Search

Impaired glucose tolerance (IGT) was introduced in 1979 as an intermediate category covering the grey area between unequivocal diabetes mellitus and risk free more normal glucose tolerance. The IGT group included those at high risk of subsequent development of non-insulin-dependent diabetes mellitus (NIDDM) but low risk of specific diabetic complications. Categorisation of subjects as IGT is hampered by the variability of the oral glucose tolerance test, but even those shown to be IGT only once are at increased risk of developing NIDDM. The relative roles of inheritance, fetal undernutrition, and environmental life style factors (physical inactivity and diet) in the aetiology and pathogenesis of IGT are discussed, with all contributing. The prevalence of IGT in different populations has now been widely studied with values ranging from 2 to 25% in adults. Rates of progression to NIDDM also vary widely from 2 to 14% per year. Risk factors for progression are discussed. IGT also carries an increased risk of development of cardiovascular disease (CVD) and forms part of the "metabolic syndrome". The role of insulin resistance as a common aetiological factor is briefly reviewed. Finally, possible means of treatment of IGT are listed with the intent of delaying the onset of diabetes and CVD, which is of obvious clinical importance.

Glucose tolerance was assessed, according to WHO diagnostic criteria, in 506 consecutive African patients admitted with sputum-positive pulmonary tuberculosis to the tuberculosis wards of Muhimbili Medical Centre, Dar es Salaam. Nine (1.8%) patients were known to have diabetes. Following a 75 g oral glucose tolerance test (OGTT) diabetes was diagnosed in a further 25 (4.9%) patients giving an overall crude diabetes prevalence rate of 6.7%. Impaired glucose tolerance (IGT) was present in 82 (16.2%) subjects. A repeat OGTT was carried out in the 25 patients found to have diabetic values after the first test. Eight (28%) of the 25 patients reverted to normal glucose tolerance after the second test, 6 (24%) to IGT, and 11 (48%) remained with blood glucose values in the diabetic range, giving a crude diabetes prevalence rate of 4%. In a survey of glucose tolerance, using similar methodology, in 693 members of an urban community in Dar es Salaam the prevalence rates of diabetes and IGT were 0.9% and 8.8% respectively. Diabetes was therefore at least four times as common in the tuberculosis patients (p less than 0.001), and IGT twice as frequent (p less than 0.0001). This study confirms the relationship between diabetes and tuberculosis. Those caring for patients with tuberculosis should be aware of the increased prevalence of diabetes in their patients, since failure to diagnose the problem may adversely affect prognosis.

Most diabetic patients will require surgery at some time in their lives, and this presents problems for glycemic control. Surgery involves a period of starvation, and also the metabolic response to trauma includes counterregulatory hormone hypersecretion (e.g., adrenaline, cortisol, etc.), which opposes insulin action. The major factor influencing the management of diabetes during surgery is assessment of the patient's insulin reserves. Thus, those on insulin treatment (whether they are true type 1, or insulin-treated type 2 patients) should be assumed to have little or no endogenous insulin reserves, and a system of continuous insulin and glucose delivery is necessary. Those not on insulin, however, clearly have their own insulin reserves and, provided they have reasonable preoperative glycemic control, for all but major surgery simple glycemic observation is usually all that is needed. Delivery of insulin and glucose may be by “separate lines” (intravenous (IV) glucose infusion with a separate IV insulin pump), or by a combined glucose–potassium–insulin infusion. The latter is easier to manage on lower-staffed general wards, and the former is particularly flexible and suitable for critical care situations with more metabolically unstable patients and higher levels of staffing. Whichever system is used, patients must have regular blood glucose and plasma electrolyte monitoring, with appropriate insulin adjustments to maintain good glycemic control. With good preoperative control and assessment, and well organized and appropriate perioperative management, the risks of surgery in diabetic patients should in most cases be similar to nondiabetic rates. Keywords: diabetes; surgery; type 1 diabetes; type 2 diabetes; insulin; postoperative complications; glucose–insulin infusion

This study demonstrates that patients who have undergone coronary artery bypass graft (CABG) surgery have a high prevalence of abnormal glucose tolerance 3 months and 1 year later. Although only 6% were known to have diabetes mellitus (DM) preoperatively, a further 4% were classified DM at two oral glucose tolerance tests (OGTTs) over the subsequent year and a further 18% were classified as having impaired glucose tolerance (IGT) at 12 months. Reproducibility of the 120-minute plasma glucose level in the 75-g OGTT was estimated from a repeat test performed within 10 days. The coefficient of variation (CV) of 120-minute glucose was between 14% and 18%. The observed changes in class of glucose tolerance observed at OGTTs repeated 6 and 12 months after surgery differed from the predicted changes based purely on the estimated variability of 120-minute glucose measurement. There was evidence of regression to the mean for the IGT group. However, there was also evidence of deteriorating glucose tolerance in some subjects. Between 4% and 9% of those with IGT 3 months after CABG surgery developed DM by 12 months, and 26 (13%) of those with initially normal glucose tolerance (NGT) developed IGT. Insulin and glucose responses in the OGTT and estimates of insulin resistance and beta-cell function from fasting samples show that insulin resistance was the principal abnormality in IGT subjects, whereas in DM subjects, both insulin resistance and beta-cell dysfunction contributed. Analysis of preoperative patient characteristics showed that the presence of either a systolic blood pressure of 140 mm Hg or body mass index (BMI) of more than 25 kg/m2 identified 51% of the subjects who would at 1 year after surgery include all those who would be classified DM and 67% of those who would have IGT. Further analyses including insulin levels identified groups at particularly high risk of DM, but no combination of readily available preoperative measures identified all those destined to be classified IGT.

We investigated awareness of hypoglycaemia and its counter-regulatory hormone response in six young (ages 24-49, mean 30 years) and seven elderly (ages 65-80, mean 72 years) healthy non-diabetic subjects. A hyperinsulinaemic glucose clamp was used to control blood glucose level on two separate occasions. During the hypoglycaemic study, blood glucose was lowered in a stepwise manner to plateaus of 3.5, 3.0 and 2.5 mmol/l. A symptom score, visual reaction time test and digit symbol substitution test was completed pre-study, and at the end of each plateau. Pulse rate, blood pressure and counter-regulatory hormone measurements were taken every 15-30 min. The euglycaemic study was identical except that blood glucose remained at the fasting level. During hypoglycaemia, the elderly group had lower symptom scores than the young group (total relative score at glucose 2.5 mmol/l, mean +/- SEM: elderly -1 +/- 2.5, young 23 +/- 6.7, p < 0.01) and fewer individual symptoms despite a similar counter-regulatory hormone response. There was no difference in deterioration of the visual reaction times or digit symbol substitution scores during hypoglycaemia between the age groups. Unlike the young group, the elderly subjects had no tachycardia in response to hypoglycaemia. Their reduced awareness of hypoglycaemia may be due to an impaired end-organ response to counter-regulatory hormones, resulting in fewer symptoms.

Summary Elevated circulating insulin levels have been reported in ischaemic heart disease, and may be of aetiological importance. Previous studies have not considered the potential influence of heart failure or of previous myocardial infarction, as opposed to stable angina. We therefore measured the insulin response to a 75 g oral glucose tolerance test in five groups with normal glucose tolerance, comparing normal male controls to men with chronic stable angina, men with recent myocardial infarction (two groups, 3 weeks and 3 months post infarction), and men with chronic severe heart failure. Only patients with chronic heart failure had fasting hyperinsulina-emia, probably reflecting associated neuroendocrine abnormalities. Stimulated hyperinsulinaemia was present in all patient groups, but was less pronounced and of shorter duration in patients with angina. At 120 min, only patients with heart failure or previous myocardial infarction were hyperinsulinaemic. The degree of stimulated hyperinsulinaemia was not influenced by the presence of heart failure or by the length of time from infarction. Hyperinsulinaemia is associated with impaired peripheral muscle glucose uptake and metabolism, and might contribute to muscular fatigue on exertion in patients with previous myocardial infarction or heart failure.

Increased thyroid hormone concentrations have been reported to have disparate effects on insulin sensitivity in man. We describe a 72-year-old lady who initially presented with episodic hypoglycemia secondary to an insulinoma that was controlled by diazoxide. She re-presented 12 months later with a recurrence of the hypoglycemia following the development of thyrotoxicosis. The diazoxide treatment was maintained and propranolol was introduced, which prevented further episodes of hypoglycemia. This appeared to be due to a direct effect of propranolol on endogenous insulin secretion, while whole body insulin sensitivity remained unchanged as assessed using the hyperinsulinemic-euglycemic clamp technique. She was later rendered biochemically euthyroid with a combination of blocking carbimazole therapy and thyroxine replacement, and this was associated with a marked decrease in insulin sensitivity. Thus, the principal effect of thyroid hormone excess in this patient was an increase in insulin sensitivity that led to the clinical relapse of the insulinoma.

Plasma adrenaline-blood glucose interrelationships in insulin-induced hypoglycaemia in man have been studied using a sensitive double-isotope derivative method for adrenaline estimation. Plasma adrenaline reached a peak of 1.24 ng/ml at 45 minutes after insulin while blood glucose reached a nadir of 22 mg/100 ml at 30 minutes. There was a strong correlation both between the rise in adrenaline and the degree of hypoglycaemia and between the rise in adrenaline and the post-hypoglycaemic rise in glucose. Plasma noradrenaline rose from 0.29 to 0.59 ng/ml, the rise correlating with the rise in adrenaline. Changes in pulse rate preceded and were unrelated to changes in plasma catecholamines. Fuel mobilisation in response to adrenaline infusion (6 mug/min. for 20 min.) in normoglycaemic man was also studied. Plasma adrenaline concentration rose from a mean of 0.02 ng/ml to 0.71 ng/ml while plasma noradrenaline concentration was unchanged. Blood glucose rose from 71 to 98 mg/100 ml while plasma insulin decreased from 11 to 8 muU/ml. Blood lactate rose by 0.85 mM while pyruvate concentration remained unchanged. Blood glycerol concentration rose twofold and ketone body concentration threefold but there was little change in the concentrations of the glucogenic amino acids, alanine, glutamate and glutamine. Both the 3-hydroxybutyrate/acetoacetate ratio and the lactate/pyruvate ratio rose implying a more reduced intracellular state due presumably to increased hepatic fatty acid oxidation. It is concluded that adrenaline enhances the recycling of lactate and spares glucose through the mobilitsation of lipids but that amino acids are little affected.

In normal man the major oxidative fuels are glucose, with its byproduct lactate, fatty acids and the ketone bodies, acetoacetate and 3-hydroxybutyrate. In the fed state catabolism is suppressed and fuel utilization will depend upon the relative proportions of fat, carbohydrate and protein consumed. Amino acids, glucose and short-chain fatty acids may all be used. In the postabsorptive and fasting states there is a more consistent pattern of fuel production and utilization. Energy stores in the average man are heavily loaded in favour of fat. Glycogen in liver amounts to only 400 to 500 kilocalories, muscle protein can yield 24000 kilocalories, although mechanisms exist to prevent this, whilst triglyceride fat can supply 135 000 kilocalories (Cahill, 1976). After an overnight fast the liver releases glucose for use mainly by obligatory glucose-using tissues such as brain, red blood cells and renal medulla. Net splanchnic glucose release amounts to 0.86 mmol/min/1.73 m 2 (220 g or 860 kilocalories over a 24-hour period), approximately 80 per cent of which derives from glycogen, the rest coming from gluconeogenesis. At the same time ketone body production, which can only occur in the liver, is 0.06 mmol/min/1.73 m 2 (Owen et al, 1980) while total fatty acid turnover is approximately 0.5 mmol/min/1.73 m 2 (calculated from Hall, Saunders and Sonksen, 1979) about one-third of which is totally oxidized, and approximately 0.13 mmol/min/1.73 m2 is taken up by liver (Owen et al, 1981). After a three-day fast the situation is reversed with glucose production falling to 0:47 mmol/min/1.73 m s, now ~11 derived from gluconeogenesis, ketone body production rising to 0.75 mmol/min/1.73 m s and fatty acid turnover increasing by 60 to 70 per cent with hepatic fatty acid extraction doubled. By this time the brain is oxidizing ketone bodies, which allows glucose utilization to fall, thereby sparing amino acids and vital tissue protein. Thus in normal man, ketone bodies serve as a minor fuel during everyday life, but as soon as fasting supervenes they become all-important as oxidative fuels, and are usable by almost every tissue. The whole process

Combination of gel-entrapped urease with a pH electrode can be used to measure plasma urea, provided that the response is related to an auxiliary pH electrode. The response time is 1–3 min and electrodes can be used for up to one week. Good correlation was obtained for routine samples with a spectrophotometric diacetyl monoxime method (y = 0.952x + 0.224, r = 0.996).

The effects of colchicine, an inhibitor of microtubule polymerization, on the maintenance of steady state binding of insulin to isolated hepatocytes was studied. Colchicine (10−5M) produced a 35–45% decrease in binding in presence of insulin (10−8M) at 37°C. This decrease in binding was time and temperature dependent. The decrease was also dependent on the amount of insulin bound to the cell. The results suggest that colchicine may prevent the maintenance of steady state binding of insulin by impairing transfer of newly synthesized or recycled receptor from within the cell to the plasma membrane.

The metabolic effect of infusing 1.0–1.5 litre of Hartmann's solution to normal subjects and those with maturity onset diabetes has been studied during surgery. It was found that the use of Hartmann's solution after operation was associated with a 7.5 mmol litre'1 increase in plasma glucose concentration compared with an increase of 2.1 mmol litre−1 in diabetic patients who received no i.v. fluids. In non-diabetic subjects who received Hartmann's solution the increase was 2.5 mmol litre−1. It is suggested that Hartmann's solution may be metabolically disadvantageous in diabetic patients.

Twelve hour metabolic rhythms have been determined in two groups of subjects during combined therapy with a sulphonylurea and phenformin 50 mg twice daily. Subjects with clinical evidence of complications of diabetes showed greater abnormalities in concentrations of blood intermediary metabolites than a group of subjects without complications despite similar mean blood glucose concentrations in the two groups (7.6 mmol/l with complications; 7.3 mmol/l without complications). Mean blood lactate (1.93 mmol/l v 1.39 mmol/l), alanine (0.56 mmol/1 v 0.43 mmol/l), total blood ketone bodies (0.20 mmol/l v 0.14 mmol/l) and several other intermediary metabolites and their ratios were significantly higher in the group with diabetic complications. It is suggested that the differences between the two groups may arise from impaired disposal of phenformin leading to higher blood concentrations in the group with diabetic complications, despite normal liver function tests and plasma creatinine concentration. It is probable that this accumulation of phenformin results in more pronounced effect upon blood glucose and other intermediary metabolites. Thus, the metabolic abnormalities previously reported in patients treated by phenformin alone are also present during combined sulphonylurea and phenformin therapy, and in the presence of diabetic microangiopathy these abnormalities are accentuated.

Most forms of liver disease are probably associated with impaired gluconeogenesis, although hypoglycaemia is rarely an important clinical feature. Blood concentrations of the gluconeogenic precursors, lactate, glycerol and alanine are elevated although, in certain situations, alanine levels may be decreased. Abnormal glucose tolerance is present in both acute and chronic liver disease, but is usually not of clinical importance. The mechanism of glucose intolerance remains uncertain, with diminished hepatocyte mass, portal diversion and insulin resistance the major postulates. Indeed, the importance of the liver in disposing of an oral glucose load, is still questioned. Both hyperinsulinism and hypoinsulinism are found in liver disease, with hyperinsulinism common in cirrhosis and acute viral hepatitis. This is accompanied by insulin resistance. The hyperinsulinism is probably due to defective hepatic clearance of insulin rather that to over-production. The cause of the insulin resistance remains to be established. Glucagon levels are raised and may contribute to this resistance. Growth hormone levels are also increased but are associated with low somatomedin levels and the role of growth hormone in insulin resistance is therefore questionable. Future developments include use of new animal models, studies of biopsy specimens and studies of hepatic hormone receptors.

Methods are described for simultaneous continuous in vivo monitoring of blood glucose, lactate, alanine and 3-hydroxybutyrate. The methods use dialysis, immobilized enzymes and measure generated reduced pyridine nucleotides fluorimetrically. The methods are accurate, sensitive, stable, and are suitable for use in emergency situations or in clinical investigation.

In 15 patients with uncontrolled diabetes mellitus, red-cell 2,3-diphos-phoglycerate (2,3-D.P.G.) levels were decreased below the normal range of 4·5±0·5 to 2·2±0·4 mmoles per litre. There was a strong positive correlation of 2,3- D.P.G. with arterial-blood pH and there were negative correlations of 2,3-D.P.G. with blood lactate/ pyruvate (L/P) and total ketone bodies. The lowered 2,3-D.P.G. level balanced the effect of pH on the oxyhaemoglobin-dissociation curve, so that there would be little net effect on tissue-oxygen supply. The raised L/P ratios before treatment could be accounted for by the change in pH without need to postulate tissue anoxia. On treatment 2,3-D.P.G. took up to 5 days to return to normal. It is suggested that this was secondary to a prolonged fall in plasma-phosphate. Patients receiving intravenous bicarbonate showed a rapid correction in blood-pH, together with an acute fall in calculated P 50 (partial pressure at which there is 50% saturation of haemoglobin) and no change in L/P ratio, indicating tissue anoxia, resulting from the low red-cell 2,3-D.P.G. concentration. It is suggested that intravenous regimens for the treatment of uncontrolled diabetes should include phosphate and exclude bicarbonate whenever possible.

A new regimen is described for the treatment of diabetic " coma " using small intramuscular doses of insulin. In fourteen patients admitted with plasma-glucose and blood-ketone-body concentrations of 637 ± 60 S.E.M. mg. per 100 ml. and 12·1 ± 1·2 mM, respectively, a mean initial dose of 16 + 2 S.E.M. units insulin was given with hourly doses of 5 to 10 units thereafter. Cumulative insulin therapy amounted to 43 ± 4, 59 ± 5, and 98 ± 6 units at 6, 12, and 24 hours. Mean serum-insulin in patients who had not previously had insulin therapy was 33 ± 10 S.E.M. μU per ml. 1 hour after the first dose, rising to 93 ± 14 μU per ml. at 5 hours. Plasma-glucose, plasma-free-fatty-acids, and blood-glycerol all fell significantly within the first hour of treatment. Plasma-glucose fell at a regular rate of 90 mg. per 100 ml. per hour. The rate of fall was not influenced by blood-ketone-body concentration or previous insulin treatment, but it was significantly reduced by infection. The new regimen was simple to follow, clinically effective, and had advantages over conventional therapy with regard to changes in plasmapotassium, blood-lactate, and plasma-growth-hormone.

To determine whether the decline in glucose tolerance with age is related to changes in peripheral tissue insulin sensitivity, 11 healthy older (65-85 years) and 12 healthy younger (20-38 years) subjects, matched for physical activity and body mass index, were studied. The glucose-insulin infusion technique, in which glucose and insulin are infused at constant rates over 150 min, was used to assess peripheral tissue insulin sensitivity. The metabolic clearance rate (MCR) of glucose, which reflects peripheral tissue insulin sensitivity, is derived from the resulting steady-state blood glucose. Fasting blood glucose was similar in the two groups, but mean blood glucose 2 h after an oral glucose load was higher in the older group. MCR of glucose expressed per kilogram lean body mass was almost identical in both groups (13.3 +/- 1.3 vs. 13.5 +/- 1.3 ml/kg lean body mass/min, respectively). Evidence of decreased glucose tolerance with normal peripheral tissue insulin sensitivity suggests that changes in other factors may be more important than previously recognised.

The changes in peripheral blood levels of insulin and glucagon and in multiple indices of carbohydrate metabolism were compared in seven dogs, made diabetic with alloxan and streptozotocin, during insulin infusion by the portal and peripheral circulations. Each experiment was carried out in unrestrained fasted dogs after insulin had been withheld for 24 hr. Increasing the rate of insulin infusion from 0.006 to 0.05 U kg −1 h −1 for 2.5 hr caused a significantly greater reduction of peripheral blood glucose only at the infusion rate of 0.05 U kg −1 h −1 when the hormone was given peripherally than intraportally. Plasma levels of lactate, alanine and glycerol rose during the 2.5 hr control infusions of the vehicle while there was no change in the levels of 3-hydroxybutyrate. Peripheral infusion of insulin reduced the levels of glycerol and 3-hydroxybutyrate at all infusion rates but lactate and alanine levels were only reduced when the infusion rate was increased to 0.025 U kg −1 h −1 , a rate which produced measurable increases in peripheral plasma insulin. Portal infusions which did not elevate peripheral insulin levels caused a reduction in all four metabolites. The glucagon levels in the diabetic dogs were markedly decreased by insulin infusion of 0.012 U kg −1 h −1 by either route and the magnitude of this fall was similar at all higher infusion rates. Thus, the difference in responses to infusion of insulin by the portal and peripheral routes cannot be accounted for by the glucagon fall. We conclude that intraportal administration of insulin at rates of 0.006–0.05 U kg −1 h −1 reduces plasma levels of glucose and the metabolites studied by a direct action on the liver and would suggest that the hepatic action of peripherally administered insulin is manifest only when the infusion rate is increased to 0.025 U kg −1 h −1 .

The consecutive measurement of 3-hydroxybutyrate and lactate by enzymatic end-point and rapid kinetic methods is described. Using end-point analysis within batch coefficient of variation (C.V.) ranged from 0.5 to 6.5% depending on concentration, and between batch C.V. was 1.7 to 9.0%. In the rapid kinetic method C.V. ranged from 4.2 to 10.6%. Both assays showed absolute agreement with conventional spectrophotometric methods. The rapid kinetic method yields plasma values within 15 min of blood sampling and is suitable for use in emergency situations.

Thirty-eight patients in diabetic coma from four different centres were treated with a continuous low-dose intravenous infusion of insulin at an average dose of 7·2 IU/hr. All patients recovered rapidly except for one profoundly shocked patient who died. The mean fall in plasma glucose was 58% four hours after the start of insulin. Blood ketone bodies and plasma free fatty acids showed a similar response. There was no significant difference in plasma glucose response according to severity of acidosis or previous treatment with insulin. Hypokalaemia was uncommon. In the treatment of diabetic coma this technique has proved simple, safe, and effective.

Juvenile diabetic patients were studied 60-72 hours after insulin withdrawal when moderate ketoacidosis had developed. Somatostatin infusion for 4 hours in five patients resulted in almost complete suppression of plasma pancreatic glucagon and growth hormone, a fall in plasma-cyclic-adenosine-monophosphate (A.M.P.) concentrations, and a large fall in plasma-glucose concentration. After infusion plasma concentrations of these substances rose again. Blood-ketone-bodies, plasma-free-fatty-acids (F.F.A.), and plasma glycerol concentrations, however, did not decrease appreciably with somatostatin administration. In three patients 2 to 3 h somatostatin infusions were twice superimposed upon a continuous 9-5 h insulin infusion (1 unit/h). An insulin effect was noticeable within 30 minutes, with pronounced falls in the concentrations of plasma glucose, pancreatic glucagon, F.F.A., and blood-ketone-bodies. There was no significant change in these patterns when somatostatin was administered or withdrawn. These results do not indicate that somatostatin infusion would be useful in the treatment of manifest diabetic ketoacidosis.