Different aetiologies of Type 2 (non-insulin-dependent) diabetes mellitus in obese and non-obese subjects

Summary Insulin responses to intravenous glucose infusion and glucose utilization during hyperinsulinaemic euglycaemic clamp were determined in a large homogeneous group of 65-year-old male subjects. Twenty-eight had untreated Type 2 (non-insulin-dependent) diabetes mellitus and the remaining 44 control subjects had a normal glucose tolerance. Diabetic patients with abdominal obesity displayed peripheral insulin resistance in combination with defective insulin secretion, whereas non-obese diabetic patients showed only a secretory defect. Thus, Type 2 diabetes in obese and non-obese elderly male subjects may take two forms where the cause of hyperglycaemia differs.     

The mechanism of exaggerated glucagon response to arginine in diabetes mellitus

As far as exaggerated arginine-induced glucagon secretion in diabetics is concerned, the authors have shown that both the restoration of blood glucose excursions and physiological insulinemia in response to arginine, obtained from an artificial endocrine pancreas (AEP) could normalize the glucagon secretory responses in diabetes mellitus. To clarify whether or not physiological glycemic excursions and/or plasma insulin profiles contribute to the normalization of the exaggerated glucagon response in diabetes mellitus, the following 4 investigations were conducted on each of 7 non-obese, non-insulin-dependent diabetic (NIDDM), and 8 insulin-dependent diabetic (IDDM) subjects, with the aid of AEP. Arginine was i.v. infused into both diabetic groups (1) in a hyperglycemic state without insulin infusion, (2) in perfect glycemic control with insulin infusion by AEP, (3) in glycemic control with AEP, but with lower plasma insulin profiles (parameters of the insulin infusion algorithm were made smaller than those of (2], (4) in a state where blood glucose levels were clamped at the same levels as obtained in (1) with the aid of glucose infusion controlled by AEP, and where physiological plasma insulin profiles were mimicked by infusing insulin at the same rates used in (2) with a pre-programmable insulin infusion system. The changes in the plasma glucagon (IRG) response in each experiment were compared with those seen in healthy subjects. For both diabetic groups it was found that: in (2) perfect normalization of glucagon response was achieved.(ABSTRACT TRUNCATED AT 250 WORDS)     

Failure of indomethacin to affect arginine-induced C-peptide and glucagon release in insulin-treated diabetics

Summary Fourteen insulin-treated diabetics were submitted to an arginine infusion test performed with either 11.7 or 5.85mg kg^-1 min^-1 arginine monohydrochloride infused during 40 min with or without previous oral administration of a low (75+50 mg) or a high (75 mg + 3 mg/kg) dose of indomethacin. Blood glucose, plasma non-esterified fatty acids, insulin, C-peptide and glucagon were determined at regular intervals before, during and after the arginine infusion. These parameters were totally unaffected by the two doses of indomethacin both in the basal state and during the arginine infusions at the two loads tested. Eight subjects had a basal C-peptide level above 0.07 pmol/ml and a mean (± SEM) maximal rise of 0.21±0.04 pmol/ml during the arginine infusion, whereas the remaining six patients had virtually zero values throughout the tests. The arginine-induced plasma glucagon rise was similar for the two rates of arginine infusion; the sum of the increments in plasma glucagon averaged 877±120 and 647±92 pg/ml (p>0.1) for the high and low rates of arginine infusion, respectively. The magnitude of the blood glucose rise appeared independent of the amount of arginine infused. Confirming previous reports, we found that the blood glucose rise after arginine was three to four times higher in subjects without C-peptide than in subjects with C-peptide. The mean glucagon response did not differ significantly between subjects with or without C-peptide. Thus, residual B cell function determines the magnitude of the blood glucose rise but not the glucagon response after intravenous arginine.     

Comparative insulinogenic effects of glucose,arginine and glucagon in patients with diabetes mellitus,endocrine disorders and liver disease

Summary In order to compare the insulinogenic effects of glucose, arginine and glucagon, plasma immunoreactive insulin levels following oral glucose loading (50 g), intravenous arginine infusion (30 g for 45 min) and intravenous glucagon injection (1 mg) were determined in patients with diabetes mellitus, various endocrine diseases and chronic hepatitis. In patients with Cushing’s syndrome, plasma insulin responses to all three stimuli were exaggerated, whereas they were low in patients with pheochromocytoma. In other diseases, certain disparities were observed in plasma insulin responses. In patients with mild diabetes mellitus, insulin secretion elicited by glucose seems to be selectively impaired, because arginine and glucagon caused a rise in plasma insulin not significantly different from that in normal subjects. In patients with hyperthyroidism, plasma insulin responses to arginine and glucagon were either absent or limited, although rather a exaggerated response was noted following oral glucose loading. On the contrary, exaggerated responses to arginine and glucagon, and limited response to glucose were observed in hypothyroidism. In patients with chronic hepatitis, the responses of plasma insulin to glucose and arginine were both exaggerated, whereas the response to glucagon was comparable to that in normal subjects. These disparate responses suggest that glucose, arginine and glucagon act on the B-cell via different mechanisms.     

Plasma islet amyloid polypeptide (Amylin) levels and their responses to oral glucose in Type 2 (non-insulin-dependent) diabetic patients

Summary Fasting plasma islet amyloid polypeptide concentrations and their responses to an oral glucose load were determined in non-diabetic control subjects and patients with abnormal glucose tolerance in relation to the responses of insulin or C-peptide. Plasma islet amyloid polypeptide was measured by radioimmunoassay. In the non-diabetic control subjects, fasting plasma islet amyloid polypeptide was 6.4±0.5 fmol/ml (mean ± SEM) and was about 1/7 less in molar basis than in insulin. The fasting islet amyloid polypeptide level rose in obese patients and fell in patients with Type 1 (insulin-dependent) diabetes mellitus. In non-obese patients with impaired glucose tolerance and Type 2 (non-insulin-dependent) diabetic patients without insulin therapy, the level was equal to that of the control subjects, but a low concentration of islet amyloid polypeptide relative to insulin or C-peptide was observed in the non-obese Type 2 diabetic group. The patterns of plasma islet amyloid polypeptide responses after oral glucose were similar to those of insulin or C-peptide. However, compared to non-obese patients, a hyper-response of islet amyloid polypeptide relative to C-peptide was noted in obese patients who had a hyper-response of insulin relative to C-peptide. This study suggests that basal hypo-secretion of islet amyloid polypeptide relative to insulin exists in non-obese Type 2 diabetes and that circulating islet amyloid polypeptide may act physiologically with insulin to modulate the glucose metabolism.     

Beta-endorphin infusion restores acute insulin responses to glucose in type-2 diabetes mellitus

To address the possibility that an abnormality in pancreatic beta-endorphin activity might contribute to abnormal insulin secretion in diabetes mellitus, we studied the effects of beta-endorphin infusion on islet function in diabetic patients. The iv infusion of human beta-endorphin at a dose of 0.5 mg/h for 2 h in type-2 non-insulin-dependent diabetic patients (n = 12) raised plasma insulin and glucagon levels and slightly but significantly lowered plasma glucose concentrations. beta-Endorphin infusion also resulted in reappearance of a clear-cut acute insulin response to glucose, while second phase insulin release was increased and glucose disposal accelerated. Acute insulin and glucagon responses to arginine were not increased by beta-endorphin, suggesting that the effect of the opioid on the B cells of the diabetic patients is specific for glucose. An intraislet abnormality of opioid peptides action and/or secretion may play a role in the disturbances of insulin secretion in patients with type-2 diabetes mellitus.     

Normalization of fasting hyperglycaemia by exogenous glucagon-like peptide 1 (7-36 amide) in Type 2 (non-insulin-dependent) diabetic patients

Summary Glucagon-like peptide 1 (GLP-1) (7-36 amide) is a physiological incretin hormone that is released after nutrient intake from the lower gut and stimulates insulin secretion at elevated plasma glucose concentrations. Previous work has shown that even in Type 2 (non-insulin-dependent) diabetic patients GLP-1 (7-36 amide) retains much of its insulinotropic action. However, it is not known whether the magnitude of this response is sufficient to normalize plasma glucose in Type 2 diabetic patients with poor metabolic control. Therefore, in 10 Type 2 diabetic patients with unsatisfactory metabolic control (HbA_lc 11.6±1.7%) on diet and sulphonylurea therapy (in some patients supplemented by metformin or acarbose), 1.2 pmol ×kg^–1×min^–1 GLP-1 (7-36 amide) or placebo was infused intravenously in the fasting state (plasma glucose 13.1±0.6 mmol/l). In all patients, insulin (by 17.4±4.7 nmol ×1^–1×min; p=0.0157) and C-peptide (by 228.0±39.1 nmol×1^–1×min; p=0.0019) increased significantly over basal levels, glucagon was reduced (by -1418±308 pmol ×1^–1×min) and plasma glucose reached normal fasting concentrations (4.9±0.3 mmol/l) within 4 h of GLP-1 (7-36 amide) administration, but not with placebo. When normal fasting plasma glucose concentrations were reached insulin returned towards basal levels and plasma glucose concentrations remained stable despite the ongoing infusion of GLP-1 (7-36 amide). Therefore, exogenous GLP-1 (7-36 amide) is an effective means of normalizing fasting plasma glucose concentrations in poorly-controlled Type 2 diabetic patients. The glucose-dependence of insulinotropic actions of GLP-1 (7-36 amide) appears to be retained in such patients.     

Evaluation of insulin resistance during inhibition of endogenous insulin and glucagon secretion by somatostatin in non-obese subjects with impaired glucose tolerance

Summary Insulin resistance was studied in seven non-obese male subjects with impaired glucose tolerance and four healthy, age and body-weight matched male control subjects by means of a continuous intravenous infusion of somatostatin, glucose and insulin over 150 min. Glucose tolerance was evaluated by means of a 2-h glucose infusion test. Endogenous insulin (C-peptide), growth hormone, and glucagon secretion were suppressed by somatostatin in both groups. Steady-state plasma insulin and glucose levels were achieved between 90–135 min. Since similar steady-state levels of exogenous insulin were achieved, the resulting steady-state plasma glucose level provided a direct estimate of the ability of insulin to dispose of the infused glucose. The glucose levels were higher in subjects with impaired glucose tolerance with values of 14.6 ± 1.8 mmol/1 compared with 5.1 ± 1.2 mmol/1 in control subjects (p < 0.01), thus indicating insulin resistance. There was a direct correlation between the steady-state plasma glucose level and glucose tolerance suggesting that the degree of glucose intolerance is proportional to the degree of insulin resistance. These results revealed that decreased insulin sensitivity is found in non-obese subjects with impaired glucose tolerance.     

Insulin secretion in patients with myotonic dystrophy and their relatives

Summary Oral glucose, i.v. tolbutamide and i.v. arginine tolerance tests were performed in 11 patients with myotonic dystrophy and 9 of their clinically unaffected relatives. Five of the myotonic patients had glucose intolerance; 7 had exaggerated immunoreactive insulin (IRI) response to glucose. One of the 9 relatives demonstrated glucose intolerance and none had exaggerated insulin response to glucose. Three relatives, all obese, hyperresponded to arginine and one of them also responded excessively to tolbutamide. The results indicate that an exaggerated IRI response to glucose is common in myotonic dystrophy. Enhanced responses to other stimuli are less frequent. Although glucose intolerance occurred in half of the patients, the fact that the highest IRI levels were seen in non-diabetic patients suggests that this excessive response may protect against glucose intolerance. Our studies in relatives do not support the potential usefulness of testing for hyperinsulinemia in the early detection of myotonic dystrophy. Presented at the 9th International Diabetes Federation Congress in New Delhi, India, November 2, 1976.     

Time-dependent inhibition of insulin release: suppression of the arginine effect by hyperglycaemia

Summary Brief stimulation of the pancreas with arginine causes a refractory state which reduces the insulin response to subsequent stimulations (time-dependent inhibition). In control subjects, a pair of arginine injections (75mg/kg) at a 30-min interval resulted in 20% reduction of peak and integrated insulin responses to the second injection. In Type 2 (non-insulin-dependent) diabetic patients and in obese subjects, the inhibitory effect of repeated arginine stimuli was abolished. Healthy subjects were made acutely hyperglycaemic (9.3±0.3 mmol/l) by the glucose clamp technique. This induced a three- to fivefold greater insulin response to arginine. Compared to the response of diabetic subjects with similar hyperglycaemia, the control subjects secreted four to nine times more insulin. When the arginine stimulation was repeated 30 min later, no inhibition was observed, the second insulin response being instead augmented 1.5- to 1.8-fold. We conclude that (1) the insulin response to arginine is markedly reduced in Type 2 diabetes; (2) arginine-induced time-dependent inhibition of insulin release is abolished in patients with minimal to moderate hyperglycaemia; (3) this is probably due to the acute synergistic action of glucose and arginine on the B cell; (4) time-dependent inhibition of insulin release may be a protective mechanism against insulin oversecretion following repetitive stimulation of the pancreas; its abolition in hyperglycaemic states may be a compensatory mechanism, allowing substantial increases in insulin output.     

Pancreatic alpha- and beta-cell function in pheochromocytoma.

Arginine infusion tests were carried out in seven patients with pheochromocytoma before and after extirpation of the tumors in order to evaluate pancreatic islet alpha- and beta-cell function during the state of endogenous catecholamine excess. Six of the patients had glucose intolerance; one did not. Preoperatively, the pancreatic glucagon response was suppressed, while the insulin response was comparable to that in normal control subjects. Plasma glucose levels decreased rapidly after the beginning of arginine infusion in all patients. Theses changes during the infusion were evident in the one patient without glucose intolerance. Postoperatively, the glucagon response and plasma glucose changes were normalized. In addition to the obvious suppression of pancreatic alpha-cell function in our patients with pheochromocytoma, it seems likely that pancreatic beta-cell function also was suppressed; there was no enhancement of the insulin response to arginine during the period of chronic hyperglycemia, a situation in which a synergistic effect between glucose and arginine might be expected.     

Xylitol: stimulation of insulin and inhibition of glucagon responses to arginine in man.

The effect of intravenous xylitol infusions on plasma glucagon and insulin responses to intravenous arginine infusions (30 g for 45 min) or arginine "pulses" (4 g for 2 min) was studied in normal subjects. Intravenous infusion of arginine caused biphasic increases in plasma glucagon and insulin in all subjects studied. The increase in plasma glucagon induced by arginine infusion was significantly reduced by xylitol infusions started at 45 min before arginine infusion, irrespective of virtually unchanged blood glucose levels. Plasma insulin response to arginine was exaggerated by xylitol infusion.Repeated arginine pulses given at 30 min intervals evoked uniphasic and almost identical rises of plasma insulin and glucagon with each pulse. Intravenous xylitol infusions significantly blunted plasma glucagon responses and augumented the plasma insulin response to arginine pulses, despite only slight elevations of plasma glucose. These results suggest that xylitol has an inhibitory effect on both basal and arginine-stimulated glucagon secretion, while it enhances insulin secretion.     

Plasma islet amyloid polypeptide levels in obesity, impaired glucose tolerance and non-insulin-dependent diabetes mellitus.

We examined the response of plasma islet amyloid polypeptide (IAPP) to an oral glucose load in non-obese and obese subjects with normal glucose tolerance or impaired glucose tolerance (IGT), and in non-obese patients with non-insulin-dependent diabetes mellitus (NIDDM). Plasma IAPP response to intravenous glucagon injection in NIDDM patients was also studied. Plasma IAPP concentration was determined by a sensitive and specific radioimmunoassay. Basal levels of plasma IAPP in non-obese subjects with normal glucose tolerance, IGT and NIDDM were not significantly different from each other. Non-obese subjects with IGT showed delayed and higher plasma IAPP response to oral glucose load compared to normal non-obese subjects. In NIDDM patients, IAPP response to glucose was delayed and lower when compared to normal non-obese subjects. Basal levels of plasma IAPP in normal obese subjects and obese subjects with IGT were significantly higher than those in normal non-obese subjects. Plasma IAPP response to glucose load in these obese subjects was higher than that in normal non-obese subjects. Plasma IAPP response was decreased in diabetic patients treated with diet, oral hypoglycemic agents and insulin in that order. We conclude that the secretion of IAPP is reduced with progression of NIDDM, although it appears to be rather augmented in IGT compared to normal non-obese subjects.     

Amylin-induced in vivo insulin resistance in conscious rats: the liver is more sensitive to amylin than peripheral tissues

Summary Amylin is a polypeptide of 37 amino acids, predominantly synthesized in pancreatic Beta cells. The peptide was suggested to be dysregulated in Type 2 (non-insulin-dependent) diabetes mellitus and it antagonized certain actions of insulin in vitro in rat muscle. This led to speculation that amylin is involved in the pathogenesis of Type 2 diabetes. We have examined the in vivo effects of rat amylin, amidated at the carboxy-terminus, on insulin-mediated carbohydrate metabolism in conscious rats, using the hyperinsulinaemic (±1 nmol/l) euglycaemic (6 mmol/l) clamp technique combined with [3-³H]-glucose infusion. Basal plasma amylin levels were 75 pmol/l. Applied amylin levels of 220±75 pmol/l (infusion rate of 12.5 pmol/min) antagonized only the insulin action on liver, resulting in a 100% increase of hepatic glucose output. Amylin levels of 4750±750 pmol/l (infusion rate of 125 pmol/min) induced a 250% increase of insulin-inhibited hepatic glucose output and, in addition, a 30% decrease of insulin-stimulated peripheral glucose uptake. Amylin did not affect: 1) the metabolic clearance rate of insulin, 2) the levels of plasma glucagon, epinephrine, norepinephrine, and corticosterone, 3) in vitro insulin binding and insulin-stimulated receptor autophosphorylation. This suggests that amylin antagonizes insulin action via binding to a yet unknown receptor. In conclusion: amylin causes in vivo insulin resistance and the liver seems the predominant organ regulated by this hormone. The in vivo effects of amylin mimic the pathophysiological abnormalities of insulin action in Type 2 diabetes.     

Hyperinsulinaemia in obesity is not accompanied by an increase in serum proinsulin/insulin ratio in groups of human subjects with and without glucose intolerance

Summary Serum proinsulin is disproportionately elevated compared to insulin in Type 2 (non-insulin-dependent) diabetes mellitus. We studied the effect of obesity on serum proinsulin with varying degrees of glucose intolerance. Serum proinsulin and insulin were measured during a 75 g oral glucose tolerance test in 73 obese and 74 non-obese subjects with normal, borderline or diabetic-type glucose tolerance. Proinsulin was assayed by a direct radioimmunoassay using proinsulin-specific antiserum. Fasting serum proinsulin and insulin and the summed values of proinsulin and insulin during oral glucose tolerance test were significantly, or tended to be, higher in obese subjects than in those without obesity in each category of glucose tolerance. However, the molar ratio of proinsulin to insulin was nearly the same between obese and non-obese groups with a similar degree of glucose tolerance. On the other hand, the proinsulin/insulin ratio increased progressively with the deterioration of glucose tolerance. We conclude that proinsulin secretion is disproportionately increased in the presence of glucose intolerance but not by obesity itself. Each Beta cell seems to function normally in obese subjects while glucose tolerance remains normal.This work was presented at the 6th International Congress on Obesity in October 1990, Kobe, Japan     

Plasma glucagon in insulinoma

Summary Nine patients with insulinoma were studied in order to investigate glucagon levels in the fasting state and the response of plasma glucagon to tolbutamide and arginine. Fasting plasma glucagon levels were within the normal range in all patients except two cases with malignant insulinoma. Although there was no correlation between blood glucose and plasma glucagon, a significant correlation between plasma glucagon and plasma insulin was observed. No detectable changes were found in glucagon levels during tolbutamide injection. In almost all patients except one an exaggerated response of plasma glucagon was demonstrated during arginine infusion test.     

Effect of a long acting glucagon selective somatostatin analogue on plasma glucose,insulin and glucagon levels in the anaesthetized rat during arginine infusion

Summary The effects of somatostatin and a long acting, glucagon selective somatostatin analog (des-Ala¹Gly²[His^4,5-D-Trp⁸]-somatostatin), were studied during arginine tolerance tests in normal anaesthetized rats. Arginine infusion in control animals resulted in a rapid increase in plasma insulin and glucagon, and an increase of 15±5 mg/dl in plasma glucose. Somatostatin infusion (1 mg/kg/h) resulted in suppression of basal insulin secretion and a decrease in arginine-induced insulin and glucagon release. Glucose levels increased rapidly during the combined arginine-somatostatin infusion reaching a peak of 72±10 mg/dl above basal levels. Similar results were obtained when somatostatin was injected SC (1 mg/kg) at times 0, 15, 30, and 45 minutes (arginine infused from 30–60 minutes). A single injection (1 mg/kg) of the long-acting somatostatin analogue resulted in significant inhibition of basal insulin and glucagon release; during arginine infusion glucagon levels rose only slightly, the insulin response was, however, nearly normal, and only a small arginine-induced increase in glucose levels was observed. Carbohydrate absorption was not influenced by either somatostatin or the analogue.     

Glucose storage and oxidation in different degrees of human obesity measured by continuous indirect calorimetry

Summary Glucose disposal of a 100 g glucose load has been determined in 26 obese compared with 10 non-obese subjects by means of a new application of continuous indirect calorimetry. The obese subjects were divided into 4 groups, according to their degree of glucose intolerance and their insulin response to the glucose load. Through this division it appeared that subjects with no glucose intolerance were moderately obese while the groups with glucose intolerance showed a higher degree of obesity, glucose intolerance increasing with age. The 10 obese subjects with no glucose intolerance (group A) presented values for glucose disposal similar to those of the control subjects. The 4 obese subjects with impaired glucose tolerance (group B) showed no significant changes in glucose storage and in basal oxidation, but a significant decrease in oxidation in response to the load (11±2g vs 19±1 g in the control group, p <0.02). The 6 obese subjects with overt diabetes and elevated insulin response to the glucose load (group C) showed a significant decrease in glucose storage (34±6 g vs 63±1 g, p < 0.001) but not in oxidation. The 6 obese subjects with overt diabetes and decreased insulin response (group D) showed a significant decrease in glucose storage (25±4 g vs 63 ±1 g, p < 0.001) and oxidation (12±1 g, vs 19+1 g, p < 0.005). These observations show that in obese diabetics, glucose intolerance results primarily from decreased glucose storage and to a lesser extent from a decrease in glucose oxidation.     

Quantitative and qualitative differences in basal and glucose- and arginine-stimulated insulin secretion in healthy subjects and different stages of NIDDM

Summary To determine quantitative and qualitative differences in insulin secretion equimolar amounts of glucose and arginine were infused in 9 healthy subjects, in 8 individuals each with obesity without and with impaired glucose tolerance, and in non-obese and obese non-insulin-dependent diabetic patients (NIDDM). Insulin secretion was calculated after individual determination of metabolic clearance rate of C-peptide (MCRcp) both as the area under the C-peptide concentration curve times MCRcp, and by a mono-compartment mathematical model, both yielding identical results. MCRcp fell consistently with increasing C-peptide infusion rate (e.g.: healthy subjects: C-peptide, 10 nmol/h, 4.2±0.4; 20 nmol/h, 3.3±0.3; 30 nmol/h, 3.1±0.2 ml/kg · min; p<0.05 to p<0.01). Basal insulin secretion was 2.1-fold greater in the obese with impaired glucose tolerance than in healthy subjects, but was unchanged in non-obese NIDDM. Glucose and arginine triggered insulin release was greater than in healthy subjects at almost identical area under the respective substrate concentration curve (AUC/kg body weight) in obese subjects without (2-fold) and with impaired glucose tolerance (4-fold), and in NIDDMs following i.v. arginine (2-fold). The mean ratio of incremental insulin release to i.v. glucose and arginine was smaller in NIDDM (normal weight, 1.3±0.4; obese, 1.0±0.2) than in healthy (2.0±0.3), or obese subjects with impaired glucose tolerance (2.8±0.7). Stimulated C-peptide/insulin ratio was reduced in all patientsvs that in healthy subjects (p<0.05). We conclude that (a) MCR of C-peptide is in part a saturable process; (b) insulin clearance may be impaired in obesity and NIDDM; and (c) insulin secretion differs in obese states and NIDDM both quantitatively and qualitatively, and thereby separates the two disorders as different entities. In addition, quantitation of insulin release in obese states may also help (d) to better define primary algorithms for insulin replacement in normal- and overweight insulin-dependent diabetic patients.     

Characterization of insulin secretion and resistance in type 2 diabetes of adolescents

CONTEXT: Type 2 diabetes (T2D) in obese children is an emerging problem, including in Europe. Its presentation at diagnosis very often differs from that in adults. OBJECTIVE: The objective of this study was to investigate the relative contributions of the two components of T2D, insulin resistance and insulin secretion, early in the history of the disease in adolescents. PATIENTS AND METHODS: Six obese adolescents with T2D were included 2 months to 4.3 yr after diagnosis (five girls and one boy; median age, 15.4 yr; median body mass index, 4.4 sd). Peripheral and hepatic insulin sensitivity was evaluated with euglycemic hyperinsulinemic (40 mU/m(2).min) clamp. First-phase insulin release was evaluated after iv glucose stimulation. A graded iv glucose infusion and an arginine test were performed to measure insulin secretion. RESULTS: All patients showed decreased peripheral glucose uptake to the same extent. Five patients showed hepatic insulin resistance. First-phase insulin release was very low in two patients. Three patients showed an exaggerated insulin response under graded glucose infusion and preserved secretion under arginine stimulation. Three other patients, with elevated fasting plasma glucose levels, demonstrated a very low insulin response under glucose stimulation and a low insulin response under arginine stimulation. CONCLUSIONS: These data emphasize that together with marked insulin resistance, the failure of beta-cell function is a major component in the course of T2D in childhood.