Search

OBJECTIVE To examine the persistence of the original treatment effects 10 years after the Diabetes Control and Complications Trial (DCCT) in the follow-up Epidemiology of Diabetes Interventions and Complications (EDIC) study. In the DCCT, intensive therapy aimed at near-normal glycemia reduced the risk of microvascular complications of type 1 diabetes mellitus compared with conventional therapy. METHODS Retinopathy was evaluated by fundus photography in 1211 subjects at EDIC year 10. Further 3-step progression on the Early Treatment Diabetic Retinopathy Study scale from DCCT closeout was the primary outcome. RESULTS After 10 years of EDIC follow-up, there was no significant difference in mean glycated hemoglobin levels (8.07% vs 7.98%) between the original treatment groups. Nevertheless, compared with the former conventional treatment group, the former intensive group had significantly lower incidences from DCCT close of further retinopathy progression and proliferative retinopathy or worse (hazard reductions, 53%-56%; P < .001). The risk (hazard) reductions at 10 years of EDIC were attenuated compared with the 70% to 71% over the first 4 years of EDIC (P < .001). The persistent beneficial effects of former intensive therapy were largely explained by the difference in glycated hemoglobin levels during DCCT. CONCLUSION The persistent difference in diabetic retinopathy between former intensive and conventional therapy ("metabolic memory") continues for at least 10 years but may be waning. TRIAL REGISTRATION (clinicaltrials.gov) Identifiers: NCT00360815 and NCT00360893.

The Diabetes Control and Complications Trial (1983–1993) of 1,441 subjects followed for an average of 6.5 years assessed the effects of intensive therapy aimed at maintaining near normal levels of blood glucose versus conventional therapy on the risks of diabetes complications of the eyes, kidneys, and nerves. The study was designed to test the hypothesis that the higher than normal blood glucose levels associated with conventional insulin therapy caused these complications. The study was terminated one year ahead of schedule by the monitoring board. This paper describes the medical, ethical, and statistical challenges faced by the study group and the monitoring board.

▪ Abstract The Diabetes Control and Complications Trial (DCCT) demonstrated that intensive treatment of patients with insulin-dependent diabetes mellitus (IDDM) can substantially reduce the onset and progression of diabetic retinopathy, nephropathy, and neuropathy. The major risk associated with intensive treatment is recurrent hypoglycemia. Implementation of intensive treatment recommendations is difficult but should be considered and probably recommended to most patients with IDDM. If intensive treatment is impractical, any improvement in glycemic control is probably beneficial. Improved glycemic control should be recommended to most patients with non-insulin-dependent diabetes mellitus (NIDDM). The use of insulin in patients with NIDDM is controversial, especially in patients who are overweight, overeating, and minimally symptomatic.

The metabolic response to exercise in insulin-dependent diabetic (IDD) man was assessed during continuous insulin infusion using the subcutaneous (CSII), intravenous (CIVII), and intraperitoneal (CIPII) routes. During the basal period, plasma glucose levels were higher with CIPII (153 +/- 17 mg/dl) than with CSII (117 +/- 13 mg/dl) or CIVII (118 +/- 17 mg/dl). Basal free insulin concentrations were similar for CSII (12.3 +/- 10 microU/ml) and CIVII (12.4 +/- 1.4 MicroU/ml) but lower in CIPII (8.5 +/- 1.0 microU/ml, P less than 0.05). Exercise on a stationary bicycle at 75 W for 60 min produced a decline of plasma glucose in each protocol that was significantly only during CIVII (55 +/- 11 mg/dl, P less than 0.01). Insulin levels remained unchanged throughout the study period in all protocols. In normals, insulin values decreased during exercise and remained below basal levels through the recovery period (P less than 0.05), while plasma glucose remained unchanged. Plasma glucagon and epinephrine levels were similar in all protocols and remained unchanged with exercise, while plasma norepinephrine tended to be higher than normal in all diabetic subjects. Significant differences between normal and diabetic subjects (P less than 0.05) were observed for blood ketone bodies, while blood lactate, glycerol, and plasma FFA were similar. Normalization of intermediary metabolites occurred only with CIVII. Continuous insulin infusion provides near-normal glycemic and metabolic control before, during and following exercise in IDD man. However, to produce normal blood concentrations of intermediary metabolites during exercise, the insulin infusion rate may be excessive in terms of its hypoglycemic effect. CSII appears to be a safe, accessible, and adequate method for treating diabetic man during exercise.

The roles of basal insulin, glucagon, and free fatty acids in the regulation of ketogenesis were studied in three-day-fasted anesthetized dogs. Four protocols were employed: (1) infusion of somatostatin, resulting in combined insulin and glucagon deficiency, (2) somatostatin and intraportal glucagon replacement (1 ng./kg./min.) to produce insulin deficiency, (3) somatostatin combined with intraportal insulin replacement (350 μU./kg./min.) to produce glucagon deficiency, and (4) saline controls. Simultaneous blood sampling from the femoral artery and portal and hepatic veins allowed determination of hepatic uptake or production of free fatty acids, glycerol, ketone bodies, and glucose. When combined deficiency of insulin and glucagon was produced, no significant effect on hepatic ketone production was noted (108 ± 6 per cent of mean basal), whereas the induction of selective Insulin deficiency (basal glucagon level maintained) resulted in a rise of net hepatic ketone production (185 ± 24 per cent of mean basal, P < 0.01). Isolated glucagon lack (basal insulin maintained) did not alter net hepatic ketone production. To assess the effect of increased substrate supply on ketogenesis, all experiments included a period of acute elevation of free fatty adds (FFA), produced by infusion of Intralipld and heparin. FFA elevation in saline controls caused only a moderate stimulation of ketone production (149 ± 16 per cent of mean basal, P < 0.01) despite a threefold increase in FFA uptake. However, the combination of elevation of FFA and selective insulin deficiency (glucagon maintained) resulted in a greatly increased hepatic ketone production (357 ± 58 per cent of mean basal, P < 0.01 vs. controls), which was also significantly higher than during combined insulin and glucagon deficiency (P < 0.05). Lipolysis, as indicated by arterial glycerol levels, as well as hepatic FFA uptake was not affected by acute suppression of pancreatic hormone levels. The study demonstrated that basal insulin plays an important role in repressing ketogenesis and that basal amounts of glucagon, while ineffective in the presence of insulin, exert a stimulatory effect on ketogenesis when insulin is deficient. Glucagon's stimulatory effect on ketogenesis was due to an effect in the liver rather than to an increase in lipolysis or in hepatic FFA uptake. Increasing FFA supply per se was associated with only limited stimulation of ketogenesis, whereas the combination of insulin deficiency, basal concentrations of glucagon, and increased FFA load produced a synergistic augmentation of hepatic ketone production.

Somatostatin has recently been used as a tool to investigate the roles of insulin and glucagon in the regulation of glucose metabolism in vivo. The alterations in glucose uptake and production that somatostatin causes have been attributed to the modifications it induces in pancreatic hormone levels rather than to effects of the peptide per se, although the direct effects of the latter on glucose turnover have not been clearly assessed. The aim of the present study therefore was to determine whether or not somatostatin directly affects either the rate of glucose uptake by fat and muscle or the rate of glucose production by the liver. Glucose uptake by the perfused rat hindquarter (skeletal muscle) was unaltered by somatostatin (10-1,000 ng./ml.) whether or not the process was stimulated by insulin (250 μU./ml.). Basal and insulin-(5-40 / μU./ml.)-stimulated glucose oxidation (uptake) by isolated rat adipocytes was also unaffected by somatostatin (5-500 ng./ml.). In addition, the peptide had no effect on basal or glucagon-(10−l0–10−9 M)-stimulated glucose output by isolated rat liver cells. Somatostatin also failed to modify the actions of these sub-maximally effective levels of glucagon on either the intracellular level of cAMP or the activity of cAMP-dependent protein kinase. The present studies support the hypothesis that the acute changes in the level and turnover of glucose induced by somatostatin are attributable to its effect on the endocrine system rather than to its direct effects on glucose metabolism.

The effect of glucagon (50 ng/kg/min) on arterial glycerol concentration and net splanchnic production of total ketones and glucose was studied after an overnight fast in four normal and five insulin-dependent diabetic men. Brachial artery and hepatic vein catheters were inserted and splanchnic blood flow determined using indocyanine green. The glucagon infusion resulted in a mean circulating plasma level of 4,420 pg/ml. In the normal subjects, the glucagon infusion resulted in stimulation of insulin secretion indicated by rising levels of immunoreactive insulin and C-peptide immunoreactivity. Arterial glycerol concentration (an index of lipolysis) declined markedly and net splanchnic total ketone production was virtually abolished. In contrast, the diabetic subjects secreted no insulin (no rise in C-peptide immunoreactivity) in response to glucagon. Arterial glycerol and net splanchnic total ketone production in these subjects rose significantly (P=

What we would like to have accomplished in this discussion, is perhaps mainly to have emphasized once more the likelihood that adipose tissue is a predominant site of insulin action, perhapsthe predominant site. Insulin would seem to be the primary hormone of energy storage, favouring both the deposition and the retention of the major energy store: fat in adipose tissue. While it may still be true, that both these effects — storage and retention — are the result of just one primary action — increased glucose transport — a separate effect directly resulting in decreased fatty acid mobilization may be an important aspect of this control, as well. We consider the availability for study of isolated adipose tissue cells, which still appear to be fully responsive to insulin, a major advance, holding the promise of arriving some day at a true understanding of the intimate mechanism of insulin action. When considering the regulation of fat release from adipose tissue, it would seem that closer scrutiny of the controlling importance of blood and plasma flow through the tissue might well be warranted. Adipose tissue is no longer considered a static tissue; it is recognized as what it is: the major site of active regulation of energy storage and mobilization, one of the primary control mechanisms responsible for the survival of any given organism. Although general recognition of the important role played by adipose tissue has abruptly increased during the past ten years, it remains for the future to provide us with a detailed understanding of many aspects of the control systems which are operative in each instance.

Suspensions of fat cells were prepared from rat epididymal adipose tissue by digestion of the tissue with crude bacterial collagenase. One portion of each suspension was exposed to p-chloromercuribenzenesulfonic acid (CMS), washed, and then incubated in Krebs-Ringer-bicarbonate buffer. A second portion, the control, was not exposed to CMS, but it was otherwise treated identically. Acceleration of glucose transport was estimated from the rate of conversion of glucose-14C to 14CO2 and 14C-labeled lipids. Inhibition of the lipolytic action of adrenocorticotropic hormone (ACTH) was evaluated by measuring (a) the cyclic 3',5'-AMP concentration of the cell suspensions and (b) the rate of release of glycerol into the incubation medium. The optimal concentration of CMS was 10-3 m and the optimal exposure time was 30 min. Glucose utilization by cells exposed to CMS was 5 to 20 times that of the control, but only 50 to 80% that of cells maximally stimulated by insulin. Most of the increased glucose utilization was due to increased glucose entry via a specific transport system since glucose utilization, when stimulated by CMS, was inhibited by 3-O-methyl glucose. Cells exposed to CMS showed a slight increase in nonspecific permeability, however, since (a) there was a slight increase in the amount of malic dehydrogenase appearing in the incubation medium, and (b) stimulation by CMS plus insulin exceeded that of insulin alone when the glucose concentration of the incubation medium was l20 mm. Incubation of CMS exposed cells in 2,3-dimercaptopropanol resulted in partial reversal of the stimulatory effect of CMS. Exposure of cells to CMS inhibited lipolysis both in the basal state and when lipolysis was stimulated by ACTH. This inhibition was overcome either by a 10-fold increase in the concentration of ACTH or by the addition of theophylline, showing that the CMS exposure had not irreversibly inhibited the lipolytic enzymes of the cell. The antilipolytic action of CMS was probably mediated through an effect on the adenyl cyclase system since the increase in cyclic 3',5'-AMP produced by ACTH was inhibited by the CMS exposure. This inhibitory effect of CMS was likewise overcome by a 10-fold increase in the ACTH concentration. We concluded that the metabolism of CMS-treated fat cells mimicked that of cells incubated with insulin in that glucose transport was stimulated and lipolysis was reversibly inhibited. Since the reaction of CMS with isolated fat cells seemed analogous to its reaction with the sulfhydryl groups of other proteins, we postulated that the insulin-like actions of CMS were mediated through a combination between CMS and certain sulfhydryl containing elements located on or near the cell surface.

With dilute suspensions of rat epididymal adipose tissue cells, glucose utilization was estimated by the conversion of glucose-14C to 14CO2 and 14C-labeled lipids. The maximal effect of insulin was observed with a glucose concentration of 1.25 mm or less. The same rate of glucose utilization was observed with and without insulin when the glucose concentration was 80 mm. Glucose utilization was stimulated with 1 microunit of insulin per ml, and insulin concentrations above 25 microunits per ml produced no further stimulation. The rate of glucose utilization was linear for 2 hours when the insulin concentration was 3 microunits per ml. With 5 microunits of insulin per ml, the glucose utilized in 2 hours did not increase when the volume of the incubation medium was doubled. The stimulatory effect of insulin was reversed by washing the cell suspension or by the addition of anti-insulin serum. These findings indicated that (a) a dilute suspension of fat cells did not inactivate insulin rapidly enough to lower the insulin concentration of the incubation medium significantly, and (b) the continued presence of insulin was required for a continuation of its biological effect. With a 30 times more concentrated suspension of fat cells, the change in immunoreactive insulin concentration of the incubation medium was measured. A rapid phase of insulin uptake occurred within the first minute, was linearly related to insulin concentration up to 10,000 microunits per ml, was not temperature-sensitive, and was abolished by a 30-sec exposure to 10-3 m maleimide. A subsequent phase of insulin uptake was slower, temperature-sensitive, and not abolished by maleimide. Cells previously incubated with insulin showed undiminished insulin uptake when reincubated in fresh medium. When insulin-treated cells were washed, no immunologically or biologically active insulin was released. It was concluded that (a) the rapid phase of insulin uptake may be the hormone-cell interaction which is necessary for the action of insulin to be manifest, and (b) a study of the factors which influence this process may lead to a better understanding of the mechanism of hormone action.

SummaryGlucose, with and without exogenous insulin, was infused into normal males under conditions which produced a submaximal fall in total plasma FAA concentrations. Analysis of the individual free plasma amino acids revealed a significant decrease in the following essential amino acids; leucine, valine, threonine, isoleucine, phenyl-alanine and tyrosine. This effect is enhanced by the presence of exogenous insulin.

The major effects of insulin on glucose, fatty acid and amino acid metabolism and on ion flux are initiated by the interaction of insulin with certain unidentified components of the cell membrane. These early events in the action of insulin were studied further with purified plasma membranes prepared from rat epididymal adipose cells. The binding of 125I-labeled insulin to fat cell plasma membranes gives results quantitatively similar to those reported previously with unlabeled insulin. In addition, intact fat cells, plasma membranes from fat cells and a soluble extract from the plasma membranes contain a highly active insulin degrading system. Both the capacity of the plasma membranes to bind insulin and the insulin degrading activity of the membrane extract can be inhibited by the addition of unlabeled insulin to the incubation medium even though albumin is present in excess. The degraded products have not been fully identified but are soluble in 10 per cent trichloroacetic acid, nonreactive with anti-insulin guinea pig serum, nonadsorbable to a 125I adsorbing resin and, therefore, probably represent fragments of the insulin molecule. When isolated fat cells are treated with trypsin they become insulin unresponsive. Plasma membranes derived from such cells have a reduced capacity to bind insulin, and the insulin degrading system of the membranes or in the soluble extract from the membranes is inactive. This suggests that both the insulin receptor and the insulin degrading system are located on the external surface of the plasma membranes of the fat cells. It is postulated that an insulin degrading system of the plasma membrane may be one of the physiological mechanisms by which the response of fat cells to insulin is terminated.

Publisher Summary Isolated fat cells have been used extensively in the study of hormone actions. This preparation offers two major advantages: a uniform cell population from which, non-adipose elements can be readily removed and sensitivity to a number of hormones at physiological concentrations. Two different types of assays have been used to measure glucose transport and the effects of various hormones (primarily insulin) upon it; of these, only one (Method 2) measures transport directly. In this method, conveniently measured end products of [ 14 C] glucose metabolism are determined in the presence and absence of the hormone being tested. The validity of the method as a measure of glucose transport rests on the demonstration in intact fat pads that, under usual circumstances, transport is a rate–limiting step in glucose metabolism and that insulin exerts its major effect on glucose metabolism at the level of membrane translocation.

Comparisons were made between New Zealand obese (NZO) and control mice with respect to the following points: (1) the age of onset of obesity, (2) the frequency and severity of glucose intolerance, and (3) the degree of insulin sensivity. Total body composition analyses, intravenous glucose tolerance tests and insulin sensitivity tests were performed on the NZO and control strains. The results indicate that NZO mice have developed a significant chemical obesity by the fourth week of life while excessive body weight does not occur until after the twelfth week. Approximately two-thirds of the NZO mice demonstrate abnormal glucose tolerance tests although the degree of the glucose intolerance is relatively mild. The NZO mice are 4 to 5 times more resistant to insulin-induced convulsions than are the control mice.

Glucagon activates hepatic adenylate cyclase, thereby increasing acutely the liver content of cyclic AMP (cAMP) as well as the release of cAMP into the hepatic vein. Insulin, on the other hand, antagonizes this glucagon-mediated cAMP production, thus providing a hypothetical mechanism through which insulin might correct some of the metabolic abnormalities of diabetes. To study this hormonal interaction in man, net splanchnic cAMP production (NScAMPP) was investigated in normal and insulin-dependent diabetic men under basal conditions and in response to intravenous glucagon, 50 ng/kg/min for 2 h. In normals (n=19), basal hepatic vein cAMP concentration was 23.6±1.1 nM and NScAMPP was 1.7±0.6 nmol/min. Glucagon stimulated NScAMPP in four normal subjects to a peak of 99.6±43 nmol/min at 25 min with a subsequent fall to 12.4±5.1 nmol/min by 90 min despite continuing glucagon infusion. Endogenous insulin secretion was stimulated as indicated by rising levels of immunoreactive insulin and C-peptide (connecting peptide) immunoreactivity, raising the possibility that endogenous insulin might be responsible for the fall in NScAMPP that followed the initial spike. In the diabetics (n=8), basal hepatic vein cAMP concentration was 24.7±1.2 nM and NScAMPP was undetectable. Glucagon stimulated NScAMPP in five diabetics to a peak of 169.9±42.6 with a subsequent fall to 17.4±3.9 nmol/min by 90 min even though endogenous insulin secretion was not stimulated (no rise in C-peptide immunoreactivity). Although the mean increase in NScAMPP was greater in the diabetics, the two groups did not differ significantly. Conclusions. In normal resting man the liver is a significant source of circulating cAMP. Diabetics do not release abnormally large amounts of hepatic cAMP under basal conditions. Glucagon markedly enhances hepatic cAMP release with a spike-decline pattern in both normal and diabetic men. The decline in hepatic cAMP release despite continuing glucagon stimulation is due to factors other than a stimulation of insulin secretion.

1. 1. The effect of insulin on the transport and metabolism of sorbitol by incubated rat adipose tissue was investigated in epipidymal fat pads and isolated adipose tissue cells. 2. 2. The production of 14 CO 2 from [ 1 C]sorbitol was stimulated by insulin when adipose tissue was incubated both in the presence and in the absence of gluocose. The stimulatory effect of insulin was also observed in isolated adipose tissue cells and was not therefore dependent on the presence of reticulo-endothelial or other non-adipose tissue cells. 3. 3. The purity of the [ 14 C]sorbitol was confirmed by demonstrating that: 14 CO 2 production was not altered by the addition of 4 mM glucose; 14 CO 2 production was linear for 6 h; 14 CO 2 production was quantitatively diminished by dilution with non-radioactive sorbitol; 14 CO 2 production was not reduced by pretreatment of the radioactive sorbitol with glucose oxidase; 8.6% of the radioactivity in the incubation vessel was recovered as 14 CO 2 , 14 C-labeled lipids or 14 C-labeled glycogen under conditions which favored maximal sorbitol utilization yet the [ 14 C]sorbitol preparation was at least 96% pure by paper radiochromatography in two separate systems. 4. 4. The oxidation of 5 mM sorbitol was inhibited by o.2 mM phloretin. Under these conditions, phloretin has been shown to be an inhibitor of the carrier system involved in sugar transport in adipose tissue but not an inhibitor of the metabolism of intracellular sugars. 5. 5. The oxidation of 5 mM sorbitol was inhibited by 40, 80, and 120 mM 3- O -methyl glucose, a sugar analogue which is not phosphorylated, but which is transported by means of the same carrier system as that involved in glucose transport. 6. 6. The distribution of sorbitol relative to the total tissue water was not altered by the presence of insulin. 7. 7. The data were interpreted in support of the view that facilitation of the transport of sorbitol across the cell membrane and into the adipose tissue cell is the rate limiting step in the metabolism of sorbitol and, therefore, the probable site of action of insulin.

Only trace amounts of insulin were detected by an immunoassay system with guinea pig antibody to pork insulin in the New World primates Cebus and Saimiri. The system found insulin levels in the Old World primates rhesus and chimpanzee which were quite like those of human beings. The findings suggest important structural differences in the insulins of the two primate divisions.

A study of patients with compensated liver disease revealed (1) inappropriately high serum insulin levels in response to a glucose load administered either orally or intravenously, and (2) a diminished glucose response to injected insulin. These observations suggested "resistance" to the regulatory action of insulin on glucose metabolism. The actions of insulin on fat and protein metabolism, as reflected by lowering of free fatty acid (FFA) and free amino acid levels after glucose loads, were not altered. Since disappearance of injected insulin proceeded at a normal rate, insulin hypersecretion, as opposed to reduced insulin degradation, was suggested as the cause of the hyperinsulinemia. Potassium depletion, FFA elevation, and elevation of growth hormone levels could not be implicated as the cause of the abnormalities noted.

An attempt was made to distinguish the glucose intolerance observed in patients with liver disease from that seen in patients with genetically determined diabetes mellitus. Serum glucose and insulin levels were measured during oral and intravenous glucose tolerance tests (GTT) in patients with liver disease, in mild diabetics, and in control subjects. Eleven of the 16 patients with liver disease had abnormal oral GTT results. The patients with liver disease and the diabetics had comparable degrees of glucose intolerance during both the oral and intravenous GTT, despite the presence of hyperinsulinemia in the former group. The mean "total insulin excess" was approximately twice as great in the patients with liver disease as in the diabetics. The mean rate constant, K g , for the decrease in serum glucose concentration after termination of infusion was 1.12% and 1.07% per minute, respectively. These observations suggest that significant insulin resistance was present in patients with