GLP‐1 receptor activated insulin secretion from pancreatic β‐cells: mechanism and glucose dependence

The major goal in the treatment of type 2 diabetes mellitus is to control the hyperglycaemia characteristic of the disease. However, treatment with common therapies such as insulin or insulinotrophic sulphonylureas (SU), while effective in reducing hyperglycaemia, may impose a greater risk of hypoglycaemia, as neither therapy is self‐regulated by ambient blood glucose concentrations. Hypoglycaemia has been associated with adverse physical and psychological outcomes and may contribute to negative cardiovascular events; hence minimization of hypoglycaemia risk is clinically advantageous. Stimulation of insulin secretion from pancreatic β‐cells by glucagon‐like peptide 1 receptor (GLP‐1R) agonists is known to be glucose‐dependent. GLP‐1R agonists potentiate glucose‐stimulated insulin secretion and have little or no activity on insulin secretion in the absence of elevated blood glucose concentrations. This ‘glucose‐regulated’ activity of GLP‐1R agonists makes them useful and potentially safer therapeutics for overall glucose control compared to non‐regulated therapies; hyperglycaemia can be reduced with minimal hypoglycaemia. While the inherent mechanism of action of GLP‐1R agonists mediates their glucose dependence, studies in rats suggest that SUs may uncouple this dependence. This hypothesis is supported by clinical studies showing that the majority of events of hypoglycaemia in patients treated with GLP‐1R agonists occur in patients treated with a concomitant SU. This review aims to discuss the current understanding of the mechanisms by which GLP‐1R signalling promotes insulin secretion from pancreatic β‐cells via a glucose‐dependent process.     

Reducing hypoglycaemia with insulin analogues

Numerous studies have demonstrated that tight glycaemic control reduces long-term complications in both Type 1 and Type 2 diabetes. However, current intensive insulin regimens increase the risk of hypoglycaemia, dissuading many attempting this approach. Normal physiological insulin profiles consist of a stable, basal component and meal-related surges in secretion. Conventional insulin regimens cannot mimic this profile accurately due to pharmacokinetic limitations. Human insulin has a slow onset of action, thus patients are advised to inject about 30 min before a meal, ensuring peak insulin concentrations coincide with postprandial glucose excursions. This is clearly impractical for many and can lead to pre-meal hypoglycaemia if the meal is delayed. Furthermore, it only partially overcomes the unphysiological insulin profile and patients experience postprandial hyperglycaemia and are vulnerable to postabsorptive hypoglycaemia. Insulin aspart and insulin lispro are rapid-acting analogues that allow a more physiological replacement of mealtime insulin secretion. They reduce postprandial glucose and usefully reduce the incidence of hypoglycaemia when used in a basal-bolus regimen in tightly controlled patients. Pre-mixed insulins, containing a combination of rapid-acting and intermediate-acting insulin, are widely used, particularly in Type 2 diabetes. They have limitations achieving tight glucose targets but early data suggest that the combination of 30% insulin aspart and 70% protaminated insulin aspart may also reduce severe hypoglycaemia. On-going clinical trials will establish whether they can be used to achieve better glycaemic control with less hypoglycaemia.     

The importance of plasma free insulin and counterregulatory hormones for the recovery of blood glucose following hypoglycaemia in type 1 diabetics

After induction of hypoglycaemia in 31 Type 1 (insulin-dependent) patients, the 10 patients with the slowest recovery of blood glucose from hypoglycaemia were arbitrarily compared with the 10 patients with the fastest recovery of blood glucose. No differences were found between the two groups regarding response of glucagon to hypoglycaemia, whereas the epinephrine (2-fold), norepinephrine (2.4-fold) and cortisol responses were significantly greater in the group with the slow recovery. The plasma free insulin concentrations were higher (2-fold) in the group with slow recovery from 30 min after stop of insulin and throughout the study. This may be explained by a 3-fold greater amount of insulin binding antibodies in this group compared to the group with fast recovery from hypoglycaemia. An inverse significant correlation was demonstrated between the rates of recovery and the amounts of insulin binding antibodies in all the patients (P less than 0.02). This implicates that enhanced counterregulatory hormone responses in the group with the slow recovery from hypoglycaemia could not compensate for the hypoglycaemic effect of a concomitant higher plasma free insulin concentration. Insulin binding antibodies, acting as a depot of circulating insulin, may be a risk factor of prolonged hypoglycaemia in Type 1 diabetics.     

Effect of pancreas transplantation on glucose counterregulation in insulin-dependent diabetic patients prone to severe hypoglycaemia.

Pancreatic transplantation was performed in three patients with insulin-dependent diabetes mellitus in whom recurrent and severe episodes of hypoglycaemia had been found to be due to defective glucose counterregulation. Thus in these patients the spontaneous blood glucose recovery after insulin-induced hypoglycaemia (0.1 U kg-1 h-1 i.v. insulin until blood glucose levels fell below 2.8 mmol l-1) was delayed, and the responses of glucagon, epinephrine and growth hormone (GH) were absent or diminished. After pancreas transplantation the patients exhibited essentially normal blood glucose control. When the insulin infusion test was repeated 3 months after the transplantation, the blood glucose level recovered rapidly after insulin withdrawal. The glucagon response was restored, and the responses of epinephrine and GH were improved. Plasma C-peptide was suppressed by approximately 50%, which is less than is observed in normal subjects. It is concluded that glucose counterregulation improves after pancreas transplantation. This appears to be mainly due to an improvement in the hypoglycaemia-induced glucagon response, but an amelioration of sympatho-adrenal and hypothalamic-pituitary regulatory mechanisms may also be involved. The apparent failure to suppress completely the insulin release from the denervated pancreas transplant indicates that inhibition of beta-cell secretion during insulin-induced hypoglycaemia may be partly under neural control.     

Human insulin use and hypoglycaemia: insights from the Pittsburgh Epidemiology of Diabetes Complications Study.

Recently, concern has arisen that human (as opposed to beef or pork) insulin may cause more frequent and/or severe hypoglycaemia in association with reduced warning symptoms. This question was examined from questionnaire data of 628 Type 1 diabetic patients (mean age 28 years and duration of diabetes 19 years) participating in the baseline examination of a follow-up study of diabetes complications. Those using human insulin (n = 73) reported an insignificantly higher frequency of hypoglycaemic reactions in the last year than those using animal insulin (66 vs 55% with reactions at least monthly) and only a weak trend was seen overall for the prevalence of human insulin use to increase with increasing frequency of hypoglycaemia (p = 0.06). Hypoglycaemic reactions resulting in unconsciousness were too rare to permit analysis by type of insulin used. The prevalence of reduced awareness of hypoglycaemia was similar among human insulin users to that seen in animal insulin users (25 vs 19%, NS). However, prevalence of reduced awareness showed a strong relationship to current blood glucose in the animal (r = -0.18, p less than 0.001) but not human (r = -0.06, NS) insulin users. Excluding patients with autonomic symptoms or neuropathy did not alter the results, nor did excluding the 34 individuals taking more than three insulin injections per day. It is concluded that human insulin use is not associated with either any substantial increased frequency of hypoglycaemia or reduction in awareness of hypoglycaemia. However, human insulin use does appear to be associated with reduced awareness of hypoglycaemia in those whose blood glucose control is relatively poor.     

Risk of hypoglycaemia with oral antidiabetic agents in patients with Type 2 diabetes.

In patients with Type 2 diabetes, the appropriate intensity of glucose control is determined by age, life expectancy, and the presence of concomitant disease. Geriatric patients are especially susceptible to hypoglycaemia and therefore particular care should be taken in this group characterized by polypharmacy, renal or hepatic dysfunction, cardiovascular multimorbidity and malnutrition. As hypoglycaemia is a significant cause of morbidity and mortality, treatment regimens for diabetes should minimize the occurrence of hypoglycaemic episodes and be tailored to the patient's individual needs. The pharmacological options for treating Type 2 diabetes have increased considerably and the risk of hypoglycaemia of the currently available drugs varies considerably. Metformin, thiazolidinediones, and acarbose, oral antidiabetic drugs that decrease insulin resistance or postprandial glucose absorption, are associated with a low risk of hypoglycaemia. These drugs can also be used effectively in various combination regimens; however, by improving insulin sensitivity, combinations of metformin and thiolidinediones with sulphonylureas or meglitinides may considerably increase the risk of hypoglycaemia. On account of its complex pharmacoprofile glibenclamide is a problematic substance carrying a high risk of hypoglycaemia. There are limited preliminary data indicating that, under routine conditions, glimepiride may be associated with a lower risk of hypoglycaemia than glibenclamide and is no more likely to cause hypoglycaemia than other shorter-acting agents such as gliclazide and glipizide. Nateglinide and repaglinide as short-acting insulin secretagogues may be associated with a reduced risk of hypoglycaemia compared with glibenclamide, in particular when dosed flexibly. Repaglinide might be beneficial in individuals with renal impairment.     

Insulin response to carbohydrate ingestion after gastric surgery with special reference to hypoglycaemia

Factors responsible for spontaneous hypoglycaemia were investigated in 23 gastrectomy patients and two patients with vagotomy and pyloroplasty. Plasma insulin and capillary blood glucose levels were followed after giving 175 ml of 50% glucose orally. In gastrectomy patients a significant correlation was found between the height of the peak blood glucose and insulin levels for the same individual. Patients with high peak glucose and insulin levels were significantly more likely to develop hypoglycaemia later in the test. These findings are compatible with the suggestion that the major factor predisposing to hypoglycaemia is a faster than average rate of emptying of the gastric remnant, but this does not explain all the results. In two cases, hypoglycaemia followed an abnormally large insulin response to oral glucose. The results of insulin injection tests in 14 patients do not exclude the further possibility that in some cases hypoglycaemia is due to excessive insulin sensitivity.     

Hypoglycaemia and cardiac arrhythmias in patients with type 2 diabetes mellitus.

Improved blood glucose control by insulin treatment in patients with Type 2 (non-insulin dependent) diabetes mellitus increases the risk for hypoglycaemic episodes. Our objective was to investigate if hypoglycaemia causes electrocardiographic changes and cardiac arrhythmias in patients with Type 2 diabetes. Six insulin-treated patients with Type 2 diabetes and no known cardiac disease took part in the study. Hypoglycaemia was induced by insulin infusion aiming at a plasma glucose less than or equal to 2.0 mmol l-1 or hypoglycaemic symptoms. All patients experienced hypoglycaemic symptoms. The median lowest arterial plasma glucose was 2.0 mmol l-1. Arterial plasma adrenaline concentration increased from 0.4 +/- 0.1 (mean +/- SE) to 6.9 +/- 0.3 nmol l-1 (p less than 0.001) while serum potassium was lowered from 4.1 +/- 0.3 mmol l-1 to 3.5 +/- 0.2 mmol l-1 (p less than 0.001). The heart rate increased significantly during hypoglycaemia except in one patient who developed hypoglycaemic symptoms and a severe bradyarrhythmia at a plasma glucose of 4.4 mmol l-1. One patient developed frequent ventricular ectopic beats during hypoglycaemia while four patients showed no arrhythmia. ST-depression in ECG leads V2 and V6 was observed during hypoglycaemia in five patients (p less than 0.05) and four patients developed flattening of the T-wave. In conclusion, the study supports the hypothesis that hypoglycaemia in patients with Type 2 diabetes may be hazardous by causing cardiac arrhythmias.     

Hypoglycaemia: the limiting factor in the glycaemic management of Type I and Type II diabetes

Hypoglycaemia is the limiting factor in the glycaemic management of diabetes. Iatrogenic hypoglycaemia is typically the result of the interplay of insulin excess and compromised glucose counterregulation in Type I (insulin-dependent) diabetes mellitus. Insulin concentrations do not decrease and glucagon and epinephrine concentrations do not increase normally as glucose concentrations decrease. The concept of hypoglycaemia-associated autonomic failure (HAAF) in Type I diabetes posits that recent antecedent iatrogenic hypoglycaemia causes both defective glucose counterregulation (by reducing the epinephrine response in the setting of an absent glucagon response) and hypoglycaemia unawareness (by reducing the autonomic and the resulting neurogenic symptom responses). Perhaps the most compelling support for HAAF is the finding that as little as 2 to 3 weeks of scrupulous avoidance of hypoglycaemia reverses hypoglycaemia unawareness and improves the reduced epinephrine component of defective glucose counterregulation in most affected patients. The mediator and mechanism of HAAF are not known but are under active investigation. The glucagon response to hypoglycaemia is also reduced in patients approaching the insulin deficient end of the spectrum of Type II (non-insulin-dependent) diabetes mellitus, and glycaemic thresholds for autonomic (including epinephrine) and symptomatic responses to hypoglycaemia are shifted to lower plasma glucose concentrations after hypoglycaemia in Type II diabetes. Thus, patients with advanced Type II diabetes are also at risk for HAAF. While it is possible to minimise the risk of hypoglycaemia by reducing risks -- including a 2 to 3 week period of scrupulous avoidance of hypoglycaemia in patients with hypoglycaemia unawareness -- methods that provide glucose-regulated insulin replacement or secretion are needed to eliminate hypoglycaemia and maintain euglycaemia over a lifetime of diabetes.     

Review of prandial glucose regulation with repaglinide: a solution to the problem of hypoglycaemia in the treatment of Type 2 diabetes?

Type 2 diabetes mellitus is characterised by abnormal beta-cell function (present at the time of diagnosis) that is often associated with insulin resistance. An important and consistent pathophysiological finding is the failure to produce adequate increments in insulin secretion in response to carbohydrate intake. Therefore, insulin secretagogue therapy, particularly when focused on prandial glucose regulation, is a logical approach to treatment because it addresses one of the most fundamental pathophysiological aspects of the disease. However, the traditional secretagogues-the sulphonylureas--have long been associated with the unwanted effect of hypoglycaemia. This is particularly likely to occur when drugs with lengthy plasma half-lives, prolonged drug-receptor interactions, active metabolites or a reliance on renal clearance are used. The problem is most prevalent in elderly patients, where sulphonylurea-induced hypoglycaemia may be related to failure to comply with strict mealtimes or the need for supplementary food intake, often in the context of compromised renal function. Data from large-scale outcome studies demonstrate that when tight glycaemic control is achieved through aggressive antidiabetic therapy, late diabetic complications can be significantly reduced. However, the pursuit of stricter HbA1c targets with more aggressive interventions may increase the risk of hypoglycaemia. This is an irony because the clinical need to avoid hypoglycaemia and patients' apprehension of it present barriers to the achievement of beneficial glycaemic targets. However, an increased risk of hypoglycaemia may not be inevitable with insulin secretagogue therapy. The recently introduced carbamoylmethyl benzoic acid derivative, repaglinide, has pharmacological properties that are well suited to its intended role as a prandial glucose regulator. When taken prior to main meals, the rapid onset and relatively short duration of action of repaglinide aid disposal of the mealtime glucose load, without continued stimulation of pancreatic beta-cells in the postprandial fasting period. Repaglinide is also characterised by hepatic metabolism and elimination, which is an advantage in the context of impaired renal function. Prandial glucose regulation with repaglinide selectively increases insulin secretion, and hence limits glucose excursions, in the prandial phase. If a meal is omitted, so too is the corresponding dose. This more flexible approach to the management of Type 2 diabetes has a number of advantages when compared with the fixed daily dosing regimens of sulphonylureas, among them a reduced risk of hypoglycaemia--a benefit that is particularly marked in the context of missed or irregular meals.     

Hypoglycaemia in Type 2 diabetes

The primary cause of hypoglycaemia in Type 2 diabetes is diabetes medication—in particular, those which raise insulin levels independently of blood glucose, such as sulphonylureas (SUs) and exogenous insulin. The risk of hypoglycaemia is increased in older patients, those with longer diabetes duration, lesser insulin reserve and perhaps in the drive for strict glycaemic control. Differing definitions, data collection methods, drug type/regimen and patient populations make comparing rates of hypoglycaemia difficult. It is clear that patients taking insulin have the highest rates of self‐reported severe hypoglycaemia (25% in patients who have been taking insulin for > 5 years). SUs are associated with significantly lower rates of severe hypoglycaemia. However, large numbers of patients take SUs in the UK, and it is estimated that each year > 5000 patients will experience a severe event caused by their SU therapy which will require emergency intervention. Hypoglycaemia has substantial clinical impact, in terms of mortality, morbidity and quality of life. The cost implications of severe episodes—both direct hospital costs and indirect costs—are considerable: it is estimated that each hospital admission for severe hypoglycaemia costs around £1000. Hypoglycaemia and fear of hypoglycaemia limit the ability of current diabetes medications to achieve and maintain optimal levels of glycaemic control. Newer therapies, which focus on the incretin axis, may carry a lower risk of hypoglycaemia. Their use, and more prudent use of older therapies with low risk of hypoglycaemia, may help patients achieve improved glucose control for longer, and reduce the risk of diabetic complications.     

Nocturnal hypoglycaemia in type 1 diabetes--consequences and assessment

Hypoglycaemia is inevitable when striving for low HbA1c values. Nocturnal hypoglycaemia often occurs without symptoms, but results in diminished next day well-being and hypoglycaemia unawareness. Frequency of nocturnal hypoglycaemia was first assessed in research ward settings, but suffered from insufficient glucose sampling frequency. This may have resulted in overestimation of the duration of hypoglycaemic episodes. The advent of the first continuous glucose sensor, the needle-type MedtronicMiniMed Continuous Glucose Measurement System, revolutionized the assessment of glucose values. However, on scrutiny, the first version of this sensor showed a drift into the hypoglycaemic area and delayed recovery from hypoglycaemia. Using the microdialysis-based GlucoDay system, our group reported a lower frequency of nocturnal hypoglycaemia in type 1 diabetes patients using an insulin pump, than that expected from the existing literature. Today, more than 80 years after the introduction of insulin for the treatment of type 1 diabetes, the associated frequency of nocturnal hypoglycaemia still awaits its definitive assessment.     

Comparing hormonal and symptomatic responses to experimental hypoglycaemia in insulin‐ and sulphonylurea‐treated Type 2 diabetes

Aims Patients with diabetes rely on symptoms to identify hypoglycaemia. Previous data suggest patients with Type 2 diabetes develop greater symptomatic and hormonal responses to hypoglycaemia at higher glucose concentrations than non‐diabetic controls and these responses are lowered by insulin treatment. It is unclear if this is as a result of insulin therapy itself or improved glucose control. We compared physiological responses to hypoglycaemia in patients with Type 2 diabetes patients treated with sulphonylureas (SUs) or insulin (INS) with non‐diabetic controls (CON). Methods Stepped hyperinsulinaemic hypoglycaemic clamps were performed on 20 subjects with Type 2 diabetes, 10 SU‐treated and 10 treated with twice‐daily premixed insulin, and 10 age‐ and weight‐matched non‐diabetic controls. Diabetic subjects were matched for diabetes duration, glycated haemoglobin (HbA_1c) and hypoglycaemia experience. We measured symptoms, counterregulatory hormones and cognitive function at glucose plateaux of 5, 4, 3.5, 3 and 2.5 mmol/l. Results Symptomatic responses to hypoglycaemia occurred at higher blood glucose concentrations in SU‐treated than INS‐treated patients [3.5 (0.4) vs. 2.6 (0.5) mmol/l SU vs. INS; P = 0.001] or controls [SU vs. CON 3.5 (0.4) vs. 3.0 (0.6) mmol/l; P = 0.05]. They also had a greater increase in symptom scores at hypoglycaemia [13.6 (11.3) vs. 3.6 (6.1) vs. 5.1 (4.3) SU vs. INS vs. CON; P = 0.017]. There were no significant differences in counterregulatory hormone responses or impairment of cognitive function among groups. Conclusions Sulphonylurea‐treated subjects are more symptomatic of hypoglycaemia at a higher glucose level than insulin‐treated subjects. This may protect them from severe hypoglycaemia but hinder attainment of glycaemic goals.     

Insulin detemir: improving the predictability of glycaemic control

The importance of strict glycaemic control in reducing diabetes-related outcomes is well recognised. Yet, despite the range of available treatment options, clinical experience suggests that individuals with diabetes frequently fail to achieve good glucose control. Instead, patients often have poorly controlled, unpredictable blood glucose levels. In insulin-treated patients, one of the reasons for this may be the inherently variable action of exogenously injected insulin. For example, the currently available longer-lasting insulin preparations that aim to provide a continuous basal level of circulating insulin are associated with markedly variable action both between and within subjects. Unpredictable insulin action contributes to variable blood glucose control, which in addition to increasing the risk of hypoglycaemia has also been shown to be an independent risk factor for mortality. It is hoped that advances in longer-lasting insulin preparations will provide significant benefits to patients, improving control and reducing variability. A new, long-lasting analogue, insulin detemir, has been shown to be significantly less variable than other basal insulin preparations in pharmacological and clinical studies. The improved predictability in insulin response offered by insulin detemir may be associated with a number of clinical benefits including a reduction in hypoglycaemia and weight gain.     

Plasma proinsulin,C-peptide and insulin in diagnostic suppression tests for insulinomas

Summary The value of plasma insulin, human C-peptide and proinsulin estimation in the diagnosis of 15 insulinomas has been investigated. Measurement of plasma proinsulin in an overnight fasting sample diagnosed all the insulinomas studied, irrespective of the plasma glucose. Patients with insulinomas had plasma proinsulin in the range 0.04–4.2 pmol/l and normal values were less than 0.01 pmol/ml. If hypoglycaemia was present, an inappropriately raised plasma immunoreactive insulin (including proinsulin) was diagnostic, but this assay was of little assistance if the plasma glucose was normal. Hypoglycaemia was induced with fish insulin in twelve patients with insulinomas and eight normal subjects. Using an antiserum which did not detect fish insulin, but cross-reacted with human proinsulin, the endogenous immunoreactive insulin was suppressed in the normal subjects, but all insulinoma patients had impaired suppression. Assay of plasma human C-peptide, or of the combined immunoreactive C-peptide and proinsulin, discriminated less well and did not clearly diagnose three insulinomas which secreted proinsulin rather than insulin and C-peptide. Plasma human proinsulin values during induced hypoglycaemia gave excellent discrimination and should detect insulinomas irrespective of their degree of histological differentiation. The assay of plasma human proinsulin allows a suppression test to be performed with hypoglycaemia induced by any type of insulin. A raised plasma proinsulin in proportion to C-peptide suggests an undifferentiated insulinoma, which may be more likely to be malignant.     

Prevention and treatment of hypoglycaemia unawareness in type 1 diabetes mellitus

Unawareness of hypoglycaemia (reduced ability/failure to recognize hypoglycaemia symptoms at the physiological threshold of 3.0 mmol/l) occurs frequently in type 1 diabetes mellitus, and patients are then at risk for severe hypoglycaemia. Unawareness of hypoglycaemia is the result of earlier frequent episodes of hypoglycaemia (iatrogenic). Likewise, a history of hypoglycaemia induces unawareness, while meticulous prevention of hypoglycaemia can reverse hypoglycaemia unawareness. Therefore, it is essential that insulin therapy regimens for type 1 diabetes mellitus be designed not only to maintain near-normoglycaemia, but also to minimize hypoglycaemia. Such a goal is feasible as long as (1) a rational plan of insulin therapy is adopted, including appropriate use of the short-acting insulin analogue lispro, (2) blood glucose is properly monitored, (3) blood glucose targets are individualized, and (4) education programs are widely implemented. Received: 8 September 1998 / Accepted: 14 September 1998     

Different awareness of hypoglycaemia induced by human or purified pork insulin in type I diabetic patients.

Recently, there have been reports on a diminished awareness of hypoglycaemia after a switch from purified pork insulin (PPI) to human insulin (HI) in insulin-dependent diabetes mellitus (IDDM). To clarify this phenomenon we investigated nine IDDM patients without signs of autonomic neuropathy. After an overnight euglycaemic clamp, blood glucose was lowered to hypoglycaemic levels by means of an artificial pancreas (Biostator) on 2 days. Insulin was used in a double-blind, randomized, cross-over fashion, either PPI or HI. The symptomatology and the hormonal counterregulation of developing hypoglycaemia was recorded. Venous concentrations of free insulin, cortisol, glucagon, growth hormone and the prevailing blood glucose were similar under both insulins. Eight out of nine IDDM patients felt more symptoms and at a higher blood glucose concentration under PPI than under HI. The first symptom of developing hypoglycaemia appeared at a mean blood glucose concentration of 61.1 +/- 5.4 mg.dl-1 under PPI and 44.4 +/- 5.3 mg.dl-1 under HI respectively (P less than 0.05). We conclude that HI may cause symptoms of hypoglycaemia to appear later and with a lesser number in comparison to PPI.     

Anti-insulin antibodies may cause hypoglycaemia following pancreas transplantation

Some pancreas transplantation (PTX) recipients experience hypoglycaemia after transplant. We studied the incidence and pattern of hypoglycaemic symptoms following PTX and later studied the glucose and pancreatic hormone response to Sustacal in patients with and without hypoglycaemia following PTX. In a survey of 54 PTX recipients who were 2 to 33 months post-transplant, 29 reported symptoms of hypoglycaemia at any time following their transplant. Of 12 patients who tried to document their blood glucose during any episode, 5 had a blood glucose less than 3.3 mmol/l, and another 5 documented a blood glucose from between 3.33 and 3.88 mmol/l. The patients reporting symptoms were less likely to be insulin resistant with longer time post-PTX, lower body mass index (BMI), and on lower doses of prednisone. Later, 12 patients with established, repeated episodes of hypoglycaemia following PTX were case-matched to PTX recipients without hypoglycaemic symptoms. Fasting glucose, free and total immunoreactive insulin (IRI), C-peptide, proinsulin and glucagon were measured following a 4 h Sustacal challenge. The hypoglycaemic group was further divided into those whose glucose rose after Sustacal (Hypo-High) and those whose blood glucose did not rise from baseline concentration (Hypo-Flat). The Hypo-High subgroup had a lower fasting free/total IRI ratio in addition to a greater increase in blood glucose after Sustacal compared with either Hypo-Flat or non-hypoglycaemic transplant recipients. The low free/total insulin ratio and hyperglycaemic response to Sustacal are consistent with the presence of anti-insulin antibodies, an established cause of hypoglycaemia. In summary, documented hypoglycaemia, while uncommon, may occur and be significant following PTX. Anti-insulin antibodies may be one of the many factors contributing to hypoglycaemia in this population. Received: 8 September 1998 / Accepted in revised form: 14 September 1998     

Glucose intolerance and post-stimulatory hypoglycemia secondary to a probably congenital intrahepatic portacaval anastomosis

Recurrent malaise in a 63 year old woman were found to be due to hypoglycaemic episodes. During a 5 hour oral glucose tolerance test, the "impaired glucose tolerance" type initial hyperglycaemic wave was followed by a post-stimulative hypoglycaemia. Serum C-peptide levels were normal during the test, but the insulin response which was initially normal became excessive, with a consequent decrease of the C-peptide/insulin ratio, similar to that usually observed in hepatic malfunction. An hepatic ultrasonography, a cavography and a selective superior mesenteric arteriography showed an intra-hepatic porto-caval anastomosis, probably congenital in origin. This vascular abnormality accounts for the blood glucose problems: the porto-caval shunt explains the early hyperglycaemia by defective liver uptake of glucose and secondary hyperinsulinism occurs because of the reduced hepatic degradation of the insulin secreted in normal quantity. The late hyperinsulinism then leads to secondary hypoglycaemia.     

Optimizing blood glucose control in type 2 diabetes: an approach based on fasting blood glucose measurements

A structured approach to the management of Type 2 diabetes, aiming to reduce fasting blood glucose levels to near-normal, can provide effective blood glucose control with minimal risk of hypoglycaemia and in a manner acceptable to most patients. When the fasting blood glucose value is maintained in the region 4-6 mmol l-1, protein glycosylation and plasma triglyceride values usually become near-normal and this may help to prevent the development of long-term diabetic complications. We propose a simple management strategy, based on 3-monthly fasting blood glucose determination, which uses not more than two therapeutic agents at any one time. If diet and maximal oral therapy fail to keep fasting blood glucose levels below 6 mmol l-1 then the addition of a basal insulin supplement, e.g. from a once daily injection of ultralente insulin, can restore near-normal fasting blood glucose levels without the need for full insulin replacement therapy. In older patients, where long-term prevention of diabetic complications is not such an immediate priority, less strict blood glucose control may be reasonable, aiming to keep the fasting blood glucose values below 10 mmol l-1 in order to prevent symptoms secondary to glycosuria. Patients can be seen at a monthly general practice morning diabetic mini-clinic or with the aid of a nurse visiting elderly patients at home.