: two had been cured, but 13 others were taking long-acting octreotide. Five patients had undergone radiotherapy and the acromegaly in two was treated primarily with long-acting octreotide. CBC, thyroid hormone, folic acid, and vit B12 levels were normal in all patients. We divided the patients into two groups according to mean GH levels after an OGTT: Group 1 (GH<2.5 ng/mL, n=10), and Group 2 (GH<2.5 ng/mL, n=8). Comparison of the two groups using Mann-Whitney U testing revealed statistically significant lower levels in Group 1 of the following parameters: GH (1.91 +/- 0.90 vs. 8.58 +/- 5.55 ng/mL, p=0.002), IGF-1 (338.30 +/- 217.90 vs. 509.60 +/- 293.58 ng/dL, p=0.06), GH after an OGTT (1.42 +/- 0.81 vs. 9.01 +/- 4.53 ng/mL, p=0.001), plasma homocysteine (12.85 +/- 4.47 vs. 18.20 +/- 4.99 micromol/L, p=0.05), total cholesterol (164.0 +/- 20.81 vs. 188.0 +/- 22.26 mg/dL, p=0.05) and LDL cholesterol (81.0 +/- 9.64 vs. 116.70 +/- 13.03 mg/dl, p=0.01). Differences between the other parameters were not significantly different. Acromegaly patients with high GH levels after an OGTT have much higher levels of homocysteine than patients with lower GH levels. The role of elevated homocysteine levels as an independent cardiovascular risk factor in the mortality of acromegaly patients should be determined in future studies.     

Effect of a single administration of somatostatin analogue (SMS 201-995) on GH, TSH and insulin secretion in patients with acromegaly

The effect of a long-acting somatostatin analogue SMS 201-995 on GH secretion was investigated. Eleven acromegalic patients received a single dose of 50 micrograms SMS 201-995 administered subcutaneously, and plasma GH, IGF-I, GRF, TSH, IRI and blood glucose were determined at regular intervals. Nine of 11 patients had elevated basal plasma GH levels above 5 ng/ml. In all patients, plasma GH levels fell immediately from 39.5 +/- 17.3 ng/ml (mean +/- SEM) to 4.3 +/- 1.6 ng/ml (P less than 0.05) with a maximal inhibition of 82.9 +/- 3.3% of the basal levels and the suppression persisted for about 6 h of the observation period. IGF-I and GRF levels were not apparently altered. TSH and IRI levels also rapidly fell. Blood glucose levels fell slightly by 0.5 h. Ten of 11 patients had pain at injection sites. Except for this, no side effects were observed. Our results show that the new somatostatin analogue SMS 201-995 may inhibit GH hypersecretion in acromegalic patients for significant periods, suggesting that this agent can be a useful clinical tool for the treatment of acromegaly.     

Effect of pentobarbital and urethane on the release of hypothalamic somatostatin and pituitary growth hormone

Hypothalamic somatostatin release was investigated in the rat to elucidate the mechanism of anesthetic action on growth hormone (GH) release from the pituitary. Intraperitoneal injection of sodium pentobarbital (5 mg/100 gm B.W.) significantly elevated serum GH levels and increased hypothalamic somatostatin concentration from basal values of 0.98 +/- 0.01 to 1.21 +/- 0.06 ng/mg wet wt. In contrast, urethane (150 mg/100 gm B.W., IP) administration lowered serum GH levels and hypothalamic somatostatin concentration (0.64 +/- 0.04 ng/mg wet wt.). However, the mean concentration of pancreatic somatostatin showed no change in either case. In rats receiving passive immunization with 0.5 ml rabbit antiserum to somatostatin (SRIF-AS), serum GH levels were significantly increased (67.5 +/- 12.3 ng/ml) and did not differ from those in the group treated with normal rabbit serum (NRS) plus pentobarbital (101.3 +/- 18.5 ng/ml). However, serum GH levels in rats injected with SRIF-AS plus pentobarbital were increased to higher values than in rats given SRIF-AS alone. When urethane was administered to rats after passive immunization with SRIF-AS, urethane-induced suppression of serum GH levels was markedly inhibited (5.5 +/- 2.0 vs. 33.5 +/- 7.5 ng/ml). These results suggest a possibility that the changes in serum GH levels observed with pentobarbital or urethane administration may be induced at least in one part by somatostatin released from the hypothalamus.     

Pancreatic B-cell function in non-insulin-dependent diabetes mellitus during successive periods of sulfonylurea and insulin treatment: serum C-peptide response to glucagon and urine C-peptide excretion

Serum C-peptide responses to glucagon and daily urine C-peptide excretion in successive periods of different treatment in two groups of patients with non-insulin-dependent diabetes mellitus (NIDDM) (mean interval between two tests less than 1 month) were compared. In group A patients (n = 8), the glycemic control was improved after transferring the treatment from sulfonylurea (SU) to insulin (fasting plasma glucose: SU: 192 +/- 47, insulin: 127 +/- 21 mg/dl, mean +/- S.D., p less than 0.01). Fasting serum C-peptide immunoreactivity (CPR) was significantly lower at the period of insulin treatment (SU: 1.93 +/- 1.01, insulin: 1.47 +/- 0.79 ng/ml, p less than 0.05), but there was no difference in the increase in serum CPR (maximal--fasting) (delta serum CPR) during glucagon stimulation in the two periods of treatment (SU: 1.70 +/- 0.72, insulin: 1.47 +/- 0.98 ng/ml). In group B patients (n = 7), there was no significant difference in glycemic control after transferring the treatment from insulin to SU (fasting plasma glucose: insulin: 127 +/- 24, SU: 103 +/- 13 mg/dl). Fasting serum CPR was significantly lower during the period of insulin treatment (insulin: 1.39 +/- 0.64, SU: 2.21 +/- 0.86 ng/ml, p less than 0.025), but delta serum CPR during glucagon stimulation still showed no significant difference between the two periods (insulin: 1.97 +/- 1.16, SU: 2.33 +/- 1.57 ng/ml).(ABSTRACT TRUNCATED AT 250 WORDS)     

Glucagon chronically impairs hepatic and muscle glucose disposal

Defects in insulin secretion and/or action contribute to the hyperglycemia of stressed and diabetic patients, and we hypothesize that failure to suppress glucagon also plays a role. We examined the chronic impact of glucagon on glucose uptake in chronically catheterized conscious depancreatized dogs placed on 5 days of nutritional support (NS). For 3 days of NS, a variable intraportal infusion of insulin was given to maintain isoglycemia (approximately 120 mg/dl). On day 3 of NS, animals received a constant low infusion of insulin (0.4 mU.kg-1.min-1) and either no glucagon (CONT), basal glucagon (0.7 ng.kg-1.min-1; BasG), or elevated glucagon (2.4 ng.kg-1.min-1; HiG) for the remaining 2 days. Glucose in NS was varied to maintain isoglycemia. An additional group (HiG+I) received elevated insulin (1 mU.kg-1.min-1) to maintain glucose requirements in the presence of elevated glucagon. On day 5 of NS, hepatic substrate balance was assessed. Insulin and glucagon levels were 10+/-2, 9+/-1, 7+/-1, and 24+/-4 microU/ml, and 24+/-5, 39+/-3, 80+/-11, and 79+/-5 pg/ml, CONT, BasG, HiG, and HiG+I, respectively. Glucagon infusion decreased the glucose requirements (9.3+/-0.1, 4.6+/-1.2, 0.9+/-0.4, and 11.3+/-1.0 mg.kg-1.min-1). Glucose uptake by both hepatic (5.1+/-0.4, 1.7+/-0.9, -1.0+/-0.4, and 1.2+/-0.4 mg.kg-1.min-1) and nonhepatic (4.2+/-0.3, 2.9+/-0.7, 1.9+/-0.3, and 10.2+/-1.0 mg.kg-1.min-1) tissues decreased. Additional insulin augmented nonhepatic glucose uptake and only partially improved hepatic glucose uptake. Thus, glucagon impaired glucose uptake by hepatic and nonhepatic tissues. Compensatory hyperinsulinemia restored nonhepatic glucose uptake and partially corrected hepatic metabolism. Thus, persistent inappropriate secretion of glucagon likely contributes to the insulin resistance and glucose intolerance observed in obese and diabetic individuals.     

Effect of twice daily subcutaneous administration of a long-acting somatostatin analog on 24-hour plasma glucose profiles in patients with insulin-dependent diabetes mellitus

To determine the effect of twice daily subcutaneous administration of a long-acting somatostatin analog on diabetic glycemic control, seven insulin-dependent diabetic subjects were treated with mixtures of insulin (regular and lente) given 30 minutes before breakfast and supper alone or along with WY-41, 747, a long-acting somatostatin analog. Postprandial hyperglycemia was markedly reduced after breakfast and supper (peak values 123 +/- 15 vs. 207 +/- 1 mg/dl, P less than 0.01 and 150 +/- 13 vs. 235 +/- 29 mg/dl, P less than 0.02, WY-41,747 + insulin vs. insulin alone, respectively). Although values after lunch and the evening snack were not significantly different, overall mean 24 hr plasma glucose concentrations were significantly less with WY-41,747 plus insulin than with insulin alone (136 +/- 9 vs. 176 +/- 13, P less than 0.05). No serious adverse effects were noted. We conclude that administration of a long-acting somatostatin analog such as WY-41,747 twice daily along with insulin may permit some diabetic patients to achieve satisfactory glycemic control without having to inject insulin 3-4 times daily prior to each meal.     

Insulin resistance in myotonic dystrophy.

The aim of the present study was to obtain a comprehensive picture of the rate of insulin secretion and of tissue sensitivity to the endogenous hormone in myotonic dystrophy patients (MyD). The minimal model approach was utilized for the analysis of frequently sampled intravenous glucose tolerance test data (FSIGT). This method provided the characteristic parameters: SI, insulin sensitivity index; SG fractional glucose disappearance independent of dynamic insulin; n, fractional insulin clearance; phi 1 and phi 2 first and second phase insulin delivery sensitivities to glucose stimulation. In MyD patients SI was reduced (p less than 0.01) by 71% to 1.4 +/- 0.3 x 10(-4) min-1/(microU/ml), whereas in controls it was 4.85 +/- 0.77; SG was within the normal range: 0.044 +/- 0.012 min-1 in MyD patients and 0.036 +/- 0.017 min-1 in controls; phi 1 increased in MyD patients (7.4 +/- 1.3 min (microU/ml)/(mg/dl) versus 4.1 +/- 1.2 in controls); phi 2 increased in MyD patients (126 +/- 47 x 10(4) min-2/(microU/ml)/(mg/dl) versus 17 +/- 6 in controls; p less than 0.05). MyD patients showed a normal tolerance with the glucose disappearance constant, KG within the normal range: 2.75 versus 2.62% min-1 in controls. In MyD patients insulin resistance was associated with a higher than normal insulin delivery for both secretory phases, although the second phase was responsible for releasing a greater amount of hormone. In conclusion MyD patients try to compensate for overall insulin resistance by a more marked pancreatic response.     

Development of insulin-sensitivity at weaning in the rat. Role of the nutritional transition.

This study was undertaken to determine the factors involved in the development of insulin-sensitivity at weaning. Glucose kinetics were studied in suckling rats and in rats weaned on to a high-carbohydrate (HC) or a high-fat (HF) diet, in the basal state and during euglycaemic-hyperinsulinaemic-clamp studies. These studies were coupled with the 2-deoxyglucose technique, allowing a measure of glucose utilization by individual tissues. In the basal state, the glycaemia was higher in HF-weaned rats (124 +/- 4 mg/dl) than in suckling (109 +/- 1 mg/dl) and HC-weaned rats (101 +/- 3 mg/dl). Glucose turnover rates were similar in the three groups of animals (14 mg/min per kg). Nevertheless, basal metabolic glucose clearance rate was 20% lower in HF-weaned rats than in the other groups. During the euglycaemic-hyperinsulinaemic experiments, hepatic glucose production was suppressed by 90% in HC-weaned rats, whereas it remained at 40% of basal value in suckling and HF-weaned rats, indicating an insulin resistance of liver of these animals. Glucose clearance rate during the clamp was 18.3 +/- 0.9 ml/min per kg in suckling rats, whereas it was 35.3 +/- 1.2 ml/min per kg in HC-weaned rats and 27.8 +/- 1.1 ml/min per kg in HF-weaned rats, indicating an insulin resistance of glucose utilization in suckling, and to a lower extent, in HF-weaned rats. The deoxyglucose technique showed that peripheral insulin resistance was localized in muscles and white adipose tissue of suckling and HF-weaned rats. These results indicate that the switch from milk to a HC diet is an important determinant of the development of insulin-sensitivity at weaning in the rat.