Search

The pathogenesis of Limited Joint Mobility (LJM) in diabetes is unknown, but the abnormality is said to be associated with an increased incidence of microangiopathy. To examine the possibility that LJM may be a manifestation of microvascular disease, fibrinolysis and blood prostacyclin metabolites were measured in diabetic patients with and without LJM, and compared with patients who have systemic sclerosis (PSS). The concentrations of plasma fibrinogen (p less than 0.01) and Plasminogen Activator Activity (PAA) (p less than 0.02) were increased in both diabetic groups, compared with non-diabetic controls, and were of similar magnitude to the changes observed in PSS. Antithrombin III (AT III) activity was significantly lower in diabetic patients with LJM than in those without LJM, but not significantly different from controls. Prostacyclin concentrations were raised in PSS patients compared to controls (p less than 0.02) but were not raised in diabetic subjects. Thus abnormal fibrinolysis was demonstrated in diabetic patients compared to non-diabetic controls, and prostacyclin concentrations in diabetics were lower than in PSS patients. The diabetic patients with LJM had lower AT III activity than diabetics without LJM, but overall a convincing difference between the two groups of diabetics was not proven. No sure inferences could be drawn on the vascular aetiology of LJM, while the impaired prostacyclin activity might contribute to the development of diabetic microangiopathy.

Following acute insulin-induced hypoglycaemia, an abnormal pattern of insulin secretion in response to a meal has been demonstrated in six healthy volunteers. This is characterized by an initial impairment in insulin secretion and late hyperinsulinaemia. Postprandial gastrointestinal hormone levels were normal following hypoglycaemia in these subjects. In four other subjects, the administration of intravenous glucose prior to the meal partially reversed the abnormal pattern of secretion. In two patients with a total pre-ganglionic sympathectomy, the pattern of blood glucose, plasma insulin and C-peptide was similar to that observed following hypoglycaemia in normal subjects. It is unlikely that an abnormal entero-insular axis or an elevation of plasma catecholamine levels are primarily responsible for this phenomenon. This effect of hypoglycaemia on postprandial insulin secretion may be caused by glucopenia of the pancreatic beta cells.

A group of five cardiovascular tests of autonomic neural function was studied in 115 asymptomatic insulin-dependent diabetic patients and nine patients with symptoms of diabetic autonomic neuropathy. In the asymptomatic group, 34 patients had one or more abnormal test. R-R variation during deep breathing was most frequently abnormal, but reliance on this test alone to examine parasympathetic function would have missed 12 of the 34 abnormal patients. The two abnormal responses to isometric exercise in this group were probably false-positive results caused by poor patient effort. In the symptomatic patients, R-R variation during deep breathing was abnormal in all nine, the blood pressure response to standing was abnormal in seven, and the response to isometric exercise was abnormal in three. The blood pressure response to isometric exercise is uncomfortable, requires special equipment, and should probably be confined to the assessment of symptomatic patients in a specialist setting.

Six diabetic patients are described who sustained serious musculoskeletal injuries during insulin-induced hypoglycemia. The convulsions were associated with nocturnal hypoglycemia, superoptimal glycemic control, pregnancy, hypoglycemic unawareness, or errors in self-management.

We have compared left ventricular function in 40 insulin-dependent diabetic patients (30 male, 10 female) aged 16-30 years (mean 23 years) with duration of diabetes 3 months to 25 years (mean 10 years) with 19 healthy age and sex-matched controls using radionuclide ventriculography at rest, during cold-pressor stimulation, isometric exercise, and dynamic exercise. There were no differences in exercise or smoking habits in the two groups. All subjects had a normal resting electrocardiogram and a normal electrocardiographic response to dynamic exercise using the Bruce protocol on the Marquette Computer Assisted System of Exercise. Left ventricular ejection fraction was similar at rest and fell in diabetics and controls, respectively, by 2.3 +/- 1.0% and 3.0 +/- 0.08% (NS) and by 2.4 +/- 0.8% and 2.2 +/- 0.8% (NS) during isometric exercise, and cold pressor testing. During supine dynamic exercise on a bicycle ergometer ejection fraction rose in both groups by 13.9 +/- 1.7% and 13.7 +/- 1.6%, respectively (NS). No statistically significant differences were found in peak ventricular filling and emptying rates and abnormalities of regional wall motion did not develop. Two female patients with diabetes for more than 10 years had an abnormal response to dynamic exercise manifested by a fall in ejection fraction. In comparison to the higher frequency of abnormalities reported in diabetics using echocardiography and the measurement of systolic time intervals, abnormal left ventricular function assessed by radionuclide ventriculography is uncommon in young patients.

The cardiovascular effects of an intravenous injection of soluble insulin and of acute hypoglycaemia were examined in six normal male subjects using multiple-gated radionuclide ventriculography. The basal left ventricular ejection fraction rose significantly from 47 +/- 3% (mean +/- SEM) to 54 +/- 3% p less than 0.01, within 5 min of the intravenous injection of insulin, and before any significant changes occurred in the blood glucose concentration. The ejection fraction subsequently rose to a peak of 72 +/- 5% coinciding with the onset of the acute hypoglycaemic reaction. This corresponded to the nadir of blood glucose and was associated with rises in heart rate, stroke volume and cardiac output. The heart rate returned to the resting value within 30 min of the acute hypoglycaemic reaction, but the ejection fraction, stroke volume and cardiac output were still elevated 90 min later. The peak ejection fraction value immediately preceded the maximal increment of plasma catecholamines released in response to hypoglycaemia. Thus, administration of intravenous insulin had a small, immediate, discernible effect on the cardiovascular system. A subsequent rise in left ventricular ejection fraction of much greater magnitude was stimulated by the development of acute hypoglycaemia, and was associated temporally with sympatho-adrenal activation. The use of radionuclide ventriculography showed that the haemodynamic changes provoked by hypoglycaemia produced a sustained effect on cardiac contractility.

Recovery from acute hypoglycaemia induced by the injection of insulin has been examined in six human subjects under control conditions, under non-selective beta blockade (propranolol) and under selective beta 1 blockade (metoprolol). The normal blood glucose recovery was biphasic with an initial rapid and a slower subsequent phase of recovery. The early recovery mechanism was unaffected by either form of beta blockade, but with propranolol the late phase of recovery was significantly prolonged. Rises in blood lactate and plasma free fatty acids following hypoglycaemia were markedly reduced by propranolol but to a much lesser degree with metoprolol. The counterregulatory hormonal responses of glucagon, cortisol and growth hormone were augmented appropriately for the prolonged hypoglycaemia associated with propranolol. Non-selective beta adrenergic blockade with propranolol is associated with an impairment of the late phase of blood glucose recovery from hypoglycaemia. The possible mechanisms of this impairment are discussed.

A 48-year-old male developed fever and sore throat while in Spain and was admitted to hospital, dehydrated, ketotic, pyrexial, and with a blood glucose of 35 mmol/l. Despite treatment with intravenous fluids, insulin, cephalosporin, potassium and bicarbonate he returned to Britain 7 days later, underhydrated and acidotic, though euglycaemic. His face was discoloured, there was painless nasal and maxillary swelling, oral candidiasis, and he went on to develop ophthalmoplegia and sudden blindness. Staphylococcus albus and mycelial fungus were isolated, the latter was identified as Mucor hiemalis, but despite treatment with amphotericin B co-ordinated with radical maxillary-facial surgery he died 31 days after the initial symptoms. Rhinocerebral mucormycosis is a recognized complication of ketoacidosis which is rarely reported in the UK. The organism invades arteries aggressively, therefore radical therapy must be started early to prevent the high mortality.

The effect of lipohypertrophy at injection sites on insulin absorption has been studied in 12 insulin-dependent diabetic patients. The clearance of 125I-insulin from sites with lipohypertrophy was significantly slower than from complementary nonhypertrophied sites (% clearance in 3 h, 43.8 +/- 3.5 +/- SEM) control; 35.3 +/- 3.9 lipohypertrophy, P less than 0.05). The degree of the effect was variable but sufficient in several patients to be of clinical importance. Injection-site lipohypertrophy is another factor that modifies the absorption of subcutaneously injected insulin.

Plasma glutathione S-transferase basic isoenzyme (GST B1) concentrations have been measured by specific radioimmunoassay in Type 1 diabetic patients and in normal subjects, before and after controlled insulin-induced hypoglycaemia, and in a further group of Type 1 diabetic patients in hypoglycaemic coma. The activities of alanine aminotransferase (ALT), aspartate amino-transferase (AST), and gamma-glutamyl transferase (gamma GT) were also measured. GST B1 concentrations were significantly increased 3 h after controlled insulin-induced hypoglycaemia, both in the diabetic patients (p less than 0.02) and in the normal group (p less than 0.05), but the magnitude of the rise did not differ between these two groups. Four of the 9 patients presenting in hypoglycaemic coma had a GST B1 concentration above the reference range. ALT, AST, and gamma GT activities did not rise following hypoglycaemia in any of the groups.

In normal humans, arginine vasopressin and oxytocin are released acutely from the posterior pituitary gland in response to hypoglycemia, and their release may assist counterregulation. The responses of these hormones to insulin-induced hypoglycemia were measured in 16 insulin-dependent diabetic patients with no autonomic neuropathy (8 patients who had been diabetic less than 5 yr and 8 patients who had been diabetic greater than 15 yr) and in 6 normal subjects. The time of the onset of hypoglycemia and the mean blood glucose nadirs were similar in all groups, but the blood glucose recovery was delayed in the diabetic patients. In the normal subjects plasma arginine vasopressin rose from a mean basal value of 0.4 +/- 0.2 (+/- SE) pmol/L to a maximum of 1.3 +/- 0.6 pmol/L, and plasma oxytocin rose from 0.7 +/- 0.1 pmol/L to a maximum of 1.2 +/- 0.2 pmol/L 30 min after the onset of hypoglycemia. The plasma arginine vasopressin and oxytocin concentrations after hypoglycemia were significantly higher in both of the diabetic groups compared with those in the normal group. Arginine vasopressin and oxytocin rose in all control subjects after hypoglycemia. The individual hormonal profiles in the diabetic patients were variable, with an exaggerated rise of oxytocin in some diabetic patients and no rise in others. The arginine vasopressin responses were exaggerated in all of the diabetic patients. There was no correlation between the hormonal responses and the duration of diabetes. The exaggerated plasma arginine vasopressin and oxytocin responses to hypoglycemia in diabetic patients may indicate the failure of a normal inhibitory mechanism which modulates hormonal secretion or a compensatory response to impaired glucose recovery.

Fasting serum trypsin concentration and pancreatic trypsin output, stimulated by secretin and cholecystokinin-pancreozymin, were measured in 18 patients with insulin-dependent diabetes, to assess a possible correlation between these two indices of exocrine pancreatic function. Serum trypsin concentration was subnormal in 13, and pancreatic trypsin output was decreased in 14 patients, but there was no significant correlation between the two measurements. There was no correlation between serum trypsin and residual beta cell function measured by plasma C-peptide immunoreactivity (CPR). After an interval of four years serum trypsin measurements were repeated in 11 subjects. All individual trypsin levels were lower than the previous results, the difference being highly significant (t = 9.0; p < 0.001). Serum trypsin concentration therefore represents a qualitative index of reduced exocrine pancreatic function in diabetes, but has no value in quantitating the degree of deficiency.

In 34 patients suspected of having pancreatic cancer glucose homeostasis and insulin secretion in response to intravenous glucagon (1 mg) were studied prospectively. Twelve patients had pancreatic cancer at laparotomy and the remainder were designated a control group. Responses were also measured in 12 healthy volunteers. There was no significant difference in the rise of blood glucose between the groups after glucagon. The mean plasma insulin concentrations rose rapidly in both groups after glucagon but were significantly lower in the pancreatic cancer group. In patients with obstructive jaundice the plasma insulin response was a better discriminator of pancreatic cancer. Abnormal pancreatic beta-cell function is detectable in patients with pancreatic carcinoma before any change in glucose homeostasis, particularly in patients with obstructive jaundice. The glucagon stimulation test may be useful in the earlier diagnosis of pancreatic carcinoma before the development of clinically overt diabetes mellitus.

To examine the associations between cigarette smoking, connective tissue changes, and diabetic retinopathy, a detailed smoking history was elicited from 150 normotensive non-diabetic subjects, and from 266 randomly selected adult patients with Type 1 diabetes, after examination for limited joint mobility, Dupuytren's contracture, and diabetic retinopathy. Mean insulin dose and current glycosylated haemoglobin concentrations were comparable in diabetic smokers and non-smokers. The historical duration of smoking correlated with the duration of diabetes (r = 0.72, p less than 0.001). In diabetic patients limited joint mobility was positively associated with retinopathy, being found in 73/147 (50%) patients with retinopathy compared with 20/114 (18%) without retinopathy (chi 2 = 28.9, p less than 0.001), and also with Dupuytren's contracture, 19/34 (56%) of patients with limited joint mobility having Dupuytren's contracture, compared with 76/232 (33%) of patients without Dupuytren's contracture (chi 2 = 7.05, p less than 0.01). Limited joint mobility was observed in 50% of diabetic smokers compared with 25% of non-smokers (odds ratio = 2.87 (corrected for diabetes duration), 95% confidence interval 1.64-5.01). Diabetic retinopathy was weakly associated with smoking (odds ratio 1.09; 95% confidence interval 0.60-1.96). There was however an increased prevalence of background retinopathy among male smokers (50% vs 29%; chi 2 = 6.88, p less than 0.01). In non-diabetic males limited joint mobility was observed in 37% of smokers but only in 11% of non-smokers (NS), while 33% of smokers and 8% of non-smokers had Dupuytren's contracture (p = 0.012). These results suggest that cigarette smoking contributes to the development of extra-articular connective tissue changes in both diabetic patients and non-diabetic subjects, and possibly to the development of diabetic retinopathy.

Convulsions may occur as a consequence of insulin-induced hypoglycaemia. We report three patients with insulin-dependent diabetes, who presented with generalized tonic-clonic seizures associated with nocturnal hypoglycaemia. None of the patients had experienced hypoglycaemia during waking hours and the convulsions were mistakenly diagnosed as idiopathic epilepsy. Recognition of the possible hypoglycaemia aetiology of these convulsions permitted appropriate alteration of the insulin regimens with no recurrence of convulsions. In one case, the seizure was associated with bilateral fractures of the neck of the humerus. Unrecognized hypoglycaemia should be considered as a possible cause of convulsions in insulin-dependent diabetic patients.