‘Gastric’ hypoglycaemia – an important concept in diabetes management

The Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) and UK Prospective Diabetes Studies (UKPDS) have established that both the development and progression of the microvascular (i.e. neuropathy, nephropathy and retinopathy), and probably macrovascular (i.e. cardiovascular disease), complications of type 1 (insulin-dependent) and type 2 (non-insulin dependent) diabetes mellitus are related to ‘average’ glycaemic control, as assessed by measurement of glycated haemoglobin.1, 2 This evidence provides a persuasive rationale for the widespread use of ‘intensive’ therapy for diabetes directed at the normalization of glycaemia. For example, in the recently reported DCCT/EDIC study which involved type 1 patients, a period of ‘intensive’, as opposed to ‘conventional’, therapy for a mean period of 6.5 years during the DCCT study (i.e. between 1983 and 1993) was associated with a reduction in the risk of a subsequent (i.e. between 1994 and 2005) cardiovascular event of 42%.1 Hence, a major goal in treating diabetes is to achieve blood glucose levels as close as possible to the normal, fasting, range. This is compatible with the American Diabetes Association recommendation of a target glycated haemoglobin of <7.0%, which may in fact be conservative given that a specific healthy ‘glycaemic threshold’ has hitherto not been demonstrated.1-3 To achieve current targets, the majority of type 2 patients will require treatment with insulin within 10 years of the diagnosis of diabetes. In practice, however, glycaemic targets are not often met in either type 1 or type 2 patients. There is a natural reluctance to initiate insulin therapy in type 2 patients because of the potentially deleterious consequences of insulin-induced hypoglycaemia, especially in those patients whose ability to recognize impending hypoglycaemia is impaired. In the DCCT study, the reduction in microvascular complications seen in the ‘intensive’ care group was accompanied by a threefold increase in the rate of severe hypoglycaemia.1 Effective management of recurrent hypoglycaemia is, of course, dependent on the correction of the underlying cause(s), which include excessive insulin dosage, inadequate carbohydrate intake, psychiatric disorders, various endocrinopathies, coeliac disease and vomiting (whether self-induced or associated with gastroparesis). However, in a substantial minority of patients a cause is not clearly identifiable. The interesting study by Lysy et al.4 reported in this issue of the Journal indicates that delayed gastric emptying represents a significant risk factor for hypoglycaemia in insulin-treated type 1 and type 2 diabetes. The authors evaluated gastric emptying in 31 insulin-treated patients who had recurrent episodes of hypoglycaemia early in the postprandial period. Gastric emptying of a scrambled egg meal was significantly slower when compared to a control group of patients with diabetes who did not have recurrent hypoglycaemia; in about 30% of patients, the magnitude of the delay in emptying was marked, even though none of them reported upper gastrointestinal symptoms such as nausea or bloating. Strengths of the study include the relatively large number of patients and the prospective design, although demographic differences (e.g. in age and body mass index) between the control and test groups may have contributed to the results. The absence of any difference in glycated haemoglobin between the two groups and the ‘reasonable’ glycated haemoglobin levels in the majority of the cohort with hypoglycaemia, are perhaps surprising. The concept that gastroparesis may result in poor glycaemic control in insulin-treated diabetic patients, even in those without upper gastrointestinal symptoms, by adversely affecting the coordination between nutrient absorption and exogenous insulin delivery, is not new.5 In his seminal paper published in 1958, Kassander coined the descriptive term ‘gastroparesis diabeticorum’ while reporting six patients with insulin-treated diabetes (of whom five had type 1 diabetes), who exhibited abnormal intragastric retention of barium in the absence of upper gastrointestinal symptoms. At least three of these patients had experienced multiple episodes of severe hypoglycaemia. Kassander concluded that the ‘retention of stomach contents in a diabetic obviously may cause confusion as far as food intake and utilization are concerned.’ In providing the title ‘Gastric’ hypoglycaemia for this editorial, we wish to draw the attention of clinicians back to this often forgotten concept. During the last c. 20 years, improvement of techniques to evaluate gastric motility has led to clarification of the significance of disturbed gastric emptying in patients with diabetes. Clinical studies suggest that gastric emptying is abnormally delayed in 30–50% of individuals with longstanding type 1 or type 2 diabetes, including adolescents and patients with so-called ‘brittle’ type 1 diabetes, although the magnitude of the delay in gastric emptying is often modest.6-8 In contrast to initial expectations, the relationship between upper gastrointestinal symptoms and the rate of gastric emptying is relatively weak.7, 8 Experimental observations on healthy volunteers and patients with diabetes have demonstrated that acute elevations in the blood glucose concentration slow gastric emptying, and there is a direct relationship between stimulus and response.6 Hence, diabetic gastroparesis may be reversible. In contrast, insulin-induced hypoglycaemia accelerates gastric emptying substantially, even in longstanding type 1 patients with gastroparesis during euglycaemia,9 and is likely to represent an important counter-regulatory response to hypoglycaemia. Thus, the hypoglycaemic episodes in the study by Lysy et al.4 would seem to be the response to, rather than the cause of, delayed gastric emptying. The central role of gastric emptying to blood glucose homeostasis has only recently been appreciated and may reflect a general lack of recognition that humans living in affluent western societies exist for much of the day in a metabolic state of ‘postabsorption’ in which nutrients are flowing into the small intestine at 2–3 kcal min−1.10 As humans ingest about 2500 kcal each day, the duration of true ‘fasting’ is probably limited to 3–4 h before breakfast.11 Thus, the traditional focus on the control of ‘fasting’ blood glucose in diabetes management appears inappropriate. It has now been clearly established that the magnitude of postprandial glycaemic excursions is closely related to the rate of entry of carbohydrate into the small intestine in healthy subjects and in type 1 and type 2 patients.12-15 Moreover, postprandial glycaemia appears to be the major determinant of glycated haemoglobin.16 These insights have led to the formulation of dietary and pharmacological strategies to minimize glycaemic excursions and optimize diabetic control by modulating gastric emptying.6, 17 In type 2 patients insulin release is delayed, whereas the overall insulin response to a meal may be relatively normal. Therefore, for those who are not on insulin, it may be anticipated that strategies which slow gastric emptying, even if it is already delayed, would prove beneficial, as long as this does not induce upper gastrointestinal symptoms. Slowing of gastric emptying is likely to be the dominant mechanism by which exogenous administration of glucagon-like peptide-1 (GLP-1) and recently developed pharmacological agents, including longer acting GLP-1 receptor agonists such as exenatide, GLP-1 analogues such as liraglutide and the amylin analogue pramlintide, reduce postprandial glycaemia in type 2 patients.17-19 Both pramlintide and exenatide have recently been approved for use in the USA. However, when type 1 or type 2 patients are treated with insulin, the rate of gastric emptying of carbohydrate must be predictable if insulin is to be administered at the appropriate time and dosage. In view of the observations by Lysy et al.4 a compelling case can now be made for routine measurement of gastric emptying, preferably by scintigraphy, in insulin-treated patients with unexplained, recurrent, hypoglycaemia. This information should improve precision in the selection and timing of both nutrient intake and exogenous insulin. Thus, dietary strategies for the management of diabetes, such as the recent ‘Dose Adjustment for Normal Eating’ (DAFNE) programme in type 1 patients, in which the dose of short-acting insulin is based on the carbohydrate content of meals,20 should take into account the rate of gastric emptying and meal composition. The existence of delayed gastric emptying may represent an indication for prokinetic therapy, even in the absence of upper gut symptoms, but the efficacy of such an approach remains to be determined by large-scale studies.21-23