Hypoglycaemia in Clinical Diabetes Second Edition
Other titles in the Wiley Diabetes in Practice Series Exercise and Sport in Diabetes Second Edition Edited by Dinesh Nagi 047002206X Complementary Therapies and the Management of Diabetes and Vascular Disease Edited by Patricia Dunning 047001458X Diabetes in Clinical Practice: Questions and Answers from Case Studies N. Katsilambros, E. Diakoumopoulou, I. Ioannidis, S. Liatis, K. Makrilakis, N. Tentolouris and P. Tsapogas 978 0470 035221 Obesity and Diabetes Edited by Anthony Barnett and Sudhesh Kumar 0470848987 Prevention of Type 2 Diabetes Edited by Manfred Ganz 0470857331 Diabetes – Chronic Complications Second Edition Edited by Kenneth Shaw and Michael Cummings 0470865972 The Metabolic Syndrome Edited by Christopher Byrne and Sarah Wild 0470025115 Psychology in Diabetes Care Second Edition Edited by Frank J. Snoek and T. Chas Skinner 0470023848 The Foot in Diabetes Fourth Edition Edited by Andrew J.M. Boulton, Peter R. Cavanagh and Gerry Rayman 0470015047 Gastrointestinal Function in Diabetes Mellitus Edited by Michael Horowitz and Melvin Samson 047189916X Diabetic Nephropathy Edited by Christoph Hasslacher 0471489921 Nutritional Management of Diabetes Mellitus Edited by Gary Frost, Anne Dornhorst and Robert Moses 0471497517 Diabetes in Pregnancy: An International Approach to Diagnosis and Management Edited by Anne Dornhorst and David R. Hadden 047196204X
Hypoglycaemia in Clinical Diabetes Second Edition
Edited by Brian M. Frier The Royal Infirmary of Edinburgh, Scotland, UK
Miles Fisher Glasgow Royal Infirmary, Scotland, UK
Copyright © 2007
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England Telephone
(+44) 1243 779777
Email (for orders and customer service enquiries):
[email protected] Visit our Home Page on www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to
[email protected], or faxed to (+44) 1243 770620. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Other Wiley Editorial Offices John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons Canada Ltd, 6045 Freemont Blvd, Mississauga, Ontario, L5R 4J3 Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Anniversary Logo Design: Richard J. Pacifico Library of Congress Cataloging in Publication Data Hypoglycaemia in clinical diabetes / edited by Brian M. Frier and Miles Fisher. — 2nd ed. p. ; cm. Includes bibliographical references and index. ISBN 978-0-470-01844-6 (cloth : alk. paper) 1. Hypoglycemia. 2. Diabetes—Treatment—Complications. 3. Hypoglycemic agents—Side effects. I. Frier, Brian M. II. Fisher, Miles. III. Title: Hypoglycemia in clinical diabetes. [DNLM: 1. Hypoglycemia—complications. 2. Hypoglycemia—physiopathology. 3. Diabetes Complications. 4. Insulin—adverse effects. WK 880 H9963 2007] RC662.2.H965 2007 616.4 66—dc22 2007012095 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 978-0-470-01844-6 Typeset in 10/12pt Times by Integra Software Services Pvt. Ltd, Pondicherry, India Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.
To Emily, Ben and Marc
Contents Preface
ix
Contributors
xi
1 Normal Glucose Metabolism and Responses to Hypoglycaemia Ian A. Macdonald and Paromita King
1
2 Symptoms of Hypoglycaemia and Effects on Mental Performance and Emotions Ian J. Deary
25
3 Frequency, Causes and Risk Factors for Hypoglycaemia in Type 1 Diabetes Mark W.J. Strachan
49
4 Nocturnal Hypoglycaemia Simon R. Heller
83
5 Moderators, Monitoring and Management of Hypoglycaemia Tristan Richardson and David Kerr
101
6 Counterregulatory Deficiencies in Diabetes David Kerr and Tristan Richardson
121
7 Impaired Awareness of Hypoglycaemia Brian M. Frier
141
8 Risks of Strict Glycaemic Control Stephanie A. Amiel
171
9 Hypoglycaemia in Children with Diabetes Krystyna A. Matyka
191
10 Hypoglycaemia in Pregnancy Ann E. Gold and Donald W.M. Pearson
217
11 Hypoglycaemia in Type 2 Diabetes and in Elderly People Nicola N. Zammitt and Brian M. Frier
239
12 Mortality, Cardiovascular Morbidity and Possible Effects of Hypoglycaemia on Diabetic Complications Miles Fisher and Simon R. Heller
265
viii
CONTENTS
13 Long-term Effects of Hypoglycaemia on Cognitive Function and the Brain in Diabetes Petros Perros and Ian J. Deary
285
14 Living with Hypoglycaemia Brian M. Frier
309
Index
333
Preface In the second edition of this book, we have continued to emphasise the clinical significance of hypoglycaemia to the person who has diabetes, particularly when receiving treatment with insulin. Since the first edition of the book was published in 1999, new therapies have emerged, including new insulin analogues and inhaled insulin, and monitoring systems are now available that can provide continuous recording of blood glucose. However, far from minimising the risk of hypoglycaemia in clinical practice, the newer treatments have been shown to be as liable to cause hypoglycaemia as before, while continuous blood glucose monitoring has revealed that this side-effect of insulin therapy is even more common than was believed previously. The frequency of severe hypoglycaemia in vulnerable groups such as children and elderly people receiving insulin therapy is unacceptably high, and presents potentially serious risks to health as well as diminishing their quality of life. Much scientific research in recent years has focused on the effects of hypoglycaemia on the brain, providing a greater understanding of the protean effects of this metabolic abnormality. New data and concepts have been incorporated in this edition, particularly where these are of importance to clinical practice. In updating and revising this book about hypoglycaemia, particular emphasis has been given to the risk factors for hypoglycaemia and how these may be reduced or avoided. New chapters have been included to discuss recognised moderators of hypoglycaemia and the role of new glucose monitoring systems, to address the increasing problem of hypoglycaemia in people with type 2 diabetes and the elderly person, and to acknowledge the major importance of nocturnal hypoglycaemia, which is frequently not identified in clinical practice but can have serious consequences, not only in its immediate morbidity, but also in promoting the development of the acquired syndromes of hypoglycaemia. We are grateful for the expert assistance and support of the colleagues who have contributed chapters, some of whom are new as authors for this edition. All have skillfully highlighted the relevance of the enhancement of scientific knowledge in this field to the everyday management of diabetes, which we hope will assist all members of the diabetes team in their efforts to prevent and manage the extremely common but unwanted scourge that is hypoglycaemia. Brian M. Frier Miles Fisher
Contributors Professor Stephanie A. Amiel, R.D. Lawrence Professor of Diabetes, Department of Medicine, King’s College Hospital, Bessemer Road, London, SE5 9PJ (e-mail:
[email protected]) Professor Ian J. Deary, Department of Psychology, University of Edinburgh, 7 George Square, Edinburgh, EH8 9JZ (e-mail:
[email protected]) Dr Miles Fisher, Consultant Physician, Glasgow Royal Infirmary, Glasgow, G4 0SF (e-mail:
[email protected]) Professor Brian M. Frier, Consultant Physician, Department of Diabetes, Royal Infirmary, 51 Little France Crescent, Edinburgh, EH16 4SA (e-mail:
[email protected]) Dr Ann E. Gold, Consultant Physician, Wards 27 and 28, Aberdeen Royal Infirmary, Foresterhill, Aberdeen, AB25 2ZN (e-mail:
[email protected]) Professor Simon R. Heller, Professor of Clinical Diabetes, Clinical Sciences Centre, Department of Diabetes & Endocrinology, Northern General Hospital, Herries Road, Sheffield, S5 7AU (e-mail:
[email protected]) Dr David Kerr, Consultant Physician, The Royal Bournemouth Hospital, Castle Lane East, Bournemouth, BH7 7DW (e-mail:
[email protected]) Dr Paromita King, Consultant Physician, Jenny O’Neill Diabetes Centre, Derbyshire Royal Infirmary, London Road, Derby, DE1 2QY (e-mail:
[email protected]) Professor Ian A. Macdonald, Professor of Metabolic Physiology, Department of Physiology and Pharmacology, Medical School, Queen’s Medical Centre, Nottingham, NG7 2UH (e-mail:
[email protected]) Dr Krystyna A. Matyka, Senior Lecturer in Paediatrics, University of Warwick Medical School, Division of Clinical Sciences, CSB Research Wing, UHCW Trust, Clifford Bridge Road, Coventry, CV2 2DX (e-mail:
[email protected]) Dr Donald W.M. Pearson, Consultant Physician, Aberdeen Royal Infirmary, Foresterhill, Aberdeen, AB25 2ZN (e-mail:
[email protected]) Dr Petros Perros, Consultant Endocrinologist, Ward 15, Freeman Hospital, Freeman Road, Newcastle-Upon-Tyne, NE7 7DN (e-mail:
[email protected]) Dr Tristan Richardson, Consultant Physician, Bournemouth Diabetes and Endocrine Centre, The Royal Bournemouth Hospital, Castle Lane East, Bournemouth, BH7 7DW (e-mail:
[email protected])
xii
CONTRIBUTORS
Dr Mark W.J. Strachan, Consultant Physician, Metabolic Unit, Western General Hospital, Crewe Road, Edinburgh, EH4 2XU (e-mail:
[email protected]) Dr Nicola N. Zammitt, Specialist Registrar, Department of Diabetes, Royal Infirmary of Edinburgh, 51 Little France Crescent, Edinburgh, EH16 4SA (e-mail:
[email protected])
1 Normal Glucose Metabolism and Responses to Hypoglycaemia Ian A. Macdonald and Paromita King
INTRODUCTION Control of blood glucose is a fundamental feature of homeostasis, i.e., the process by which the internal environment of the body is maintained stable allowing optimal function. Blood glucose concentrations are regulated within a narrow range (which in humans is known as normoglycaemia or euglycaemia) despite wide variability in carbohydrate intake and physical activity. Teleologically, the upper limit is defended because high glucose concentrations cause microvascular complications, and the lower limit, because the brain cannot function without an adequate supply of glucose. In this chapter the mechanisms that protect against hypoglycaemia in healthy individuals and the physiological consequences of low glucose concentrations are discussed.
NORMAL GLUCOSE HOMEOSTASIS Humans evolved as hunter-gatherers and, unlike people today, did not consume regular meals. Mechanisms therefore evolved for the body to store food when it was in abundance, and to use these stores to provide an adequate supply of energy, in particular in the form of glucose when food was scarce. Cahill (1971) originally described the ‘rules of the metabolic game’ which humans had to follow to ensure their survival. These rules were modified by Tattersall (personal communication) and are as follows: 1. Maintain glucose within very narrow limits. 2. Maintain an emergency energy source (glycogen) which can be tapped quickly for fleeing or fighting. 3. Waste not want not, i.e., store (fat and protein) in times of plenty. 4. Use every trick in the book to maintain protein reserves.
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
NORMAL GLUCOSE METABOLISM AND RESPONSES
2
Insulin and glucagon are the two hormones controlling glucose homeostasis, and therefore the mechanisms enabling the ‘rules’ to be followed. The most important processes governed by these hormones are: • Glycogen synthesis and breakdown (glycogenolysis): Glycogen, a carbohydrate, is an energy source stored in the liver and skeletal muscle. Liver glycogen is broken down to provide glucose for all tissues, whereas the breakdown of muscle glycogen results in lactate formation. • Gluconeogenesis: This is the production of glucose in the liver from precursors: glycerol, lactate and amino acids (in particular alanine). The process can also occur in the kidneys, but this site is not important under physiological conditions. • Glucose uptake and metabolism (glycolysis) by skeletal muscle and adipose tissue. The actions of insulin and glucagon are summarised in Boxes 1.1 and 1.2. Insulin is an anabolic hormone, reducing glucose output by the liver (hepatic glucose output), increasing the uptake of glucose by muscle and adipose tissue (increasing peripheral uptake) and increasing protein and fat formation. Glucagon opposes the actions of insulin in the liver. Thus insulin tends to reduce, and glucagon to increase, blood glucose concentrations.
Box 1.1
Actions of insulin
Liver ↑ Glycogen synthesis (↑ glycogen synthetase activity) ↑ Glycolysis ↑ Lipid formation ↑ Protein formation ↓ Glycogenolysis (↓ phosphorylase activity) ↓ Gluconeogenesis ↓ Ketone formation Muscle ↑ Uptake of glucose amino acids ketone potassium ↑ Glycolysis ↑ Synthesis of glycogen protein ↓ Protein catabolism ↓ Release of amino acids Adipose tissue ↑ Uptake glucose potassium Storage of triglyceride
NORMAL GLUCOSE HOMEOSTASIS
Box 1.2
3
Actions of glucagon
Liver ↑ Glycogenolysis ↑ Gluconeogenesis ↑ Extraction of alanine ↑ Ketogenesis No significant peripheral action
The metabolic effects of insulin and glucagon and their relationship to glucose homeostasis are best considered in relationship to fasting and the postprandial state (Siegal and Kreisberg, 1975). In both these situations it is the relative and not absolute concentrations of these hormones that are important.
Fasting (Figure 1.1a) During fasting, insulin concentrations are reduced and glucagon increased, which maintains blood glucose concentrations in accordance with rule 1 above. The net effect is to reduce peripheral glucose utilisation, to increase hepatic glucose production and to provide non-glucose fuels for tissues not entirely dependent on glucose. After a short (for example overnight) fast, glucose production needs to be 5–6 g/h to maintain blood glucose concentrations, with the brain using 80% of this. Glycogenolysis provides 60–80% and gluconeogenesis 20–40% of the required glucose. In prolonged fasts, glycogen becomes depleted and glucose production is primarily from gluconeogenesis, with an increasing proportion from the kidney compared to the liver. In extreme situations renal gluconeogenesis can contribute as much as 45% of glucose production. Thus glycogen is the short term or ‘emergency’ fuel source (rule 2), with gluconeogenesis predominating during more prolonged fasts. The following metabolic alterations enable this increase in glucose production to occur: • Muscle: Glucose uptake and oxidative metabolism are reduced and fatty acid oxidation increased. Amino acids are released. • Adipose tissue: There are reductions in glucose uptake and triglyceride storage. The increase in the activity of the enzyme hormone-sensitive lipase results in hydrolysis of triglyceride to glycerol (a gluconeogenic precursor) and fatty acids, which can be metabolised. • Liver: Increased cAMP concentrations result in increased glycogenolysis and gluconeogenesis thus increasing hepatic glucose output. The uptake of gluconeogenic precursors (i.e. amino acids, glycerol, lactate and pyruvate) is also increased. Ketone bodies are produced in the liver from fatty acids. This process is normally inhibited by insulin and stimulated by glucagon, thus the hormonal changes during fasting lead to an increase in ketone production. Fatty acids are also a metabolic fuel used by the liver and provide a source of energy for the reactions involved in gluconeogenesis.
NORMAL GLUCOSE METABOLISM AND RESPONSES
4
glucagon, insulin (a)
Insulin, (b)
FASTING
lactate
POST PRANDIAL
Glucose
Glycogen
Amino acids
glucagon
Glycogen
Protein
Amino acids
Protein
Muscle
TG Glucose FA
Glycerolphosphate FA
Free FA
Glycerol
Glycerolphosphate
Free FA Triglyceride Triglyceride
Adipose Tissue Amino acids
Glycogen Glucose
lactate Ketones
Ketones
FA Glycerol
Glucose FFA, TG, lipoprotein
Liver Hepatic glucose output
Glycogen
FA Glycerol
FFA, TG, Hepatic glucose output
Figure 1.1 Metabolic pathways for glucose homeostasis in muscle, adipose tissue and liver during fasting (left) and postprandially (right). FA = fatty acids; TG = triglyceride (associated CO2 production excluded for clarity)
The reduced insulin : glucagon ratio favours a catabolic state, but the effect on fat metabolism is greater than protein, and thus muscle is relatively preserved (rule 4). These adaptations meant that not only did hunter-gatherers have sufficient muscle power to pursue their next meal, but also that brain function was optimally maintained to help them do this.
EFFECTS OF GLUCOSE DEPRIVATION
5
Fed state (Figure 1.1b) In the fed state, in accordance with the rules of the metabolic game, excess food is stored as glycogen, protein and fat (rule 3). The rise in glucose concentrations results in an increase in insulin and reduction in glucagon secretion. This balance favours glucose utilisation, reduction of glucose production and increases glycogen, triglyceride and protein formation. The following changes enable these processes to occur: • Muscle: Insulin increases glucose transport, oxidative metabolism and glycogen synthesis. Amino acid release is inhibited and protein synthesis is increased. • Adipose tissue: In the fat cells, glucose transport is increased, while lipolysis is inhibited. At the same time the enzyme lipoprotein lipase, located in the capillaries, is activated and causes triglyceride to be broken down to fatty acids and glycerol. The fatty acids are taken up into the fat cells and re-esterified to triglyceride (using glycerol phosphate derived from glucose) before being stored. • Liver: Glucose uptake is increased in proportion to plasma glucose, a process which does not need insulin. However, insulin does decrease cAMP concentrations, which results in an increase in glycogen synthesis and the inhibition of glycogenolysis and gluconeogenesis. These effects ‘retain’ glucose in the liver and reduce hepatic glucose output. This complex interplay between insulin and glucagon maintains euglycaemia and enables the rules of the metabolic game to be followed, ensuring not only the survival of the hunter-gatherer, but also of modern humans.
EFFECTS OF GLUCOSE DEPRIVATION ON CENTRAL NERVOUS SYSTEM METABOLISM The brain constitutes only 2% of body weight, but consumes 20% of the body’s oxygen and receives 15% of its cardiac output (Sokaloff, 1989). It is almost totally dependent on carbohydrate as a fuel and since it cannot store or synthesise glucose, depends on a continuous supply from circulating blood. The brain contains the enzymes needed to metabolise fuels other than glucose such as lactate, ketones and amino acids, but under physiological conditions their use is limited by insufficient quantities in the blood or slow rates of transport across the blood-brain barrier. When arterial blood glucose falls below 3 mmol/l, cerebral metabolism and function decline. Metabolism of glucose by the brain releases energy, and also generates neurotransmitters such as gamma amino butyric acid (GABA) and acetylcholine, together with phospholipids needed for cell membrane synthesis. When blood glucose concentration falls, changes in the synthesis of these products may occur within minutes because of reduced glucose metabolism, which can alter cerebral function. This is likely to be a factor in producing the subtle changes in cerebral function detectable at blood glucose concentrations as high as 3 mmol/l, which is not sufficiently low to cause a major depletion in ATP or creatine phosphate, the brain’s two main sources of energy (McCall, 1993).
6
NORMAL GLUCOSE METABOLISM AND RESPONSES
Isotope techniques and Positron Emission Tomography (PET) allow the study of metabolism in different parts of the brain and show regional variations in metabolism during hypoglycaemia. The neocortex, hippocampus, hypothalamus and cerebellum are most sensitive to hypoglycaemia, whereas metabolism is relatively preserved in the thalamus and brainstem. Changes in cerebral function are initially reversible, but during prolonged severe hypoglycaemia, general energy failure (due to the depletion of ATP and creatine phosphate) can cause permanent cerebral damage. Pathologically this is caused by selective neuronal necrosis most likely due to ‘excitotoxin’ damage. Local energy failure induces the intrasynaptic release of glutamate or aspartate, and failure of reuptake of the neurotransmitters increases their concentrations. This leads to the activation of N-methyl-D-aspartate (NMDA) receptors causing cerebral damage. One study in rats has shown that an experimental compound called AP7, which blocks the NMDA receptor, can prevent 90% of the cerebral damage associated with severe hypoglycaemia (Wieloch, 1985). In humans with fatal hypoglycaemia, protracted neuroglycopenia causes laminar necrosis in the cerebral cortex and diffuse demyelination. Regional differences in neuronal necrosis are seen, with the basal ganglia and hippocampus being sensitive, but the hypothalamus and cerebellum being relatively spared (Auer and Siesjö, 1988; Sieber and Traysman, 1992). The brain is very sensitive to acute hypoglycaemia, but can adapt to chronic fuel deprivation. For example, during starvation, it can metabolise ketones for up to 60% of its energy requirements (Owen et al., 1967). Glucose transport can also be increased in the face of hypoglycaemia. Normally, glucose is transported into tissues using proteins called glucose transporters (GLUT) (Bell et al., 1990). This transport occurs down a concentration gradient faster than it would by simple diffusion and does not require energy (facilitated diffusion). There are several of these transporters, with GLUT 1 being responsible for transporting glucose across the blood-brain barrier and GLUT 3 for transporting glucose into neurones (Figure 1.2). Chronic hypoglycaemia in animals (McCall et al., 1986) and in humans (Boyle et al., 1995) increases cerebral glucose uptake, which is thought to be promoted by an increase in the production and action of GLUT 1 protein. It has not been
Figure 1.2 Transport of glucose into the brain across the blood–brain barrier
COUNTERREGULATION DURING HYPOGLYCAEMIA
7
established whether this adaptation is of major benefit in protecting brain function during hypoglycaemia.
COUNTERREGULATION DURING HYPOGLYCAEMIA The potentially serious effects of hypoglycaemia on cerebral function mean that not only are stable blood glucose concentrations maintained under physiological conditions, but also if hypoglycaemia occurs, mechanisms have developed to combat it. In clinical practice, the principal causes of hypoglycaemia are iatrogenic (as side-effects of insulin and sulphonylureas used to treat diabetes) and excessive alcohol consumption. Insulin secreting tumours (such as insulinoma) are rare. The mechanisms that correct hypoglycaemia are called counterregulation, because the hormones involved oppose the action of insulin and therefore are the counterregulatory hormones. The processes of counterregulation were identified in the mid 1970s and early 1980s, using either a bolus injection or continuous infusion of insulin to induce hypoglycaemia (Cryer, 1981; Gerich, 1988). The response to the bolus injection of 0.1 U/kg insulin in a normal subject is shown in Figure 1.3. Blood glucose concentrations decline within minutes of the administration of insulin and reach a nadir after 20–30 minutes, then gradually rise to near normal by two hours after the insulin was administered. The fact
Figure 1.3 (a) Glucose and (b) insulin concentrations after intravenous injection of insulin 0.1 U/kg at time 0. Reproduced from Garber et al. (1976) by permission of the Journal of Clinical Investigation
8
NORMAL GLUCOSE METABOLISM AND RESPONSES
that blood glucose starts to rise when plasma insulin concentrations are still ten times the baseline values means that it is not simply the reduction in insulin that reverses hypoglycaemia, but active counterregulation must also occur. Many hormones are released when blood glucose is lowered (see below), but glucagon, the catecholamines, growth hormone and cortisol are regarded as being the most important. Several studies have determined the relative importance of these hormones by producing isolated deficiencies of each hormone (by blocking its release or action) and assessing the subsequent response to administration of insulin. These studies are exemplified in Figure 1.4 which assesses the relative importance of glucagon, adrenaline (epinephrine) and growth hormone in the counterregulation of short term hypoglycaemia. Somatostatin infusion blocks glucagon and growth hormone secretion and significantly impairs glucose recovery (Figure 1.4a). If growth hormone is replaced in the same model to produce isolated glucagon deficiency (Figure 1.4b), and glucagon replaced to produce isolated growth hormone deficiency (Figure 1.4c), it is clear that it is glucagon and not growth hormone that is responsible for acute counterregulation. Combined alpha and beta adrenoceptor blockade using phentolamine and propranolol infusions or adrenalectomy (Figure 1.4d), can be used to evaluate the role of the catecholamines. These and other studies demonstrate that glucagon is the most important counterregulatory hormone whereas catecholamines provide a backup if glucagon is deficient (for example in type 1 diabetes, see Chapters 6 and 7). Cortisol and
Figure 1.4 Glucose recovery from acute hypoglycaemia. Glucose concentration following an intravenous injection of insulin of 0.05 U/kg at time 0; after (a) saline infusion (continuous line) and somatostatin, (b) somatostatin and growth hormone (GH), (c) somatostatin and glucagon, (d) combined alpha and beta blockade with phentolamine and propranolol infusions or adrenalectomy, (e) somatostatin with alpha and beta blockade, and (f) somatostatin in adrenalectomised patients. Saline infusion = continuous lines; experimental study = broken lines. Reproduced from Cryer (1981) courtesy of the American Diabetes Association (epinephrine = adrenaline)
COUNTERREGULATION DURING HYPOGLYCAEMIA
9
growth hormone are important only in prolonged hypoglycaemia. Therefore if glucagon and catecholamines are both deficient, as in longstanding type 1 diabetes, counterregulation is seriously compromised, and the individual is defenceless against acute hypoglycaemia (Cryer, 1981). Glucagon and catecholamines increase glycogenolysis and stimulate gluconeogenesis. Catecholamines also reduce glucose utilisation peripherally and inhibit insulin secretion. Cortisol and growth hormone increase gluconeogenesis and reduce glucose utilisation. The role of the other hormones (see below) in counterregulation is unclear, but they are unlikely to make a significant contribution. Finally, there is evidence that during profound hypoglycaemia (blood glucose below 1.7 mmol/l), hepatic glucose output is stimulated directly, although the mechanism is unknown. This is termed hepatic autoregulation. The depth, as well as the duration, of hypoglycaemia is important in determining the magnitude of the counterregulatory hormone response. Studies using ‘hyperinsulinaemic clamps’ show a hierarchical response of hormone production. In this technique, insulin is infused at a constant rate and a glucose infusion rate varied to maintain blood glucose concentrations within ±02 mmol/l of target concentrations. This permits the controlled evaluation of the counterregulatory hormone response at varying degrees of hypoglycaemia. It also demonstrates that glucagon, catecholamines and growth hormone start to be secreted at a blood glucose concentration of 3.5–3.7 mmol/l, with cortisol produced at a lower glucose of 3.0 mmol/l (Mitrakou et al., 1991). The counterregulatory response is initiated before impairment in cerebral function commences, usually at a blood glucose concentration of approximately 3.0 mmol/l (Heller and Macdonald, 1996). The magnitude of the hormonal response also depends on the length of the hypoglycaemic episode. The counterregulatory hormonal response commences up to 20 minutes after hypoglycaemia is achieved and continues to rise for 60 minutes (Kerr et al., 1989). In contrast, this response is attenuated as a result of a previous episode of hypoglycaemia (within a few days) (reviewed by Heller and Macdonald, 1996) and even by prolonged exercise the day before hypoglycaemia is induced. Galassetti et al. (2001) showed that in non-diabetic subjects three hours of moderate intensity exercise the previous day markedly decreased the counterregulatory response to hypoglycaemia induced by the infusion of insulin, and that the reduced counterregulatory response was more marked in men than in women. Although the primary role of the counterregulatory hormones is on glucose metabolism, any effects on fatty acid utilisation can have an indirect effect on blood glucose. Thus, the increase in plasma epinephrine (adrenaline) (and activation of the sympathetic nervous system) that is seen in hypoglycaemia can stimulate lipolysis of triglyceride in adipose tissue and muscle and release fatty acids which can be used as an alternative fuel to glucose, making more glucose available for the CNS. Enoksson et al. (2003) demonstrated that patients with type 1 diabetes, who had lower plasma epinephrine responses to hypoglycaemia than nondiabetic controls, also had reduced rates of lipolysis in adipose tissue and skeletal muscle, making them more dependent on glucose as a fuel and therefore at risk of developing a more severe hypoglycaemia. The complex counterregulatory and homeostatic mechanisms described above are thought to be mostly under the control of the central nervous system. Evidence for this comes from studies in dogs, where glucose was infused into the carotid and vertebral arteries to maintain euglycaemia in the brain. Despite peripheral hypoglycaemia, glucagon did not increase and responses of the other counterregulatory hormones were blunted. This, and
10
NORMAL GLUCOSE METABOLISM AND RESPONSES
other studies in rats, led to the hypothesis that the ventromedial nucleus of the hypothalamus (VMH), which does not have a blood–brain barrier, acts as a glucose-sensor and co-ordinates counterregulation (Borg et al., 1997). However, evidence exists that other parts of the brain may also be involved in mediating counterregulation. It is now clear that glucose-sensing neurones can involve either glucokinase or ATPsensitive K+ channels (Levin et al., 2004). In rats, the VMH has ATP-sensitive K+ channels which seem to be involved in the counterregulatory responses to hypoglycaemia, as injection of the sulphonylurea, glibenclamide, directly into the VMH suppressed hormonal responses to systemic hypoglycaemia (Evans et al., 2004). The existence of hepatic autoregulation suggests that some peripheral control should exist. Studies producing central euglycaemia and hepatic portal venous hypoglycaemia in dogs have provided evidence for hepatic glucose sensors and suggest that these sensors, as well as those in the brain, are important in the regulation of glucose (Hamilton-Wessler et al., 1994). However, this topic is somewhat controversial and more recent studies on dogs have failed to demonstrate an effect of hepatic sensory nerves on the responses to hypoglycaemia (Jackson et al., 2000). Moreover, studies in humans by Heptulla et al. (2001) showed that providing glucose orally rather than intravenously during a hypoglycaemic hyperinsulinaemic clamp actually enhanced the counterregulatory hormone responses rather than reduced them.
HORMONAL CHANGES DURING HYPOGLYCAEMIA Hypoglycaemia induces the secretion of various hormones, some of which are responsible for counterregulation, many of the physiological changes that occur as a consequence of lowering blood glucose and contribute to symptom generation (see Chapter 2), The stimulation of the autonomic nervous system is central to many of these changes.
Activation of the Autonomic Nervous System The autonomic nervous system comprises sympathetic and parasympathetic components (Figure 1.5). Fibres from the sympathetic division leave the spinal cord with the ventral roots from the first thoracic to the third or fourth lumbar nerves to synapse in the sympathetic chain or visceral ganglia, and the long postganglionic fibres are incorporated in somatic nerves. The parasympathetic pathways originate in the nuclei of cranial nerves III, VII, IX and X, and travel with the vagus nerve. A second component, the sacral outflow, supplies the pelvic viscera via the pelvic branches of the second to fourth spinal nerves. The ganglia in both cases are located near the organs supplied, and the postganglionic neurones are therefore short. Selective activation of both components of the autonomic system occurs during hypoglycaemia. The sympathetic nervous system in particular is responsible for many of the physiological changes during hypoglycaemia and the evidence for its activation can be obtained indirectly by observing functional changes such as cardiovascular responses (considered below), measuring plasma catecholamines which gives a general index of sympathetic activation, or by directly recording sympathetic activity.
HORMONAL CHANGES DURING HYPOGLYCAEMIA
11
Figure 1.5 Anatomy of the autonomic nervous system. Pre = preganglionic neurones; post = postganglionic neurones; RC = ramus communicans
Direct recordings are possible from sympathetic nerves supplying skeletal muscle and skin. Sympathetic neural activity in skeletal muscle involves vasoconstrictor fibres which innervate blood vessels and are involved in controlling blood pressure. During hypoglycaemia (induced by insulin), the frequency and amplitude of muscle sympathetic activity are increased as blood glucose falls, with an increase in activity eight minutes after insulin is injected intravenously, peaking at 25–30 minutes coincident with the glucose nadir, and persisting
12
NORMAL GLUCOSE METABOLISM AND RESPONSES
Figure 1.6 (a) Muscle sympathetic activity during euglycaemia and hypoglycaemia. Reproduced from Fagius et al. (1986) courtesy of the American Diabetes Association. (b) Skin sympathetic activity during euglycaemia and hypoglycaemia. Reproduced from Berne and Fagius (1986), with kind permission from Springer Science and Business Media
for 90 minutes after euglycaemia is restored (Figure 1.6a) (Fagius et al., 1986). During hypoglycaemia, a sudden increase in skin sympathetic activity is seen, which coincides with the onset of sweating. This sweating leads to vasodilatation of skin blood vessels, which is also contributed to by a reduction in sympathetic stimulation of the vasoconstrictor components of skin arterio-venous anastomoses (Figure 1.6b) (Berne and Fagius, 1986). These effects (at least initially) increase total skin blood flow and promote heat loss from the body. Activation of both muscle and skin sympathetic nerve activity are thought to be centrally mediated. Tissue neuroglycopenia can be produced by 2-deoxy-D-glucose, a glucose analogue, without increasing insulin. Infusion of this analogue causes stimulation of muscle and skin sympathetic activity demonstrating that it is the hypoglycaemia per se, and not the insulin used to induce it, which is responsible for the sympathetic activation (Fagius and Berne, 1989). The activation of the parasympathetic nervous system (vagus nerve) during hypoglycaemia cannot be measured directly. The most useful index of parasympathetic function is the measurement of plasma pancreatic polypeptide, the peptide hormone secreted by the PP cells of the pancreas, which is released in response to vagal stimulation.
HORMONAL CHANGES DURING HYPOGLYCAEMIA
13
Neuroendocrine Activation (Box 1.3) Insulin-induced hypoglycaemia was used to study pituitary function as early as the 1940s. The development of assays for adrenocorticotrophic hormone (ACTH) and growth hormone (GH) allowed the direct measurement of pituitary function during hypoglycaemia in the 1960s, and many of the processes governing these changes were unravelled before elucidation of the counterregulatory system. The studies are comparable to those evaluating counterregulation, in that potential regulatory factors are blocked to measure the hormonal response to hypoglycaemia with and without the regulating factor.
Box 1.3
Neuroendocrine activation
Hypothalamus
↑ ↑
Corticotrophic releasing hormone Growth hormone releasing hormone
Anterior Pituitary
↑ ↑ ↑ ↑ ↔ ↔
Adrenocorticotrophic hormone Beta endorphin Growth hormone Prolactin Thyrotrophin Gonadotrophins
Posterior pituitary
↑ ↑ ↑ ↑ ↑
Vasopressin Oxytocin Glucagon Pancreatic polypeptide Insulin
↑ ↑ ↑ ↑ ↑ ↑
Cortisol Epinephrine (adrenaline) Aldosterone Parathyroid hormone Gastrin Somatostatin (28)
Pancreas
Adrenal
Others
Hypothalamus and anterior pituitary ACTH, GH and prolactin concentrations increase during hypoglycaemia, but there is no change in thyrotrophin or gonadotrophin secretion. The secretion of these pituitary hormones is controlled by releasing factors which are produced in the median eminence of the hypothalamus, secreted into the hypophyseal portal vessels and then pass to the pituitary gland (Figure 1.7). The mechanisms regulating the releasing factors are incompletely understood, but may involve the ventromedial nucleus, one site where brain glucose sensors are situated (Fish et al., 1986).
NORMAL GLUCOSE METABOLISM AND RESPONSES
14
Figure 1.7 Anatomy of the hypothalamus and pituitary gland
• ACTH: Secretion is governed by release of corticotrophin releasing hormone (CRH) from the hypothalamus; alpha adrenoceptors stimulate CRH release, and beta adrenoceptors have an inhibitory action. A variety of neurotransmitters control the release of CRH into the portal vessels, including serotonin and acetylcholine which are stimulatory and GABA which is inhibitory. The increase in ACTH causes cortisol to be secreted from the cortices of the adrenal glands. • Beta endorphins are derived from the same precursors as ACTH and are co-secreted with it. The role of endorphins in counterregulation is uncertain, but they may influence the secretion of the other pituitary hormones during hypoglycaemia. • GH: Growth hormone secretion is governed by two hypothalamic hormones: growth hormone releasing hormone (GHRH) which stimulates GH secretion, and somatostatin which is inhibitory. GHRH secretion is stimulated by dopamine, GABA, opiates and through alpha adrenoceptors, whereas it is inhibited by serotonin and beta adrenoceptors. A study in rats showed that bioassayable GH and GHRH are depleted in the pituitary and hypothalamus respectively after insulin-induced hypoglycaemia (Katz et al., 1967). • Prolactin: The mechanisms underlying its secretion are not established. Prolactin secretion is normally under the inhibitory control of dopamine, but evidence also exists for releasing factors being produced during hypoglycaemia. Prolactin does not contribute to counterregulation. Posterior pituitary Vasopressin and oxytocin both increase during hypoglycaemia (Fisher et al., 1987). Their secretion is under hormonal and neurotransmitter control in a similar way to the hypothalamic
PHYSIOLOGICAL RESPONSES
15
hormones. Vasopressin has glycolytic actions and oxytocin increases hepatic glucose output in dogs, but their contribution to glucose counterregulation is uncertain. Pancreas • Glucagon: The mechanisms of glucagon secretion during hypoglycaemia are still not fully understood. Although activation of the autonomic nervous system stimulates its release, this pathway has been shown to be less important in humans. A reduction in glucose concentrations may have a direct effect on the glucagon-secreting pancreatic alpha cells, or the reduced beta cell activity (reduced insulin secretion), which also occurs with low blood glucose, may release the tonic inhibition of glucagon secretion. However, such mechanisms would be disturbed in type 1 diabetes, where hypoglycaemia is normally associated with high plasma insulin levels and there is no direct effect of beta cell-derived insulin on the alpha cells. • Somatostatin: This is thought of as a pancreatic hormone produced from D cells of the islets of Langerhans, but it is also secreted in other parts of the gastrointestinal tract. There are a number of structurally different polypeptides derived from prosomatostatin: the somatostatin-14 peptide is secreted from D cells, and somatostatin-28 from the gastrointestinal tract. The plasma concentration of somatostatin-28 increases during hypoglycaemia (Francis and Ensinck, 1987). The normal action of somatostatin is to inhibit the secretion both of insulin and glucagon, but somatostatin-28 inhibits insulin ten times more effectively than glucagon, and thus may have a role in counterregulation by suppressing insulin release. • Pancreatic polypeptide: This peptide has no known role in counterregulation, but its release during hypoglycaemia is stimulated by cholinergic fibres through muscarinic receptors and is a useful marker of parasympathetic activity. Adrenal and Renin–Angiotensin system The processes governing the increase in cortisol during hypoglycaemia are discussed above. The rise in catecholamines, in particular epinephrine from the adrenal medulla, which occurs when blood glucose is lowered, is controlled by sympathetic fibres in the splanchnic nerve. The increase in renin, and therefore angiotensin and aldosterone, during hypoglycaemia is stimulated primarily by the intra-renal effects of increased catecholamines, mediated through beta adrenoceptors, although the increase in ACTH and hypokalaemia due to hypoglycaemia contributes (Trovati et al., 1988; Jungman et al., 1989). These changes do not have a significant role in counterregulation, although angiotensin II has glycolytic actions in vitro.
PHYSIOLOGICAL RESPONSES Haemodynamic Changes (Box 1.4) The haemodynamic changes during hypoglycaemia (Hilsted, 1993) are mostly caused by the activation of the sympathetic nervous system and an increase in circulating epinephrine.
NORMAL GLUCOSE METABOLISM AND RESPONSES
16
Box 1.4 ↑ ↑ ↑ ↓ ↑
Haemodynamic changes
Heart rate Systolic blood pressure Cardiac output Peripheral resistance Myocardial contractility
An increase in heart rate (tachycardia), myocardial contractility and cardiac output occurs, which is mediated through beta1 adrenoceptors, but increasing vagal tone counteracts this effect so the increase is transient. Peripheral resistance, estimated from mean arterial pressure divided by cardiac output, is reduced. A combination of the increase in cardiac output and reduction in peripheral resistance results in an increase in systolic and a decrease in diastolic pressure, i.e. a widening of pulse pressure without a change in mean arterial pressure.
Changes in Regional Blood Flow (Box 1.5 and Figure 1.8) • Cerebral blood flow: Early work produced conflicting results, but these studies were in subjects receiving insulin shock therapy, and the varying effects of convulsions and altered level of consciousness may have influenced the outcome. Subsequent studies have consistently shown an increase in cerebral blood flow during hypoglycaemia despite the use of different methods of measurement (isotopic, single photon emission computed tomography (SPECT) and Doppler ultrasound). In most of the studies blood glucose concentration was less than 2 mmol/l before a change was observed. In animals, hypoglycaemia is associated with loss of cerebral autoregulation (the ability of the brain to maintain cerebral blood flow despite variability in cardiac output) through beta adrenoceptor stimulation, but the exact mechanisms are unknown (Bryan, 1990; Sieber and Traysman, 1992). • Gastrointestinal system: Total splanchnic blood flow (supplying the intestines, liver, spleen and stomach) is increased and splanchnic vascular resistance reduced as assessed by the bromosulphthalein extraction technique (Bearn et al., 1952). Superior mesenteric artery blood flow measured using Doppler ultrasound increases during hypoglycaemia due to beta adrenoceptor stimulation (Braatvedt et al., 1993). Radioisotope scanning has
Box 1.5 ↑ ↑ ↓ ↑ ↓
Changes in regional blood flow
Cerebral flow Total splanchnic flow Splenic flow Skin flow variable (early ↑, late ↓) Muscle flow Renal flow
PHYSIOLOGICAL RESPONSES
17
Figure 1.8 Changes in regional blood flow during hypoglycaemia
demonstrated a reduction in splenic activity during hypoglycaemia (Fisher et al., 1990), which is thought to be a consequence of alpha adrenoceptor-mediated reduction in blood flow. These changes would all be expected to increase hepatic blood flow. • Skin: The control of blood flow to the skin is complex and different mechanisms predominate in different areas. Studies of the effect of hypoglycaemia on skin blood flow are inconsistent partly because different methods have been used for blood flow measurement and induction of hypoglycaemia, as well as differences in the part of the body studied. Definitive conclusions are therefore not possible. Studies using the dorsum of the foot and the face (cheek and forehead) have consistently shown an initial vasodilatation and increase in blood flow followed by later vasoconstriction at a blood glucose of 2.5 mmol/l (Maggs et al., 1994). These findings are consistent with the clinical picture of initial flushing and later pallor, with an early rise in skin blood flow followed by a later fall. • Muscle blood flow: A variety of techniques have been used to study muscle blood flow (including venous occlusion plethysmography, isotopic clearance techniques and the use of thermal conductivity meters). All studies have consistently shown an increase in muscle
NORMAL GLUCOSE METABOLISM AND RESPONSES
18
blood flow during hypoglycaemia irrespective of skin blood flow. This change is mediated by beta2 adrenoceptors (Abramson et al., 1966; Allwood et al., 1959). • Kidney: Inulin and sodium hippurate clearance can be used to estimate glomerular filtration rate and renal blood flow respectively. Both decrease during hypoglycaemia (Patrick et al., 1989) and catecholamines and renin are implicated in initiating the changes. The changes in blood flow in various organs, like the haemodynamic changes, are mostly mediated by the activation of the sympathetic nervous system or circulating epinephrine. The majority either protect against hypoglycaemia or increase substrate delivery to vital organs. The increase in cerebral blood flow increases substrate delivery to the brain. Increasing muscle flow enhances the release and washout of gluconeogenic precursors. The increase in splanchnic blood flow and reduction in splenic blood flow serve to increase hepatic blood flow to maximise hepatic glucose production. Meanwhile, blood is diverted away from organs such as the kidney and spleen, which are not required in the acute response to the metabolic stress.
Functional Changes (Box 1.6) • Sweating: Sweating is mediated by sympathetic cholinergic nerves, although other neurotransmitters such as vasoactive intestinal peptide and bradykinin may also be involved. The activation of the sympathetic innervation of the skin as described above results in the sudden onset of sweating. Sweating is one of the first physiological responses to occur during hypoglycaemia and can be demonstrated within ten minutes of achieving a blood glucose of 2.5 mmol/l (Maggs et al., 1994). It coincides with the onset of other measures of autonomic activation such as an increase in heart rate and tremor (Figure 1.9). • Tremor: Trembling and shaking are characteristic features of hypoglycaemia and result from an increase in physiological tremor. The rise in cardiac output and vasodilatation occurring during hypoglycaemia increase the level of physiological tremor and this is exacerbated by beta adrenoceptor stimulation associated with increased epinephrine concentrations (Kerr et al., 1990). Since adrenalectomy does not entirely abolish tremor, other components such as the activation of muscle sympathetic activity must be involved.
Box 1.6 ↑ ↑ ↓ ↓ ↑ ↑
Functional changes
Sweating (sudden onset) Tremor Core temperature Intraocular pressure Jejunal activity Gastric emptying
PHYSIOLOGICAL RESPONSES
19
We do not have rights to reproduce this figure electronically
Figure 1.9 Sudden onset of sweating, tremor and increase in heart rate during the induction of hypoglycaemia. Reproduced from Hypoglycaemia and Diabetes: Clinical and Physiological Aspects (eds B.M. Frier and M. Fisher), © 1993 Edward Arnold, by permission of Edward Arnold (Publishers) Ltd
• Temperature: Despite a beta adrenoceptor-mediated increase in metabolic rate, core temperature falls during hypoglycaemia. The mechanisms by which this occurs depend on whether the environment is warm or cold. In a warm environment, heat is lost because of sweating and increased heat conduction from vasodilatation. Hypoglycaemia reduces core temperature by 03 C and skin temperature up to 2 C (depending on the part of the body measured) after 60 minutes (Maggs et al., 1994). Shivering is reduced in the cold, and together with vasodilatation and sweating this causes a substantial reduction in core
20
NORMAL GLUCOSE METABOLISM AND RESPONSES
temperature (Gale et al., 1983). In rats, mortality was increased in animals whose core temperature was prevented from falling during hypoglycaemia (Buchanan et al., 1991). In humans there is anecdotal evidence from subjects undergoing insulin shock therapy that those who had a rise in body temperature showed delayed neurological recovery (Ramos et al., 1968). These findings support the hypothesis that the fall in core temperature reduces metabolic rate, allowing hypoglycaemia to be better tolerated, and thus the changes in body temperature are of survival value. The beneficial effects are likely to be limited, particularly in a cold environment, where the impairment of cerebral function means subjects may not realise they are cold, causing them to be at risk of severe hypothermia. • Other functional changes include a reduction in intraocular pressure, greater jejunal but not gastric motility and inconsistent abnormalities of liver function tests. An increase in gastric emptying occurs during hypoglycaemia (Schvarcz et al., 1995), which may be protective in that carbohydrate delivery to the intestine is increased, enabling faster glucose absorption and reversal of hypoglycaemia.
CONCLUSIONS • Homeostatic mechanisms exist to maintain glucose concentration within narrow limits despite a wide variety of circumstances. • The dependence of the central nervous system on glucose has led to a complex series of biochemical, functional and haemodynamic changes aimed at restoring glucose concentrations, producing symptoms and protecting the body in general, and central nervous system in particular, against the effects of a low blood glucose (Figure 1.10). • Many symptoms of hypoglycaemia result from the activation of the autonomic nervous system and help to warn the individual that blood glucose is low. This encourages the ingestion of carbohydrate, so helping to restore glucose concentrations in addition to counterregulation. • Faster gastric emptying and the changes in regional blood flow which also occur as a result of the activation of the autonomic nervous system increase substrate delivery. • The greater cerebral blood flow increases glucose delivery to the brain (although loss of autoregulation is undesirable), and the increased splanchnic flow results in a greater delivery of gluconeogenic precursors to the liver. • Activation of the autonomic nervous system also increases sweating, and together with the inhibition of shivering, this predisposes to hypothermia, which may be neuroprotective.
ACKNOWLEDGEMENTS We would like to thank Professor Robert Tattersall for reading the chapter and for his helpful suggestions.
REFERENCES
21
NORMAL GLUCOSE HOMEOSTASIS INSULIN : GLUCAGON RATIO Excess Insulin HYPOGLYCAEMIA
DETECTION BY BRAIN GLUCOSE SENSORS
PANCREAS
PITUITARY
ADRENAL
GLUCAGON
(GH and CORTISOL)
CATECHOLAMINES
GASTRIC EMPTYING AND CHANGES IN BLOOD FLOW
COUNTERREGULATION
SYMPTOMS (Autonomic)
(increased substrate production)
ACTIVATION OF AUTONOMIC NERVOUS SYSTEM
CEREBRAL BLOOD FLOW
SWEATING
SPLANCHNIC BLOOD FLOW HYPOTHERMIA
(eating)
RESTORE BLOOD GLUCOSE
INCREASE SUBSTRATE DELIVERY
PROTECTION OF VITAL ORGANS
Figure 1.10 Glucose homeostasis and the correction of hypoglycaemia
REFERENCES Abramson EA, Arky RA, Woeber KA (1966). Effects of propranolol on the hormonal and metabolic responses to insulin induced hypoglycaemia. Lancet ii: 1386–9. Allwood MJ, Hensel H, Papenberg J (1959). Muscle and skin blood flow in the human forearm during insulin hypoglycaemia. Journal of Physiology 147: 269–73. Auer RN, Siesjö BK (1988). Biological differences between ischaemia, hypoglycaemia and epilepsy. Annals of Neurology 24: 699–707.
22
NORMAL GLUCOSE METABOLISM AND RESPONSES
Bearn AG, Bulling BH, Sherlock S (1952). The response of the liver to insulin in normal subjects and in diabetes mellitus: hepatic vein catheterisation studies. Clinical Science 11: 151–64. Bell GI, Karyano T, Buse JB, Burant CF, Takeda T, Lin D et al. (1990). Molecular biology of mammalian glucose transporters. Diabetes Care 13: 198–208. Berne C, Fagius J (1986). Skin sympathetic activity during insulin induced hypoglycaemia. Diabetologia 29: 855–60. Borg MA, Sherwin RS, Borg WP, Tamborlane WV, Shulman GI (1997). Local ventromedial hypothalamus glucose perfusion blocks counterregulation during systemic hypoglycemia in awake rats. Journal of Clinical Investigation 99: 361–5. Boyle PJ, Kempers SE, O’Conner AM, Nagy RJ (1995). Brain glucose uptake and unawareness of hypoglycemia in patients with insulin-dependent diabetes mellitus. New England Journal of Medicine 333: 1726–31. Braatvedt GD, Flynn MD, Stanners A, Halliwell M, Corrall RJM (1993). Splanchnic blood flow in man: evidence for mediation via a -adrenergic mechanism. Clinical Science 84: 201–7. Bryan RM (1990). Cerebral blood flow and energy metabolism during stress. American Journal of Physiology 259: H269–80. Buchanan TA, Cane P, Eng CC, Sipos GF, Lee C (1991). Hypothermia is critical for survival during prolonged insulin induced hypoglycemia in rats. Metabolism 40: 330–4. Cahill GF (1971). Physiology of insulin in man. Diabetes 20: 785–99. Cryer PE (1981). Glucose counterregulation in man. Diabetes 30: 261–4. Enoksson S, Caprio SK, Rife F, Shulman GI, Tamborlane WV, Sherwin RS (2003). Defective activation of skeletal muscle and adipose tissue lipolysis in type 1 diabetes mellitus during hypoglycemia. Journal of Clinical Endocrinology and Metabolism 88: 1503–11. Evans ML, McCrimmon RJ, Flanagan DE, Keshavarz T, Fan X, McNay EC et al. (2004). Hypothalamic ATP-sensitive K+ channels play a key role in sensing hypoglycemia and triggering counterregulatory epinephrine and glucagon responses. Diabetes 53: 2542–51. Fagius J, Niklasson F, Berne C (1986). Sympathetic outflow in human muscle nerves increases during hypoglycemia. Diabetes 35: 1124–9. Fagius J, Berne C (1989). Changes in sympathetic nerve activity induced by 2-deoxy-D-glucose infusion in humans. American Journal of Endocrinology 256: E714–21. Fish HR, Chernow B, O’Brian JT (1986). Endocrine and neurophysiologic responses of the pituitary of insulin-induced hypoglycemia: a review. Metabolism 35: 763–80. Fisher M, Baylis PH, Frier BM (1987). Plasma oxytocin, arginine vasopressin and atrial natriuretic peptide response to insulin-induced hypoglycaemia in man. Clinical Endocrinology 26: 179–85. Fisher M, Gillen G, Hepburn DA, Dargie HJ, Barnett E, Frier BM (1990). Splenic responses to acute insulin-induced hypoglycaemia in humans. Clinical Science 78: 469–74. Francis BH, Ensinck JW (1987). Differential alterations of the circulating prosomatostatin-derived peptides during insulin induced hypoglycemia in man. Journal of Clinical Endocrinology and Metabolism 65: 880–4. Gale EAM, Bennet J, Macdonald IA, Holst JJ, Mathews JA (1983). The physiological effects of insulin-induced hypoglycaemia in man: responses at differing levels of blood glucose. Clinical Science 65: 262–71. Galassetti P, Neill AR, Tate D, Ertl AC, Wasserman DH, Davis SN (2001). Sexual dimorphism in counterregulatory responses to hypoglycemia after antecedent exercise. Journal of Clinical Endocrinology and Metabolism 86: 3516–24. Garber AJ, Cryer PE, Santiago JV, Hammond MW, Pagliara AS, Kipnis DM (1976). The role of adrenergic mechanisms in the substrate and hormonal response to insulin-induced hypoglycemia in man. Journal of Clinical Investigation 58: 7–15. Gerich JE (1988). Glucose counterregulation and its impact in diabetes mellitus. Diabetes 37: 1608–17.
REFERENCES
23
Hamilton-Wessler M, Bergman RN, Halter JB, Watanabe RM, Donovan CM (1994). The role of liver glucosensors in the integrated sympathetic response induced by deep hypoglycemia in dogs. Diabetes 43: 1052–60. Heller SR, Macdonald IA (1996). The measurement of cognitive function during acute hypoglycaemia: experimental limitations and their effect on the study of hypoglycaemia unawareness. Diabetic Medicine 13: 607–15. Heptulla RA, Tamborlane WV, Ma TY-Z, Rife F, Sherwin RS (2001). Oral glucose augments the counterregulatory hormone response during insulin-induced hypoglycemia in humans. Journal of Clinical Endocrinology and Metabolism 86: 645–8. Hilsted J (1993). Cardiovascular changes during hypoglycaemia. Clinical Physiology 13: 1–10. Jackson PA, Cardin S, Coffey CS, Neal DW, Allen EJ, Penazola AR et al. (2000). Effect of hepatic denervation on the counterregulatory response to insulin-induced hypoglycemia in the dog. American Journal of Endocrinology 279: E1249–57. Jungman E, Konzog C, Holl E, Fassibinder W, Schoffling K (1989). Effect of a human atrial naturetic peptide on blood glucose concentrations and hormone stimulation during insulin-induced hypoglycaemia in healthy man. European Journal of Clinical Pharmacology 36: 593–7. Katz SH, Dhariwal APS, McCann SM (1967). Effects of hypoglycaemia on the content of pituitary growth hormone (GH) and hypothalamic growth hormone releasing factor (GNRH) in the rat. Endocrinology 81: 333–9. Kerr D, Macdonald IA, Tattersall RB (1989). Influence of duration of hypoglycemia on the hormonal counterregulatory response in normal subjects. Journal of Clinical Endocrinology and Metabolism 68: 118–22. Kerr D, Macdonald IA, Heller SR, Tattersall RB (1990). A randomised double-blind placebo controlled trial of the effects of Metoprolol CR, Atenolol and Propranolol LA on the physiological responses to hypoglycaemia in the non-diabetic. British Journal of Clinical Pharmacology 29: 685–94. Levin BE, Routh VH, Kang L, Sanders NM, Dunn-Meynell AA (2004). Neuronal glucosensing: What do we know after 50 years? Diabetes 53: 2521–8. Maggs DG, Scott AR, Macdonald IA (1994). Thermoregulatory responses to hyperinsulinemic hypoglycemia and euglycemia in humans. American Journal of Physiology 267: R1266–72. McCall AL, Fixman LB, Fleming N, Tornheim K, Chick W, Ruderman ND (1986). Chronic hypoglycemia increases brain glucose transport. American Journal of Endocrinology 251: E442–5. McCall AL (1993). Effects of glucose deprivation on glucose metabolism in the central nervous system. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M, eds. Edward Arnold, London: 56–71. Mitrakou A, Ryan C, Veneman T, Mokan M, Jenssen T, Kiss I et al. (1991). Hierarchy of glycemic thresholds for counterregulatory hormone secretion, symptoms and cerebral dysfunction. American Journal of Endocrinology 266: E67–74. Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrara MG, Cahill GF Jr (1967). Brain metabolism during fasting. Journal of Clinical Investigation 46: 1589–95. Patrick AW, Hepburn DA, Craig KJ, Thompson I, Swainson CD, Frier BM (1989). The effects of acute insulin-induced hypoglycaemia on renal function in normal human subjects. Diabetic Medicine 6: 703–8. Ramos E, Zorilla E, Hadley WB (1968). Fever as a manifestation of hypoglycemia. Journal of the American Medical Association 205: 590–2. Schvarcz E, Palmér M, Åman J, Berne C (1995). Hypoglycemia increases the gastric emptying rate in healthy subjects. Diabetes Care 18: 674–6. Sieber FE, Traysman RJ (1992). Special Issues: Glucose and the brain. Critical Care Medicine 20: 104–14. Siegal AM, Kreisberg RA (1975). Metabolic homeostasis: insulin-glucagon interactions. In: Diabetes Mellitus (4th edition). Sussman KE and Metz RJS, eds. American Diabetes Association, New York: 29–35.
24
NORMAL GLUCOSE METABOLISM AND RESPONSES
Sokaloff (1989). Circulation and energy metabolism of the brain. In: Basic Neurochemistry. Siegel G, Agranoff B, Albers RW and Molinoff P, eds. Raven Press, New York: 565–90. Trovati M, Massucco P, Mularoni E, Cavalot F, Anfossi G, Matiello L, Emanuelli G (1988). Insulininduced hypoglycaemia increases plasma concentrations of angiotensin II and does not modify atrial naturetic polypeptide secretion in man. Diabetologia 31: 816–20 Wieloch T (1985). Hypoglycemia induced neuronal damage prevented by an N-Methyl D-Aspartate antagonist. Science 230: 681–3.
2 Symptoms of Hypoglycaemia and Effects on Mental Performance and Emotions Ian J. Deary
INTRODUCTION This chapter describes the symptoms that are perceived during acute hypoglycaemia, and the changes in mental functions and emotions that occur during this metabolic state. The most obvious benefit to a person of knowing about the symptoms of hypoglycaemia is the ability to recognise the onset of a hypoglycaemic episode as early as possible. This is of key importance in informing and educating people with diabetes. Moreover, if a person with diabetes understands which mental functions are affected by hypoglycaemia he or she can judge which activities may be most threatened in this state.
SYMPTOMS OF HYPOGLYCAEMIA Identifying the Symptoms The physiological responses to hypoglycaemia are described in Chapter 1. The response to hypoglycaemia results in physical symptoms, which raises several questions (McAulay et al., 2001b). Can we compile a comprehensive list of symptoms of hypoglycaemia? Which are the more common symptoms of hypoglycaemia? Are there early warning symptoms of hypoglycaemia? Do people differ in how quickly and accurately they detect or recognise hypoglycaemia? Do people differ in the set of symptoms of hypoglycaemia they experience? How can individuals distinguish the symptoms of hypoglycaemia from other bodily changes? The total symptom complex The most basic question is: what symptoms do people report when they develop hypoglycaemia? In humans the symptoms associated with hypoglycaemia were first recorded when insulin became available for the treatment of diabetes (Fletcher and Campbell, 1922). A list of characteristic symptoms was described (Table 2.1). It was noted: that some symptoms
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
47–84 32–78 28–71 24–60 39–49 8–62 7–41 10–39 11–41 24–36 10–44
5–20 31–75 38–46 16–33 13–53
Tingling around the mouth Dizziness
Headache Anxiety
Nausea Difficulty concentrating Tiredness Drowsiness Confusion
Percentage (minimum–maximum) of people reporting the given symptom as associated with hypoglycaemia (after Hepburn, 1993)
Sweating Trembling Weakness Visual disturbance Hunger Pounding heart Difficulty with speaking
Symptoms associated with hypoglycaemia as derived from population studies (after Hepburn, 1993)
Table 2.1 Common symptoms associated with hypoglycaemia
Nervousness, anxiety, excitement, emotional upset – – – – Confusion, disorientation, ‘goneness’ Pallor Incoordination Feeling of heat or cold Emotional instability
Sweating Tremulousness Weakness Diplopia Excessive hunger Change in pulse rate Dysarthria, sensory and motor aphasia – Vertigo, faintness, syncope
Symptoms of hypoglycaemia as noted by Fletcher and Campbell (1922)
Slowed thinking (70)
Uncoordinated (75) Cold sweats (40)
Drowsy–sleepy (40)
– Difficulty concentrating (80)
Headache (30) Nervous/tense (65)
Numb lips (50) Light-headed/dizzy (60)
Sweating (80) Trembling (65) Fatigue/weak (70) Blurred vision (20) Hunger (60) Pounding heart (55) Slurred speech (40)
Symptoms (and percentage [to nearest 5%]) of people endorsing symptoms as associated with hypoglycaemia; after Cox et al., 1993a, Figure 2)
SYMPTOMS OF HYPOGLYCAEMIA
27
appeared before others during hypoglycaemia; that the blood glucose level at which subjects became aware of hypoglycaemia was characteristic for the individual; that there were large individual differences in the levels of blood glucose at which awareness of hypoglycaemia commenced; and that the preceding blood glucose concentration could affect the onset of symptoms. Lists of common symptoms of hypoglycaemia have been compiled from more recent research. Hepburn (1993) summarised eight population studies of the symptoms of hypoglycaemia experienced by adults and children with insulin-treated diabetes, and Cox et al. (1993a) also produced a list of symptoms (Table 2.1). It is evident that the three lists of symptoms do not differ greatly, and that Fletcher and Campbell’s early report (1922) had captured many of the symptoms found in subsequent, more structured investigations. However, their report omitted to mention some symptoms such as tiredness, drowsiness and difficulty concentrating, though it did include others – such as pallor (a sign rather than a symptom), incoordination and feelings of temperature change – that are emphasised by other researchers as regularly perceived symptoms. Table 2.1 establishes a useful group of symptoms that are commonly reported in hypoglycaemia. The way we ask people to describe their symptoms of hypoglycaemia can alter what they tell us. The rank order of symptoms alters considerably if patients are asked to indicate the relevance of each symptom rather than merely to identify that the symptom is associated with hypoglycaemia (Cox et al., 1993a). With regard to the criterion of relevance, the most useful symptoms in detecting hypoglycaemia are as follows: • sweating; • trembling; • difficulty concentrating; • nervousness, tenseness; • light-headedness, dizziness.
The initial symptoms Another important question is: which hypoglycaemic symptoms appear early during an episode? The symptoms of hypoglycaemia that appear first and offer early warning of the onset of hypoglycaemia (Hepburn, 1993) are as follows: • trembling; • sweating; • tiredness; • difficulty concentrating; • hunger. This knowledge is obviously useful for the prompt detection and treatment of hypoglycaemia.
28
SYMPTOMS OF HYPOGLYCAEMIA
The Validity of Symptom Beliefs The individuality of hypoglycaemic symptom clusters A great deal of the interest in symptoms of hypoglycaemia has been stimulated by concerns about patient education. It is helpful to let patients know the range of symptoms found in hypoglycaemia and to inform them of the early warning symptoms reported by other people with diabetes, much as we all tend to know the range of symptoms that are experienced with the common cold. Many surveys and laboratory studies have shown that people differ considerably in the symptoms of hypoglycaemia they experience (Cox et al., 1993a). In addition to learning the generally reported symptoms, individuals with diabetes should be encouraged to learn about their own typical symptoms of hypoglycaemia. Correctly interpreting symptoms as representing hypoglycaemia Symptoms of hypoglycaemia do not appear on top of the bodily equivalent of a blank sheet of paper. Sometimes we experience symptoms when there is nothing wrong with our bodily functions; on the other hand, sometimes we fail to notice any symptoms when the body is malfunctioning. The alert person with diabetes who is on the lookout for hypoglycaemia must make two sorts of decisions. First, symptoms of hypoglycaemia must be detected and correctly identified. It would be dangerous for a patient to ignore symptoms of hypoglycaemia because he or she thought they were related to something else. Second, symptoms that have nothing to do with hypoglycaemia must be excluded. Unwanted hyperglycaemia could occur if patients treated themselves for hypoglycaemia when the symptoms had another cause. These two main types of error are a failure to treat hypoglycaemia when blood glucose is low, and inappropriate treatment when blood glucose is acceptable or high (Cox et al., 1985; 1993a) (Figure 2.1). Blood glucose concentration – symptom report correlations Do patients’ reports of symptoms of hypoglycaemia bear any relation to their concurrent blood glucose concentrations? After all, the principal aim of educating people to be
Figure 2.1 Consequences of correct and incorrect perception of hypoglycaemic symptoms
SYMPTOMS OF HYPOGLYCAEMIA
29
aware of symptoms of hypoglycaemia is that they become alert to low and potentially dangerous levels of blood glucose. To answer the above question some researchers have employed a field study approach where people are invited to list any symptoms they are experiencing, and then measure and record their blood glucose concentration several times a day for weeks. As a result of these studies it is known that each person has some symptoms that are most reliably associated with their actual blood glucose concentrations (Pennebaker et al., 1981). Some of the symptoms that people report during hypoglycaemia are more closely related to their actual blood glucose concentrations than are others, and if we can identify each individual’s most informative symptoms, we can instruct people to pay more attention to them. The following symptoms are most consistently associated with actual blood glucose concentrations (Pennebaker et al., 1981): • hunger (in 53% of people); • trembling (in 33%); • weakness (in 27%); • light-headedness (in 20%); • pounding heart and fast heart rate (both 17%). The same symptoms are not informative for everyone. There were 27% of people for whom weakness was significantly associated with hypoglycaemia, but there were 7% in whom it was a good symptom of hyperglycaemia! Most people reported more than three symptoms that were strongly associated with the measured blood glucose concentration. It is evident that an individual’s symptoms are idiosyncratic. If we can help a patient to identify the symptoms of hypoglycaemia peculiar to him or her, which relate to actual blood glucose concentrations, then, by attending to these symptoms, the person should be especially accurate in recognising hypoglycaemia. People who have one or more reliable symptom(s) of hypoglycaemia correctly recognise half of their episodes of hypoglycaemia (defined as a blood glucose less than 3.95 mmol/l [Cox et al., 1993a]). Those who have four or more reliable symptoms recognise a blood glucose below 3.95 mmol/l three-quarters of the time. The field study method has suggested that attention to the following symptoms was particularly useful in detecting actual low blood glucose concentrations: • nervousness/tenseness; • slowed thinking; • trembling; • light-headedness/dizziness; • difficulty concentrating; • pounding heart; • lack of co-ordination.
30
SYMPTOMS OF HYPOGLYCAEMIA
Classifying Symptoms of Hypoglycaemia Until now the symptoms of hypoglycaemia have been treated as a homogeneous whole. Can these symptoms be divided into different groups? Hypoglycaemia has effects on more than one part of the body, and the symptoms of hypoglycaemia reflect this. First, the direct effects of a low blood glucose concentration on the brain – especially the cerebral cortex – cause neuroglycopenic symptoms. Second, autonomic symptoms result from activation of parts of the autonomic nervous system. Finally, there may be some non-specific symptoms that are not directly generated by either of these two mechanisms. It is only recently that scientific investigations have taken place to confirm the idea that these separable groups of hypoglycaemic symptoms exist. As suggested above, there are at least two distinct groups of symptoms during the body’s reaction to hypoglycaemia (Hepburn et al., 1991): • Autonomic, with symptoms such as trembling, anxiety, sweating and warmness. • Neuroglycopenic, with symptoms such as drowsiness, confusion, tiredness, inability to concentrate and difficulty speaking. This information can assist with patient education by supplying evidence for separable groups of symptoms, and by indicating which symptoms belong to each group. Some neuroglycopenic symptoms, such as the inability to concentrate, weakness and drowsiness, are among the earliest detectable symptoms, but patients tend to rely more on autonomic symptoms when detecting the onset of hypoglycaemia. Paying more attention to the potentially useful, early neuroglycopenic symptoms could help with the early detection of hypoglycaemia. Similar groups of symptoms of hypoglycaemia have been discovered by asking people to recall the symptoms they typically noticed during hypoglycaemia. However, in addition to the two groups described above, a general feeling of malaise is added (Deary et al., 1993): • Autonomic: e.g. sweating, palpitations, shaking and hunger. • Neuroglycopenic: e.g. confusion, drowsiness, odd behaviour, speech difficulty and incoordination. • General malaise: e.g. headache and nausea. These 11 symptoms are so reliably reported by people and so clearly separable into these three groups, that they are used as the ‘Edinburgh Hypoglycaemia Scale’ (Deary et al., 1993). Table 2.2 shows how different researchers have found similar groups of autonomic and hypoglycaemic symptoms. In addition to the above studies that used patients’ self-reported symptoms, physiological studies have also confirmed that the symptoms of hypoglycaemia can be divided into autonomic and neuroglycopenic groups. Symptoms such as sweating, hunger, pounding heart, tingling, nervousness and feeling shaky/tremulous (autonomic symptoms) can be reduced or even prevented by drugs that block neurotransmission within the autonomic nervous system (Towler et al., 1993), confirming that these symptoms are caused by the autonomic response to hypoglycaemia. Symptoms such as warmth, weakness, difficulty thinking/confusion,
SYMPTOMS OF HYPOGLYCAEMIA
31
Table 2.2 Different authors’ lists of autonomic and neuroglycopenic symptoms of hypoglycaemia Autonomic Deary et al. (1993)
Towler et al. (1993)
Neuroglycopenic Weinger et al. (1995)
Deary et al. (1993)
Sweating
Sweaty
Sweating
Confusion
Palpitation
Heart pounding
Drowsiness
Shaking
Shaky/tremulous
Pounding heart, fast pulse Trembling
Hunger
Hungry Tingling Nervous/anxious
Tense Breathing hard
Odd behaviour Speech difficulty Incoordination
Towler et al. (1993) Difficulty thinking/ confused Tired/drowsy
Difficulty speaking
Weak Warm
feeling tired/drowsy, feeling faint, difficulty speaking, dizziness and blurred vision (neuroglycopenic symptoms) are not prevented by drugs that block the autonomic nervous system. Therefore, neuroglycopenic symptoms are not mediated via the autonomic nervous system and are thought to be caused by the direct effect of glucose deprivation on the brain. This type of research has also observed that people tend to rely on autonomic symptoms to detect hypoglycaemia, even when neuroglycopenic symptoms are just as prominent (Towler et al., 1993). Once more, this suggests that more emphasis should be placed on education of the potential importance of neuroglycopenic symptoms for the early warning of hypoglycaemia. Symptoms might gather into slightly different groupings depending on the situation. The symptom groupings in the Edinburgh Hypoglycaemia Scale were developed from diabetic patients’ retrospective reports. However, when people are asked to rate the same group of symptoms during acute, experimentally-induced moderate hypoglycaemia, a slightly different pattern emerges (McCrimmon et al., 2003). In Table 2.3 there is an autonomic grouping, and the single neuroglycopenia group has divided into two symptom groups: one with mostly cognitive symptoms and the other with more general symptoms. This division probably arose because the subjects in the studies used to form Table 2.3 were engaged in cognitive tasks Table 2.3 Symptom groupings of the Edinburgh Hypoglycaemia Scale during experimentally-induced hypoglycaemia Neuroglycopenic symptoms Cognitive dysfunction
Neuroglycopenia
Autonomic symptoms
Inability to concentrate Blurred vision Anxiety Confusion Difficulty speaking Double vision
Drowsiness Tiredness Hunger Weakness
Sweating Trembling Warmness
SYMPTOMS OF HYPOGLYCAEMIA
32
during the period of hypoglycaemia. Therefore, they would be especially aware of cognitive shortcomings, making this group of symptoms more prominent and coherent.
Symptoms in Children and Older People Children often have difficulty in recognising symptoms of hypoglycaemia, and they show marked variability in symptoms between episodes of hypoglycaemia (Macfarlane and Smith, 1988). Trembling and sweating are often the first symptoms recognised by children. From interviews with the parents of children (aged up to 16 years) with type 1 diabetes, and with some of their children, more is known about the frequency of symptoms of hypoglycaemia in children (McCrimmon et al., 1995; Ross et al., 1998) (Table 2.4). The most frequently reported sign that parents observed was pallor (noted by 88%). The parents Table 2.4 Symptoms of hypoglycaemia in children (derived from Ross et al., 1998) Frequency of rating (%) Symptom
Parents’ reports
Tearful Headache Irritable Uncoordinated Naughty Weak Aggressive Trembling Sleepiness Nightmares Sweating Slurred speech Blurred vision Tummy pain Feeling sick Hungry Yawning Odd behaviour Warmness Restless Daydreaming Argumentative Pounding heart Confused Tingling lips Dizziness Tired Feeling awful
73 73 85 62 47 79 75 79 63 33 76 53 52 67 63 74 48 65 57 61 70 64 21 75 20 66 83 92
Children’s reports 47 65 65 56 31 83 62 88 69 19 73 45 55 41 53 84 45 50 68 57 48 50 44 70 24 87 76 79
Correlation between parents’ and children’s intensity ratingsa 040d 033d 016 018 023b 021b 026c 025b 027c 033d 028c 028c 030c 036d 032c 019 020b 022b 013 021b 014 021b 002 041d −001 028c 026c 020
a
Correlations: p of z (corrected for ties) 1.0 represents perfect agreement; 0 represents no agreement.
b
p < 005, c p < 001, d p < 0001.
SYMPTOMS OF HYPOGLYCAEMIA
33
frequently reported symptoms of behavioural disturbance such as irritability, argumentativeness and aggression. This latter group of symptoms is not prominent in adults, although the Edinburgh Hypoglycaemia Scale includes ‘odd behaviour’ as an adult neuroglycopenic symptom. Others had previously noted the prominence of symptoms such as irritability, aggression and disobedience in the parents’ reports of their children’s symptoms of hypoglycaemia (Macfarlane and Smith, 1988; Macfarlane et al., 1989). Parents tend to under-report the subjective symptoms of hypoglycaemia, such as weakness and dizziness, but generally there is good agreement between parents and their children about the most prominent symptoms of childhood hypoglycaemia (McCrimmon et al., 1995; Ross et al., 1998). Separate groups of autonomic and neuroglycopenic symptoms were not found in children with type 1 diabetes (McCrimmon et al., 1995; Ross et al., 1998). These symptoms are reported together by children and are not distinguished as separate groups, whereas the group of symptoms related to behavioural disturbance is clearly reported as a distinct group. In a refinement of the earlier study by McCrimmon et al. (1995), Ross et al. (1998) found that parents could distinguish between autonomic and neuroglycopenic symptoms. People with insulin-treated type 2 diabetes report symptoms during hypoglycaemia that separate into autonomic and neuroglycopenic groups (Henderson et al., 2003). Elderly patients with type 2 diabetes treated with insulin commonly report neurological symptoms of hypoglycaemia which may be misinterpreted as features of cerebrovascular disease, such as transient ischaemic attacks (Jaap et al., 1998). The age-specific differences in the groups of hypoglycaemic symptoms, classified using statistical techniques (Principal Component Analysis), are shown in Table 2.5. Health professionals and carers who are involved in the treatment and education of diabetic patients should be aware of which symptoms are common at either end of the age spectrum.
From Symptom Perception to Action People with diabetes are better at estimating their blood glucose in natural, everyday situations, as opposed to clinical laboratory settings (Cox et al., 1985). In some ways this is surprising as natural hypoglycaemia often occurs at a time when it is unexpected. In this situation, attention toward symptoms will not be as actively directed toward detection as in the laboratory setting where it is usually anticipated. Furthermore, hypoglycaemia in everyday life occurs on the background of other bodily feelings and must be separated from other causes of the same symptoms. For example, exercise and various acute illnesses can provoke sweating in people with diabetes, independently of their association with hypoglycaemia. In a real-life situation a person must detect symptoms of hypoglycaemia and then interpret them. Failure to detect symptoms can lead to a failure to treat hypoglycaemia, but detection Table 2.5 Classification of symptoms of hypoglycaemia using Principal Components Analysis in patients with insulin-treated diabetes depending on age group Children (pre pubertal)
Adults
Elderly
Autonomic/neuroglycopenic
Autonomic Neuroglycopenic
Autonomic Neuroglycopenic
Behavioural
Non-specific malaise
Neurological
34
SYMPTOMS OF HYPOGLYCAEMIA
without the correct interpretation of the cause of the symptoms is equally dangerous. Furthermore, it is obvious that someone who does interpret symptoms correctly as being caused by hypoglycaemia, but who does not take action to treat the low blood glucose, will be at the same risk. These and other steps toward the avoidance of severe hypoglycaemia demonstrate the key role of education about symptoms in people with diabetes (Gonder-Frederick et al., 1997). The psycho-educational programmes of Blood Glucose Awareness Training (BGAT; Schachinger et al., 2005) and Hypoglycemia Anticipation, Awareness and Treatment Training (HAATT; Cox et al., 2004) have led to better recognition of hypoglycaemic states and reduced frequency of hypoglycaemia. Symptom generation Figure 2.2 outlines the stages that intervene between low blood glucose occurring in an individual and the implementation of effective treatment (Gonder-Frederick et al., 1997). It is interesting to note the importance of behavioural factors in generating states of low blood glucose. Episodes of low blood glucose are most likely to come about because of changes in routine aspects of diabetes management, such as taking extra insulin, eating less food or taking more exercise (Clarke et al., 1997). These factors predict more than 85% of episodes of hypoglycaemia in people with diabetes. In the presence of an intact physiological response to low blood glucose, autonomic and neuroglycopenic symptoms, and symptoms of general malaise, are generated (Figure 2.2). The degree of hypoglycaemia, the person’s quality of glycaemic control and any recent episodes of hypoglycaemia may all affect the magnitude of the body’s physiological response. Recent, preceding hypoglycaemia can reduce the symptomatic and counterregulatory hormonal responses to subsequent hypoglycaemia, resulting in a diminished awareness of symptoms. This effect of ‘antecedent’ hypoglycaemia is described in Chapter 7. Gender does not appear to influence the symptomatic response to hypoglycaemia (Geddes et al., 2006). At the second stage in Figure 2.2 comes the actual generation of physical symptoms. Among the variables that can influence this stage is the prior ingestion of caffeine, which has been shown to enhance the intensity of the autonomic and neuroglycopenic symptoms of experimentally induced hypoglycaemia (Debrah et al., 1996) (see Chapter 5). Caffeine may act by increasing the intensity of symptoms of hypoglycaemia to perceptible levels, much as a magnifying glass enables one to read otherwise too-small print. Symptom detection The occurrence of physiological changes in the body does not guarantee that a person will detect symptoms (Gonder-Frederick et al., 1997). If attention is directed to physical changes, people are more likely to detect symptoms than if their attention is held elsewhere. Everyone has had the experience of feeling less discomfort, and being less likely to detect a physical symptom, when being distracted by something diverting. The personal relevance of the symptom may affect detection; for example, a person with heart disease may be very likely to detect palpitations (Cox et al., 1993a). The activity of the person at the time of the physiological change is obviously important. Hypoglycaemic symptoms will be more obvious to the person engaged in active mental effort (McCrimmon et al., 2003), such as sitting an examination, than to the person relaxing and watching television. A doctor engaged in microsurgery may be very sensitive to the onset of tremor. In one laboratory study the
SYMPTOMS OF HYPOGLYCAEMIA
35
Figure 2.2 A model for the occurrence and avoidance of severe hypoglycaemia (after GonderFrederick et al., 1997). Note on the right-hand side of each stage the factors that affect its occurrence
people who had higher anxiety levels were better at detecting symptoms of hypoglycaemia (Ryan et al., 2002). The autonomic symptoms of hypoglycaemia are often emphasised in the detection of hypoglycaemia. However, a strong case can be made for an equal emphasis on neuroglycopenic symptoms (Gonder-Frederick et al., 1997; McCrimmon et al., 2003) because: • performance on mental tasks deteriorates during hypoglycaemia, and subjective awareness of this decrement begins at very mild levels of hypoglycaemia • the difference in glycaemic thresholds for the onset of autonomic and neuroglycopenic symptoms is so small that it is unlikely to be detected when blood glucose declines rapidly
SYMPTOMS OF HYPOGLYCAEMIA
36
• neuroglycopenic symptoms are as strongly related to actual blood glucose concentrations as are autonomic symptoms (Cox et al., 1993a) • people with insulin-treated diabetes cite autonomic and neuroglycopenic symptoms with equal frequency as the primary warning symptoms of hypoglycaemia (Hepburn et al., 1991). Low blood glucose detection – symptom interpretation The correct detection of a symptom of hypoglycaemia does not always lead to correct interpretation (Figure 2.2). After correctly detecting relevant symptoms, people fail to detect about 26% of episodes of low blood glucose (Gonder-Frederick et al., 1997). There are several factors that could break a perfect relationship between detection and recognition, and some are discussed below. However, it should be appreciated that symptom detection (internal cues) is not mandatory for the detection of low blood glucose. Self-testing of blood glucose or the information of family members (external cues) can lead to the successful recognition of hypoglycaemia without the patient having detected the episode by symptomatic perception (see Box 2.1). Correct knowledge about symptoms of hypoglycaemia is necessary for the detection of low blood glucose. The lack of such knowledge among elderly people with diabetes in particular gives cause for concern (Mutch and Dingwall-Fordyce, 1985). Of 161 diabetic people between ages 60 and 87, all of whom were injecting insulin or taking a sulphonylurea, only 22% had ever been told the symptoms of hypoglycaemia and 9% knew no symptoms at all! The percentages of the insulin-treated diabetic patients who knew that the following symptoms were associated with hypoglycaemia were as follows: • sweating (82%); • palpitations (62%); • confusion (53%); • hunger (51%); • inability to concentrate (50%); • speech problems (41%); • sleepiness (33%).
Box 2.1
Identification of hypoglycaemia
• Internal cues – autonomic, neuroglycopenic and non-specific symptoms • External cues – relationship of insulin injection to meals, exercise and experience of self-management • Blood glucose monitoring • Information from observers (e.g. relatives, friends, colleagues)
ACUTE HYPOGLYCAEMIA AND COGNITIVE FUNCTIONING
37
In the midst of this ignorance, much hypoglycaemia may not be treated because of a lack of knowledge of the symptoms of hypoglycaemia which would aid their recognition. Most, if not all, of the symptoms of hypoglycaemia can be explained by other physical conditions. Therefore, correct symptom detection may be usurped by incorrect attribution of the cause. For example, having completed some strenuous activity, an athlete may attribute the symptoms of sweating and palpitations to physical exertion. An obvious problem in detecting a low blood glucose is the fact that the organ responsible for the detection and interpretation of symptoms – the brain, especially the cerebral cortex – is impaired. Thus, impaired concentration and lowered consciousness levels can beget even more severe hypoglycaemia.
Symptom scoring systems The controversy about the effect of human insulin on symptom awareness (Chapter 7) stimulated the development of scoring systems for hypoglycaemia to allow comparative studies between insulin species. This produced scoring systems such as the Edinburgh Hypoglycaemia Scale (Deary et al., 1993), and any such system must be validated for research application. It is important to note that the nature and intensity of individual symptoms are as important as, if not more important than, the number of symptoms generated by hypoglycaemia. The concepts involved are discussed in detail by Hepburn (1993). More information on the symptoms of hypoglycaemia is provided by McAulay et al. (2001b).
ACUTE HYPOGLYCAEMIA AND COGNITIVE FUNCTIONING Symptoms are subjective reports of bodily sensations. With respect to hypoglycaemia some of these reports – especially neuroglycopenic symptoms – pertain to altered cognitive (mental ability) functioning. Do reports of ‘confusion’ and ‘difficulty thinking’ (Table 2.2) concur with objective mental test performance in hypoglycaemia? Before experimental hypoglycaemia became an accepted investigative tool in diabetes, expert clinical observers noted impairments of cognitive functions despite clear consciousness during hypoglycaemia (Fletcher and Campbell, 1922; Wilder, 1943). Cognitive functions include the following sorts of mental activity: orientation and attention, perception, memory (verbal and non-verbal), language, construction, reasoning, executive function and motor performance. Early studies (Russell and Rix-Trot, 1975) established that the following abilities become disrupted below blood glucose levels of about 3.0 mmol/l: • fine motor co-ordination; • mental speed; • concentration; • some memory functions.
38
SYMPTOMS OF HYPOGLYCAEMIA
The hyperinsulinaemic glucose clamp technique allows more controlled experiments of acute hypo- and hyperglycaemia. However, although this technique is used in most studies of cognitive function in hypoglycaemia, it does not mimic the physiological or temporal characteristics of ‘natural’ or intercurrent episodes of hypoglycaemia experienced by people with type 1 diabetes. From laboratory experiments using the glucose clamp technique it was found that blood glucose concentrations between 3.1 and 3.4 mmol/l caused the following effects (Holmes, 1987; Deary, 1993): • Slowed reaction times (this experiment involves making a fast response when a light appears on a computer screen. Hypoglycaemia had more effect on reaction times when the reaction involved making a decision). • Slowed mental arithmetic. • Impaired verbal fluency (in this test one has to think of words beginning with a given letter, probably involving the frontal lobes of the brain). • Impaired performance in parts of the Stroop test (in this test one has to read aloud a series of ink colours when words are printed in a different colour from that of the name, e.g. the word RED printed in green ink). Some mental functions were spared during hypoglycaemia, for example: • simple motor (like the speed of tapping) and sensory skills; • the speed of reading words aloud. By 1993 over 16 studies had investigated cognitive functions during acute and mild– moderate hypoglycaemia (Deary, 1993). The levels of blood glucose ranged from 2.0 to 3.7 mmol/l. The way that hypoglycaemia was induced varied among studies, as did the methods of blood sampling (e.g. arterialised or venous blood). Moreover, the ability levels of the people in different studies varied, and there was much heterogeneity in the test batteries used to assess mental performance. An authoritative statement as to the mental functions disrupted during hypoglycaemia is still not possible. However, in at least one or more of the studies a number of tests were significantly impaired during hypoglycaemia (Box 2.2). Few areas of mental function are preserved at normal levels during acute hypoglycaemia. There is a general dampening of many abilities that involve conscious mental effort. In the face of so many deleterious effects, what mental functions remain intact during acute hypoglycaemia? At blood glucose concentrations similar to those indicated above, the following mental tests are not significantly impaired: • finger tapping; • forward digit span (repeating back a list of numbers in the same order); • simple reaction time; • elementary sensory processing.
ACUTE HYPOGLYCAEMIA AND COGNITIVE FUNCTIONING
Box 2.2
39
Cognitive function tests impaired during acute hypoglycaemia
• Trail making (involving visual scanning and mental flexibility) • Digit symbol (speed of replacing a list of numbers with abstract codes) • Reaction time (especially involving a decision) • Mental arithmetic • Verbal fluency • Stroop test • Grooved pegboard (a test of fine manual dexterity) • Pursuit rotor (a test of eye–hand co-ordination) • Letter cancellation (striking out occurrences of a given letter in a page of letters) • Delayed verbal memory • Backward digit span (repeating back a list of numbers backwards) • Story recall
Thus tests which involve speeded responses and which are more cognitively complex and attention-demanding tend to show impairment during hypoglycaemia (Deary, 1993). Heller and Macdonald (1996) have concluded that: • even quite severe degrees of hypoglycaemia do not impair simple motor functions; • choice reaction time (where a mental decision of some kind is needed before reacting to a stimulus) is affected at higher blood glucose concentrations more than simple reaction time; • speed of responding is sometimes slowed in a task in which accuracy is preserved; • many aspects of mental performance become impaired when blood glucose falls below about 3.0 mmol/l; • there are important individual differences; some people’s mental performance is already impaired above a blood glucose of 3.0 mmol/l, whereas others continue to function well at lower levels; • the speed of response of the brain in making decisions slows down during hypoglycaemia (Tallroth et al., 1990; Jones et al., 1990); • it can take as long as 40 to 90 minutes after blood glucose returns to normal for the brain to recover fully (Blackman et al., 1992; Lindgren et al., 1996).
40
SYMPTOMS OF HYPOGLYCAEMIA
Figure 2.3 Cognitive effects of hypoglycaemia. After Deary (1998). This Article was published in Diabetes Annual 11, SM Marshall, PD Home and RA Rizza (eds), 97–118, Copyright Elsevier 1998
Determining whether mental performance is impaired at all during mild to moderate hypoglycaemia, while a person is still fully conscious, is only the beginning of this line of investigation. The next question to ask is whether some particular functions are more susceptible and some less so? Figure 2.3 encapsulates this problem and illustrates three other important questions about the cognitive effects of acute hypoglycaemia: 1. What factors affect the degree of cognitive impairment during hypoglycaemia, other than the level of blood glucose? 2. Do impairments in laboratory cognitive tasks have a bearing on mental performance in real life? 3. Which basic brain functions are disturbed during acute hypoglycaemia?
Influences on the Degree of and Threshold for Cognitive Dysfunction During Acute Hypoglycaemia Although, on average, impairment of mental performance is worse during hypoglycaemia, some people do not change or may even improve (Pramming et al., 1986; Hoffman et al.,
ACUTE HYPOGLYCAEMIA AND COGNITIVE FUNCTIONING
41
1989). It is not yet certain whether such individual differences in responses are stable (Gonder-Frederick et al., 1994; Driesen et al., 1995). The following factors might increase a person’s degree of cognitive impairment during acute hypoglycaemia: • male sex (Draelos et al., 1995; but this is disputed for people with type 2 diabetes by Bremer et al., 2006); • impaired hypoglycaemia awareness (Gold et al., 1995b); • type 1 diabetes (Wirsen et al., 1992); • high IQ (Gold et al., 1995a). Does glycaemic control affect the cognitive impact of hypoglycaemia? People with type 1 diabetes on intensified insulin therapy attain glycaemic control that is nearer to normal than most people treated with conventional insulin treatment. As a result, the frequency of severe hypoglycaemia is increased and is associated with a greater risk of impaired hypoglycaemia awareness. Neuroendocrine responses to hypoglycaemia are reduced in magnitude and begin at lower absolute blood glucose concentrations than in people with less strict glycaemic control. However, the hypoglycaemic threshold for cognitive dysfunction may not change in a similar fashion. Diabetic patients on intensive insulin therapy reported autonomic and neuroglycopenic symptoms at blood glucose concentrations of about 2.4 and 2.3 mmol/l respectively, whereas in those with less strict glycaemic control and in nondiabetic individuals, these symptoms commenced at between 2.8 and 3.0 mmol/l. However, in all three groups, the accuracy and speed in a reaction time test deteriorated significantly at blood glucose concentrations between 2.8 and 3.0 mmol/l (Amiel et al., 1991; Maran et al., 1995). Therefore, people with insulin-treated diabetes who have strict glycaemic control have the misfortune that the deterioration in their mental performance begins before the onset of warning symptoms of hypoglycaemia. By contrast, neurophysiological responses (P300 event-related potentials), which have been linked to various measures of cognitive function, occur at lower blood glucose concentrations, suggesting that cerebral adaptation has occurred (Ziegler et al., 1992). The effects of quality of glycaemic control, antecedent hypoglycaemia and impaired hypoglycaemia awareness on the mental performance responses to hypoglycaemia, and the relation of these responses to the perceptions of the symptoms of hypoglycaemia, are important topics still under study (see Chapters 7 and 8). The interrelation of these factors makes the field complex, and progress is further hampered by the lack of consensus agreement on a validated battery of cognitive tests for use in hypoglycaemia.
Are the Cognitive Changes During Acute Hypoglycaemia Important and Valid? Do the impairments of mental test performance actually have implications for real-life functions? In addition, are the mental changes during hypoglycaemia a result of impairments in basic brain functions? One common, important and potentially dangerous area of real-life functioning is driving (see Chapter 14), which involves many cognitive abilities including psychomotor control and divided attention. Cox and colleagues (1993b; 2000) employed a sophisticated driving simulator and had people ‘drive’ on this during controlled hypoglycaemia using a glucose
42
SYMPTOMS OF HYPOGLYCAEMIA
clamp technique. With very mild hypoglycaemia (blood glucose below 3.8 mmol/l) the diabetic drivers committed significant driving errors, and during hypoglycaemia the patients often drove very slowly, possibly using a compensatory mechanism to avoid errors. Despite this, more global errors of driving were committed and about half of the participants, despite demonstrating a seriously impaired ability to drive, said they felt competent to drive irrespective of their low blood glucose! It cannot be stated with certainty that the findings obtained in a driving simulator will apply to real-life driving. However, studies that examine the practical cognitive effects of hypoglycaemia are invaluable and more are required. Just as more studies that examine the practical cognitive aspects of hypoglycaemia would be useful, so would more studies of the brain’s processing efficiency. Cognitive tests typically involve a melange of inseparable mental processes, and yet very specific aspects of the human brain’s activities can be measured in the clinical laboratory (Massaro, 1993). Studies of the cognitive effects of hypoglycaemia have thus begun to address the impairments to various cognitive domains in more detail. Basic, specific aspects of visual and auditory processing have been examined during acute hypoglycaemia in non-diabetic humans. Standard tests of visual acuity – those that are measured by an optometrist – are not affected by hypoglycaemia, but other aspects of vision are affected (McCrimmon et al., 1996). These include: • contrast sensitivity (the ability to discriminate faint patterns); • inspection time (the ability to see what is in a pattern when it is shown for a very brief period of time); • visual change detection (the ability to spot a small, quick change in a pattern); • visual movement detection (the ability to spot brief movement in a pattern). This means that the ability to see the environment changes in important ways during hypoglycaemia. Visual acuity is preserved, as tested by the ability to read black letters on a white background. However, most visual activity is not like that; many of the things we see happen quickly and in relatively poor light. When the level of contrast falls, or discriminations must be made under pressure of time, visual processing is impaired during hypoglycaemia. However, at about the same degree of hypoglycaemia, the ability to distinguish one colour from another does not appear to be impaired (Hardy et al., 1995). Speed of auditory processing also appears to be impaired by hypoglycaemia, and the ability to discriminate the loudness of two tones is disrupted (McCrimmon et al., 1997; Strachan et al., 2003). This suggests that the ability to understand language may be compromised during hypoglycaemia. However, despite there being disruption to central nervous system processing during hypoglycaemia, no disturbance has been detected in peripheral nerve conduction (Strachan et al., 2001). If basic information processing provides a fundamental limitation to how well the brain is operating, then at a higher level of function, attention is important in carrying out a number of cognitive functions. A detailed study of a number of different aspects of attention during hypoglycaemia found that the abilities to attend selectively and to switch attention as necessary both deteriorated (McAulay et al., 2001a; 2005). In turn, attention is necessary in order to learn and form new memories. There has now been detailed study of the different aspects of memory during acute, insulin-induced hypoglycaemia (Sommerfield et al., 2003a; 2003b; Deary et al., 2003). Most memory
ACUTE HYPOGLYCAEMIA AND EMOTIONS
43
systems are disrupted during hypoglycaemia. However, some are especially badly affected: long-term memory, which is the ability to retain new information after many minutes and much distraction; working memory, which is the ability to retain and manipulate information at the same time; and prospective memory, which is the ability to remember to do things (as in a shopping list) (Warren et al., 2007). Indeed, the ability to perform one tricky working memory task was obliterated during hypoglycaemia (Deary et al., 2003). That is, no matter how good the person was at performing the task during euglycaemia, the same task could not be done during moderate hypoglycaemia. Further information on hypoglycaemia and cognitive function is available in a review by Warren and Frier (2005).
ACUTE HYPOGLYCAEMIA AND EMOTIONS Mood change is part of the experience of hypoglycaemia. Moods are emotion-like experiences that are quite general rather than applied to specific situations. Psychologists recognise three basic moods: • energetic arousal (a tendency to feel lively and active rather than tired and sluggish); • tense arousal (a tendency to feel anxious and nervous versus relaxed and calm); • hedonic tone (a tendency to feel happy versus sad). When people are asked to rate their mood states during hypoglycaemia induced in the laboratory, changes occur in all of these basic mood states. People feel less energetic, more tense and less happy (Gold et al., 1995c; McCrimmon et al., 1999a; Hermanns et al., 2003). During hypoglycaemia the emotional arousal in response to stimuli becomes more intense (Hermanns et al., 2003). In addition, some people become more irritable and have angry feelings during hypoglycaemia (Merbis et al., 1996; McCrimmon et al., 1999b). The feeling of low energy takes over half an hour to be restored to normal levels, whereas the feelings of tenseness and unhappiness disappear when blood glucose returns to normal. The prolonged feeling of low energy after hypoglycaemia may affect work performance, so that when hypoglycaemia has been treated, an immediate return to the normal state should not be expected. In addition to some people experiencing a low, tense, washed-out, angry mood state, hypoglycaemia alters the way some people look at their life problems. When junior doctors were asked to assess their career prospects during controlled hypoglycaemia, they were more pessimistic (McCrimmon et al., 1995) and if a general state of pessimism is common during hypoglycaemia, it would be a poor state from which to make personal decisions. It is possible that the change in mood states during hypoglycaemia is one of the causes of adults admitting to more ‘odd behaviour’ (Deary et al., 1993). Altered mood may also account in part for symptoms of behavioural disturbance that are so prominent in the responses to hypoglycaemia of children with diabetes (McCrimmon et al., 1995). In addition to emotional responses as a result of hypoglycaemia, some people have emotional responses in anticipation of hypoglycaemia. In Edinburgh one young man with insulin-treated diabetes developed a phobic anxiety state; his phobia related to becoming comatose as a result of hypoglycaemia (Gold et al., 1997a). Such a case is exceptional, but many people with diabetes are frightened of hypoglycaemia (see Chapter 14). The
44
SYMPTOMS OF HYPOGLYCAEMIA
Hypoglycaemia Fear Survey (HFS), which comes in two parts, measures this tendency (Cox et al., 1987). The first part asks people several questions concerning how much they worry about hypoglycaemia (e.g. ‘Do you worry if you have no one around you during a [hypoglycaemic] reaction?’). The second part asks several questions about what people do to avoid hypoglycaemia (e.g. ‘Do you eat large snacks at bedtime?’). People with greater fear of hypoglycaemia (Polonsky et al., 1992; Hepburn et al., 1994) • have more anxious personalities in general; • are more likely to confuse symptoms of anxiety for those of hypoglycaemia; • report having had more episodes of hypoglycaemia. It is not yet known whether people who experience more hypoglycaemia become worriers about it, or whether people who are worriers in general just worry more about hypoglycaemia as well. Perhaps both are true. However, it does seem likely that the experience of more severe hypoglycaemia in the past and the development of impaired awareness of hypoglycaemia lead to increased worry about subsequent hypoglycaemia (Gold et al., 1997b).
CONCLUSIONS • Because people with diabetes are closely involved in their own treatment it is important that they and their educators know about the main side-effects and sequelae of the disorder and its treatments. • Accurate knowledge of the symptoms of hypoglycaemia may be used to avoid the dangers of hypoglycaemia. • The progressively more serious impairment in cognitive function that occurs as blood glucose declines provides knowledge about the brain’s compromised state during hypoglycaemia: basic functions such as visual processing deteriorate and driving becomes dangerously error-prone. Performance on a host of mental tests becomes worse during hypoglycaemia. • Some of the neuroglycopenic symptoms of hypoglycaemia are thought to be subjective impressions of impaired cognitive function: these impressions are fully supported by the results of objective cognitive testing. • Moderate hypoglycaemia may induce a state of anxious tension, unhappiness and low energy, and even irritability and anger. Thus hypoglycaemia importantly touches the emotions as well as inducing bodily symptoms and affecting mental performance.
REFERENCES Amiel SA, Pottinger RC, Archibald HR, Chusney G, Cunnah DTF, Prior PF, Gale EAM (1991). Effect of antecedent glucose control on cerebral function during hypoglycemia. Diabetes Care 14: 109–18. Blackman JD, Towle VL, Sturis J, Lewis GF, Spire J-P, Polonsky KS (1992). Hypoglycemic thresholds for cognitive dysfunction in IDDM. Diabetes 41: 392–9.
REFERENCES
45
Bremer JP, Baron M, Peters H, Oltmanns KM, Kern W, Fehm HL et al. (2006). Hormonal, subjective and neurocognitive responses to brief hypoglycemia in postmenopausal women and age-matched men with type 2 diabetes mellitus. Metabolism 55: 331–8. Clarke WL, Cox DJ, Gonder-Frederick LA, Julian D, Schlundt D, Polonsky W (1997). The relationship between nonroutine use of insulin, food, and exercise and the occurrence of hypoglycemia in adults with IDDM and varying degrees of hypoglycemia awareness and metabolic control. Diabetes Education 23: 55–8. Cox DJ, Clarke WL, Gonder-Frederick L, Pohl S, Hoover C, Snyder A et al. (1985). Accuracy of perceiving blood glucose in IDDM. Diabetes Care 8: 529–36. Cox DJ, Irvine A, Gonder-Frederick L, Nowacek G, Butterfield J (1987). Fear of hypoglycemia: quantification, validation and utilization. Diabetes Care 10: 617–21. Cox DJ, Gonder-Frederick L, Antoun B, Cryer PE, Clarke WL (1993a). Perceived symptoms in the recognition of hypoglycemia. Diabetes Care 16: 519–27. Cox D, Gonder-Frederick L, Clarke W (1993b). Driving decrements in type 1 diabetes during moderate hypoglycemia. Diabetes 42: 239–43. Cox DJ, Gonder-Frederick LA, Kovatchev BP, Julian DM, Clarke WL (2000). Progressive hypoglycemia’s impact on driving simulation performance. Occurrence, awareness and correction. Diabetes Care 23: 163–70. Cox DL, Kovatchev B, Koev D, Koeva L, Dachev S, Tcharaktchiev D et al. (2004). Hypoglycemia anticipation, awareness and treatment training (HAAT) reduces occurrence of severe hypoglycemia among adults with type 1 diabetes. International Journal of Behavioral Medicine 11: 212–18. Deary IJ (1993). Effects of hypoglycaemia on cognitive function. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M, eds. Edward Arnold, London: 80–92. Deary IJ, Hepburn DA, MacLeod KM, Frier M (1993). Partitioning the symptoms of hypoglycaemia using multi-sample confirmatory factor analysis. Diabetologia 36: 771–7. Deary IJ (1998). The effects of diabetes on cognitive function. In: Diabetes Annual 11. Marshall SM, Home PD and Rizza RA eds. Elsevier, London: 97–118. Deary IJ, Sommerfield AJ, McAulay V, Frier BM (2003). Moderate hypoglycaemia obliterates working memory in humans with and without insulin treated diabetes. Journal of Neurology, Neurosurgery and Psychiatry 74: 277–82. Debrah T, Sherwin RS, Murphy J, Kerr D (1996). Effect of caffeine on recognition of and physiological responses to hypoglycaemia in insulin-dependent diabetes. Lancet 347: 19–24. Draelos MT, Jacobson AM, Weinger K, Widom B, Ryan CM, Finkelstein DM, Simonson DC (1995). Cognitive function in patients with insulin-dependent diabetes mellitus during hyperglycemia and hypoglycemia. American Journal of Medicine 98: 135–44. Driesen NR, Cox DJ, Gonder-Frederick L, Clarke W (1995). Reaction time impairment in insulindependent diabetes: task complexity, blood glucose levels, and individual differences. Neuropsychology 9: 246–54. Fletcher AA, Campbell WR (1922). The blood sugar following insulin administration and the symptom complex-hypoglycemia. Journal of Metabolic Research 2: 637–49. Geddes J, Warren, RE, Sommerfield AJ, McAulay V, Strachan MWJ, Allen KV et al. (2006). Absence of sexual dimorphism in the symptomatic responses to hypoglycemia in adults with and without type 1 diabetes. Diabetes Care 29: 1667–9. Gold AE, Deary IJ, MacLeod KM, Frier BM (1995a). The effect of IQ level on the degree of cognitive deterioration experienced during acute hypoglycemia in normal humans. Intelligence 20: 267–90. Gold AE, MacLeod KM, Deary IJ, Frier BM (1995b). Hypoglycemia-induced cognitive dysfunction in diabetes mellitus: effect of hypoglycemia unawareness. Physiology and Behavior 58: 501–11. Gold AE, MacLeod KM, Frier BM, Deary IJ (1995c). Changes in mood during acute hypoglycemia in healthy subjects. Journal of Personality and Social Psychology 68: 498–504. Gold AE, Deary IJ, Frier BM (1997a). Hypoglycaemia and non-cognitive aspects of psychological function in insulin-dependent (type 1) diabetes mellitus (IDDM). Diabetic Medicine 14: 111–18.
46
SYMPTOMS OF HYPOGLYCAEMIA
Gold AE, Frier BM, MacLeod KM, Deary IJ (1997b). A structural equation model for predictors of severe hypoglycaemia in patients with insulin-dependent diabetes mellitus. Diabetic Medicine 14: 309–15. Gonder-Frederick L, Cox D, Driesen NR, Ryan CM, Clarke W (1994). Individual differences in neurobehavioral disruption during mild and moderate hypoglycemia in adults with IDDM. Diabetes 43: 1407–12. Gonder-Frederick L, Cox D, Kovatchev B, Schlundt D, Clarke W (1997). A biopsychobehavioral model of risk of severe hypoglycemia. Diabetes Care 20: 161–9. Hardy KJ, Scase MO, Foster DH, Scarpello JH (1995). Effect of short term changes in blood glucose on visual pathway function in insulin dependent diabetes. British Journal of Ophthalmology 79: 38–41. Heller SR, Macdonald IA (1996). The measurement of cognitive function during acute hypoglycaemia: experimental limitations and their effects on the study of hypoglycaemia unawareness. Diabetic Medicine 13: 607–15. Henderson JN, Allen KV, Deary IJ, Frier BM (2003). Hypoglycaemia in insulin-treated type 2 diabetes: frequency, symptoms and impaired awareness. Diabetic Medicine 20: 1016–21. Hepburn DA, Deary IJ, Frier BM, Patrick AW, Quinn JD, Fisher M (1991). Symptoms of acute insulininduced hypoglycemia in humans with and without IDDM. Factor-analysis approach. Diabetes Care 14: 949–57. Hepburn DA (1993). Symptoms of hypoglycaemia. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M eds. Edward Arnold, London: 93–103. Hepburn DA, Deary IJ, MacLeod KM, Frier BM (1994). Structural equation modeling of symptoms, awareness and fear of hypoglycemia, and personality in patients with insulin-treated diabetes. Diabetes Care 17: 1273–80. Hermanns N, Kubiak T, Kulzer B, Haak T (2003). Emotional changes during experimentally-induced hypoglycaemia in type 1 diabetes. Biological Psychology 63: 15–44. Hoffman RG, Speelman DJ, Hinnen DA, Conley KL, Guthrie RA, Knapp RK (1989). Changes in cortical functioning with acute hypoglycemia and hyperglycemia in type 1 diabetes. Diabetes Care 12: 193–7. Holmes CS (1987). Metabolic control and auditory information processing at altered glucose levels in insulin dependent diabetes. Brain and Cognition 6: 161–74. Jaap AJ, Jones GC, McCrimmon RJ, Deary IJ, Frier BM (1998). Perceived symptoms of hypoglycaemia in elderly type 2 diabetic patients treated with insulin. Diabetic Medicine 15: 398–401. Jones TW, McCarthy G, Tamborlane WV, Caprio S, Roessler E, Kraemer D et al. (1990). Mild hypoglycemia and impairment of brainstem and cortical evoked potentials in healthy subjects. Diabetes 39: 1550–5. Lindgren M, Eckert B, Stenberg G, Agardh C-D (1996). Restitution of neurophysiological functions, performance, and subjective symptoms after moderate insulin-induced hypoglycaemia in non-diabetic men. Diabetic Medicine 13: 218–25. Maran A, Lomas J, Macdonald IA, Amiel SA (1995). Lack of preservation of higher brain function during hypoglycaemia in patients with intensively-treated IDDM. Diabetologia 38: 1412–18. Massaro DW (1993). Information processing models: microscopes of the mind. Annual Review of Psychology 44: 383–425. Macfarlane PI, Smith CS (1988). Perceptions of hypoglycaemia in childhood diabetes mellitus: a questionnaire study. Practical Diabetes 5: 56–8. Macfarlane PI, Walters M, Stutchfield P, Smith CS (1989). A prospective study of symptomatic hypoglycaemia in childhood diabetes. Diabetic Medicine 6: 627–30. McAulay V, Deary IJ, Ferguson SC, Frier BM (2001a). Acute hypoglycemia in humans causes attentional dysfunction while nonverbal intelligence is preserved. Diabetes Care 24: 1745–50.
REFERENCES
47
McAulay V, Deary IJ, Frier BM (2001b). Symptoms of hypoglycaemia in people with diabetes. Diabetic Medicine 18: 690–705. McAulay V, Deary IJ, Sommerfield AJ, Frier BM (2005). Attentional functioning is impaired during acute hypoglycaemia in people with type 1 diabetes. Diabetic Medicine 23: 26–31. McCrimmon RJ, Gold AE, Deary IJ, Kelnar CJH, Frier BM (1995). Symptoms of hypoglycemia in children with IDDM. Diabetes Care 18: 858–61. McCrimmon RJ, Deary IJ, Huntly BJP, MacLeod KJ, Frier BM (1996). Visual information processing during controlled hypoglycaemia in humans. Brain 119: 1277–87. McCrimmon RJ, Deary IJ, Frier BM (1997). Auditory information processing during acute insulininduced hypoglycaemia in non-diabetic human subjects. Neuropsychologia 35: 1547–53. McCrimmon RJ, Deary IJ, Frier BM (1999a). Appraisal of mood and personality during hypoglycaemia. Physiology and Behavior 67: 27–33. McCrimmon RJ, Ewing FME, Frier BM, Deary IJ (1999b). Anger-state during acute insulin-induced hypoglycaemia. Physiology and Behavior 67: 35–9. McCrimmon RJ, Deary IJ, Gold AE, Hepburn DA, MacLeod KM, Ewing FME, Frier BM (2003). Symptoms reported during experimental hypoglycaemia: effect of method of induction of hypoglycaemia and of diabetes per se. Diabetic Medicine 20: 507–9. Merbis MAE, Snoek FJ, Kanc K, Heine RJ (1996). Hypoglycaemia induces emotional disruption. Patient Education and Counseling 29: 117–22. Mutch WJ, Dingwall-Fordyce I (1985). Is it a hypo? Knowledge of symptoms of hypoglycaemia in elderly diabetic patients. Diabetic Medicine 2: 54–6. Pennebaker JW, Cox DJ, Gonder-Frederick L, Wunsch MG, Evans WS, Pohl S (1981). Physical symptoms related to blood glucose in insulin-dependent diabetics. Psychosomatic Medicine 43: 489–500. Polonsky WH, Davis CL, Jacobson AM, Anderson BJ (1992). Correlates of hypoglycemic fear in type 1 and type 2 diabetes mellitus. Health Psychology 11: 199–202. Pramming S, Thorsteinsson B, Theilgaard A, Pinner EM, Binder C (1986). Cognitive function during hypoglycaemia in type 1 diabetes mellitus. British Medical Journal 292: 647–50. Ross LA, McCrimmon RJ, Frier BM, Kelnar CJH, Deary IJ (1998). Hypoglycaemic symptoms reported by children with type 1 diabetes mellitus and by their parents. Diabetic Medicine 15: 836–43. Russell PN, Rix-Trot HM (1975). An exploratory study of some behavioural consequences of insulininduced hypoglycaemia. New Zealand Medical Journal 81: 337–40. Ryan CM, Dulay D, Suprasongsin C, Becker DJ (2002). Detection of symptoms by adolescents and young adults with type 1 diabetes during experimental induction of mild hypoglycemia: role of hormonal and psychological variables. Diabetes Care 25: 852–8. Schachinger H, Hegar K, Hermanns N, Straumann M, Keller U, Fehm-Wolfsdorf G et al. (2005). Randomized controlled clinical trial of Blood Glucose Awareness Training (BGAT III) in Switzerland and Germany. Journal of Behavioral Medicine 28: 587–94. Sommerfield AJ, Deary IJ, McAulay V, Frier BM (2003a). Moderate hypoglycemia impairs multiple memory functions in healthy adults. Neuropsychology 17: 125–32. Sommerfield AJ, Deary IJ, McAulay V, Frier BM (2003b). Short-term, delayed, and working memory are impaired during hypoglycemia in individuals with type 1 diabetes. Diabetes Care 26: 390–6. Strachan MWJ, Deary IJ, Ewing FME, Ferguson SC, Young MJ, Frier BM (2001). Acute hypoglycemia impairs the functioning of the central but not peripheral nervous system. Physiology & Behavior 72: 83–92. Strachan MWJ, Ewing FME, Frier BM, McCrimmon RJ, Deary IJ (2003). Effects of acute hypoglycaemia on auditory information processing in adults with type 1 diabetes. Diabetologia 46: 97–105. Tallroth G, Lindgren M, Stenberg G, Rosen I, Agardh C-D (1990). Neurophysiological changes during insulin-induced hypoglycaemia and in the recovery period following glucose infusion in type 1 (insulin-dependent) diabetes mellitus and in normal man. Diabetologia 33: 319–23.
48
SYMPTOMS OF HYPOGLYCAEMIA
Towler DA, Havlin CE, Craft S, Cryer P (1993). Mechanism of awareness of hypoglycemia: perception of neurogenic (predominantly cholinergic) rather than neuroglycopenic symptoms. Diabetes 42: 1791–8. Warren RE, Frier BM (2005). Hypoglycaemia and cognitive function. Diabetes, Obesity and Metabolism 7: 493–503. Warren RE, Zammitt NN, Deary IJ, Frier BM (2007). The effects of acute hypoglycaemia on memory acquisition and recall and prospective memory in type 1 diabetes. Diabetologia 50: 178–85. Weinger K, Jacobson AM, Draelos MT, Finkelstein DM, Simonson DC (1995). Blood glucose estimation and symptoms during hyperglycemia and hypoglycemia in patients with insulin-dependent diabetes mellitus. American Journal of Medicine 98: 22–31. Wilder J (1943). Psychological problems in hypoglycemia. American Journal of Digestive Diseases 10: 428–35. Wirsen A, Tallroth G, Lindgren M, Agardh C-D (1992). Neuropsychological performance differs between type 1 diabetic and normal men during insulin-induced hypoglycaemia. Diabetic Medicine 9: 156–65. Ziegler D, Hubinger A, Muhlen H, Gries FA (1992). Effects of previous glycaemic control on the onset and magnitude of cognitive dysfunction during hypoglycaemia in type 1 (insulin-dependent) diabetic patients. Diabetologia 35: 828–34.
3 Frequency, Causes and Risk Factors for Hypoglycaemia in Type 1 Diabetes Mark W.J. Strachan
INTRODUCTION Hypoglycaemia was first described in humans in the early years of the 20th century, but did not become firmly established as a pathophysiological entity until the discovery of insulin in 1922. Despite the substantial advances in insulin therapy and blood glucose monitoring that have occurred in the subsequent 80 years, hypoglycaemia remains the most common complication of type 1 diabetes (The Diabetes Control and Complications Trial Research Group, 1993) and generates as much anxiety in patients as the threat of advanced diabetic complications, such as blindness or renal failure (Pramming et al., 1991). Few people with type 1 diabetes escape intermittent exposure and, as a result, hypoglycaemia is the principal limiting factor in achieving good glycaemic control (Cryer, 1994; Cryer et al., 2003). The magnitude of the psychological and physical consequences of hypoglycaemia cannot be overestimated and is considered in detail in other chapters of this book. In this chapter the frequency of hypoglycaemia is described in people with type 1 diabetes, along with its underlying causes and risk factors.
DEFINITIONS OF HYPOGLYCAEMIA Any attempt to consider critically the frequency of hypoglycaemia in clinical practice, requires definitive criteria for what constitutes an episode of hypoglycaemia. This poses an immediate difficulty because researchers of the epidemiology of hypoglycaemia have not employed common definitions with shared specifications.
Biochemical Definitions of Hypoglycaemia At first glance, it would seem sensible to employ a biochemical definition of hypoglycaemia, specifying a given blood glucose concentration, below which hypoglycaemia would be deemed to occur. However, it is not possible to provide such a precise biochemical criterion for the diagnosis of hypoglycaemia (Service, 1995). As blood glucose concentrations
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
50
FREQUENCY, CAUSES AND RISK FACTORS
decline, a hierarchy of events occur at individual glycaemic thresholds, commencing with counterregulation (arterialised blood glucose ∼38 mmol/l), impairment of different cognitive functions (∼32–26 mmol/l) and the onset of symptoms and neurophysiological changes (∼32–24 mmol/l). In clinical practice, however, it is usual for venous or capillary blood glucose levels to be measured, and these are lower than contemporaneous arterialised blood glucose concentrations (which are usually measured in research studies of hypoglycaemia) (Heller and Macdonald, 1996). In the non-diabetic individual, venous blood glucose concentrations below 3.0 mmol/l may occur following an overnight fast or during the course of a prolonged oral glucose tolerance test (Service, 1995). Moreover, as is discussed later, the blood glucose thresholds for the onset of symptoms and counterregulation in patients with type 1 diabetes may vary depending on the preceding or prevailing glycaemic control. Thus, patients with poor glycaemic control may experience symptoms of hypoglycaemia at venous plasma glucose concentrations substantially higher than 3.0 mmol/l (Boyle et al., 1988). Patients with preceding strict glycaemic control may not experience the onset of symptoms of hypoglycaemia until venous plasma glucose concentrations have declined to below 2.0 mmol/l (Boyle et al., 1995). On a pragmatic basis, in routine clinical practice, Diabetes UK has recommended that individuals with diabetes should try to ensure that their blood glucose concentrations do not fall below 4.0 mmol/l (O’Neill, 1997), but this does not define hypoglycaemia. The American Diabetes Association has proposed a blood glucose concentration of 3.9 mmol/l as representing hypoglycaemia (ADA Workgroup on Hypoglycemia, 2005), but this has been challenged as being too high.
Clinical Definitions of Hypoglycaemia The inability to agree on a biochemical definition for hypoglycaemia requires instead the application of clinical criteria (Box 3.1). The difficulty here is that because the symptoms of hypoglycaemia are not specific, and vary between individuals (see Chapter 2), the use of symptomatology alone may be unreliable and may result in the inclusion of episodes that are not true hypoglycaemia. In one prospective study, where capillary blood glucose was measured whenever a patient had symptoms suggestive of hypoglycaemia, only 29% of such episodes were accompanied by evidence of biochemical hypoglycaemia (i.e., blood glucose <30 mmol/l) (Pramming et al., 1990). For that reason, hypoglycaemia is usually defined as an episode in which typical symptoms occur and the symptoms are reversed by appropriate treatment to raise blood glucose. Ideally, corroborative evidence should be provided by the contemporaneous demonstration of a low capillary or venous blood glucose concentration but, as this may be neither available nor possible, it need not be regarded as an absolute requirement to identify hypoglycaemia. The definition of severe hypoglycaemia used in the Diabetes Control and Complications Trial (DCCT) has gained widespread acceptance and is now regarded as the standard definition for clinical research (The Diabetes Control and Complications Trial Research Group, 1997). Severe hypoglycaemia was defined as an episode of hypoglycaemia in which assistance from a third party was required to effect treatment and recovery. The clear advantage of this definition is its inherent simplicity, which allows it to be used in a variety of different clinical and research settings. The disadvantages of the definition are that it does not specify a threshold level below which the blood glucose concentration must fall
FREQUENCY OF HYPOGLYCAEMIA
Box 3.1
51
Clinical definitions of hypoglycaemia
• Asymptomatic hypoglycaemia – low blood glucose identified on routine blood test, with no associated symptoms. • Mild symptomatic hypoglycaemia∗ – symptoms suggestive of hypoglycaemia; episode successfully treated by the patient alone. • Severe hypoglycaemia∗ – assistance from a third party is required to effect treatment. • Profound hypoglycaemia∗ – associated with permanent neurological deficits or death. ∗
Contemporaneous demonstration of a low blood glucose concentration is not mandatory but, particularly in the case of mild episodes, if it is available it does make the definition more robust.
before an episode of hypoglycaemia can be considered to be severe, nor does it specify any particular method of treating the low blood glucose. There is a substantial difference in the degree of neuroglycopenia during an episode of hypoglycaemia that has resolved following the oral administration of carbohydrate, and one that has resulted in coma and required the administration of intramuscular glucagon or intravenous glucose. Mild hypoglycaemia is considered to occur when an episode is self-treated by the affected individual. Although the recording of episodes of severe hypoglycaemia is generally regarded as being a fairly robust measure, mild hypoglycaemia is a much ‘softer’ end-point. Symptom perception by the patient may be difficult and the glycaemic thresholds at which individuals develop symptoms of hypoglycaemia are very variable. In many clinical research studies, it is not uncommon to include a composite definition of mild hypoglycaemia, which includes symptomatic episodes that respond to self-treatment (with or without a confirmatory blood test) and asymptomatic episodes below an arbitrary biochemical blood glucose value (e.g. 3.5 mmol/l) that have been identified through routine blood glucose monitoring (Anderson et al., 1997). The introduction by some investigators of a further category of ‘moderate’ hypoglycaemia is not helpful and confuses the distinction between mild and severe episodes.
FREQUENCY OF HYPOGLYCAEMIA The true frequency of hypoglycaemia in people with type 1 diabetes is difficult to estimate accurately and not simply because of differences in the definitions of hypoglycaemia. Most episodes occur at home, at work or during leisure activities, without involving medical, nursing or paramedical staff, and the subsequent recall of such episodes by patients is generally poor, particularly with regard to mild hypoglycaemia. Prospective studies therefore have a strong advantage over retrospective studies in their respective abilities to document the frequency of hypoglycaemia with accuracy. Another crucially important feature is the nature of the population of people with type 1 diabetes under study – factors such as the
52
FREQUENCY, CAUSES AND RISK FACTORS
quality of glycaemic control (The Diabetes Control and Complications Trial Research Group, 1997) and the presence of impaired awareness of hypoglycaemia (Gold et al., 1997) have a major influence on the frequency of hypoglycaemia. Participants in clinical interventional studies are often not representative of the wider body of people with type 1 diabetes. For example, in the DCCT (The Diabetes Control and Complications Trial Research Group, 1993), the subjects were young, motivated and received substantial professional support (particularly those in the intensive group). They were pre-selected depending on their history of hypoglycaemia. Thus, individuals were excluded if, in the previous two years, they had experienced more than one episode of severe hypoglycaemia causing neurological impairment, without experiencing warning symptoms, or more than two episodes of seizure or coma, regardless of attributed cause. Such exclusions will have inevitably influenced the baseline frequency of severe hypoglycaemia in the study groups. Small studies estimating the frequency of hypoglycaemia are likely to be biased because chance variations in the prevalence of risk factors for hypoglycaemia can magnify (or diminish) the frequency of hypoglycaemic events to a much greater extent than would be observed in larger investigations. There is also likely to be a major period effect in studies of the prevalence of hypoglycaemia. In the post-DCCT era, it might be anticipated that the incidence of hypoglycaemia would rise as patients and diabetes healthcare professionals sought to tighten glycaemic control (Johnson et al., 2002). This trend may be counterbalanced by the increased use of home blood glucose monitoring, improved educational programmes and the greater use of insulin analogues and continuous subcutaneous insulin infusion therapy (Chase et al., 2001). Thus, estimates of the frequency of hypoglycaemia must be considered in relation to the definitions employed and the characteristics of the patients studied. Comparisons between present-day and historical studies must be made in the knowledge that treatment targets, methods of monitoring blood glucose and insulin therapy have changed greatly in recent years.
Frequency of Mild, Symptomatic Hypoglycaemia The features of several studies that have examined the frequency of mild hypoglycaemia, either prospectively or retrospectively, in adults with type 1 diabetes are shown in Table 3.1. Rates of mild hypoglycaemia vary substantially, ranging from 8 to 160 episodes per patient per year. However, it is extremely difficult to make direct comparisons between individual studies because of differences in methodology and patient characteristics. Retrospective studies Two retrospective studies by a Danish group (Pedersen-Bjergaard et al., 2001; PedersenBjergaard et al., 2004) were very similar in their patient characteristics, most of whom were administering four or more injections of insulin per day and had moderate glycaemic control. On being asked to recall the number of episodes of mild, symptomatic hypoglycaemia experienced in the preceding week, the subjects reported a frequency of two episodes per patient per week. The strength of these studies lies in the large numbers of patients examined and the short period of recall. This should have minimised inaccuracy, but may have been an unrepresentative time frame.
1 week (P) 6 weeks (P) 1 week (R) 12 months (P) 1 week (R) 4 weeks (P) 12 months (P) 9–12 months
170
1076
94 243 (27 T2)e 46 (< 5 yrs) 54 (> 15 yrs)
Follow-upa
411 31 201
Number of patients
86 83 91 73 78
49% S&If 80% multiple All ≥ 2 injections
84
87% ≥ 4 inj 72% multiple
87 72 86
HbA1c c (%)
78% twice daily All multiple 86% multiple
Treatmentb
Symptomatic Symptomatic Symptomatic and/or BG < 30 mmol/l
Symptomatic
Symptomatic
Symptomatic BG < 35 mmol/l Symptomatic
Definition of hypoglycaemia
42 8 35 29
104
88
94 160 104
Frequencyd (number/pt/yr)
This table does not include early studies which tended to focus on rates of mild hypoglycaemia in particular subsets of patients, e.g. with impaired awareness of hypoglycaemia or individuals using porcine versus human insulin. a Duration of follow-up: P = prospective estimation of hypoglycaemia frequency; R = retrospective. b Multiple refers to preprandial injections (inj) of soluble (or analogue) insulin and bedtime isophane insulin. c Mean HbA1c for the subjects under study. d Frequency of hypoglycaemia in each study has been adjusted to represent an annual rate per person. e 27 Patients in this study had type 2 diabetes f Patients were on soluble and isophane insulins, but frequency of injections not specified. BG = capillary blood glucose concentration.
Pramming et al., 1991 Janssen et al., 2000 Pedersen-Bjergaard et al., 2001 Pedersen-Bjergaard et al., 2003a Pedersen-Bjergaard et al., 2004 Donnelly et al., 2005 Leckie et al., 2005 UK Hypoglycaemia Study Group, 2007
Authors
Table 3.1 Frequency of mild hypoglycaemia in adults with type 1 diabetes
54
FREQUENCY, CAUSES AND RISK FACTORS
Prospective studies Prospective studies offer the potential to provide more convincing data on frequency of mild hypoglycaemia, but substantial differences in prevalence were again reported. In an earlier study of 441 patients with type 1 diabetes, managed principally with a twice daily insulin regimen containing soluble and isophane insulins, the weekly average was 1.8 episodes of mild symptomatic hypoglycaemia (Pramming et al., 1991). These patients had moderate glycaemic control and the period of assessment was one week. This study may be of less relevance today in view of the current use of intensive insulin therapy and insulin analogues, yet it is interesting that the rate of mild hypoglycaemia is unchanged today. A Danish prospective study (Pedersen-Bjergaard et al., 2003a) included patients with similar characteristics to those in two retrospective studies by the same group (PedersenBjergaard et al., 2001; Pedersen-Bjergaard et al., 2004). Hypoglycaemia was recorded monthly with episodes of mild symptomatic hypoglycaemia being reported for the preceding week. Mild hypoglycaemia occurred on average 1.7 times per patient per week. Subjects were also asked to perform a monthly five-point blood glucose profile and to record in addition any blood glucose value below 3.0 mmol/l. Measurements demonstrating biochemical hypoglycaemia represented 3.7% of all blood glucose readings. In a community-based study in Tayside, Scotland, 94 adults with type 1 diabetes were selected at random from a regional diabetes database and were asked to record episodes of hypoglycaemia prospectively over one month (Donnelly et al., 2005). Their median age was 40 years, median duration of diabetes was 18 years and median HbA1c was 8.3%. Biphasic insulin was used by 35% of participants and 49% used a combination of intermediate and short-acting insulins (although the frequency of injections was not reported). A total of 325 episodes of mild hypoglycaemia occurred, representing a rate of 41.5 episodes per person per year, i.e., approximately half the rate reported by Pramming et al., (1991) and Pedersen-Bjergaard et al., (2003a). The study has weaknesses; it was relatively small and data on frequency of blood glucose monitoring were limited. The precise criteria for defining mild hypoglycaemia were not clearly described and, in particular, the role of contemporaneous monitoring data was not specified. Nevertheless, the subjects were probably very representative of the population of people with type 1 diabetes in that region. Similar data have been reported from a multicentre study from the United Kingdom (UK Hypoglycaemia Study Group, 2007). The primary aim of this study was to compare the frequencies of hypoglycaemia in individuals with different types and durations of diabetes, receiving different treatment modalities. As part of the study, 50 adults with type 1 diabetes of duration less than five years and 57 adults with type 1 diabetes of greater than 15 years duration were recruited. All subjects used two or more injections of insulin per day and their glycaemic control was good (mean HbA1c < 80%). The participants were followed for between 9–12 months (mean 10 months) and were asked to report all episodes of symptomatic, self-treated hypoglycaemia and episodes where blood glucose was less 3.0 mmol/l, regardless of symptomatology. Subjects were given forms to record such episodes and were encouraged to record contemporaneous blood glucose levels. To maximise compliance, subjects were asked to send in completed forms every month to the local research centre, including when no episodes of hypoglycaemia had occurred. If no forms were received, telephone contact was made with the subjects. Using this robust methodology, mean rates of hypoglycaemia of 35 and 29 episodes per person per year were reported for the short and long duration groups respectively. The distribution of episodes was much skewed, with
FREQUENCY OF HYPOGLYCAEMIA
55
some individuals reporting no events and others in excess of 200 per year; overall, approximately 85% of individuals experienced at least one episode of mild hypoglycaemia. The authors did not report the relative proportion of symptomatic and asymptomatic episodes, or the proportion of symptomatic episodes with a corroborative blood test, but these data are extremely informative and indicate that duration of diabetes has little impact on the frequency of mild hypoglycaemia, an observation that has been made before (Pedersen-Bjergaard et al., 2004). In a 12 month prospective study of 243 insulin-treated adults, Leckie and colleagues reported that mild, symptomatic hypoglycaemia occurred with a frequency of only eight episodes per patient per year (Leckie et al., 2005). This is one of the lowest rates to be reported in a large group of people, most of whom had type 1 diabetes. However, several reasons may account for this result. The subjects had suboptimal glycaemic control, with a mean HbA1c of 9.1%. A small number of people with insulin-treated type 2 diabetes were included, who might be expected to have had a lower overall frequency of hypoglycaemia. The prevalence of impaired awareness of hypoglycaemia, a recognised risk factor for hypoglycaemia (see Chapter 7), was exceptionally low in this cohort at 3%. The proportion of subjects using insulin analogues was not reported, but it would certainly have been higher than in studies from the early 1990s. The main strength of the study, namely a long period of follow up, may have been an inadvertent weakness by causing ‘patient fatigue’, i.e., participants may have been less assiduous in recording episodes of mild hypoglycaemia as the study progressed. Finally, this was primarily a study of the impact of hypoglycaemia occurring in the work place. Thus, all of the participants were in full-time employment and so represented an atypical group of individuals who may have adopted strategies to reduce the frequency of hypoglycaemia because of the potentially adverse effects that this could have on their jobs.
Frequency of Asymptomatic, Biochemical Hypoglycaemia Aside from the problem of specifying a biochemical threshold for hypoglycaemia, there can be little doubt that asymptomatic hypoglycaemia is even more common than symptomatic, mild hypoglycaemia. However, a major determinant of the frequency of documented asymptomatic hypoglycaemia is, necessarily, the frequency with which blood glucose is measured. Traditionally, two different approaches have been used to investigate this phenomenon: (a) taking multiple sequential blood samples in a controlled, experimental setting, over a prespecified time period; and (b) inviting patients in the community to perform capillary blood tests at multiple, pre-set time points. Providing the sampling frame is sufficiently frequent, the advantage of the former strategy is that it will capture all episodes of biochemical hypoglycaemia, however brief, during the time period under study. The disadvantage is that it is labour intensive for investigators and so only a small number of subjects can be studied over a relatively short time period, such as 12 or 24 hours. By contrast, the use of home blood glucose monitoring, particularly with meters that store the results electronically, allows larger numbers of subjects to be studied over very prolonged time periods. Typically, subjects are asked to perform periodic seven-point profiles, i.e., blood glucose estimations before each meal, two hours after food and during the night. The obvious disadvantages of this mode of investigation are that subjects may forget to perform the relevant monitoring,
56
FREQUENCY, CAUSES AND RISK FACTORS
or may do so at the wrong times, and that episodes of asymptomatic hypoglycaemia may occur outside the sampling time-frames and, thus, be missed. Thorsteinsson et al., (1986) examined seven-point capillary blood glucose profiles in 99 adults with type 1 diabetes and demonstrated that the frequency of biochemical hypoglycaemia was inversely related, in a curvilinear manner, to the median blood glucose concentration. Thus, for example, in patients who had a median blood glucose concentration of 5.0 mmol/l, 10% of blood glucose levels were less than 3.0 mmol/l. By contrast, only 2.5% of blood glucose levels were below 3.0 mmol/l, in patients whose median blood glucose concentration was 10 mmol/l (Figure 3.1; Thorsteinsson et al., 1986). In a separate study, nocturnal blood glucose profiles were examined in 31 patients with type 1 diabetes using multiple injection therapy with soluble and isophane (NPH) insulin (mean HbA1c 8.6%). Venous blood samples were taken every 30 minutes from 11 p.m. until 7.30 a.m.. Nocturnal hypoglycaemia (blood glucose less than 3.0 mmol/l) occurred on 29% of occasions and 67% of these episodes were asymptomatic (Vervoort et al., 1996). Six individuals were studied on two separate nights and the blood glucose profiles, perhaps predictably, showed considerable intra-individual variation.
Figure 3.1 Correlation between the median blood glucose concentration and the frequencies of blood glucose concentrations below 4.0, 3.0, 2.5 and 2.0 mmol/l in adults with type 1 diabetes. Solid lines represent data from 70 adults on twice daily insulin therapy and dotted lines represent data from 20 adults treated with continuous subcutaneous insulin. Approximately 10% of readings were below 3.0 mmol/l, when median blood glucose concentration was 5.0 mmol/l. Reproduced with permission from Thorsteinsson et al. (1986) © John Wiley & Sons, Ltd
FREQUENCY OF HYPOGLYCAEMIA
57
Janssen et al., (2000) have studied, prospectively, the frequency of biochemical hypoglycaemia over a six week period in 31 people with type 1 diabetes with a mean HbA1c of 7.2% (all had a HbA1c ≤ 83%). Subjects were all using a multiple injection regimen with soluble and isophane insulins and performed one seven-point blood glucose profile and six four-point profiles each week. No overnight readings were performed. Patients completed a mean of 82% of the required monitoring schedule and overall experienced a mean of 18.7 episodes of hypoglycaemia (i.e., approximately 160 episodes per year). The range was wide, however, at between zero and 41 episodes per individual, over the six weeks of the study. The more widespread application of continuous glucose monitoring systems may help to elucidate the frequency of mild hypoglycaemia with greater accuracy. In one study, 65% of people with type 1 diabetes monitored over three days had an episode of asymptomatic hypoglycaemia (interstitial glucose < 33 mmol/l) (Bode et al., 2005). However, much work still needs to be done to clarify the relationship between interstitial glucose and blood glucose concentrations, before this can be regarded as a robust tool for detecting hypoglycaemia (see Chapter 5).
Frequency of Severe Hypoglycaemia As with mild hypoglycaemia, direct comparisons between individual studies on the frequency of severe hypoglycaemia are not straightforward. However, it is a more robust end-point than mild hypoglycaemia and, because episodes typically have a more profound effect on individuals, retrospective recall is much more reliable. At the conclusion of their prospective study, Pedersen-Bjergaard et al. (2003b), showed that 90% of subjects recalled correctly their experience of severe hypoglycaemia over the preceding year (Figure 3.2). Subjects who had experienced a high incidence of severe hypoglycaemia (prospectively recorded), retrospectively underestimated the overall rate by around 15%. Table 3.2 lists some of the large surveys (each in excess of 100 participants) that have examined the frequency of severe hypoglycaemia. The table focuses primarily on studies examining unselected groups of individuals with type 1 diabetes, and so excludes intervention trials (The Diabetes Control and Complications Trial Research Group, 1993; Reichard and Pihl, 1994; MacLeod et al., 1995) and studies that examined particular sub-groups of patients, such as people with impaired awareness of hypoglycaemia (Gold et al., 1994; MacLeod et al., 1994; Clarke et al., 1995), differing durations of diabetes (UK Hypoglycaemia Study Group, 2007) or subjects who have received intensive therapy or education (Muhlhauser et al., 1985; Pampanelli et al., 1996; Bott et al., 1997). A study that examined a mixed group of children, adolescents and adults (Allen et al., 2001) has also been excluded. The frequency of severe hypoglycaemia (defined as episodes requiring third party assistance) in the studies listed in Table 3.2 is remarkably consistent at 1.0 to 1.6 episodes per patient per year. However, it is important to note that the frequency of severe hypoglycaemia in unselected populations does not follow a Gaussian distribution (Figure 3.3). The distribution is heavily skewed such that the majority of individuals do not experience any severe hypoglycaemia in a given year, while a small number of individuals have recurrent episodes. In the studies highlighted in Table 3.2, between 30–40% of individuals experienced at least one episode of severe hypoglycaemia over the period in question. The proportion affected
FREQUENCY, CAUSES AND RISK FACTORS
58
Recalled severe hypoglycaemia (episodes per patient-year)
15
10
5
0 0
5 10 15 Prospectively recorded severe hypoglycaemia (episodes per patient-year)
Figure 3.2 Correlation between prospectively recorded and retrospectively recalled rate of severe hypoglycaemia over the same one year period in 230 people with type 1 diabetes. Marker sizes are weighted by the number of cases. R2 = 066; p < 0001. Reproduced with permission from PedersenBjergaard et al. (2003b) © John Wiley & Sons, Ltd
in the study by Pramming et al., (1991) was much lower, but the period of follow up was only one week. This skewed distribution serves to emphasise further the importance of patient selection in ascertaining the frequency of severe hypoglycaemia with accuracy, as the exclusion of a relatively small number of people at high risk would substantially reduce the overall risk. In three of the studies of unselected adults with type 1 diabetes, severe hypoglycaemia was further subdivided to examine episodes associated with more significant neuroglycopenia, i.e., those resulting in coma and/or seizures (Table 3.2) (ter Braak et al., 2000; PedersenBjergaard et al., 2003a; Pedersen-Bjergaard et al., 2004). Furthermore, in the study by Muhlhauser et al., (1998) only episodes treated with intra-muscular glucagon or intravenous glucose were addressed. Predictably, such events were rarer and represented about one quarter of all episodes of severe hypoglycaemia. Emergency and hospital services will occasionally be involved in the management of severe hypoglycaemia and there are data on the use of such agencies. This clearly has to be interpreted with caution, as the majority of all hypoglycaemia is managed in the community, without involvement of (para)clinical staff. Individuals admitted to hospital with hypoglycaemia are probably atypical, and have an increased prevalence of alcohol dependence and mental illness (Hart and Frier, 1998). An early study from Australia reported that, over one year, 3.5% of people attending an urban diabetes clinic had an episode of hypoglycaemia severe enough to warrant referral to hospital (Moses et al., 1985). More recent data from Tayside, Scotland, demonstrated that 7.1% of people with type 1 diabetes
24 months (R) 12 months (P) 12 months (R) 12 months (P) 9–12 months
170
1076
243 (27 T2)e 46 (<5 years) 54 (>15 years)
12 months (R)
195
207
1 week (P) 12 months (R) 12 months (R)
Follow-upa
411 600 (56 T2)e 684
Number of patients
84 86
87% ≥ 4 inj 72% ≥ 4 inj 91 7.3 7.8
86
86% ≥ 4 inj
80% multiple All ≥ 2 injections
78
87 107 A1 80
HbA1c c (%)
82% intensive
78% twice daily 76% twice daily 70% > 2 inj
Treatmentb
Third party Coma/seizure Third party Coma/seizure Third party Third party
Third party Coma/seizure Third party
Third party Third party Glucagon/glucose
Definition of hypoglycaemia
11 03 13 035 098 1.1 3.2
41
15 04 11
34 22 46
37
39
NR
3 29 13
14 16 021
Frequencyd Proportion (episodes/pt/yr) Affected (%)
This table only considers studies examining in excess of 100 subjects and does not include early studies which tended to focus on rates of severe hypoglycaemia in particular subsets of patients, e.g. with impaired awareness of hypoglycaemia or individuals using porcine versus human insulin. a Duration of follow-up: P = prospective estimation of hypoglycaemia frequency; R = retrospective. b Multiple refers to preprandial injections (inj) of soluble (or analogue) insulin and bedtime isophane insulin. c Mean HbA1c for the subjects under study; A1 = figure is HbA1 . d Frequency of hypoglycaemia in each study has been adjusted to represent an annual rate per person. NR = data not reported. e 56 include patients with type 2 diabetes.
Pedersen-Bjergaard et al., 2001 Pedersen-Bjergaard et al., 2003a Pedersen-Bjergaard et al., 2004 Leckie et al., 2005 UK Hypoglycaemia Study Group
Pramming et al., 1991 MacLeod et al., 1993 Muhlhauser et al., 1998 ter Braak et al., 2000
Authors
Table 3.2 Frequency of severe hypoglycaemia in adults with type 1 diabetes
FREQUENCY, CAUSES AND RISK FACTORS
60 90 80 70
% of patients
60 50 40 30 20 10 0 0
2
4
6
8
10
Severe hypoglycaemia, episodes per year
Figure 3.3 Distribution of self-reported number of episodes of severe hypoglycaemia during the preceding year in 1049 unselected patients with type 1 diabetes (light bars) and 209 patients selected by criteria to mimic the characteristics of the DCCT cohort (dark bars). Reproduced with permission from Pedersen-Bjergaard et al. (2004) © John Wiley & Sons, Ltd
had an episode of hypoglycaemia that required contact with emergency medical teams over one year, with an incidence rate of 0.12 per patient per year (Leese et al., 2003). Thus, only about one in 10 of all episodes of severe hypoglycaemia result in contact with emergency services.
Frequency of Hypoglycaemia: Summary and Conclusions The literature on frequency of hypoglycaemia is heterogeneous and inconsistent and, thus there are considerable methodological limitations in our ability to ascertain accurately the frequency of hypoglycaemia. The definitive study, a prolonged, prospective evaluation of a large number of unselected people with type 1 diabetes, outside ‘trial’ conditions, has yet to be performed. Existing data on severe hypoglycaemia are probably fairly accurate, but estimates of mild hypoglycaemia should be regarded with caution. However, the ‘average’ patient with type 1 diabetes will probably experience about 1–2 episodes of mild hypoglycaemia per week and be exposed to several thousand episodes over a lifetime with diabetes. Overall, severe hypoglycaemia may be expected to occur once or twice each year, but the distribution is heavily skewed, such that most individuals will be unaffected while a small number will have multiple episodes. In the early 1980s, Robert Tattersall’s group in Nottingham reported that admission to hospital as a consequence of hypoglycaemia represented the ‘tip of an iceberg’ of all episodes of hypoglycaemia (Potter et al., 1982), and this observation does not appear to have changed.
CAUSES OF HYPOGLYCAEMIA
61
CAUSES OF HYPOGLYCAEMIA Although advances in insulin therapy have been made over the last 80 years, the administration of exogenous insulin remains a very crude means of managing type 1 diabetes. The time-action profiles of the modern insulin analogues do not mimic the physiological changes in plasma insulin concentrations that occur in non-diabetic individuals (Figure 3.4) and, crucially, concentrations of exogenous insulin cannot respond to changes in blood glucose concentration. Therefore, at its most fundamental level, hypoglycaemia in people with type 1 diabetes is the result of an imbalance between insulin-mediated glucose efflux from the blood stream and the amount of glucose entering the circulation from ingested carbohydrate and from the liver. Cryer et al. (2003), have grouped the causes of hypoglycaemia in type 1 diabetes into six categories (Box 3.2) depending on their relative effects on insulin
Figure 3.4 Mean 24 hour plasma glucose and insulin profiles in 12 healthy non-diabetic individuals. Shaded areas represent 95% confidence intervals. Glucose levels remain within tight limits, while there is considerable variation in insulin concentrations, particularly around meal times. Reprinted from The Lancet, 358, Owens et al., Insulins today and beyond 739–746 (2001), with permission from Elsevier
Box 3.2
Causes of hypoglycaemia in type 1 diabetes
1. Inappropriate insulin injection – e.g. excessive dose, inappropriate time, inappropriate insulin formulation. 2. Inadequate exogenous carbohydrate – e.g. missed meal or snack, overnight fast. 3. Increased carbohydrate utilisation – e.g. exercise. 4. Decreased endogenous glucose production – e.g. excessive alcohol consumption. 5. Increased insulin sensitivity – e.g. night time, exercise, weight loss. 6. Decreased insulin clearance – e.g. renal failure. Modified from Cryer et al., 2003.
62
FREQUENCY, CAUSES AND RISK FACTORS
concentrations or sensitivity and on glucose entry into the circulation. Although this classification is not perfect, it is a useful starting point for considering the causes of an episode of hypoglycaemia.
Patient Error It is common after an episode of hypoglycaemia for the person with diabetes or, indeed a member of the diabetes team, to try to ascertain why the episode occurred. Although there is always a risk of ‘spurious attribution’, in many instances an obvious cause can be identified and this is often the result of an error of judgement. Carbohydrate intake may have been inadequate because a meal was missed or delayed, or simply contained an insufficient content of carbohydrate. Alternatively, the patient may have injected too much insulin relative to the amount of carbohydrate in the meal. It is also not uncommon for insulin to be administered at an inappropriate time or for the ‘wrong’ insulin to be injected accidentally, e.g. a rapid-acting insulin analogue is administered at a time when basal insulin should have been given. An often unrecognised problem is the inadequate re-suspension of isophane (or lente) insulins or of fixed mixtures of insulin. If these insulins are not fully resuspended prior to subcutaneous injection, the insulin may be absorbed at variable rates resulting in unpredictable insulin levels and, thus, an increased risk of hypoglycaemia (Owens et al., 2001). Deliberate overdose of insulin is rare, but may result in protracted hypoglycaemia.
Alcohol The relationship of alcohol to hypoglycaemia is considered in detail in Chapter 5. Surveys based on patient interviews have implicated alcohol in up to one fifth of episodes of severe hypoglycaemia requiring hospital admission (Potter et al., 1982; Moses et al., 1985; Feher et al., 1989; Hart and Frier, 1998). In a recent study of 141 people treated for severe hypoglycaemia in three emergency centres in Copenhagen, alcohol was detected in the blood of 17% of the subjects (Pedersen-Bjergaard et al., 2005). The median blood alcohol concentration was 11 mmol/l. Alcohol inhibits gluconeogenesis and so may directly contribute to the development of hypoglycaemia. In addition, there are some data to suggest that alcohol attenuates the counterregulatory response to hypoglycaemia (Avogaro et al., 1993; Turner et al., 2001). However, the main impact of alcohol probably centres on its ability to impair awareness of hypoglycaemia and so hinder the ability of individuals to take appropriate corrective therapy (Kerr et al., 1990). Thus, an individual under the influence of alcohol may not recognise that he or she is hypoglycaemic, and even if the symptoms are recognised, the person may not have the capacity to self-treat. An episode of mild hypoglycaemia may therefore be converted rapidly into a severe episode. Moreover, friends, colleagues or bystanders may presume that the neuroglycopenic symptoms and signs exhibited by the individual are a consequence of alcohol intoxication and so again appropriate treatment and medical help may not be provided. It is for these reasons that alcohol is implicated in many instances of profound and protracted insulin-induced hypoglycaemia associated with permanent neurological damage (Arky et al., 1968).
RISK FACTORS FOR SEVERE HYPOGLYCAEMIA
63
Exercise Exercise is recommended for people with type 1 diabetes because of its positive physiological and psychological effects. However, exercise can increase the risk of hypoglycaemia both during the physical activity itself and in the recovery period (see Chapter 14). The reasons for the high incidence of exercise-induced hypoglycaemia in adults with type 1 diabetes have not been fully elucidated. Moderate exercise in non-diabetic individuals causes a fall in plasma insulin to 40–50% of pre-exercise levels (Galassetti et al., 2003). This normal physiological decline cannot occur when insulin is being administered exogenously, unless the dose is reduced before exercise commences (Sonnenberg et al., 1990). An acute increase in insulin sensitivity following exercise also increases the risk of hypoglycaemia (Sonnenberg et al., 1990). The metabolic and counterregulatory hormonal responses to acute hypoglycaemia and exercise are qualitatively very similar. Antecedent exercise blunts the counterregulatory response to subsequent acute hypoglycaemia (Galassetti et al., 2001a; Galassetti et al., 2001b). The inverse situation also applies, i.e., antecedent hypoglycaemia diminishes the counterregulatory response to subsequent exercise (Galassetti et al., 2003). This means that at the very time that metabolic substrate requirements are increasing, there is the potential for an acute failure of endogenous glucose production. Thus, impaired counterregulatory responses may be an important mechanism in promoting exercise-related hypoglycaemia in type 1 diabetes.
RISK FACTORS FOR SEVERE HYPOGLYCAEMIA In a relatively high proportion of cases – in some series as high as 40% (Potter et al., 1982; Feher et al., 1989) – it is not possible to identify the precipitating cause of an episode of hypoglycaemia. Indeed, this figure may be even higher, because the phenomenon of ‘spurious attribution’ means that in some instances the perceived precipitant of a given episode may have been an innocent bystander. It has, therefore, been increasingly recognised that diabetes healthcare professionals must look beyond conventional precipitating factors and consider other phenomena which may be associated with an increased risk of hypoglycaemia, namely, the risk factors for hypoglycaemia (Table 3.3). Many of these are discussed in more detail in other chapters.
Intensive Insulin Therapy Randomised trials, most notably the DCCT, have provided substantial data on the epidemiology of hypoglycaemia in adults with type 1 diabetes and, in particular, on the impact of intensive insulin therapy. The Diabetes Control and Complications Trial (DCCT) The DDCT was a landmark study and provided diabetes specialists with the long-awaited proof that strict glycaemic control limited the incidence and severity of microvascular complications in people with type 1 diabetes. A total of 1441 patients with type 1 diabetes
−
− − +
+
− −
Intensive therapy
+ + − − +∗ − − − − −
+
− − −
Low HbA1c
+ + + +
+
Previous episode
+ + + − + − + +∗ − + +
+
−
−
Duration
+ + − +
+ + + +
+
+
+
Impaired awareness
− −
+ +
+
C-peptide negative
− − − − F
−
+ M
Sex
− + − − −
+ + −
−
Age
− −
+
−
+
−
Insulin dose
+
+
ACE activity
This table examines the risk factors for severe hypoglycaemia that have been most commonly examined in adults with type 1 diabetes. Studies examining predominantly mild hypoglycaemia were not included. + = positive association between risk factor and severe hypoglycaemia; − = no association between risk factor and severe hypoglycaemia. ∗ = only a risk factor if awareness of hypoglycaemia not included. M = severe hypoglycaemia more common in men; F = severe hypoglycaemia more common in women.
Muhlhauser et al., 1985 Muhlhauser et al., 1987 MacLeod et al., 1993 Gold et al., 1994 EURODIAB, 1994 MacLeod et al., 1994 SDIS, 1994 Clarke et al., 1995 Pampanelli, et al., 1996 Bott et al., 1997 DCCT, 1997 Gold et al., 1997 Muhlhauser et al., 1998 Pedersen-Bjergaard et al., 2001 ter Braak et al., 2000 Leese et al., 2003 Pedersen-Bjergaard et al., 2003 Pedersen-Bjergaard et al., 2004 Leckie et al., 2005 UK Hypoglycaemia Study Group, 2007
Study
Table 3.3 Risk factors for severe hypoglycaemia in adults with type 1 diabetes
RISK FACTORS FOR SEVERE HYPOGLYCAEMIA
65
were randomly assigned either to intensive insulin therapy (based on multiple injection insulin regimens or continuous subcutaneous insulin infusion therapy) or to conventional insulin therapy (one or two insulin injections daily) (The Diabetes Control and Complications Trial Research Group, 1993). In the conventional group, patients did not generally perform home blood glucose monitoring, clinical reviews were undertaken every three months and patients were not informed about their HbA1c result, unless it was in excess of 13%. By contrast, in the intensive therapy group, subjects performed frequent home blood glucose monitoring, had monthly visits with the study team and also had frequent telephone contact to achieve as strict glycaemic control as possible. Over a mean follow-up of 6.5 years, the average HbA1c concentration in the intensive group was approximately 7.0% and in the conventional group approximately 8.8% (The Diabetes Control and Complications Trial Research Group, 1993). The subjects recruited to this study were not typical of the wider population of people with type 1 diabetes. The participants had greater motivation and individuals were not permitted to take part if, in the preceding two years, they had experienced more than one episode of severe neurological impairment without warning symptoms of hypoglycaemia, or more than two episodes of seizure or coma, regardless of attributed cause. Moreover, during the study itself, the occurrence of an episode of severe hypoglycaemia in an individual patient prompted a review of conventional risk factors and, in instances where probable causes were identified, corrective actions such as re-educating the patient were undertaken (The Diabetes Control and Complications Trial Research Group, 1997). Despite all these factors, 3788 episodes of severe hypoglycaemia occurred in the 1441 patients over the course of the study and, of these, 1027 were associated with coma and/or seizure (The Diabetes Control and Complications Trial Research Group, 1997). The rate of severe hypoglycaemia in the intensively-treated patients was 0.61 per patient per year, while that in the conventionally-treated group was 0.19 per patient per year, i.e., a three-fold difference. Over the mean of 6.5 years of follow-up, 65% of the intensive group experienced at least one episode of severe hypoglycaemia, compared with 35% of the individuals in the conventional group. The Stockholm Diabetes Intervention Study (SDIS) The SDIS was a much smaller study than the DCCT, but had similar aims and objectives. A group of 102 patients with type 1 diabetes were recruited and 89 remained after 7.5 years of follow-up (Reichard and Pihl, 1994). The mean HbA1c was 7.1% in the intensive group and 8.5% in the conventional group. Severe hypoglycaemia, defined as episodes requiring third party assistance, occurred in 80% of subjects in the intensive group and 58% of those in the conventional group over the follow-up period. Overall, the rate of severe hypoglycaemia was 1.1 per patient per year in the intensive group and 0.4 per patient per year in the intensive group. Other risk factors for severe hypoglycaemia were not reported. The Bucharest-Dusseldorf Study The DCCT was a multicentre study; 27 out of the 29 centres reported that intensive therapy was associated with an increased risk of severe hypoglycaemia, but in two centres no increased risk was observed (The Diabetes Control and Complications Trial Research Group, 1997). This led some investigators to claim that with appropriate education and training
66
FREQUENCY, CAUSES AND RISK FACTORS
intensive insulin therapy need not necessarily be associated with an increased risk of hypoglycaemia (Plank et al., 2004). In the Bucharest-Dusseldorf Study, 300 individuals with type 1 diabetes were randomised to: (a) conventional therapy for one year and then to one year of intensive therapy; (b) two years of intensive therapy; or (c) a four-day in-patient group teaching programme with conventional insulin therapy for one year (Muhlhauser et al., 1987). Glycated haemoglobin (HbA1 ), remained unchanged at around 12–13% during conventional therapy, but fell to ∼95% during intensive therapy. In the first year of the study, severe hypoglycaemia occurred in 6% of the intensively-treated patients and 12% of the conventionally-treated patients, and in year two the proportion of patients experiencing severe hypoglycaemia in both intensive groups fell to 3–4%. Observational data The Dusseldorf team reported observational data on the impact of intensive therapy on the frequency of severe hypoglycaemia (Bott et al., 1997). A total of 636 people with type 1 diabetes who had participated in a structured five-day in-patient treatment and teaching programme for intensification of insulin therapy in one of ten different hospitals in Germany, were re-examined at intervals over six years. The mean HbA1c fell from 8.3% to 7.6%. Severe hypoglycaemia, defined as episodes treated with intramuscular glucagon or intravenous glucose, decreased from 0.28 episodes per patient per year in the year preceding the programme to 0.17 episodes per patient per year afterwards (Bott et al., 1997). The variation in incidence of severe hypoglycaemia between different centres ranged from 0.05–0.27 episodes per patient per year. Pampanelli and colleagues (1996) from Perugia reported retrospective data on 112 individuals who had been commenced on intensive insulin therapy (preprandial soluble insulin and bedtime isophane insulin) at diagnosis of diabetes and who were attending clinic at least four times per year. Mean HbA1c was 7.2% and mean duration of diabetes was 7.8 years. Frequency of mild hypoglycaemia was estimated by a review of the patients’ blood glucose monitoring diaries in the nine months prior to study, and the overall rate was 35.6 episodes per patient per year. Severe hypoglycaemia, defined as episodes requiring third party assistance, was assessed retrospectively over the duration of diabetes and was recalled by only six patients (representing an overall rate of 0.001 episodes per patient per year). Summary Thus, in the DCCT and the SDIS, intensive insulin therapy was associated with a nearly three-fold increase in the risk of severe hypoglycaemia. By contrast, in other studies, the incidence of severe hypoglycaemia fell when intensive therapy was coupled with a detailed education programme. The teams from Dusseldorf and Perugia would argue that the high quality of their education programmes resulted in the low frequencies of observed severe hypoglycaemia. Although this may be the case, the importance of patient selection in all of these studies cannot be over-emphasised. In their study of 1076 adults with type 1 diabetes from the United Kingdom and Denmark, Pedersen-Bjergaard et al. (2004) examined a subset of 230 individuals whose clinical characteristics were similar to patients enrolled in the DCCT. This sub-group accounted for only 5.4% of all reported episodes of severe hypoglycaemia, with an overall rate of 0.35 episodes per patient per year, i.e., approximately
RISK FACTORS FOR SEVERE HYPOGLYCAEMIA
67
one quarter that of the entire group. Risk factors for severe hypoglycaemia in this group (impaired awareness and retinopathy) were different from that of the study population as a whole (see Table 3.3). Thus, the DCCT patients were not representative of people with type 1 diabetes, whereas the education programme undertaken in Dusseldorf is not something that is widely replicated in mainstream diabetes practice. Risk factors for hypoglycaemia may differ according to the specific characteristics of individuals being studied. Thus, in conclusion, although most specialists would accept that severe hypoglycaemia is more common in patients receiving intensive insulin therapy, it is not inevitable and better patient education may actually reduce the incidence.
Strict Glycaemic Control Strict glycaemic control, as evidenced by glycated haemoglobin concentrations that approach the upper end of the non-diabetic range, is closely linked to intensive insulin therapy. Since the DCCT results were published, glycated haemoglobin concentrations at or around this level have become the usual target for many people with type 1 diabetes. In the DCCT, there was a quadratic relationship between HbA1c and risk of severe hypoglycaemia (Figure 3.5), with the risk of hypoglycaemia increasing as HbA1c decreased (The Diabetes Control and Complications Trial Research Group, 1997). However, the attained HbA1c did not account for all the difference in risk of severe hypoglycaemia between the two arms of the study, as subjects in the intensive group still had an excess risk of severe hypoglycaemia after 100
Rate per 100 Patient Years
80
60
* 40
* * *
20
* *
* *
0 5
6
7
8
9
** *
*
10
11
* 12
13
14
HbA1c (%) During Study
Figure 3.5 Risk of severe hypoglycaemia as a function of monthly updated HbA1c in the Diabetes Control and Complications Trial. The circles represent data from the intensive group and asterisks data from the conventional group. The bold solid and bold dashed lines represent the regression plots for each group, and the non-bold dashed lines on either side show the upper and lower 95% confidence bands; DCCT (1997). Copyright © 1997 American Diabetes Association. Reprinted with permission from The American Diabetes Association
68
FREQUENCY, CAUSES AND RISK FACTORS
statistical adjustment for HbA1c concentration. Indeed, in the intensive group, only about 5% of the variation in frequency of severe hypoglycaemia could be explained by the glycated haemoglobin concentration (The Diabetes Control and Complications Trial Research Group, 1997). Other studies have reported variable relationships between glycaemic control and frequency of severe hypoglycaemia. In the study by Bott et al., (1997) a lower mean HbA1c was associated with severe hypoglycaemia, but there was no linear or quadratic relationship between HbA1c and severe hypoglycaemia. In the EURODIAB IDDM Complications Study, severe hypoglycaemia (defined as episodes requiring third party assistance) occurred in 32% of individuals over 12 months. A clear relationship to glycated haemoglobin was evident, in that 40% of individuals with an HbA1c < 54% were affected compared with 24% of individuals with a HbA1c > 78% (The EURODIAB IDDM Complications Study Group, 1994). In the study of workplace hypoglycaemia, recurrent severe hypoglycaemia was associated with strict glycaemic control, but no link was found in individuals who had experienced only one episode (Leckie et al., 2005). In several other studies, no relationship was observed between severe hypoglycaemia and glycated haemoglobin (Muhlhauser et al., 1985; Muhlhauser et al., 1987; MacLeod et al., 1993; Gold et al., 1997; Muhlhauser et al., 1998; ter Braak et al., 2000; Pedersen-Bjergaard et al., 2001; Leese et al., 2003; Pedersen-Bjergaard et al., 2003a; Pedersen-Bjergaard et al., 2004) after adjustment for other risk factors. Glycated haemoglobin does not, of course, provide the entire picture about an individual’s glycaemic control and although HbA1c may not predict risk of hypoglycaemia, low mean home blood glucose concentrations and excessive variability in blood glucose can identify individuals more prone to hypoglycaemia (Cox et al., 1994; Janssen et al., 2000). Thus, a straightforward relationship does not exist between severe hypoglycaemia and glycaemic control. For an episode of mild hypoglycaemia to progress to one that causes significant neuroglycopenia and impairs consciousness, other factors must operate, which negate the normal symptomatic and hormonal responses to hypoglycaemia.
Acquired Hypoglycaemia Syndromes Within each treatment group of the DCCT, the number of previous episodes of severe hypoglycaemia was the strongest predictor of risk of future episodes (The Diabetes Control and Complications Trial Research Group, 1997). Moreover, approximately 30% of patients in each group experienced a second episode of severe hypoglycaemia within four months following an initial episode. The importance of a previous history of severe hypoglycaemia in predicting future risk has been replicated in several other studies (MacLeod et al., 1993; Bott et al., 1997; Gold et al., 1997; Muhlhauser et al., 1998). Moreover, as demonstrated in Table 3.3, many studies have also linked increased duration of diabetes with an increased risk of severe hypoglycaemia. Cryer has suggested that the integrity of the glucose counterregulatory system may be a pivotal factor in determining whether the relative or absolute hyperinsulinism that frequently occurs in insulin-treated diabetes ultimately results in the development of hypoglycaemia (Cryer, 1994; Cryer et al., 2003). Three acquired hypoglycaemia syndromes are associated with an increased risk of severe hypoglycaemia in people with type 1 diabetes. These are considered in greater detail in Chapters 6 and 7.
RISK FACTORS FOR SEVERE HYPOGLYCAEMIA
69
Counterregulatory hormonal deficiencies Hypoglycaemia-induced secretion of glucagon declines in most patients within five years of developing type 1 diabetes (Gerich et al., 1973; Bolli et al., 1983). A defective epinephrine response to hypoglycaemia may develop some years later (Bolli et al., 1983; Hirsch and Shamoon, 1987; Dagogo-Jack et al., 1993). As with the glucagon response, the impaired epinephrine response is hypoglycaemia-specific but, in contrast to glucagon, it exhibits a threshold effect – i.e., an epinephrine response can still be elicited by hypoglycaemia, but only at a lower blood glucose concentration (Dagogo-Jack et al., 1993). If hypoglycaemia develops in patients who have this combined counterregulatory hormonal deficiency, glucose recovery may be severely compromised (see Chapter 6). Subjecting such patients to intensified insulin therapy increased the risk of severe hypoglycaemia by 25 times, compared with subjects who had an intact epinephrine response (White et al., 1983). Impaired awareness of hypoglycaemia In many patients with insulin-treated diabetes, the hypoglycaemia symptom profile alters with time, resulting in impaired perception of the onset of hypoglycaemia (see Chapter 7). Commonly, autonomic warning symptoms are diminished and neuroglycopenic symptoms predominate. Around 25% of people with type 1 diabetes develop impaired awareness of hypoglycaemia and the prevalence of this problem increases with the duration of insulin treatment (Hepburn et al., 1990; Gerich et al., 1991; Pramming et al., 1991). Prospective studies have demonstrated that the frequency of severe hypoglycaemia is increased up to six-fold in patients with impaired awareness compared to those with normal hypoglycaemia awareness (Gold et al., 1994; Clarke et al., 1995). Hypoglycaemia-associated central autonomic failure The above acquired hypoglycaemia syndromes tend to segregate together clinically. Patients with glycated haemoglobin concentrations close to the non-diabetic range are at greater risk of developing impaired awareness (Boyle et al., 1995; Kinsley et al., 1995; Pampanelli et al., 1996), while the glycaemic thresholds for the onset of symptoms and responses are altered in patients with impaired awareness (Grimaldi et al., 1990; Hepburn et al., 1991; Mokan et al., 1994; Bacatselos et al., 1995). Cryer has suggested that these acquired abnormalities represent a form of central ‘Hypoglycaemia-Associated Autonomic Failure’ (HAAF) in type 1 diabetes, speculating that recurrent severe hypoglycaemia is the primary cause (Cryer, 1992; Cryer et al., 2003). If hypoglycaemia is the precipitant, then it is possible to see how a vicious cycle may become established with the development of the acquired hypoglycaemia syndromes promoting further episodes of severe hypoglycaemia. Summary There is a well-known adage that ‘hypoglycaemia begets hypoglycaemia’. People who have experienced one episode of severe hypoglycaemia are much more likely to develop further episodes, and the greatest risk occurs in the weeks and months after the index event. Severe hypoglycaemia becomes a more common problem in people with long-standing type 1
70
FREQUENCY, CAUSES AND RISK FACTORS
diabetes (UK Hypoglycaemia Group, 2007). This is the legacy of the burden of hypoglycaemia that such individuals have experienced over many years with diabetes. The impaired symptomatic and counterregulatory responses to hypoglycaemia dramatically increase the likelihood that an episode of mild hypoglycaemia will progress to a more severe event.
Genetic Predisposition to Hypoglycaemia Most of the precipitants and risk factors for hypoglycaemia have been known about for many years. In 2001, Pedersen-Bjergaard et al. (2001) reported a novel risk factor for severe hypoglycaemia in adults with type 1 diabetes and raised the notion that some individuals may have an inherent genetic susceptibility to hypoglycaemia. The Danish investigators noted the similarity between endurance exercise and hypoglycaemia in that both are states of limited metabolic fuel availability. Previous studies had linked exercise performance to a particular polymorphism of the angiotensin-converting enzyme (ACE) gene. Specifically, the insertion (I) allele, which resulted in low serum ACE activity, was associated with superior performance capacity compared with the deletion (D) allele. In an initial retrospective survey of 207 adults with type 1 diabetes, patients with the DD genotype had a 3.2-fold increased risk of severe hypoglycaemia in the preceding two years, compared with individuals with the II genotype. There was also a significant relationship between serum ACE activity, with a 1.4 increment in risk of severe hypoglycaemia for every 10 U/l rise in serum ACE concentration (Figure 3.6). The serum ACE activity was directly linked to ACE genotype, and it remained a significant risk factor even after adjustment for conventional risk factors. Moreover, serum ACE activity was a stronger risk factor for severe hypoglycaemia in C-peptide negative individuals with impaired awareness, than in other groups (relative risk 1.7 per 10 U/l; Figure 3.7). No significant relationship was observed between serum ACE activity or genotype and frequency of mild hypoglycaemia.
Figure 3.6 Risk of severe hypoglycaemia according to serum ACE activity in 207 patients with type 1 diabetes, untreated with ACE inhibitors or angiotensin-2 receptor antagonists. Broken lines represent 95% confidence intervals. Reprinted from The Lancet, Pedersen-Bjergaard et al. (2001) with permission from Elsevier
RISK FACTORS FOR SEVERE HYPOGLYCAEMIA
71
Figure 3.7 Association between severe hypoglycaemia and serum ACE activity according to C-peptide status and self-estimated awareness of hypoglycaemia. Reprinted from The Lancet, PedersenBjergaard et al. (2001) with permission from Elsevier
These findings were replicated in a prospective study for one year in 107 adults (PedersenBjergaard et al., 2003a). Serum ACE activity in the fourth quartile was associated with a 2.7-fold increased risk of severe hypoglycaemia compared to activity in the lowest quartile. Compared to subjects with the II genotype, individuals with the DD genotype had a 1.8-fold increased risk of severe hypoglycaemia, although this did not reach statistical significance. Higher serum ACE concentrations were also associated with an increased risk of severe hypoglycaemia in Swedish children and adolescents with type 1 diabetes (Nordfeldt and Samuelsson, 2003). The Danish group have put forward a number of possible mechanisms to explain their observations (Pedersen-Bjergaard et al., 2001; Pedersen-Bjergaard et al., 2003a). They speculate that low levels of serum ACE may be associated with less cognitive deterioration during acute hypoglycaemia, thereby increasing the likelihood that remedial action to correct hypoglycaemia is taken. Alternatively, low serum ACE activity might enhance counterregulation or reduce production of toxic substances, e.g. reactive oxygen species, during hypoglycaemia. All these putative mechanisms remain highly speculative, but the authors also raise one other intriguing possibility: namely that ACE inhibition might reduce the frequency of hypoglycaemia. This may seem counterintuitive, because previous population-based studies suggested an association between severe hypoglycaemia and use of ACE inhibitors (Herings et al., 1996; Morris et al., 1997). However, these studies are flawed and a re-examination of the role of ACE inhibitors in ameliorating the impact of hypoglycaemia seems warranted. Recent data from one centre in Scotland have not demonstrated such a strong link between serum ACE and severe hypoglycaemia in adults with type 1 diabetes (Zammitt et al., 2007), nor has a study of children and adolescents with type 1 diabetes in Australia (Bulsara et al., 2007). This should, therefore, serve as a reminder that genetic susceptibility to any biological variable may not be the same in different populations and reinforces the need for the Danish observations to be examined in other countries and ethnic groups. However, the studies by Pedersen-Bjergaard and colleagues raise the intriguing prospect that there may be yet other genetic factors that influence susceptibility to hypoglycaemia. Identification of
72
FREQUENCY, CAUSES AND RISK FACTORS
these may help stratify risk in individuals with type 1 diabetes and may ultimately lead to the development of novel therapeutic interventions to prevent or ameliorate the impact of hypoglycaemia.
Absent Endogenous Insulin Secretion Several studies have demonstrated the importance of endogenous insulin secretion in defining risk of hypoglycaemia. Individuals who are C-peptide negative, i.e., who have no endogenous insulin production, have an approximately two- to fourfold increased risk of severe hypoglycaemia compared to people with detectable C-peptide (Bott et al., 1997; The Diabetes Control and Complications Trial Research Group, 1997; Muhlhauser et al., 1998; Pedersen-Bjergaard et al., 2001). These data mirror the clinical experience that severe hypoglycaemia is rare in the 12 months after the diagnosis of type 1 diabetes (Davis et al., 1997), when significant concentrations of endogenous insulin can be measured. The presumption is that the ability of endogenous insulin levels to fall in the face of a declining blood glucose concentration provides an additional layer of protection in the defences against hypoglycaemia and thus reduces overall risk.
Dose and Type of Insulin Few diabetes specialists who were practising in the mid-1980s will forget the controversy that surrounded the introduction of human insulin. A substantial and vocal minority of patients with type 1 diabetes claimed that the change from animal-derived to human insulin was associated with a loss of warning symptoms of hypoglycaemia and thus an increased risk. These claims were subject to considerable scientific scrutiny and ultimately a systematic review of the evidence found no evidence to support the premise that treatment with human, as opposed to animal, insulin was associated with an increased risk of hypoglycaemia (Airey et al., 2000). Since that time, several insulin analogues have been developed and their introduction into clinical practice has been accompanied by the publication of studies that purport to demonstrate that use of the insulin analogues is associated with a lower frequency of hypoglycaemia, in the face of stable or improved glycaemic control (Anderson et al., 1997; Garg et al., 2004; Hermansen et al., 2004). However, recent systematic reviews suggest that neither short-acting (Siebenhofer et al., 2004) nor long-acting (Warren et al., 2004) insulin analogues are associated with clinically significant lower rates of hypoglycaemia. The usual caveats of such clinical trials apply in terms of patient characteristics and recording of hypoglycaemia. In one observational study from Colorado, the frequency of severe hypoglycaemia rose in clinic patients in the immediate aftermath of the DCCT as efforts to intensify insulin therapy were instituted, but from 1996 onwards, the rates of severe hypoglycaemia declined and the authors linked this to the introduction of insulin lispro (Chase et al., 2001). This study was subject to the effects of numerous confounders, but further data from routine clinical practice are required to help clarify the impact of insulin analogues on risk of hypoglycaemia. The introduction of inhaled insulin (Exubera) has not been associated with a lower risk of hypoglycaemia in either type 1 or type 2 diabetes; when compared with insulin administered by the subcutaneous route, rates of severe hypoglycaemia were either equivalent (Hollander et al., 2004) or higher (Skyler et al., 2005).
RISK FACTORS FOR SEVERE HYPOGLYCAEMIA
73
Higher doses of insulin have also been associated with an increased risk of hypoglycaemia in some studies (The Diabetes Control and Complications Trial Research Group, 1997; ter Braak et al., 2000), but not in others (Table 3.3). Several possible explanations may account for this relationship – high insulin doses may be a sign of less endogenous insulin production, or of efforts to achieve strict glycaemic control. It may also reflect sub-optimal compliance with insulin therapy and so indicate a pattern of behaviour that predisposes to marked fluctuations in glucose control.
Sleep As has already been discussed in the section on asymptomatic hypoglycaemia, nocturnal hypoglycaemia is very common (Vervoort et al., 1996). In the DCCT, 43% of episodes of severe hypoglycaemia occurred between midnight and 8.00 a.m., and 55% of episodes occurred when individuals were asleep (The DCCT Research Group, 1991). The potential for nocturnal hypoglycaemia engenders significant anxiety among patients, particularly in individuals who live alone. Patients worry that they will not awaken when hypoglycaemia occurs and that they may be left incapacitated or die as a consequence. Many individuals, as a result, maintain higher blood glucose concentrations at bedtime to reduce the risk of nocturnal hypoglycaemia. Hypoglycaemia appears to be more common at night because counterregulatory hormone responses are blunted during sleep in people with type 1 diabetes (Jones et al., 1998; Banarer and Cryer, 2003). Moreover, hypoglycaemia awareness is also reduced during sleep (Banarer and Cryer, 2003), and so ultimately sleep impairs both the physiological and behavioural responses to hypoglycaemia. Unsurprisingly, considerable effort has been directed at developing strategies to reduce the risk of nocturnal hypoglycaemia and these are discussed in Chapter 4.
Microvascular Complications In a retrospective study of 44 patients with type 1 diabetes with impaired renal function (serum creatinine ≥ 133 umol/l and proteinuria), the incidence of severe hypoglycaemia was five times higher than in matched subjects with normal renal function (Muhlhauser et al., 1991). In other studies, severe hypoglycaemia was linked with nephropathy (ter Braak et al., 2000), peripheral neuropathy and retinopathy (Pedersen-Bjergaard et al., 2004). Although it is generally recognised that insulin requirement declines in advanced renal disease, with reduced clearance of insulin, this association between hypoglycaemia and nephropathy (and other microvascular complications) could be confounded by many other factors, e.g. concomitant drug therapy (ter Braak et al., 2000) and the co-existence of acquired hypoglycaemia syndromes which, like microvascular complications, are linked with increasing duration of diabetes. The role of peripheral autonomic neuropathy in increasing the risk of severe hypoglycaemia has been considered in several studies and deserves mention. In the EURODIAB IDDM Complications Study, the presence of abnormal cardiovascular reflexes was associated with a 1.7-fold increased risk of severe hypoglycaemia (Stephenson et al., 1996). Gold et al. (1997) also demonstrated that autonomic neuropathy was associated with a small increased risk of severe hypoglycaemia in 60 patients with type 1 diabetes, but no such relationship was demonstrated in the DCCT (The DCCT Research Group, 1991). The
74
FREQUENCY, CAUSES AND RISK FACTORS
mechanism underlying this association remains to be fully elucidated. Peripheral autonomic neuropathy often coexists with impaired hypoglycaemia awareness in patients with type 1 diabetes, presumably because both conditions are associated with diabetes of long duration (Hepburn et al., 1990), but impaired awareness can readily occur in the absence of peripheral autonomic neuropathy (Hepburn et al., 1990; Ryder et al., 1990; Bacatselos et al., 1995). It is well established that severe autonomic neuropathy is associated with ‘gastroparesis diabeticorum’, which can cause marked swings in blood glucose concentration. Although such severe gastroparesis is now rare, delayed gastric emptying is relatively common (Kong et al., 1996) and may explain at least part of the association between hypoglycaemia and autonomic neuropathy.
Social and Psychological Factors Psychological factors clearly play a crucial role in determining an individual’s likelihood of developing severe hypoglycaemia. Low mood (Gonder-Frederick and Cox, 1997), emotional coping (Bott et al., 1997) and socio-economic status (Muhlhauser et al., 1998; Leese et al., 2003) have been linked to risk of severe hypoglycaemia and so too have other more straightforward behavioural factors such an individual’s propensity to carry a supply of carbohydrate for emergency use (Bott et al., 1997) and their determination to achieve normoglycaemia (Muhlhauser et al., 1998). Since the 1980s, Cox and colleagues at the University of Virginia, USA, have carried out seminal investigations to explore the psychological impact of hypoglycaemia on people with type 1 diabetes. They developed a Fear of Hypoglycaemia scale, which sought to quantify the anxieties that people with type 1 diabetes have with respect to hypoglycaemia and the extent to which individuals take steps to avoid experiencing such episodes (Cox et al., 1987). Unsurprisingly, there is a close association between fear of hypoglycaemia and perceived risk of future severe hypoglycaemia (Gonder-Frederick et al., 1997). In many instances this is appropriate, in that ‘fear’ ratings are often high in people who have impaired awareness of hypoglycaemia and/or have experienced multiple episodes in the past (Gold et al., 1996). However, in other people, fear of hypoglycaemia may be high while absolute risk is low – such individuals often display high levels of trait anxiety or have had a traumatic previous experience of hypoglycaemia. It is often extremely difficult to persuade such individuals to maintain strict glycaemic control. Conversely, there are people who have a very low fear of hypoglycaemia, despite a propensity to recurrent episodes. Such individuals may fail to take appropriate precautionary measures, thereby putting themselves and others at risk if hypoglycaemia occurs, for example, when that individual is driving a car.
Endocrinopathies Endocrine disorders, such as Addison’s disease and hypopituitarism, which are associated with a deficiency of counterregulatory hormones, can be associated with an increased risk of hypoglycaemia in adults with type 1 diabetes. These are uncommon in everyday diabetes practice, but clinicians should maintain a high index of suspicion particularly in the patient who simultaneously demonstrates a marked and otherwise unexplained decline in insulin requirements.
CONCLUSIONS
75
Other Risk Factors The list of other possible risk factors for hypoglycaemia is a long one. Smoking is a relatively novel marker of increased risk (Pedersen-Bjergaard et al., 2004), but may be confounded by other diabetes-related and lifestyle factors, e.g. regular use of alcohol (ter Braak et al., 2000). In the DCCT, men and adolescents were at increased risk (The Diabetes Control and Complications Trial Research Group, 1997), but these associations have not been replicated with any consistency in other studies (Table 3.3). Risks associated with pregnancy are addressed in Chapter 10.
Causes and Risk Factors for Hypoglycaemia: Summary and Conclusions Many of the risk factors described above are inter-related and any given individual may have more than one, which clearly will increase their overall risk. However, at its most fundamental level, hypoglycaemia in adults with type 1 diabetes is a consequence of the inability of exogenous insulin levels to fall in response to a declining blood glucose concentration. In people who have had type 1 diabetes for several years, the situation is exacerbated because of a failure of the normal physiological counterregulatory defence mechanisms, which in non-diabetic individuals serve to increase exogenous glucose production and generate warning symptoms. This counterregulatory failure worsens with increasing duration of diabetes, particularly if glycaemic control is strict and the individual has experienced recurrent episodes of severe hypoglycaemia. Other factors may come in to play, for example particular behavioural patterns and the time-action profiles of the exogenous insulin. Recent data also raise the possibility that there may be inherent genetic susceptibility to hypoglycaemia, but this needs to be affirmed in wider populations and the underlying mechanisms more clearly dissected. Novel strategies to reduce overall risk of hypoglycaemia are urgently required.
CONCLUSIONS • There are no definitive criteria for what constitutes hypoglycaemia, but most specialists distinguish mild from severe episodes depending on whether or not the individual is able to self-treat. • The ‘average’ adult with type 1 diabetes will experience many thousands of episodes of mild hypoglycaemia over a lifetime, with a typical frequency of one to two episodes per week. • Severe hypoglycaemia is less common, and on average occurs once or twice every year, with an annual prevalence of around 30%. However, the distribution is heavily skewed, such that many individuals are unaffected over a calendar year, while a small number experience recurrent episodes. • Severe hypoglycaemia requiring treatment with intramuscular glucagon and/or intravenous glucose is even less common, and the majority of all episodes of hypoglycaemia are managed in the community by the patient and/or relatives and friends, without recourse to emergency services.
76
FREQUENCY, CAUSES AND RISK FACTORS
• Hypoglycaemia occurs when there is an imbalance between insulin-mediated glucose disposal and glucose influx into the circulation from the liver and exogenous carbohydrate. Typical precipitants include patient error in insulin dosage, alcohol and exercise. • Plasma concentrations of exogenous insulin cannot decline in response to falling blood glucose and the time action profiles of current insulins do not accurately mimic the normal physiological variation in insulin. In a high proportion of cases, no underlying precipitant of a given episode of hypoglycaemia can be identified. • The DCCT and other intervention studies have provided substantial information of the epidemiology of severe hypoglycaemia, but individuals participating in such studies may not be representative of the wider population who have type 1 diabetes. • A major determinant of increased risk of severe hypoglycaemia is HypoglycaemiaAssociated Autonomic Failure, which is a consequence of exposure to recurrent episodes of hypoglycaemia. This is a feature of individuals with long-standing diabetes, intensive insulin therapy and strict glycaemic control. • Recent data suggest that there may be specific genetic factors that predispose to an increased risk of hypoglycaemia.
REFERENCES Airey CM, Williams DRR, Martin PG, Bennett CMT, Spoor PA (2000). Hypoglycaemia induced by exogenous insulin – ‘human’ and animal insulin compared. Diabetic Medicine 17: 416–32. Allen C, LeClaire T, Palta M, Daniels K, Meredith M, D’Alessio DJ, for the Wisconsin Diabetes Registry Project (2001). Risk factors for frequent and severe hypoglycemia in type 1 diabetes. Diabetes Care 24: 1878–81. American Diabetes Association Workgroup on Hypoglycemia (2005). Defining and Reporting Hypoglycemia in Diabetes. Diabetes Care 28: 1245–9. Anderson JH, Brunelle RL, Koivisto VA, Pfutzner A, Trautmann ME, Vignati L, DiMarchi R, The Multicenter Insulin Lispro Study Group (1997). Reduction of postprandial hyperglycemia and frequency of hypoglycemia in IDDM patients on insulin-analog treatment. Diabetes 46: 265–70. Arky RA, Veverbrants E, Abramson EA (1968). Irreversible hypoglycemia. A complication of alcohol and insulin. Journal of the American Medical Association 206: 575–8. Avogaro A, Beltramello P, Gnudi L, Maran A, Valerio A, Miola M et al. (1993). Alcohol intake impairs glucose counterregulation during acute insulin-induced hypoglycemia in IDDM patients. Evidence for a critical role of free fatty acids. Diabetes 42: 1626–34. Bacatselos SO, Karamitsos DT, Kourtoglou GI, Zambulis CX, Yovos JG, Vyzantiadis AT (1995). Hypoglycaemia unawareness in type 1 diabetic patients under conventional insulin treatment. Diabetes Nutrition and Metabolism 8: 267–75. Banarer S, Cryer PE (2003). Sleep-related hypoglycemia-associated autonomic failure in type 1 diabetes. Diabetes 52: 1195–1203. Bode BW, Schwartz S, Stubbs HA, Block JE (2005). Glycemic characteristics in continuously monitored patients with type 1 and type 2 diabetes. Normative values. Diabetes Care 28: 2361–6. Bolli G, De Feo P, Compagnucci P, Cartechini MG, Angeletti G, Santeusanio F et al. (1983). Abnormal glucose counterregulation in insulin-dependent diabetes mellitus. Interaction of antiinsulin antibodies and impaired glucagon and epinephrine secretion. Diabetes 32: 134–41.
REFERENCES
77
Bott S, Bott U, Berger M, Muhlhauser I (1997). Intensified insulin therapy and the risk of severe hypoglycaemia. Diabetologia 40: 926–32. Boyle PJ, Schwartz NS, Shah SD, Clutter WE, Cryer PE (1988). Plasma glucose concentrations at the onset of hypoglycemic symptoms in patients with poorly controlled diabetes and in non-diabetics. New England Journal of Medicine 318: 1487–92. Boyle PJ, Kempers SF, O’Connor AM, Nagy RJ (1995). Brain glucose uptake and unawareness of hypoglycemia in patients with insulin-dependent diabetes mellitus. New England Journal of Medicine 333: 1726–31. Bulsara MK, Holman CDJ, van Bockxmeer FM, Davis EA, Gallego PH, Beilby JP, Palmer LJ et al. (2007). The relationship between ACE genotype and risk of severe hypoglycemia in a large population-based cohort of children and adolescents with type 1 diabetes. Diabetologia 50: 965–71. Chase HP, Lockspeiser T, Peery B, Shepherd M, MacKenzie T, Anderson J, Garg SK (2001). The impact of the Diabetes Control and Complications Trial and Humalog insulin on glycohemoglobin levels and severe hypoglycemia in type 1 diabetes. Diabetes Care 24: 430–4. Clarke WL, Cox DJ, Gonder-Frederick LA, Julian D, Schlundt D, Polonsky W (1995). Reduced awareness of hypoglycemia in adults with IDDM. A prospective study of hypoglycemic frequency and associated symptoms. Diabetes Care 18: 517–22. Cox DJ, Irvine A, Gonder-Frederick L, Nowacek G, Butterfield J (1987). Fear of hypoglycemia: quantification, validation and utilization. Diabetes Care 10: 617–21. Cox DJ, Kovatchev BP, Julian DM, Gonder-Frederick LA, Polonsky WH, Schlundt DG, Clarke WL (1994). Frequency of severe hypoglycemia in insulin-dependent diabetes mellitus can be predicted from self-monitoring blood glucose data. Journal of Clinical Endocrinology and Metabolism 79: 1659–62. Cryer PE (1992). Iatrogenic hypoglycemia as a cause of hypoglycemia-associated autonomic failure in IDDM. A vicious cycle. Diabetes 41: 255–260. Cryer PE (1994). Hypoglycemia: the limiting factor in the management of IDDM. Diabetes 43: 1378–89. Cryer PE, Davis SN, Shamoon H (2003). Hypoglycemia in diabetes. Diabetes Care 26: 1902–12. Dagogo-Jack SE, Craft S, Cryer PE (1993). Hypoglycemia-associated autonomic failure in insulindependent diabetes mellitus. Recent antecedent hypoglycemia reduces autonomic responses to, symptoms of, and defense against subsequent hypoglycemia. Journal of Clinical Investigation 91: 819–28. Davis EA, Keating B, Byrne GC, Russell M, Jones TW (1997). Hypoglycemia: Incidence and clinical predictors in a large population-based sample of children and adolescents with IDDM. Diabetes Care 20: 22–5. Donnelly LA, Morris AD, Frier BM, Ellis JD, Donnan PT, Durrant R et al., for the DARTS/MEMO Collaboration (2005). Frequency and predictors of hypoglycaemia in type 1 and insulin-treated type 2 diabetes: a population-based study. Diabetic Medicine 22: 749–55. Feher MD, Grout P, Kennedy A, Elkeles RS, Touquet R (1989). Hypoglycaemia in an inner-city accident and emergency department: a 12-month survey. Archives of Emergency Medicine 6: 183–8. Galassetti P, Mann S, Tate D, Neill RA, Costa F, Wasserman DH, Davis SN (2001a). Effects of antecedent prolonged exercise on subsequent counterregulatory responses to hypoglycemia. American Journal of Endocrinology 280: E908–17. Galassetti P, Neill AR, Tate D, Ertl AC, Wasserman DH, Davis SN (2001b). Sexual dimorphism in counterregulatory responses to hypoglycemia after antecedent exercise. Journal of Clinical Endocrinology and Metabolism 86: 3516–24. Galassetti P, Tate D, Neill RA, Morrey S, Wasserman DH, Davis SN (2003). Effect of antecedent hypoglycemia on counterregulatory responses to subsequent euglycemic exercise in type 1 diabetes. Diabetes 52: 1761–9. Garg SK, Gottlieb PA, Hisatomi ME, D’Souza A, Walker AJ, Izuora KE, Chase HP (2004). Improved glycemic control without an increase in severe hypoglycemic episodes in intensively treated patients
78
FREQUENCY, CAUSES AND RISK FACTORS
with type 1 diabetes receiving morning, evening, or split dose insulin glargine. Diabetes Research and Clinical Practice 66: 49–56. Gerich J, Langlois M, Noacco C, Karam J, Forsham P (1973). Lack of glucagon response to hypoglycemia in diabetes: evidence for an intrinsic pancreatic alpha-cell defect. Science 182: 171–3. Gerich JE, Mokan M, Veneman T, Korytkowski M, Mitrakou A (1991). Hypoglycemia unawareness. Endocrine Reviews 12: 356–71. Gold AE, MacLeod KM, Frier BM (1994). Frequency of severe hypoglycemia in patients with type 1 diabetes with impaired awareness of hypoglycemia. Diabetes Care 17: 697–703. Gold AE, Frier BM, MacLeod KM, Deary IJ (1997). A structural equation model for predictors of severe hypoglycaemia in patients with insulin-dependent diabetes mellitus. Diabetic Medicine 14: 309–15. Gonder-Frederick L, Clarke WL, Cox DJ (1997). The emotional, social, and behavioural implications of insulin-induced hypoglycemia. Seminars in Clinical Neuropsychiatry 2: 57–65. Gonder-Frederick LA, Cox D (1997). The relationship between psychological depression and frequency of severe hypoglycemia (SH) in IDDM. Diabetes 46: 268A(Abstract) Grimaldi A, Bosquet F, Davidoff P, Digy JP, Sachon C, Landault C et al. (1990). Unawareness of hypoglycemia by insulin-dependent diabetics. Hormone and Metabolic Research 22: 90–5. Hart SP, Frier BM (1998). Causes, management and morbidity of acute hypoglycaemia in adults requiring hospital admission. Quarterly Journal of Medicine 91: 505–10. Heller SR, Macdonald IA (1996). The measurement of cognitive function during acute hypoglycaemia: experimental limitations and their effect on the study of hypoglycaemia unawareness. Diabetic Medicine 13: 607–15. Hepburn DA, Patrick AW, Eadington DW, Ewing DJ, Frier BM (1990). Unawareness of hypoglycaemia in insulin-treated diabetic patients: prevalence and relationship to autonomic neuropathy. Diabetic Medicine 7: 711–17. Hepburn DA, Patrick AW, Brash HM, Thomson I, Frier BM (1991). Hypoglycaemia unawareness in type 1 diabetes: a lower plasma glucose is required to stimulate sympatho-adrenal activation. Diabetic Medicine 8: 934–45. Herings RMC, de Boer A, Stricker BHC, Leufkens HGM, Porsius A (1996). Hypoglycemia associated with the use of inhibitors of angiotensin-converting enzyme. Lancet 345: 1195–8. Hermansen K, Fontaine P, Kukolja KK, Peterkova V, Leth G, Gall M-A (2004). Insulin analogues (insulin detemir and insulin aspart) versus traditional human insulins (NPH insulin and regular human insulin) in basal-bolus therapy for patients with type 1 diabetes. Diabetologia 47: 622–9. Hirsch BR, Shamoon H (1987). Defective epinephrine and growth hormone responses in type 1 diabetes are stimulus specific. Diabetes 36: 20–6. Hollander PA, Blonde L, Rowe R, Mehta AE, Milburn JL, Hershon KS et al., the Exubera Phase III Study Group (2004). Efficacy and safety of inhaled insulin (Exubera) compared with subcutaneous insulin therapy in patients with type 2 diabetes. Diabetes Care 27: 2356–62. Janssen MMJ, Snoek FJ, de Jongh RT, Casteleijn S, Deville W, Heine RJ (2000). Biological and behavioural determinants of the frequency of mild, biochemical hypoglycaemia in patients with Type 1 diabetes on multiple insulin injection therapy. Diabetes Metabolism Research and Reviews 16: 157–63. Johnson ES, Koepsell TD, Reiber G, Stergachis A, Platt R (2002). Increasing incidence of serious hypoglycemia in insulin users. Journal of Clinical Epidemiology 55: 253–9. Jones TW, Porter P, Sherwin RS, Davis EA, O’Leary P, Frazer F et al. (1998). Decreased epinephrine responses to hypoglycemia during sleep. New England Journal of Medicine 338: 1657–62. Kerr D, Macdonald IA, Heller SR, Tattersall RB (1990). Alcohol causes hypoglycaemic unawareness in healthy volunteers and patients with type 1 (insulin-dependent) diabetes. Diabetologia 33: 216–21. Kinsley BT, Widom B, Simonson DC (1995). Differential regulation of counterregulatory hormone secretion and symptoms during hypoglycemia in IDDM. Diabetes Care 18: 17–26. Kong M-F, Macdonald IA, Tattersall RB (1996). Gastric emptying in diabetes. Diabetic Medicine 13: 112–9.
REFERENCES
79
Leckie AM, Graham MK, Grant JB, Ritchie PJ, Frier BM (2005). Frequency, severity, and morbidity of hypoglycemia occurring in the workplace in people with insulin-treated diabetes. Diabetes Care 28: 1333–8. Leese GP, Wang J, Broomhall J, Kelly P, Marsden A, Morrison W et al., for the DARTS/MEMO Collaboration (2003). Frequency of severe hypoglycemia requiring emergency department treatment in type 1 and type 2 diabetes. A population-based study of health service resource use. Diabetes Care 26: 1176–80. MacLeod KM, Hepburn DA, Frier BM (1993). Frequency and morbidity of severe hypoglycaemia in insulin-treated diabetic patients. Diabetic Medicine 10: 238–45. MacLeod KM, Deary IJ, Graham KS, Hepburn DA, Frier BM (1994). Hypoglycaemia unawareness in adult patients with type 1 diabetes: relationship to severe hypoglycaemia and cognitive impairment. Diabetes, Nutrition and Metabolism 7: 205–11. MacLeod KM, Gold AE, Frier BM (1995). Frequency, severity and symptomatology of hypoglycaemia: a comparative trial of human and porcine insulins in type 1 diabetic patients. Diabetic Medicine 12: 134–41. Mokan M, Mitrakou A, Veneman T, Ryan C, Korytkowski M, Cryer PE, Gerich J (1994). Hypoglycemia unawareness in IDDM. Diabetes Care 17: 1397–403. Morris AD, Boyle DIR, McMahon AD, Pearce H, Evans JMM, Newton RW et al., for the DARTS/MEMO Collaboration (1997). ACE inhibitor use is associated with hospitalization for severe hypoglycemia in patients with diabetes. Diabetes Care 20: 1363–7. Moses RG, Hubert PA, Lewis-Driver DJ (1985). Severe hypoglycaemia. A one-year prospective survey in Wollongong. Medical Journal of Australia 142: 294–6. Muhlhauser I, Berger M, Sonnenberg G, Koch J, Jorgens V, Schernthaner G, Scholz V (1985). Incidence and management of severe hypoglycemia in 434 adults with insulin-dependent diabetes mellitus. Diabetes Care 8: 268–73. Muhlhauser I, Bruckner I, Berger M, Cheta D, Jorgens V, Ionescu-Tirgoviste C et al. (1987). Evaluation of an intensified insulin treatment and teaching programme as routine management of Type 1 (insulin-dependent) diabetes. The Bucharest-D¨usseldorf Study. Diabetologia 30: 681–90. Muhlhauser I, Toth G, Sawicki PT, Berger M (1991). Severe hypoglycemia in type 1 diabetic patients with impaired kidney function. Diabetes Care 14: 344–6. Muhlhauser I, Overmann H, Bender R, Bott U, Berger M (1998). Risk factors for severe hypoglycaemia in adult patients with type 1 diabetes. Diabetologia 41: 1274–82. Nordfeldt S, Samuelsson U (2003). Serum ACE predicts severe hypoglycemia in children and adolescents with type 1 diabetes. Diabetes Care 26: 274–8. O’Neill, S (1997). Make four the floor. Balance Jan/Feb: 20–3. Owens DR, Zinman B, Bolli GB (2001). Insulins today and beyond. Lancet 358: 739–46. Pampanelli S, Fanelli C, Lalli C, Ciofetta M, Del Sindaco P, Lepore M et al. (1996). Long-term intensive insulin therapy in IDDM: effects on HbA1c , risk for severe and mild hypoglycaemia, status of counterregulation and awareness of hypoglycaemia. Diabetologia 39: 677–86. Pedersen-Bjergaard U, Agerholm-Larsen B, Pramming S, Hougaard P, Thorsteinsson B (2001). Activity of angiotensin-converting enzyme and risk of severe hypoglycaemia in type 1 diabetes mellitus. Lancet 357: 1248–53. Pedersen-Bjergaard U, Agerholm-Larsen B, Pramming S, Hougaard P, Thorsteinsson B (2003a). Prediction of severe hypoglycaemia by angiotensin-converting enzyme activity and genotype in type 1 diabetes. Diabetologia 46: 89–96. Pedersen-Bjergaard U, Pramming S, Thorsteinsson B (2003b). Recall of severe hypoglycaemia and self-estimated state of awareness in type 1 diabetes. Diabetes Metabolism Research and Reviews 19: 232–40. Pedersen-Bjergaard U, Pramming S, Heller SR, Wallace TM, Jorgensen HV, Matthews DR et al. (2004). Severe hypoglycaemia in 1076 adult patients with type 1 diabetes: influence of risk markers and selection. Diabetes Metabolism Research and Reviews 20: 479–86.
80
FREQUENCY, CAUSES AND RISK FACTORS
Pedersen-Bjergaard U, Reubsaet JLE, Nielsen SL, Pedersen-Bjergaard S, Perrild H, Pramming S, Thorsteinsson B (2005). Psychoactive drugs, alcohol, and severe hypoglycemia in insulin-treated diabetes: Analysis of 141 cases. American Journal of Medicine 118: 307–10. Plank J, Kohler G, Rakovac I, Semlitsch BM, Horvath K, Bock G, et al. (2004). Long-term evaluation of a structured out-patient education programme for intensified insulin therapy in patients with type 1 diabetes: a 12 year follow-up. Diabetologia 47: 1370–5. Potter J, Clarke P, Gale EAM, Dave SH, Tattersall RB (1982). Insulin-induced hypoglycaemia in an accident and emergency department: the tip of an iceberg? British Medical Journal 285: 1180–2. Pramming S, Thorsteinsson B, Bendtson I, Binder C (1990). The relationship between symptomatic and biochemical hypoglycaemia in insulin-dependent diabetic patients. Journal of Internal Medicine 228: 641–6. Pramming S, Thorsteinsson B, Bendtson I, Binder C (1991). Symptomatic hypoglycaemia in 411 type 1 diabetic patients. Diabetic Medicine 8: 217–22. Reichard P, Pihl M (1994). Mortality and treatment side-effects during long-term intensified conventional insulin treatment in the Stockholm Diabetes Intervention Study. Diabetes 43: 313–7. Ryder REJ, Owens DR, Hayes TM, Ghatei MA, Bloom SR (1990). Unawareness of hypoglycaemia and inadequate hypoglycaemic counterregulation: no causal relationship with diabetic autonomic neuropathy. British Medical Journal 301: 783–7. Service FJ (1995). Hypoglycemic disorders. New England Journal of Medicine 332: 1144–52. Siebenhofer A, Plank J, Berghold A, Horvath K, Sawicki PT, Beck P, Pieber TR (2004). Meta-analysis of short-acting insulin analogues in adult patients with type 1 diabetes: continuous insulin infusion versus injection therapy. Diabetologia 47: 1895–1905. Skyler JS, Weinstock RS, Raskin P, Yale J-F, Barrett E, Gerich JE, Gerstein HC, the Inhaled Insulin Phase III type 1 Diabetes Study Group (2005). Diabetes Care 28: 1630–35. Sonnenberg GE, Kemmer FW, Berger M (1990). Exercise in type 1 (insulin-dependent) diabetic patients treated with continuous subcutaneous insulin infusion. Prevention of exercise induced hypoglycaemia. Diabetologia 33: 696–703. Stephenson JM, Kempler P, Cavallo Perin P, Fuller JH, the EURODIAB IDDM Complications Study Group (1996). Is autonomic neuropathy a risk factor for severe hypoglycaemia? The EURODIAB IDDM Complications Study. Diabetologia 39: 1372–6. ter Braak EWMT, Appelman AMMF, van de Laak MF, Stolk RP, Van Haeften TW, Erkelens DW (2000). Clinical characteristics of type 1 diabetic patients with and without severe hypoglycemia. Diabetes Care 23: 1467–71. The DCCT Research Group (1991). Epidemiology of severe hypoglycemia in the Diabetes Control and Complications Trial. American Journal of Medicine 90: 450–9. The Diabetes Control and Complications Trial Research Group (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New England Journal of Medicine 329: 977–86. The Diabetes Control and Complications Trial Research Group (1997). Hypoglycemia in the Diabetes Control and Complications Trial. Diabetes 46: 271–86. The EURODIAB IDDM Complications Study Group (1994). Microvascular and acute complications in IDDM patients: the EURODIAB IDDM Complications Study. Diabetologia 37: 278–85. Thorsteinsson B, Pramming S, Lauritzen T, Binder C (1986). Frequency of daytime biochemical hypoglycaemia in insulin-treated diabetic patients: relation to daily median blood glucose concentrations. Diabetic Medicine 3: 147–51. Turner BC, Jenkins E, Kerr D, Sherwin RS, Cavan DA (2001). The effect of evening alcohol consumption on next-morning glucose control in type 1 diabetes. Diabetes Care 24: 1888–93. UK Hypoglycaemia Study Group (2007). Risk of hypoglycaemia in types 1 and 2 diabetes: effects of treatment modalities and their duration. Diabetologia 50: 1140–47.
REFERENCES
81
Vervoort G, Goldschmidt HMG, van Doorn LG (1996). Nocturnal blood glucose profiles in patients with type 1 diabetes mellitus on multiple (> 4) daily insulin injection regimens. Diabetic Medicine 13: 794–9. Warren E, Weatherley-Jones E, Chilcott J, Beverley C (2004). Systematic review and economic evaluation of a long-acting insulin analogue, insulin glargine. Health Technology Assessment 8 (45): iii, 1–57. White NH, Skor DA, Cryer PE, Levandoski LA, Bier DM, Santiago JV (1983). Identification of type 1 diabetic patients at increased risk for hypoglycemia during intensive therapy. New England Journal of Medicine 308: 485–91. Zammitt NN, Geddes J, Warren RE, Marioni R, Ashby JP, Frier BM (2007). Serum AngiotensinConverting Enzyme and frequency of severe hypoglycaemia in type 1 diabetes: does a relationship exist? Diabetic Medicine, in press.
4 Nocturnal Hypoglycaemia Simon R. Heller
INTRODUCTION People with type 1 diabetes, who owe their lives to insulin, worry as much about hypoglycaemia as they do about the prospect of developing serious diabetic complications, and episodes occurring during sleep are feared more than any other. Several studies have confirmed how commonly hypoglycaemia occurs during the night, and the possibility of the development of severe nocturnal episodes can drive parents to sleep in their child’s bedroom for many years. However, nocturnal hypoglycaemia may have consequences beyond a domestic emergency. Asymptomatic episodes may contribute to impaired hypoglycaemia awareness and deficient counterregulation, while nocturnal hypoglycaemia may also be associated with cognitive impairment and is implicated in the ‘Dead in Bed’ syndrome (see Chapter 12) with the increased risk of sudden death during sleep in young people with type 1 diabetes. In this chapter, the reasons why nocturnal hypoglycaemia occurs so frequently in type 1 diabetes are discussed. Also, some clinical approaches are described that might help to reduce its frequency and severity and some controversial topics are considered such as the Somogyi effect.
EPIDEMIOLOGY – HOW COMMON IS NOCTURNAL HYPOGLYCAEMIA? Nocturnal hypoglycaemia has always been of major concern to individuals with diabetes, with many patients experiencing severe episodes, but it was not until the 1970s that this problem was studied systematically. Patients with type 1 diabetes on routine insulin therapy were admitted for overnight monitoring using intermittent venous blood sampling (Gale and Tattersall, 1979). Nocturnal hypoglycaemia was reported in 22 of 39 adults, who had what would now be considered poorly controlled diabetes and were being treated with one or two injections of insulin each day. Since then a number of investigators have examined the frequency of nocturnal hypoglycaemia, either using patients’ self-reports or by measuring blood glucose at intervals overnight. Although some studies have reported relatively low rates, others have demonstrated how common nocturnal hypoglycaemia can be. Reported rates of hypoglycaemia vary between 7 and 60% both in children and adults (Table 4.1).
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
A
A
A
C
A
C
A
C C
C
C C
82
58
71
23 25 135
31
61 150
29
56 47
Adults (A) or children (C)
39
Number
MIT/CSII MIT/CSII
BD
BD BD/MIT
Basal bolus
BD insulin OD insulin MDI CSII Conventional
Conventional
BD insulin
BD insulin
Treatment
45
35
CGM CGM
17
BG BG BG
29
34 10 30 44 14
29
3.0
BG
BG
BG
BG
BG
BG
BG
Sampling (blood glucose (BG) or continuous (CGM))
70 27
13 47
9 36 7
22
2.5
8
9 12 2
11
9
27
56
2.0
Percentage with glucose below: (mmol/l)
100 100
100
49
67
57 66 100
9
27
64
Proportion asymptomatic
8.6
8.8 (A1c)
8.6 (A1c)
8.3 (A1) 7.0 (A1c) 10.3 (A1)
9.1 (A1)
9.5 (A1)
Mean HbA1c
BD = twice daily; CSII = continuous subcutaneous insulin infusion; OD = once daily; MDI = multiple doses of insulin; CGM = continuous glucose monitoring; MIT = multiple injection therapy; BG = blood glucose
Gale and Tattersall, 1979 Dornan et al., 1981 Pramming et al., 1985 Whincup and Milner, 1987 Bendtson et al., 1988 Shalwitz et al., 1990 Vervoort et al., 1996 Porter et al., 1997 Beregszaszi et al., 1997 Matyka et al., 1999b Boland et al., 2001 Kaufman et al., 2002
Study
Table 4.1 Frequency of nocturnal hypoglycaemia in insulin-treated diabetes
EPIDEMIOLOGY OF NOCTURNAL HYPOGLYCAEMIA?
85
7 6
Glucose (mmol/l)
5 4 3 2 1 0
21:00
23:00
01:00
03:00
05:00
07:00
Time (hours)
Figure 4.1 Overnight glucose profiles of one child with type 1 diabetes who was hypoglycaemic on both study nights. The shaded area represents the range of glucose values from the overnight profiles of the children without diabetes
Although this is clearly dependent upon the blood glucose level that is designated as representing hypoglycaemia, it is interesting that the studies that reported the lowest rates of hypoglycaemia had sampled blood glucose less frequently, and it is conceivable that hypoglycaemia at other times of the night was simply not identified. Furthermore, most early studies involved patients with poor glycaemic control. It seems likely that with intensive insulin therapy now being offered to most patients, rates of nocturnal hypoglycaemia are even higher. This certainly appears to be true in children. In one alarming study of children treated with multiple injection therapy, the rates of nocturnal hypoglycaemia exceeded 50% (Beregszaszi et al., 1997). It is noteworthy that children have been directly observed to sleep through most nocturnal episodes despite having a blood glucose value below 2 mmol/l (Matyka et al., 1999b) (Figure 4.1). The introduction of continuous blood glucose monitoring with glucose being measured every few minutes has allowed more frequent recording during sleep. Initial reports using these methods have suggested that hypoglycaemic events may be even more common than was proposed previously (Table 4.1). However, although continuous blood glucose recording appears to have revealed a high rate of nocturnal episodes in some studies, it may over-estimate the overall frequency (McGowan et al., 2002). The technology measures glucose in the extra cellular space and not in the blood vessels, and the relationship between the glucose concentrations in these sites is not clear (Kulcu et al., 2003). It is also possible that the technology itself produces an inbuilt bias (Wentholt et al., 2005). Nevertheless, not only is nocturnal hypoglycaemia common when measured by conventional blood glucose sampling but it is also often of long duration with some episodes lasting for more than three hours (Gale and Tattersall, 1979; Matyka et al., 1999b).
NOCTURNAL HYPOGLYCAEMIA
86
CAUSES OF NOCTURNAL HYPOGLYCAEMIA The Limitations of Therapeutic Insulin Delivery The ability of people with insulin-treated diabetes to maintain strict glucose targets and prevent long-term tissue damage is compromised by the deficiencies of currently available methods of insulin delivery. In non-diabetic individuals, endogenous secretion of insulin precisely meets demand. When food is eaten, insulin secretion increases rapidly to match nutrient intake, particularly when the ingested food contains carbohydrate. In between meals, basal insulin secretion falls to a low but consistent level to maintain basal metabolism, without the risk of hypoglycaemia, even during prolonged fasting that lasts for hours or days. In contrast, the limitations of delivering insulin by subcutaneous injection to patients who can no longer produce endogenous insulin, leads not only to inadequate plasma insulin concentrations during eating and the immediate postprandial period, but also to inappropriately raised plasma insulin levels in the post-absorptive phase (Rizza et al., 1980). This results in high postprandial blood glucose concentrations in the hour or so after eating and a tendency to cause hypoglycaemia in the period before the next meal. Individuals are particularly likely to be affected during the night, when the interval between ingestion of food may be several hours (Box 4.1). As discussed below, developments in insulin delivery by using insulin analogues or continuous subcutaneous insulin infusion offer some benefit over conventional insulin, although these approaches mitigate rather than cure the problem of nocturnal hypoglycaemia.
Impaired Counterregulatory Responses to Hypoglycaemia during Nocturnal Hypoglycaemia Various mechanisms contribute to the additional risk of developing hypoglycaemia during the night. Although an earlier study suggested that hormonal responses to hypoglycaemia might be increased during nocturnal episodes (Bendtson et al., 1993), recent work has indicated that counterregulatory defences are generally impaired. Matyka et al. (1999a) studied 29 prepubertal children overnight in their own homes on two separate occasions. They confirmed that not only was nocturnal hypoglycaemia common, but also that prolonged episodes of hypoglycaemia were not accompanied by increases in epinephrine (adrenaline) and other
Box 4.1
Factors contributing to the development of nocturnal hypoglycaemia
1. Long period between meals (especially in children). 2. Inconsistency of conventional subcutaneous basal insulin delivery – nocturnal hyperinsulinaemia. 3. Unawareness of early symptoms of hypoglycaemia when asleep. 4. Diminished counterregulatory hormone release and symptomatic response in supine posture and effect of sleep per se.
CAUSES OF NOCTURNAL HYPOGLYCAEMIA
87
protective endocrine responses. Since protective counterregulatory responses mitigate the severity of hypoglycaemia by increasing hepatic glucose output and reducing peripheral glucose uptake, as well as enhancing some of the warning autonomic symptoms, defective responses during the night may have a major effect in increasing the risk of nocturnal hypoglycaemia. One important question that emerges from this work is why nocturnal hypoglycaemia should provoke different physiological responses to those that occur during the day. It now appears that impaired responses are the result of additional factors at night, in particular a supine posture and also sleep per se.
Supine Posture Two studies have investigated the effect of posture on counterregulatory hormonal defences to hypoglycaemia. Hirsch et al. (1991) compared the physiological responses to hypoglycaemia induced using a hyperinsulinaemic clamp in young people with type 1 diabetes, who were either in an upright or a horizontal position. Increments in hypoglycaemic symptom scores were more than 50% lower when patients remained supine compared to the usual increase that was observed when they were standing erect. These findings have been confirmed by other researchers, who also demonstrated that plasma epinephrine levels were three times lower during hypoglycaemia in the supine position compared to concentrations when subjects were standing (Robinson et al., 1994). The precise physiological explanation is not entirely clear but may be related to the recruitment of adrenoreceptors which are primed by an upright posture. Whatever the cause, these observations suggest that patients are more vulnerable to progressing to a severe episode of hypoglycaemia when lying horizontal in bed, because of a reduction in symptom intensity and in magnitude of hormonal counterregulation.
Sleep Most studies have reported suppression of physiological protection by counterregulatory mechanisms during hypoglycaemia. Jones et al. (1998) lowered blood glucose to 2.8 mmol/l both during the daytime and at night, and demonstrated diminished epinephrine responses in diabetic and non-diabetic adolescents while they were asleep when compared to the brisk responses while they were awake, whether during daytime hours or during the night (Figure 4.2). Banerer and Cryer (2003) confirmed these observations in patients with type 1 diabetes but, interestingly, in non-diabetic subjects no difference in epinephrine response was observed between these states. Diminished counterregulatory responses, including epinephrine, have also been demonstrated during spontaneous nocturnal hypoglycaemic episodes in children with diabetes (Matyka et al., 1999a). Therefore, sleep appears to be associated with diminished catecholamine and symptomatic responses to hypoglycaemia with a reduction in wakening during a hypoglycaemic episode. The diminished response may occur because the glycaemic thresholds for activation of these responses have been reset to a lower glucose level (Gais et al., 2003), so that more profound hypoglycaemia is necessary to provoke a similar response to that observed in subjects when they are awake. As the investigators involved in these studies have pointed out, this increases the risk of a
NOCTURNAL HYPOGLYCAEMIA
88 Patients with Diabetes
Normal Subjects
Plasma Epinephrine (pg/ml)
350
450 Daytime, awake Nighttime, asleep Nighttime, awake
300
Daytime, awake Nighttime, asleep
400 350
250
300
200
250
150
200 150
100
100 50
50
0
0 –60
–40
–20
0
20
40
60
–60
–40
–20
0
20
40
60
Time (minutes)
Figure 4.2 Mean (± SE) plasma epinephrine concentrations in eight patients with type 1 diabetes and six normal subjects during periods of hypoglycaemia when they were awake during the day, awake at night and asleep at night. (To convert plasma epinephrine values to picomoles per litre, multiply by 5.458. The zero on the x-axis indicates the beginning of the hypoglycaemic period). Reproduced with permission from Jones et al. (1998). Copyright © 1998 Massachusetts Medical Society
severe hypoglycaemic episode occurring at night. However, the mechanisms that disturb the physiological responses to hypoglycaemia during sleep remain unknown. Sleep is not a unitary process (Oswald, 1987). Sleep is dominated by Slow Wave Sleep (SWS) during the first third of the night, and by the cyclical appearance of Rapid Eye Movement (REM) sleep during the latter two thirds. Autonomic activity during SWS is relatively steady but in REM sleep (desynchronisation of the EEG, with absence of activity in the anti-gravity and periodic eye movements) modulations in respiratory and cardiovascular events occur with other changes in the autonomic nervous system. These differences in autonomic activity between SWS and REM sleep suggest that the effect of hypoglycaemia on autonomic responses may vary depending on which stage of sleep is being experienced. However, to date no studies have been published that have explored the effect of different phases of sleep on counterregulatory responses to hypoglycaemia.
CONSEQUENCES OF NOCTURNAL HYPOGLYCAEMIA Impaired Awareness of Hypoglycaemia The demonstration in the early 1990s that repeated exposure to hypoglycaemia leads to impaired physiological defences to subsequent episodes and a reduction in the intensity of symptoms (Heller and Cryer, 1991; Dagogo-Jack et al., 1993) identified a mechanism that explains why some individuals lose their hypoglycaemia warning symptoms. Further studies demonstrated that such episodes did not need to be symptomatic to produce alterations in physiological responses. Veneman et al. (1993) induced hypoglycaemia in 10 non-diabetic subjects overnight and tested their physiological responses to hypoglycaemia the following morning. They reported lower symptomatic and hormonal responses when compared to a non-hypoglycaemia control night. These relatively mild levels of hypoglycaemia are rarely
CONSEQUENCES OF NOCTURNAL HYPOGLYCAEMIA
89
reported by patients because they remain asleep, but may contribute to the development of impaired awareness of hypoglycaemia.
The ‘Dead in Bed’ Syndrome The possible contribution of hypoglycaemia to cardiac arrhythmias and sudden death is discussed in detail in Chapter 12. However, since it has been proposed that nocturnal hypoglycaemia may be a specific precipitant, it merits a brief mention here. A detailed survey of unexpected deaths in young people with type 1 diabetes in the early 1990s highlighted a rare but distinct mode of death in which young people were found lying in an undisturbed bed. Autopsy failed to reveal a structural cause but circumstantial evidence implicated nocturnal hypoglycaemia as a precipitant. Not all affected individuals had strict glycaemic control, but many were known to have been susceptible to developing nocturnal hypoglycaemia. This type of sudden and unexpected death has since been confirmed in epidemiological surveys in other countries, possibly accounting for between 5–10% of all deaths under the age of 40 in people with diabetes. Hypoglycaemia causes an increase in the QT interval and one possible explanation is that an episode of nocturnal hypoglycaemia triggers a ventricular arrhythmia in a susceptible individual. However, further work is required to establish this plausible hypothesis as the cause of these deaths.
Neurological Consequences of Nocturnal Hypoglycaemia on Cerebral Function The possibility that recurrent exposure to hypoglycaemia, particularly occurring during sleep, might insidiously damage cerebral function and cause permanent cognitive impairment was raised by the finding that children who had developed type 1 diabetes before the age of five years, exhibited cognitive impairment when compared with non-diabetic controls (Ryan et al., 1985). This observation was replicated in different studies using a range of cognitive tests (Bjorgaas et al., 1997; Rovet and Ehrlich, 1999). However, the relationship to hypoglycaemia, particularly during the night, was unclear (Golden et al., 1989); hypoglycaemia-induced convulsions have also been implicated, and it is possible that other factors associated with early-onset diabetes may contribute to cognitive impairment. Several studies have explored the extent to which nocturnal episodes may affect performance on the following day. Bendtson et al. (1992) found no difference in cognitive performance among adults with type 1 diabetes when tested on the morning after an episode of nocturnal hypoglycaemia in comparison with a night with no hypoglycaemia. Similar findings have been reported after the induction of experimental hypoglycaemia during the night (King et al., 1998) although the subjects were more fatigued on the following morning. It seems that even children are relatively unaffected by a single episode of nocturnal hypoglycaemia. Matyka et al. (1999b) tested pre-pubertal children after episodes of prolonged, spontaneously occurring, hypoglycaemia that occurred during sleep, many of which lasted several hours, and found no deleterious effect on cognitive function on the following morning, although mood was adversely affected. In summary, although it seems plausible that recurrent nocturnal hypoglycaemia might contribute to cognitive decline, on the available evidence the verdict remains unproven.
90
NOCTURNAL HYPOGLYCAEMIA
CAN NOCTURNAL HYPOGLYCAEMIA BE PREDICTED? The ability to predict whether a nocturnal hypoglycaemic episode is likely, by measuring glucose concentrations at bedtime, has generally been tested in studies using intermittent blood glucose sampling to detect hypoglycaemia. It is likely that such a design will fail to identify all episodes. The results of two early studies that observed patients who were receiving one or two daily injections of insulin, suggested that a bedtime glucose concentration of 6.0 mmol/l or less indicated an 80% chance of a period of biochemical hypoglycaemia during the night (Pramming et al., 1985; Whincup and Milner, 1987). Furthermore, these early studies were generally undertaken among patients who were taking two injections of conventional insulins each day. Studies of children have suggested that blood glucose measurements at bedtime are less predictive of hypoglycaemia in the first half of the night, although a value of < 75 mmol/l does indicate an increased risk (Matyka et al., 1999b). That study and others have also shown that a low fasting blood glucose is a strong indicator that hypoglycaemia has occurred in the latter half of the night (Matyka, 2002). Since severe nocturnal hypoglycaemia is relatively rare, it is much more difficult to establish to what extent bedtime blood glucose values are predictive. A previous history of severe hypoglycaemia, impaired awareness of hypoglycaemia or strict control with low HbA1c values, will confer a greater chance of a severe episode but are a poor guide to the risk of developing a severe nocturnal episode on any particular night.
THE SOMOGYI PHENOMENON: THE CONCEPT OF REBOUND HYPERGLYCAEMIA In the late 1930s, a Hungarian biochemist, Michael Somogyi, working in St Louis, USA, suggested that nocturnal hypoglycaemia might provoke rebound hyperglycaemia on the following morning, and he supported his hypothesis with a demonstration that reducing evening doses of insulin led to a reduction in fasting urinary glycosuria (Somogyi, 1959). He proposed that nocturnal hypoglycaemia provokes a counterregulatory response with rises in plasma epinephrine, cortisol and growth hormone resulting in the release of glucose from the liver and inhibition of the effects of insulin over the next few hours. The logical conclusion from his hypothesis was that this ‘rebound’ elevated fasting blood glucose in the morning should be treated, not by an increase in the evening dose of insulin, but paradoxically by a reduction. The idea of ‘rebound hyperglycaemia’ following nocturnal hypoglycaemia, (also known as the Somogyi phenomenon) as an explanation for a high fasting blood glucose in insulin-treated patients, has proved to be very attractive to many diabetes healthcare professionals who firmly believe in its existence. The consequences are important as patients are often advised to reduce their evening insulin dose, particularly if they complain of nocturnal hypoglycaemia. However, its clinical relevance was challenged over 20 years ago and repeated studies have established that fasting hyperglycaemia is largely a result of falling plasma insulin concentrations during the night, as the subcutaneous depot of insulin that was injected the day before is dissipated. Gale et al. (1980) demonstrated that periods of hypoglycaemia during the night were often prolonged and were not accompanied by a large rise in counterregulatory hormones.
Probability of hypoglycaemic episodes in the night
THE SOMOGYI PHENOMENON
91
1.0
0.8
0.6
0.4
0.2
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21
Morning blood glucose (mmol/I)
Figure 4.3 Risk of nocturnal hypoglycaemia according to fasting morning blood glucose (95% Cl) in 594 nights. Black bars = hypoglycaemic nights; shaded bars = possibly hypoglycaemic nights. Reproduced from Hoi-Hansen et al. (2005). With kind permission from Springer Science and Business Media
Although fasting glucose concentrations were frequently high in the patients they studied, this was related directly to a waning of circulating plasma insulin concentrations. Some investigators have demonstrated that when hypoglycaemia is experimentally-induced during the night, this can raise blood glucose on the following morning, even if circulating plasma insulin concentrations are maintained (Perriello et al., 1988). However, the additional increase in the fasting blood glucose concentration is modest (around 2.0 mmol/l) and its clinical relevance is questionable. Other researchers have found no effect on daytime concentrations of blood glucose after lowering blood glucose to hypoglycaemic levels during the night (Hirsch et al., 1990). Careful analysis of data collected both by self-monitoring of blood glucose (Havlin and Cryer, 1987) and by continuous glucose monitoring (Hoi-Hansen et al., 2005) (Figure 4.3) during everyday activities, has also shown that nocturnal hypoglycaemia does not provoke rebound fasting hyperglycaemia. Many insulin-treated diabetic patients experience high fasting blood glucose levels but this common clinical problem is essentially a consequence of inadequate basal insulin replacement overnight. The important mechanisms that contribute to fasting hyperglycaemia appear to be a combination of waning plasma insulin levels and glucose release from the liver secondary to nocturnal spikes of growth hormone secretion (Campbell et al., 1985), a physiological process termed the ‘dawn phenomenon’. High blood glucose levels following symptomatic nocturnal hypoglycaemia may also result from excessive intake of oral carbohydrate, ingested as treatment of the hypoglycaemia, rather than a powerful counterregulatory response, which is usually suppressed at night. The clinical message is therefore clear: fasting hyperglycaemia indicates a need for adjusting basal insulin in terms of type or timing rather than reducing the dose. Some useful clinical steps to be undertaken in patients presenting with this problem are listed in Box 4.2.
NOCTURNAL HYPOGLYCAEMIA
92
Box 4.2 Clinical approach to high fasting blood glucose complicated by nocturnal hypoglycaemia 1. Measure blood glucose at 2–3 a.m. over a few days. 2. If nocturnal hypoglycaemia is present, ensure that basal insulin is taken at bedtime (i.e., split pre-mixed evening insulin). 3. Progressively increase bedtime long-acting insulin in doses of 2–4 units while checking with 3 a.m. blood glucose measurements that this is not precipitating nocturnal hypoglycaemia 4. Use a long-acting insulin analogue, either glargine or detemir. 5. Teach patients to take an appropriate (but not excessive) quantity of a high energy glucose drink, orange juice or glucose as sweets or tablets to treat nocturnal hypoglycaemic episodes. 6. When available, consider obtaining a continuous glucose monitoring profile.
CLINICAL SOLUTIONS (BOX 4.3) Dietary Measures A time honoured approach to reducing the frequency of nocturnal hypoglycaemia has been to counteract the effect of insulin by ensuring that patients eat a bedtime snack. This seemed particularly important for children who go to bed early and sleep for many hours between their evening insulin dose and breakfast on the following day. It clearly makes sense for all individuals taking insulin to measure their blood glucose before bed and to take additional food if their blood glucose is low. However, the extent to which protective eating can prevent nocturnal hypoglycaemia in patients taking intermittent injections of insulin is limited.
Box 4.3
Potential remedies for problematical nocturnal hypoglycaemia
1. Long and rapid-acting insulin analogues. 2. -agonists (terbutaline or salbutamol). 3. Appropriate snacks: – uncooked cornstarch – high protein foods. 4. CSII.
CLINICAL SOLUTIONS
93
The effectiveness of different approaches has generally been assessed by measuring the effectiveness of different snacks in preventing biochemical or symptomatic episodes of hypoglycaemia. Before the advent of continuous blood glucose monitoring, such studies demanded regular blood sampling for measurement of glucose and since this generally required admission to hospital for study, these investigations were not exactly examining a typical clinical situation. Furthermore, the number of subjects studied has generally been relatively small, and although sufficient to measure changes in blood glucose concentration or frequency of biochemical hypoglycaemia, the studies have been inadequately powered to determine effects on the frequency of severe episodes. Some studies have examined the potential of carbohydrate foods that are absorbed slowly and thus able to counter the hypoglycaemic effect of insulin over longer periods. Uncooked cornstarch, which has a very low glycaemic index, has been studied intensively, particularly because it is used successfully to prevent severe hypoglycaemia in glycogen storage diseases (Goldberg and Slonim, 1993). In one study, when young people were given uncooked cornstarch incorporated into a normal bedtime snack, the incidence of symptomatic and biochemical nocturnal hypoglycaemia was three-fold lower (Kaufman and Devgan, 1996). A blinded randomised trial reported a similarly lower frequency of nocturnal episodes (Kaufman et al., 1997). Other work has examined the effect of snacks rich in protein or fat. Different types of snack were compared against placebo in a trial using a crossover design in 15 adults with type 1 diabetes (Kalergis et al., 2003). When patients retired to bed with blood glucose greater than 10.0 mmol/l, nocturnal episodes were not observed. At a pre-bedtime glucose below 7.0 mmol/l, a high protein snack prevented any episode of nocturnal hypoglycaemia by contrast with either a standard snack or one containing cornstarch (Figure 4.4). Thus the clinical data measuring the effectiveness of cornstarch are
100%
% Nights
80%
60%
40%
20%
0%
Pl
S CS Prot <7 mmol/L
Pl
S CS Prot 7–10 mmol/L Bedtime blood glucose category
Pl
S CS Prot >10 mmol/L
Figure 4.4 Effect of snacks and bedtime blood glucose concentration on frequency of nocturnal hypoglycaemia. Open box = neither nocturnal hypoglycaemia nor morning hypoglycaemia; solid black box = hypoglycaemia; striped box = morning hypoglycaemia only. Snack type: Pl = placebo; S = standard diet; CS = cornstarch; prot = protein
94
NOCTURNAL HYPOGLYCAEMIA
conflicting. Since uncooked cornstarch is not easy to prepare in a digestible form, it is not surprising that it is not used widely to prevent nocturnal hypoglycaemia. An alternative approach to dietary supplements is to reduce the rate of absorption of carbohydrates using a disaccharidase inhibitor such as acarbose. Three studies of these agents have examined their effects in patients with type 1 diabetes. McCulloch et al. (1983) studied the effect of acarbose on the risk of overnight hypoglycaemia, and found that the risk of symptomatic nocturnal hypoglycaemia was lowered by 39%. Taira et al. (2000) reported similar benefits using voglibose. However, a recent study reported no benefit of acarbose over placebo when both pharmaceutical and snacking interventions were compared with respect to their effect on preventing hypoglycaemia (Raju et al., 2006). In the light of these limited and conflicting data it seems unlikely that acarbose will never be widely used, particularly as sucrose-containing products cannot be used as a treatment for hypoglycaemia when acarbose is being taken.
Pharmaceutical Interventions There are indications that -agonists may have some use in reducing the risk of nocturnal hypoglycaemia. For some years inhaled terbutaline has been proposed as a method of elevating blood glucose. In the early 1990s, Wiethop and Cryer (1993) demonstrated that its oral or subcutaneous delivery following induced hypoglycaemia, led to a rise in blood glucose compared to placebo, an effect that lasted for some hours. More recently, Wright and Wales (2003) reported that children with type 1 diabetes who were receiving treatment for asthma had fewer episodes of nocturnal hypoglycaemia when compared to a non-asthmatic group of children in a survey lasting three months, and they implicated a beneficial effect of -agonist therapy. Raju et al. (2006) also compared the effect of inhaled terbutaline with other therapeutic remedies such as cornstarch, standard snacks and acarbose on nocturnal blood glucose levels in patients with type 1 diabetes. They found that terbutaline prevented nocturnal hypoglycaemia in all 15 subjects. However, although this treatment offered the greatest protection against hypoglycaemia at night, it also led to the highest fasting blood glucose among the different remedies, indicating that additional work needs to be done to establish this treatment as a realistic therapeutic option.
Timing and Type of Insulin, Including Insulin Analogues The introduction of insulin analogues with pharmacokinetic properties that bear more resemblance to physiological insulin profiles in non-diabetic individuals highlighted the potential of such preparations to lower the risk of hypoglycaemia. Over a full 24 hours, the overall frequency of symptomatic hypoglycaemia in trials of rapid-acting insulin analogues has been modestly lower than with conventional insulins, but a consistent finding has been a lower rate of nocturnal hypoglycaemia. It appears that the tendency of conventional soluble insulin to self-associate into hexamers at therapeutic concentrations leads to increasing plasma insulin levels with repeated injection. Since rapid-acting insulin analogues separate into single molecules much more readily, accumulation of insulin is less likely and the risk of nocturnal hypoglycaemia is subsequently lower. The frequency of nocturnal hypoglycaemia observed in clinical trials has been variable. Rates of nocturnal hypoglycaemia have been over 50% lower in some trials involving patients with type 1 diabetes with strict
CONCLUSIONS
95
glycaemic control (Heller et al., 1999; Heller et al., 2004), and although this has generally been demonstrated for symptomatic episodes, these data might reflect a lower frequency of severe hypoglycaemia (Holleman et al., 1997). The long-acting analogues – insulins glargine and detemir – which provide basal insulin replacement, also appear to lower the risk of nocturnal hypoglycaemia. They have a longer duration of action when compared to isophane (NPH) insulin, together with less of a peak in their time-action profile and a more consistent duration of action (Barnett, 2003). Both of these properties probably contribute to the lower rates of nocturnal hypoglycaemia that have been observed during clinical trials. Relative risk reductions of around 30% have been reported for long-acting preparations in trials involving patients with type 1 (De Leeuw et al., 2005; Pieber et al., 2000) and with type 2 diabetes (Hermansen et al., 2006; Yki-Jarvinen et al., 2000). The combination of both rapid and long-acting insulin analogues might be expected to have a particularly powerful effect in lowering the risk of nocturnal hypoglycaemia. In the few studies comparing combinations of insulin analogues to conventional insulins this appears to be the case. Hermansen et al. (2004) reported a 55% lower rate of symptomatic nocturnal hypoglycaemia when using an insulin detemir/aspart combination as basal-bolus therapy in patients with type 1 diabetes, which was accompanied by a modest but significant fall in HbA1c of 0.2%. Ashwell et al. (2006) observed a fall in HbA1c of 0.5% using insulin glargine and lispro in a basal-bolus regimen in patients with type 1 diabetes, while nocturnal hypoglycaemia was 44% lower in frequency. However, nocturnal hypoglycaemia was not eradicated, leading to the conclusion that although insulin analogues may help to reduce the side-effects of insulin therapy, they do not approach the requirements of therapeutic insulin delivery.
Continuous Subcutaneous Insulin Infusion (CSII) Since nocturnal hypoglycaemia is largely the result of inadequate basal insulin replacement, one would expect that the most effective method of basal insulin delivery currently available, CSII, could limit the frequency of nocturnal hypoglycaemia (Pickup and Keen, 2002). However, early studies reported surprisingly little effect on hypoglycaemia, perhaps because of a failure to train patients in the essential related skills of carbohydrate and insulin dose adjustment. More recent work has indicated that CSII can reduce overall rates of hypoglycaemia (Bode et al., 1996; Boland et al., 1999; Kanc et al., 1998), but specific reporting on rates of nocturnal episodes is unusual. Furthermore, few trials have used a randomised design, suggesting that a lower risk might relate to other characteristics of those who use CSII rather than to the technology itself. Thus the amount to which modern pump therapy lowers the risk of nocturnal hypoglycaemia has still be to be established. Nevertheless, in those who have experienced recurrent nocturnal episodes and who have not improved after a trial of insulin analogues, it seems worthwhile undertaking a trial of CSII.
CONCLUSIONS • Nocturnal hypoglycaemia remains an unresolved clinical side-effect of insulin therapy preventing the attainment of strict glycaemic control for many people. It contributes to morbidity, and perhaps mortality, in patients with type 1 diabetes.
96
NOCTURNAL HYPOGLYCAEMIA
• Nocturnal hypoglycaemia is caused chiefly by the limitations of current methods of insulin delivery. The inability of subcutaneous insulin therapy, particularly conventional basal insulins, to maintain low-level stable insulin concentrations overnight, leads both to nocturnal hypoglycaemia and fasting hyperglycaemia. • In addition to the limitations of insulin delivery, hypoglycaemia is also more common as a result of diminished counterregulatory responses overnight, associated in part with sleep, which has a specific inhibitory effect on physiological defences to hypoglycaemia, and a supine posture which suppresses autonomic responses. • Nocturnal hypoglycaemia is a particular problem in children, partly as a consequence of the long period of fasting between their evening meal and their breakfast. • The risk of nocturnal hypoglycaemia is greatest in those who have a bedtime glucose below 7.0 mmol/l. Bedtime snacks can reduce the risk during the early part of the night. Specific foods (uncooked cornstarch or protein rich snacks) reduce the risk of nocturnal hypoglycaemia in some studies. • Rebound hyperglycaemia (the ‘Somogyi phenomenon’) is rarely caused by counterregulatory hormone release provoked by an overnight hypoglycaemic episode, since hormonal secretion is generally suppressed. It is mainly a consequence of waning of circulating plasma insulin levels and should be treated by an adjustment in the timing and type of insulin used rather than a reduction in insulin dose. • The problem of nocturnal hypoglycaemia may respond to the use of insulin analogues (both rapid and long-acting) or to CSII with an insulin pump. However, its eradication will depend upon new methods of insulin delivery.
REFERENCES Ashwell SG, Amiel SA, Bilous RW, Dashora U, Heller SR, Hepburn DA et al. (2006). Improved glycaemic control with insulin glargine plus insulin lispro: a multicentre, randomized, cross-over trial in people with type 1 diabetes. Diabetic Medicine 23: 285–92. Banarer S, Cryer PE (2003). Sleep-related hypoglycemia-associated autonomic failure in type 1 diabetes: reduced awakening from sleep during hypoglycemia. Diabetes 52: 1195–203. Barnett AH (2003). A review of basal insulins. Diabetic Medicine 20: 873–85. Bendtson I, Kverneland A, Pramming S, Binder C (1988). Incidence of nocturnal hypoglycaemia in insulin-dependent diabetes. Acta Endocrinologica 223: 543–8. Bendtson I, Gade J, Theilgard A, Binder C (1992). Cognitive function in type 1 IDD patients after nocturnal hypoglycaemia. Diabetologia 35: 898–903. Bendtson I, Rosenfalck AM, Binder C (1993). Nocturnal versus diurnal hormonal counterregulation to hypoglycemia in type 1 (insulin-dependent) diabetic patients. Acta Endocrinologica 128: 109–15. Beregszaszi M, Tubiana-Rufi N, Benali K, Noel M, Bloch J, Czernichow P (1997). Nocturnal hypoglycemia in children and adolescents with insulin-dependent diabetes mellitus: prevalence and risk factors. Journal of Pediatrics 131: 27–33. Bjorgaas M, Gimse R, Vik T, Sand T (1997). Cognitive function in type 1 diabetic children with and without episodes of severe hypoglycaemia. Acta Paediatrica 86: 148–53. Bode BW, Steed RD, Davidson PC (1996). Reduction in severe hypoglycemia with long-term continuous subcutaneous insulin infusion in type I diabetes. Diabetes Care 19: 324–7.
REFERENCES
97
Boland EA, Grey M, Oesterle A, Fredrickson L, Tamborlane WV (1999). Continuous subcutaneous insulin infusion. A new way to lower risk of severe hypoglycemia, improve metabolic control, and enhance coping in adolescents with type 1 diabetes. Diabetes Care 22: 1779–84. Boland E, Monsod T, Delucia M, Brandt CA, Fernando S, Tamborlane WV (2001). Limitations of conventional methods of self-monitoring of blood glucose: lessons learned from 3 days of continuous glucose sensing in pediatric patients with type 1 diabetes. Diabetes Care 24: 1858–62. Campbell PJ, Bolli GB, Cryer PE, Gerich JE (1985). Pathogenesis of the dawn phenomenon in patients with insulin-dependent diabetes mellitus. Accelerated glucose production and impaired glucose utilization due to nocturnal surges in growth hormone secretion. New England Journal of Medicine 312: 1473–9. Dagogo-Jack SE, Craft S, Cryer PE (1993). Hypoglycemia-associated autonomic failure in insulindependent diabetes mellitus. Journal of Clinical Investigation 91: 819–28. De Leeuw I, Vague P, Selam JL, Skeie S, Lang H, Draeger E, Elte JW (2005). Insulin detemir used in basal-bolus therapy in people with type 1 diabetes is associated with a lower risk of nocturnal hypoglycaemia and less weight gain over 12 months in comparison to NPH insulin. Diabetes Obesity and Metabolism 7: 73–82. Dornan TL, Orde Peckar C, Mayon-White VA, Knight AH, Moore RA, Hockaday TDR et al. (1981). Unsuspected hypoglycaemia, haemoglobin A1 and diabetic control. Quarterly Journal of Medicine 197: 31–8. Gais S, Born J, Peters A, Schultes B, Heindl B, Fehm HL, Werner K (2003). Hypoglycemia counterregulation during sleep. Sleep 26: 55–9. Gale EA, Tattersall RB (1979). Unrecognised nocturnal hypoglycaemia in insulin-treated diabetics. Lancet 1: 1049–52. Gale EA, Kurtz AB, Tattersall RB (1980). In search of the Somogyi effect. Lancet 2: 279–82. Goldberg T, Slonim AE (1993). Nutrition therapy for hepatic glycogen storage diseases. Journal of the American Dietetic Association 93: 1423–30. Golden MP, Ingersoll GM, Brack CJ, Russell BA, Wright JC, Huberty TJ (1989). Longitudinal relationship of asymptomatic hypoglycemia to cognitive function in IDDM. Diabetes Care 12: 89–93. Havlin CE, Cryer PE (1987). Nocturnal hypoglycemia does not commonly result in major morning hyperglycemia in patients with diabetes mellitus. Diabetes Care 10: 141–7. Heller SR, Cryer PE (1991). Reduced neuroendocrine and symptomatic responses to subsequent hypoglycemia after 1 episode of hypoglycemia in non-diabetic humans. Diabetes 40: 223–6. Heller SR, Amiel SA, Mansell P (1999). Effect of the fast-acting insulin analog lispro on the risk of nocturnal hypoglycemia during intensified insulin therapy. Diabetes Care 22: 1607–11. Heller SR, Colagiuri S, Vaaler S, Wolffenbuttel BH, Koelendorf K, Friberg HH et al. (2004). Hypoglycaemia with insulin aspart: a double-blind, randomised, crossover trial in subjects with type 1 diabetes. Diabetic Medicine 21: 769–75. Hermansen K, Fontaine P, Kukolja KK, Peterkova V, Leth G, Gall MA (2004). Insulin analogues (insulin detemir and insulin aspart) versus traditional human insulins (NPH insulin and regular human insulin) in basal-bolus therapy for patients with type 1 diabetes. Diabetologia 47: 622–9. Hermansen K, Davies M, Derezinski T, Martinez Ravn G, Clauson P, Home P (2006). A 26-week, randomized, parallel, treat-to-target trial comparing insulin detemir with NPH insulin as add-on therapy to oral glucose-lowering drugs in insulin-naive people with type 2 diabetes. Diabetes Care 29: 1269–74. Hirsch IB, Smith LJ, Havlin CE, Shah SD, Clutter WE, Cryer PE (1990). Failure of nocturnal hypoglycemia to cause daytime hyperglycemia in patients with IDDM. Diabetes Care 13: 133–42. Hirsch IB, Heller SR, Cryer PE (1991). Increased symptoms of hypoglycaemia in the standing position in insulin-dependent diabetes mellitus. Clinical Science 80: 583–6. Hoi-Hansen T, Pedersen-Bjergaard U, Thorsteinsson B (2005). The Somogyi phenomenon revisited using continuous glucose monitoring in daily life. Diabetologia 48: 2437–8.
98
NOCTURNAL HYPOGLYCAEMIA
Holleman F, Schmitt H, Rottiers R, Rees A, Symanowski S, Anderson JH, The Benelux-UK Insulin Lispro Study Group (1997). Reduced frequency of severe hypoglycemia and coma in well-controlled IDDM patients treated with insulin lispro. Diabetes Care 20: 1827–32. Jones TW, Porter P, Sherwin RS, Davis EA, O’Leary P, Frazer F et al. (1998). Decreased epinephrine responses to hypoglycemia during sleep. New England Journal of Medicine 338: 1657–62. Kalergis M, Schiffrin A, Gougeon R, Jones PJ, Yale JF (2003). Impact of bedtime snack composition on prevention of nocturnal hypoglycemia in adults with type 1 diabetes undergoing intensive insulin management using lispro insulin before meals: a randomized, placebo-controlled, crossover trial. Diabetes Care 26: 9–15. Kanc K, Janssen MM, Keulen ET, Jacobs MA, Popp-Snijders C, Snoek FJ, Heine RJ (1998). Substitution of night-time continuous subcutaneous insulin infusion therapy for bedtime NPH insulin in a multiple injection regimen improves counterregulatory hormonal responses and warning symptoms of hypoglycaemia in IDDM. Diabetologia 41: 322–9. Kaufman FR, Devgan S (1996). Use of uncooked cornstarch to avert nocturnal hypoglycemia in children and adolescents with type I diabetes. Journal of Diabetes and Its Complications 10: 84–7. Kaufman FR, Halvorson M, Kaufman ND (1997). Evaluation of a snack bar containing uncooked cornstarch in subjects with diabetes. Diabetes Research and Clinical Practice 35: 27–33. Kaufman FR, Austin J, Neinstein A, Jeng L, Halvorson M, Devoe DJ, Pitukcheewanont P (2002). Nocturnal hypoglycemia detected with the Continuous Glucose Monitoring System in pediatric patients with type 1 diabetes. Journal of Pediatrics 141: 625–30. King P, Kong MF, Parkin H, Macdonald IA, Tattersall RB (1998). Well-being, cerebral function, and physical fatigue after nocturnal hypoglycemia in IDDM. Diabetes Care 21: 341–5. Kulcu E, Tamada JA, Reach G, Potts RO, Lesho MJ (2003). Physiological differences between interstitial glucose and blood glucose measured in human subjects. Diabetes Care 26: 2405–9. Matyka KA, Crowne EC, Havel PJ, Macdonald IA, Matthews D, Dunger DB (1999a). Counterregulation during spontaneous nocturnal hypoglycemia in prepubertal children with type 1 diabetes. Diabetes Care 22: 1144–50. Matyka KA, Wigg L, Pramming S, Stores G, Dunger DB (1999b). Cognitive function and mood after profound nocturnal hypoglycaemia in prepubertal children with conventional insulin treatment for diabetes. Archives of Disease in Childhood, 81: 138–42. Matyka KA (2002). Sweet dreams? Nocturnal hypoglycemia in children with type 1 diabetes. Pediatric Diabetes 3: 74–81. McCulloch DK, Kurtz AB, Tattersall RB (1983). A new approach to the treatment of nocturnal hypoglycemia using alpha-glucosidase inhibition. Diabetes Care 6: 483–7. McGowan K, Thomas W, Moran A (2002). Spurious reporting of nocturnal hypoglycemia by CGMS in patients with tightly controlled type 1 diabetes. Diabetes Care 25: 1499–503. Oswald I (1987). The normal record of sleep. In: A Textbook of Clinical Neurophysiology. Halliday AM, Butler SR and Paul R, eds. John Wiley & Sons Inc., New York: 173–85. Perriello G, De Feo P, Torlone E, Calcinaro F, Ventura MM, Basta G et al. (1988). The effect of asymptomatic nocturnal hypoglycemia on glycemic control in diabetes mellitus. New England Journal of Medicine 319: 1233–9. Pickup J, Keen H (2002). Continuous subcutaneous insulin infusion at 25 years: evidence base for the expanding use of insulin pump therapy in type 1 diabetes. Diabetes Care 25: 593–8. Pieber TR, Eugene-Jolchine I, Derobert E (2000). Efficacy and safety of HOE 901 versus NPH insulin in patients with type 1 diabetes. The European Study Group of HOE 901 in type 1 diabetes. Diabetes Care 23: 157–62. Porter PA, Keating B, Byrne G, Jones TW (1997). Incidence and predictive criteria of nocturnal hypoglycemia in young children with insulin-dependent diabetes mellitus. Journal of Pediatrics 130: 366–72. Pramming S, Thorsteinsson B, Bendtson I, Ronn B, Binder C (1985). Nocturnal hypoglycaemia in patients receiving conventional treatment with insulin. British Medical Journal 291: 376–9.
REFERENCES
99
Raju B, Arbelaez AM, Breckenridge SM, Cryer PE (2006). Nocturnal hypoglycemia in type 1 diabetes: an assessment of preventive bedtime treatments. Journal of Clinical Endocrinology and Metabolism 91: 2087–92. Rizza RA, Gerich JE, Haymond MW, Westland RE, Hall LD, Clemens AH, Service FJ (1980). Control of blood sugar in insulin-dependent diabetes: comparison of an artificial endocrine pancreas, continuous subcutaneous insulin infusion and intensified conventional insulin therapy. New England Journal of Medicine 303: 1313–18. Robinson AM, Parkin HM, Macdonald IA, Tattersall RB (1994). Physiological response to postural change during mild hypoglycaemia in patients with IDDM. Diabetologia 37: 1241–50. Rovet JF, Ehrlich RM (1999). The effect of hypoglycemic seizures on cognitive function in children with diabetes: a 7-year prospective study. Journal of Pediatrics 134: 503–6. Ryan C, Vega A, Drash A (1985). Cognitive deficits in adolescents who developed diabetes early in life. Pediatrics 75: 921–7. Shalwitz RA, Farkas-Hirsch R, White NH, Santiago JV (1990). Prevalence and consequences of nocturnal hypoglycemia among conventionally treated children with diabetes mellitus. Journal of Pediatrics 116: 685–9. Somogyi M (1959). Exacerbation of diabetes by excess insulin action. American Journal of Medicine 26: 169–91. Taira M, Takasu N, Komiya I, Taira T, Tanaka H (2000). Voglibose administration before the evening meal improves nocturnal hypoglycemia in insulin-dependent diabetic patients with intensive insulin therapy. Metabolism 49: 440–3. Veneman T, Mitrakou A, Mokan M, Cryer P, Gerich J (1993.) Induction of hypoglycemia unawareness by asymptomatic nocturnal hypoglycemia. Diabetes 42: 1233–7. Vervoort G, Goldschmidt HM, van Doorn LG (1996). Nocturnal blood glucose profiles in patients with type 1 diabetes mellitus on multiple (> or = 4) daily insulin injection regimens. Diabetic Medicine 13: 794–9. Wentholt IM, Vollebregt MA, Hart AA, Hoekstra JB, DeVries JH (2005). Comparison of a needletype and a microdialysis continuous glucose monitor in type 1 diabetic patients. Diabetes Care 28: 2871–6. Whincup G, Milner RD (1987). Prediction and management of nocturnal hypoglycaemia in diabetes. Archives of Disease in Childhood 62: 333–7. Wiethop BV, Cryer PE (1993). Alanine and terbutaline in treatment of hypoglycemia in IDDM. Diabetes Care 16: 1131–6. Wright NP, Wales JK (2003). The incidence of hypoglycaemia in children with type 1 diabetes and treated asthma. Archives of Disease in Childhood 88: 155–6. Yki-Jarvinen H, Dressler A, Ziemen M (2000). Less nocturnal hypoglycemia and better post-dinner glucose control with bedtime insulin glargine compared with bedtime NPH insulin during insulin combination therapy in type 2 diabetes. HOE 901/3002 Study Group. Diabetes Care 23: 1130–6.
5 Moderators, Monitoring and Management of Hypoglycaemia Tristan Richardson and David Kerr
INTRODUCTION Despite advances in insulin pharmacology and delivery and in patient education, the lifetime frequency of symptomatic hypoglycaemia remains substantial, with the average patient likely to experience thousands of episodes over the course of their life with insulin-treated diabetes. Furthermore, there are a number of everyday factors that continue to influence (moderate) the risk, presentation and rate of recovery from low blood glucose concentrations (Table 5.1, and see Chapter 3). An understanding of the predisposing factors that influence hypoglycaemia is important to allow appropriate advice and education to be given to patients, and to appreciate the relative importance of different moderators and reduce their risks (real and perceived), particularly of causing recurrent severe hypoglycaemia.
RISK FACTORS FOR THE DEVELOPMENT OF HYPOGLYCAEMIA In physiological terms, current insulin regimens are less than ideal. Imperfect insulin is often given at the ‘wrong’ time, in the ‘wrong’ place and at the ‘wrong’ dose. Unsurprisingly, a mismatch between insulin and carbohydrate absorption frequently occurs, leading to overinsulinisation and the potential risk of hypoglycaemia. Several other factors can influence insulin absorption after subcutaneous injection including: • depth of the injection; • angle of the needle for giving the injection; • site of the injection; • presence of lipohypertrophy at injection site; • time of day that injection is given; • phase of the menstrual cycle; • relationship with exercise;
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
102
MODERATORS, MONITORING AND MANAGEMENT
Table 5.1 Causes of hypoglycaemia in patients with diabetes Change in insulin sensitivity ‘Honeymoon period’ in newly diagnosed type 1 diabetes Post-partum Menstruation Alcohol Renal failure
Change in insulin pharmaco-dynamics
Altered insulin: carbohydrate ratio
Change in injection site
Unplanned exercise
Change in insulin formulation Change in temperature
Change in social circumstances Breastfeeding Gastroparesis Malabsorption e.g. Coeliac Disease
Other related conditions Endocrine dysfunction (Addison’s disease, hypopituitarism) Psychological
Exercise
• ambient temperature; • psychological factors such as mood; • use of other medicines affecting skin blood flow; • for pre-mixed insulin preparations – whether the insulin has been shaken adequately before the injection. Although there is now greater awareness of the importance of educating patients about the nuances of dietary carbohydrate counting, it is important to remember that the absorption of food is also variable within individuals, being affected by the constituents and size of a meal, and the speed with which it is eaten. In addition, gastric emptying is variable within an individual with diabetes and is influenced by the status of the autonomic nervous system (Feldman and Schiller, 1983; Vinik et al., 2003). In addition, improving glycaemic control and lowering HbA1c per se are also associated with a trebling of risk for hypoglycaemia (The Diabetes Control and Complications Trial Research Group, 1995). However, this alone does not wholly predict the occurrence of hypoglycaemia. In the intensively treated group in the DCCT, HbA1c only accounted for 60% of the risk of severe hypoglycaemia (The Diabetes Control and Complications Trial Research Group, 1997). Additional risk factors have also been identified, many of which are discussed in detail in Chapter 3: • previous episodes of severe hypoglycaemia (Bott et al., 1997; The Diabetes Control and Complications Trial Research Group, 1997); • long duration of type 1 diabetes (Cox et al., 1994; The Diabetes Control and Complications Trial Research Group, 1997); • intensive insulin therapy (The DCCT Research Group, 1991); • strict glycaemic control (The Diabetes Control and Complications Trial Research Group, 1997);
LIFESTYLE MODERATORS
103
• absolute insulin deficiency (Bott et al., 1997; The Diabetes Control and Complications Trial Research Group, 1997); • sleep (Pramming et al., 1990; The Diabetes Control and Complications Trial Research Group, 1997); • impaired hypoglycaemia awareness (Gold et al., 1994); • alcohol (Richardson et al., 2005b); • exercise (MacDonald, 1987); • pregnancy (Rosen et al., 1995); • impaired renal function (Muhlhauser et al., 1991). Despite these risk factors being recognised, it is often difficult to untangle the major precipitating factors for a given episode of severe hypoglycaemia. Patient recall is often vague, preconceptions may have been applied and amnesia for an event is common. Anecdotally, around a third to half of episodes remain unexplained in routine clinical practice, although patient ‘error’ is still perceived (by healthcare professionals) as the most likely ‘cause’ of a hypoglycaemic episode (The DCCT Research Group, 1991).
LIFESTYLE MODERATORS Numerous lifestyle influences predispose towards hypoglycaemia, and some of these, such as pregnancy (Chapter 10) and exercise (Chapter 14), are discussed elsewhere in this book.
Alcohol and Hypoglycaemia Alcohol is an important risk factor for hypoglycaemia for individuals treated with insulin, with estimates suggesting that up to 20% of severe hypoglycaemic events are attributable to its use (Potter et al., 1982; Nilsson et al., 1988). However, there is nothing to suggest that (in general terms) people with type 1 diabetes adopt a different approach to their use of alcohol than the rest of the population. Alcohol has been associated with hypoglycaemia in several ways: • Ingestion of even small amounts may impair the ability of the individual to detect the onset of hypoglycaemia at a stage when they are still able to take appropriate action, i.e., eat some carbohydrate. • Hypoglycaemia per se may be mistaken for intoxication by observers, with legal and health consequences. • Alcohol has been shown in some studies to impact directly on gluconeogenesis and/or the counterregulatory responses to hypoglycaemia (Turner et al., 2001; Kerr et al., 2007). • Recent data indicate that small amounts of alcohol can augment the cognitive deficit associated with hypoglycaemia in individuals with type 1 diabetes (Cheyne et al., 2004).
104
MODERATORS, MONITORING AND MANAGEMENT
In the past, biochemical hypoglycaemia (in non-diabetic individuals) associated with alcohol intoxication was attributed to the toxic effects of impurities associated with the production of illicit drinks (‘hooch’ or ‘moonshine’). These included methanol, gasoline and ethyl acetate (Brown and Harvey, 1941). More recently, it was thought that the biochemical effects of alcohol were associated with hypoglycaemia in three ways: • inhibition of gluconeogenesis; • potentiation of the effects of exercise and glucose-lowering agents; • causing reactive hypoglycaemia in susceptible individuals. More than 90% of an ethanol load is metabolised by the liver, being catalysed by alcohol dehydrogenase into acetate. The rate-limiting step is ultimately dependent upon the availability of nicotinamide di-nucleotide (NAD+) in a glycogen-replete state. In conditions where glycogen stores are depleted and blood glucose is being maintained by hepatic glucose production (e.g. in malnourished individuals and after prolonged exercise), alcohol ingestion may lead directly to a fall in blood glucose. In contrast, studies have failed to show any short-term effect of alcohol consumed with a meal (Kerr et al., 1990; Avogaro et al., 1993), or when given intravenously after an overnight fast (Kolaczynski et al., 1988). This is probably because the suppression of gluconeogenesis by ethanol has little effect on hepatic glucose output in well-fed subjects (Gin et al., 1992) and may in fact be counterbalanced by a reduction in peripheral glucose uptake (Koivisto et al., 1993). In well-nourished, nondiabetic subjects, very little evidence is available to suggest that alcohol has any significant effect on glucose homeostasis (Trojan et al., 1999). However, alcohol has been shown to suppress lipolysis acutely (Avogaro et al., 1993). Following ingestion of alcohol, a reduction in plasma levels of free fatty acids is associated with a reduction in gluconeogenesis and an increased risk of hypoglycaemia in type 1 diabetes (Avogaro et al., 1993). In other words, any potential effect of alcohol upon prevailing glucose is likely to be maximal at a time when glucose homeostasis is dependent upon free fatty acid production, e.g. overnight, when lipolysis increases to promote gluconeogenesis (Hagstrom-Toft et al., 1997). The alcohol-induced suppression of lipolysis may then predispose to hypoglycaemia the next morning (Figures 5.1 and 5.2). The predisposition to delayed hypoglycaemia in type 1 diabetes may be augmented by relative hyperinsulinaemia, which in turn further suppresses lipolysis. However, it is likely that other factors are involved, as in well-fed subjects the suppression of gluconeogenesis by alcohol per se may have little effect on hepatic glucose output (Gin et al., 1992), and is considered to be counterbalanced by a reduction in peripheral glucose uptake (Koivisto et al., 1993). In a clinical context the main concern for insulin-treated individuals is that moderate alcohol intake (6–9 units) can acutely diminish hypoglycaemia awareness (Moriarty et al., 1993) and impair counterregulatory responses to insulin-induced hypoglycaemia (YkiJarvinen and Nikkila, 1985; Pukakainen et al., 1991). Glucagon release has been shown to be suppressed by alcohol in some studies (Rasmussen et al., 2001), but not in others (Kerr et al., 1990). However, the prolonged hypoglycaemic effect of alcohol following its ingestion is more likely to implicate either cortisol or growth hormone responses to hypoglycaemia (Figure 5.3). In animal studies, alcohol has been shown to stimulate corticotrophin-releasing hormone and thus increase plasma cortisol (Rivier et al., 1984). A rise in circulating glucocorticoid can inhibit growth hormone release (Wehrenberg et al., 1990), which in turn may
LIFESTYLE MODERATORS
20
22
24
2
4
6
8
105
10
12
14
Time Adrenaline
FFA
Growth Hormone
Glucose
Figure 5.1 Pictoral showing steady nocturnal and next-day glucose concentrations, a normal rise in growth hormone and free fatty acids overnight, and a normal counterregulatory response should there be a predisposition to hypoglycaemia
Alcohol Potential for hypoglycaemia
20
22
24
2
4
6
8
10
12
14
Time Adrenaline
FFA
GH
Glucose
Figure 5.2 Pictoral indicating a slow decline in plasma glucose following evening alcohol ingestion, suppression of free fatty acids and overnight growth hormone release, and the inability to counterregulate an increased predisposition to hypoglycaemia and impaired awareness of hypoglycaemia
be further reduced through the direct inhibition of growth hormone release through the acute ingestion of alcohol (Conway and Mauceri, 1991). In type 1 diabetes, the influence of alcohol ingestion on the immediate counterregulatory responses has been investigated by a hyperinsulinaemic clamp study (Kerr et al., 2007). Ingestion of modest amounts of alcohol (to plasma levels of less than 50 mg/dl)
MODERATORS, MONITORING AND MANAGEMENT
106
Cortisol (nmol/l)
Free insulin (pmol/l) 500 400
Wine Water
300 200 100 0
800 700 600 500 400 300 200 100 0
20 22 24 02 04 06 08 10 12 Time (24 h)
20 22 24 02 04 06 08 10 12 Time (24 h)
Growth hormone (µg/l) 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0
Wine Water
20 22 24 02 04 06 08 10 12 Time (24 h)
Wine Water
Glucagon (ng/l) 90 80 70 60 50 40 30 20 10 0
Wine Water 20 22 24 02 04 06 08 10 12 Time (24 h)
Figure 5.3 Counterregulatory hormone responses following earlier ingestion of alcohol indicating suppression of overnight growth hormone secretion. Reproduced from Turner et al., 2001, with permission from The American Diabetes Association
appears to attenuate the usual growth hormone response to mild hypoglycaemia (Figure 5.4). Other investigators have reported that acute and sustained alcohol ingestion in non-diabetic subjects can suppress growth hormone release in response to insulin-induced hypoglycaemia (Kolaczynski et al., 1988; Kerr et al., 1990), and also in individuals suffering from reactive hypoglycaemia (Avogaro et al., 1993). In association with blunting of the hormonal counterregulatory responses to hypoglycaemia following alcohol ingestion, insulin sensitivity also appears to be increased (Ting and Lautt, 2006), thus perhaps directly suppressing hepatic glucose production. Recent studies have also reported blunting of the epinephrine response to hypoglycaemia after alcohol (Figure 5.5). The reduction is catecholamine response was mirrored by a reduction in hypoglycaemia awareness, which has been described previously with alcohol (Kerr et al., 1990). This blunting of the adreno-medullary response could result from a number of different mechanisms: • Antecedent nocturnal hypoglycaemia could predispose to delayed hypoglycaemia (Veneman et al., 1993) (i.e., hypoglycaemia leads to more hypoglycaemia). • Sleep patterns are altered by alcohol, which may increase the time spent in ‘deep’ non-REM sleep, thereby increasing the predisposition towards hypoglycaemia through a reduction in the ability to counterregulate (Jones et al., 1998) (see Chapter 4).
LIFESTYLE MODERATORS Growth hormone (µg / l)
107
Growth hormone (mean and SE)
35
Glc = 4.5
Glc = 4.5
Glc = 4.5 / 2.8
30 25
alcohol
E+P E+A
20
H+P H+A
15 10 5 0 0
30
60
90
120
150
180
210
240
Time (mins)
Figure 5.4 Growth hormone concentrations during the four study conditions (E = euglycaemia, P = placebo, H = hypoglycaemia, A = alcohol and Glc = glucose)
Plasma adrenaline during hyperinsulinaemic hypoglycaemic clamp 1200
Adrenaline (ng/ml)
1000 800 600 400 200 0 Baseline euglycemia Initial hypoglycaemia 4.5mmol / l 2.5mmol / l
Alcohol
End hypoglycaemia 2.5mmol / l
Euglycaemia 4.5mmol / l
Placebo
Figure 5.5 Epinephrine (adrenaline) response during a hypoglycaemic clamp following alcohol ingested 12 hours previously (unpublished data of authors)
MODERATORS, MONITORING AND MANAGEMENT
108
• Other mechanisms, which could account for a failure to counterregulate through an inadequate catecholamine response, could be explained by a reduction in GH-associated priming of the catecholamine response. The blunted catecholamine response to hypoglycaemia may explain why symptom scores are lower; this may account for the previously described, alcohol-induced impairment of hypoglycaemia awareness that is associated with delayed hypoglycaemia following consumption of alcohol at an earlier time (Kerr et al., 1990). The predisposition of alcohol to cause prolonged, recurrent hypoglycaemia with blunting of the counterregulatory response, may lead to delayed hypoglycaemia and impaired awareness of hypoglycaemia. Knowledge of the increased risks and dangers involved with an increased risk of severe hypoglycaemia is important so that patients can make appropriate adjustments to their lifestyles. Although patients often ask for guidance about alcohol and diabetes, the advice offered can be variable and conflicting. For people treated with insulin, it is often recommended that dietary carbohydrate should not be omitted and alcohol should be taken with, or shortly before, food. Patients should be advised that the risk of hypoglycaemia may extend for ‘several hours’ after drinking. It is an important clinical observation that, at blood alcohol levels that remain within the statutory limits for driving in the UK, autonomic and neuroglycopenic warning symptoms of early hypoglycaemia can be impaired and the cognitive deficits that are usually associated with mild hypoglycaemia are augmented (Cheyne et al., 2004). As is recommended for all non-diabetic drivers, the advice should be to consume no alcohol if an individual is planning to drive. A further consideration is the ‘morning after the night before’ phenomenon in terms of hypoglycaemia risk. In a laboratory-based study, Turner et al. (2001) reported that ingestion of alcohol with an evening meal increased the risk of hypoglycaemia the next morning (Figure 5.6). More recently, a study of people with type 1 diabetes during their normal daily lives and using a continuous glucose monitoring system (CGMS), confirmed a predisposition 20 Wine Water
16 12 Glucose (mmol / l) 8 4 0 18
20
22
24
02 04 06 Time (24 h)
08
10
12
Figure 5.6 Change in overnight plasma glucose following alcohol or placebo (Turner et al., 2001). The period of drinking is indicated by the shaded bar and the times of symptomatic hypoglycaemia are indicated by the vertical arrows. Reprinted with permission from The American Diabetes Association
mean change in interstitial glucose (mmol / l)
LIFESTYLE MODERATORS
109
3.0 2.0 1.0
period of increased risk of hypoglycaemia
0.0 18:00 20:00 22:00 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00
–1.0 –2.0 –3.0 –4.0 –5.0
Figure 5.7 Mean difference in interstitial glucose between alcohol and placebo beverages following standardised meal at 19:00–20:00
to delayed hypoglycaemia (Richardson et al., 2005b). In this study alcohol was shown to reduce average interstitial tissue glucose (Figure 5.7) and increase the risk of hypoglycaemia over the following 24 hours, with patients reporting more than twice as many hypoglycaemic episodes per day. In the study of Richardson et al. (2005b), the average interstitial glucose level was 1– 2 mmol/l lower with alcohol, but the rate of hypoglycaemia the next day depended on the prevailing levels of glucose overnight. Patients who had a lower baseline glucose at the beginning of the night or in the morning, were at a greater risk of developing alcohol-induced hypoglycaemia than those with preceding hyperglycaemia. This increased risk persisted for nearly a full day after alcohol ingestion (Richardson et al., 2005b).
Caffeine The consumption of caffeine has occurred for over 8000 years. Its impact on health – both good and bad – has been widely reported and disputed. Coffee (and caffeine) is the most widely used stimulant in the world and is consumed by more than 50% of Britons regularly. This amounts to about 75 million cups of coffee consumed every day. Caffeine is also consumed in a variety of different formats such as tea and soft drinks and in ‘over the counter’ remedies for coughs and colds. During the period from 1960 to 1982, the level of consumption of caffeine-containing products rose by 231% (Gilbert, 1984), and in the UK, the average caffeine consumption by adults was estimated to be 400 mg daily (Gilbert, 1984). Children, who do not drink coffee, may consume equivalent amounts in soft drinks. Caffeine exerts a variety of pharmacological actions at diverse sites, both centrally and peripherally, principally through adenosine receptor antagonism. Although it has been suggested that the amount of caffeine consumed is related to the risk of developing or protecting against type 2 diabetes (Pereira et al., 2006), the discussion here is focused on the influence that caffeine has on the physiological responses to a fall in blood glucose.
MODERATORS, MONITORING AND MANAGEMENT Minutes of interstitial hypoglycaemia
110
200 160 120 80 40 0
Day placebo
Day caffeine
Night placebo
Night caffeine
Figure 5.8 Diurnal variation in time spent in hypoglycaemic range (interstitial glucose < 35 mmol/l) comparing caffeine and placebo. Error bars indicate the confidence intervals of the means
Previous studies have shown that ingestion of moderate amounts of caffeine may be useful by augmenting the symptomatic and hormonal responses to mild hypoglycaemia. Caffeine (3–4 cups of drip-brewed coffee each day) enhances the symptomatic and sympathoadrenal responses to hypoglycaemia in healthy non-diabetic volunteers and in patients with type 1 diabetes (Kerr et al., 1993; Debrah et al., 1996). This may enhance the ability of individuals to perceive the onset of symptoms and take appropriate action by ingesting carbohydrate before neuroglycopenia develops. The beneficial effect of caffeine on hypoglycaemia risk is independent of a change in glycaemic control (Watson et al., 2000). Putative mechanisms have included an increase in counterregulatory hormones, including epinephrine, growth hormone and cortisol, but not norepinephrine (Debrah et al., 1996). More recently the influence of caffeine on frequency of hypoglycaemia in patients with long-standing type 1 diabetes has been studied using CGMS. By using continuous monitoring, caffeine appears to reduce the duration of nocturnal hypoglycaemia in subjects with type 1 diabetes by almost 50% (Richardson et al., 2005a) (Figure 5.8). It has been suggested that a beneficial effect of the caffeine-associated reduction in nocturnal hypoglycaemia may be to reduce the risk of developing impaired awareness of hypoglycaemia on the next day. The caffeine-associated reduction in ‘antecedent’ nocturnal hypoglycaemia seen in these studies may explain the augmentation in the symptomatic and hormonal responses to mild daytime hypoglycaemia described previously (Debrah et al., 1996). The relationship between autonomic dysfunction and the development of impaired awareness of hypoglycaemia was unclear for many years (see Chapter 7). Most studies have suggested that peripheral autonomic neuropathy is not associated with an increased risk of severe hypoglycaemia (Polinsky et al., 1980; Bjork et al., 1990; Ryder et al., 1990; The DCCT Research Group, 1991) although central autonomic dysfunction may be important (Evans et al., 2003). Although caffeine improves parasympathetic autonomic function (Richardson et al., 2004), no correlation was found with the observed reduction in nocturnal hypoglycaemia associated with caffeine. As mentioned earlier, it is possible that caffeine uncouples cerebral blood flow and glucose utilisation via antagonism of adenosine receptors (Laurienti et al., 2003), attenuating the glucose supply to the brain (reduced cerebral blood
MONITORING
111
flow) while simultaneously increasing glucose demand, thus resulting in relative neuroglycopenia and earlier release of counterregulatory hormones. Caffeine may also act through an alteration in sleep pattern as adenosine has been implicated in the physiological regulation of sleep. Caffeine reduces non-rapid eye movement (REM) sleep (Landolt et al., 1995), a stage of sleep that is associated with attenuated counterregulatory responses to hypoglycaemia (Jones et al., 1998). Therefore, it could be hypothesised that caffeine reduces the time spent in non-REM sleep, lessening the period during which counterregulatory responses are suppressed. This may protect against prolonged hypoglycaemia and may explain the findings of fewer and shorter moderate episodes of nocturnal hypoglycaemia in people using caffeine. The suggested beneficial effects of caffeine on nocturnal hypoglycaemia would support the notion that caffeine use should be encouraged in a population that is prone to the risks of severe nocturnal hypoglycaemia.
MONITORING It is a sine qua non that the best defence against hypoglycaemia is the ability to recognise it at an early stage and take appropriate action. Nevertheless, some people lose their ability to detect hypoglycaemia: • Individuals may fail to develop appropriate warning symptoms. • Individuals may fail to recognise the warning symptoms as being related to hypoglycaemia. • Individuals may recognise the warning symptoms but may be unable to take appropriate action because of neuroglycopenia. Memory impairment is commonly associated with hypoglycaemia (see Chapter 2) and thus patient recall may underestimate the true frequency of the problem. Furthermore the definition of hypoglycaemia needs to be agreed by the patient, their relatives and their health professionals especially for more modest events, which are often ‘accepted’ by patients as an inevitable part of insulin treatment. As a consequence, it is important that patients have additional methods of detecting low blood glucose levels. Invariably this involves finger stick measurements of blood glucose levels using glucose meters. However, this method has problems per se related to: • poor technical performance, e.g. inadequate samples; • contamination (actually rare in routine practice); • technical limitations of the devices (Melki et al., 2006). Recently novel methods for measuring glucose levels have been introduced although at present none fulfils the ‘holy grail’ of non-invasive glucose monitoring. It is noteworthy that the newer methods of measuring interstitial glucose have highlighted the fact that hypoglycaemia remains a common problem in type 1 diabetes management and that many episodes remain unrecognised (Cheyne and Kerr, 2002). These methods can, however, be a useful aid, allowing patients (and their healthcare professionals) to determine the modulating
MODERATORS, MONITORING AND MANAGEMENT
112
Sensor Profile – after pump therapy 25.0
20.0
20.0 sensor value
Sensor value (mmol/l)
Sensor Profile – before pump therapy 25.0
15.0 10.0
15.0 10.0 5.0
5.0 0.0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00
Time
0.0 0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 Time
Glucose Concentration (mmol/L)
Figure 5.9 CGMS profile before and after conversion to CSII in a patient with > 15 years of poorly controlled type 1 diabetes
20.0 15.0 10.0 5.0 0.0 0:00
4:00
8:00
12:00
16:00
20:00
24:00
Time
Figure 5.10 CGMS profile indicating the extent of postprandial hyperglycaemia. Capillary blood glucose measurements are shown in square boxes
influence of a number of factors on glycaemic control and insulin use (Figures 5.9 and 5.10). Unfortunately they have a number of limitations, as detailed in the next section.
Continuous Glucose Monitoring Systems (CGMS) There are limitations to the use of traditional blood glucose measurements for detecting low blood glucose levels. Consequently, a great deal of time, effort and money has been spent on developing new methods for the detection of low blood glucose levels. The most popular system at present is the use of a continuous glucose monitoring system (CGMS) (Hoi-Hansen et al., 2005). However, technical considerations influence the use of interstitial glucose monitoring for the detection of hypoglycaemia, specifically: • There is a physiological lag between changes in blood and interstitial glucose levels which suggest that CGMS may overestimate the duration of a hypoglycaemic event. • It is unclear as to the significance of changes at the level of interstitial fluid compared to fluctuations in blood glucose level.
MONITORING
113
• It is unclear whether changes in interstitial glucose levels, measured in the anterior abdominal wall, mirror those occurring at the level of glucose sensing neurones in the hypothalamus and elsewhere. • The recordings from the CGMS are markedly influenced by the accuracy and frequency of the results of fingerstick blood testing for calibration purposes. The first commercially available device, the CGMS (Medtronic Minimed, Minneapolis, USA), was designed to monitor glucose levels continuously in tissue fluid. The system comprised a disposable subcutaneous glucose sensor connected by a cable to a pager-sized glucose monitor. The sensor measured interstitial glucose levels every 10 seconds and averaged over 5 minutes, i.e., 288 measurements each day, and provided measurements of glucose in the range of 2.2–22 mmol/l. The first version provided retrospective data. The latest generation sensor transfers data to an on-screen display in real-time for the patient to adjust his or her own medication (Bode et al., 2004). One problem has been in defining hypoglycaemia using these devices, although some groups have defined interstitial hypoglycaemia according to the level at which there is activation of the counterregulatory hormone cascade and onset of neuroglycopenic symptoms respectively (Fanelli et al., 1994) (Table 5.2). There are also other limitations (including expense) in the use of continuous interstitial monitoring devices. Continuous glucose monitoring is a technique in its infancy and as such there is debate whether it has the ability to detect ‘true’ hypoglycaemia (McGowan et al., 2002). This uncertainty relates to a number of factors including:
Table 5.2 Guidelines for interpreting interstitial hypoglycaemia using CGMS as developed by the UK Hypoglycaemia Study Group (2007), and published in Department for Transport Road Safety Research Report No. 61, “Stratifying hypoglycemic event risk in insulin-treated diabetes” (2006), pp. 68–9 Defining an episode of hypoglycaemia There must be four consecutive readings of 3.5 mmol/l or lower for an episode to be classified as hypoglycaemia. The first of the readings (of 3.5 mmol/l or less) signifies the start of hypoglycaemia. The first reading of 3.5 mmol/l or above signifies the end of hypoglycaemia. Hypoglycaemia ends fully when there are four or more consecutive readings above 3.5 mmol/l. If during hypoglycaemia the sensor value is 3.5 mmol/l or higher for 1–3 readings and then goes back down below 3.5 mmol/l, then the whole episode is counted as one hypoglycaemic event and the short period above 3.5 mmol/l is included in the duration of hypoglycaemia. To be labelled as moderate hypoglycaemia, the sensor has to read 3.0 mmol/l or below for at least four consecutive readings. If during mild hypoglycaemia the reading falls to 3.0 mmol/l or below for four or more consecutive readings, the whole hypoglycaemic episode will be considered as moderate hypoglycaemia. Prolonged hypoglycaemia is defined as lasting > 2 hours.
MODERATORS, MONITORING AND MANAGEMENT
114
Sensor value compared to blood glucose 6.00
Glucose (mmol/l)
5.00 4.00 3.00 2.00 sensor average average BG
1.00 0.00 0:00
1:00
2:00 Time (hours)
3:00
4:00
Figure 5.11 Comparison between interstitial (CGMS) and plasma glucose under hyperinsulinaemic hypoglycaemic clamp conditions
• the potential error in recording glucose values at the limit of the detection range; • the potential for artefact (a flat line might indicate a low glucose value or a technical problem with the recording); • differences between blood and extracellular interstitial glucose. Continuous glucose monitoring methodology does not directly sample blood glucose but records glucose values in the extracellular interstitial space. Glucose diffuses across the capillary wall into the interstitial space before being transported into cells where it is metabolised or stored. The relationship between blood and interstitial glucose is not well understood and there are physiological and pharmacological reasons why the two may differ. Laboratory studies have demonstrated a variable relationship between blood and interstitial glucose concentrations according to whether blood glucose is falling or rising, and circulating insulin concentrations may also be important (Caplin et al., 2003). Finally, it is important to note that the use of CGMS is best used to corroborate the more traditional method of recording hypoglycaemia events, which is relying on patients’ self-reports. Even taking into account the uncertainty of which CGMS value represents true hypoglycaemia, rates of low glucose are comparable whether measured by prospective collection of self-reported episodes or by continuous glucose monitoring. Although there is increasing experience with the CGMS system for detection of hypoglycaemia, the relationship between interstitial glucose levels measured from the anterior abdominal wall and cerebral interstitial levels is unknown. Furthermore, in non-diabetic individuals, the CGMS may overestimate the duration of hypoglycaemia as there appears to be a time lag between sensor-measured interstitial tissue glucose and peripheral blood glucose levels during recovery from hypoglycaemia (Cheyne et al., 2002) (Figure 5.11). Overestimation of nocturnal hypoglycaemia in patients with strictly controlled type 1 diabetes has also been reported (McGowan et al., 2002). However, in more customary populations where capillary glucose sensor calibration was more widely dispersed, accurate prediction of hypoglycaemia with the CGMS has been demonstrated (McGowan et al., 2002; Caplin et al., 2003).
MANAGEMENT OF HYPOGLYCAEMIA
115
MANAGEMENT OF HYPOGLYCAEMIA It goes without saying that prevention is better than cure and patients with diabetes need to be thoroughly and continuously educated about the potential risks of hypoglycaemia. Diabetes UK recommends a policy of 4 is the floor as a capillary blood glucose level for intervention with regard to treating hypoglycaemia. The treatment of hypoglycaemia is best regarded as a spectrum of increasing intervention with at one end the simple ingestion of oral carbohydrate and at the other, acute medical therapy in an intensive care unit (Table 5.3). The simplest treatment, when the patient recognises the early warning symptoms (see Chapter 2), is to eat carbohydrate, which must be palatable, concentrated and portable. Glucose tablets (Dextrosol) are usually recommended in the UK, barley sugar in the USA and, in France, lumps of sugar (sucrose). Beverages such as soft drinks or orange juice with a high glucose content are also suitable. The important factor is that short-acting carbohydrate should be followed by some form of longer-acting carbohydrate such as bread or biscuits. The second level of treatment is when the patient is clearly hypoglycaemic but cannot or will not take oral fast-acting carbohydrate. People on insulin will often not admit to being hypoglycaemic and may react adversely to attempts to give them carbohydrate. Liquid glucose solutions are often unsatisfactory because the patient can spit them out. It is better to use a commercially-available glucose gel such as GlucoGel (Diabetic Bio-diagnostics), which can be squeezed like toothpaste into the mouth and is absorbed through the buccal mucosa. Although some doubts have been expressed about the effectiveness of oral glucose gels, relatives often seem to prefer to try this before resorting to injecting glucagon. It should not be used in semi-comatose patients or those at risk of aspirating. Jam or honey may be just as effective. Glucagon promotes hepatic glycogenolysis and the glycaemic response to a dose of 1 mg is essentially the same whether it is injected subcutaneously, intramuscularly or intravenously (Muhlhauser et al., 1985a). The advantage of glucagon is that it can be given
Table 5.3 The therapeutic spectrum of hypoglycaemia; complexity of treatment depends primarily on duration (adapted from MacCuish, 1993) Duration of hypoglycaemia Initial Management (minutes) By patient Oral carbohydrate (> 20 g)
By family
By paramedics
Oral carbohydrate (liquid/solid)
Glucagon 1mg im or iv
Glucose gel
25 g glucose iv
Glucagon 1 mg im
Ongoing management (hours) In hospital A&E department
In intensive care
25 g glucose iv
Mannitol (20%, 20 ml)
Glucagon 1 mg iv
Dexamethasone Dextrose/insulin infusion Oxygen Anticonvulsants Sedation
116
MODERATORS, MONITORING AND MANAGEMENT
by relatives or friends after minimal training. Paramedics can also use it at the patient’s home or in an ambulance. The disadvantages are that it takes longer (approximately 10 minutes) than intravenous glucose to restore consciousness and does not work in patients who have deficient or absent hepatic glycogen stores (alcoholics or people with cachexia). Unfortunately, even where glucagon is available, relatives or friends may not use it. In one study (Muhlhauser et al., 1985b), 53 of 123 episodes of severe hypoglycaemia were treated by relatives or friends with glucagon, 30 by assisting physicians and 44 required hospital admission. When glucagon was available but not used, it was because those who knew how to use it were not present (20 cases) or were too anxious to do so (24 cases). In children with diabetes, Daneman et al. (1989) found that glucagon was used in only a third of households in which it was available – presumably because relatives were either too frightened or poorly educated. Another problem is the limited shelf life; Ward et al. (1990) found that nearly three-quarters of patients knew about glucagon but only 20% had a supply that was in date. Its effect is less certain if coma has been prolonged. In a study of 100 patients brought to a hospital outpatient department, glucagon was immediately effective in only 41% of patients whose mean estimated duration of coma was 50 minutes (MacCuish et al., 1970). Within five minutes the traditional dose of 50 ml of a 50% glucose (dextrose) solution raises blood glucose from below 1 to > 12 mmol/l (Collier et al., 1987). This dose is now considered to be too large by using too concentrated a solution, and 20% dextrose will suffice. The main problem is the difficulty of giving an intravenous injection to an uncooperative patient who may be refusing or resisting treatment or who has to be physically restrained to receive an intravenous injection. Extravasation of the concentrated glucose solution outside the vein causes painful phlebitis and, because of its hypertonicity, even an intravascular injection can cause phlebitis or thrombosis. It is not therefore used much outside hospitals, and few GPs carry glass vials of dextrose for this purpose. When a patient known to have type 1 diabetes is admitted to hospital in hypoglycaemic coma, but fails to recover after being given intravenous glucose, other causes of coma such as excessive ingestion of alcohol, self-poisoning with opiates or other drugs, acute vascular events such as subarachnoid haemorrhage or stroke, head injury or other intracranial catastrophes must be excluded. Cerebral oedema is a recognised sequel of severe hypoglycaemia, and urgent neuroimaging is required to establish its presence; treatment (Table 5.3) is usually in an intensive care unit as this complication has a high mortality. The most difficult decision is to know for how long to continue treatment. Patients, who have made a full recovery after being unconscious for several days, may (anecdotally) appear subsequently to have significant cognitive impairment or permanent brain damage. It is beyond the scope of this chapter to discuss the investigation and management of hypoglycaemia in non-diabetic individuals. In conclusion, hypoglycaemia continues to be a common problem in the management of individuals with type 1 diabetes. The use of newer technologies of continuous glucose monitoring has highlighted that it is almost impossible to eliminate hypoglycaemia completely with present insulin therapy, although understanding moderating factors such as alcohol and including them as a component of education programmes for people with insulin-treated diabetes may help to alleviate some of the anxiety associated with the risk of living constantly with the threat of hypoglycaemia.
REFERENCES
117
CONCLUSIONS • Hypoglycaemia is associated with multiple risk factors. • Alcohol is a frequent cause of severe hypoglycaemia. • Alcohol is associated with delayed hypoglycaemia through impaired hypoglycaemia awareness and an abnormal counterregulatory response. • Understanding of the nature of alcohol-associated hypoglycaemia is the first step to reducing the risks associated with alcohol. • Caffeine may augment the symptomatic and hormonal responses to hypoglycaemia and reduce nocturnal hypoglycaemia in individuals with type 1 diabetes. • Continuous Glucose Monitoring is a useful adjunct to the management of the patient with type 1 diabetes. • Treatment of hypoglycaemia can be regarded as a spectrum of increasing therapeutic complexity depending on the severity of the hypoglycaemia and the clinical status of the patient.
REFERENCES Avogaro A, Beltramello P, Gnudi L, Maran A, Valerio A, Miola M et al. (1993). Alcohol intake impairs glucose counterregulation during acute insulin-induced hypoglycemia in IDDM patients. Evidence for a critical role of free fatty acids. Diabetes 42: 1626–34. Bjork E, Palmer M, Schvarcz E, Berne C (1990). Incidence of severe hypoglycaemia in an unselected population of patients with insulin-treated diabetes mellitus, with special reference to autonomic neuropathy. Diabetes Nutrition and Metabolism 4: 303–9. Bode B, Gross K, Rikalo N, Schwartz S, Wahl T, Page C et al. (2004). Alarms based on real-time sensor glucose values alert patients to hypo- and hyperglycemia: the guardian continuous monitoring system. Diabetes Technology and Therapeutics 6: 105–13. Bott S, Bott U, Berger M, Muhlhauser I (1997). Intensified insulin therapy and the risk of severe hypoglycaemia. Diabetologia 40: 926–32. Brown TM, Harvey AM (1941). Spontaneous hypoglycemia in ‘smoke’ drinkers. Journal of the American Medical Association 17: 12–5. Caplin NJ, O’Leary P, Bulsara M, Davis EA, Jones TW (2003). Subcutaneous glucose sensor values closely parallel blood glucose during insulin-induced hypoglycaemia. Diabetic Medicine 20: 238–41. Cheyne EH, Cavan DA, Kerr D (2002). Performance of a continuous glucose monitoring system during controlled hypoglycemia in healthy volunteers. Diabetes Technology and Therapeutics 4: 607–13. Cheyne E, Kerr D (2002). Making ‘sense’ of diabetes: using a continuous glucose sensor in clinical practice. Diabetes Metabolism Research and Reviews 18 Suppl 1: S43–8. Cheyne EH, Sherwin RS, Lunt MJ, Cavan DA, Thomas PW, Kerr D (2004). Influence of alcohol on cognitive performance during mild hypoglycaemia; implications for type 1 diabetes. Diabetic Medicine 21: 230–7. Collier A, Steedman DJ, Patrick AW, Nimmo GR, Matthews DM, MacIntyre CA et al. (1987). Comparison of intravenous glucagon and dextrose in treatment of severe hypoglycemia in an Accident and Emergency Department. Diabetes Care 10: 712–5. Conway S, Mauceri H (1991). The influence of acute ethanol exposure on growth hormone release in female rats. Alcohol 8: 159–64.
118
MODERATORS, MONITORING AND MANAGEMENT
Cox DJ, Gonder-Frederick L, Julian DM, Clarke W (1994). Long-term follow-up of blood glucose awareness training. Diabetes Care 17: 1–5. Daneman D, Frank M, Perlman K, Tamm J, Ehrlich R (1989). Severe hypoglycemia in children with insulin-dependent diabetes mellitus: frequency and predisposing factors. Journal of Pediatrics 115: 681–5. Debrah K, Sherwin RS, Murphy J, Kerr D (1996). Effect of caffeine on recognition of and physiological responses to hypoglycaemia in insulin-dependent diabetes. Lancet 347: 19–24. Evans SB, Wilkinson CW, Gronbeck P, Bennett JL, Taborsky GJ Jr, Figlewicz DP (2003). Inactivation of the PVN during hypoglycemia partially simulates hypoglycemia-associated autonomic failure. American Journal of Physiology. 284: R57–65. Fanelli C, Pampanelli S, Epifano L, Rambotti AM, Ciofetta M, Moda F et al. (1994). Relative roles of insulin and hypoglycaemia on induction of neuroendocrine responses to, symptoms of, and deterioration of cognitive function in hypoglycaemia in male and female humans. Diabetologia 37: 797–807. Feldman M, Schiller LR (1983). Disorders of gastrointestinal motility associated with diabetes mellitus. Annals of Internal Medicine 98: 378–84. Gilbert, RM (1984). Caffeine consumption. In: The Methylxanthines Beverages and Foods: Chemistry, Consumption and Health Effects. Spiller GA, ed. Alan R Liss, New York: 235–301. Gin H, Morlat P, Ragnaud JM, Aubertin J (1992). Short-term effect of red wine (consumed during meals) on insulin requirement and glucose tolerance in diabetic patients. Diabetes Care 15: 546–8. Gold AE, MacLeod KM, Frier BM (1994). Frequency of severe hypoglycemia in patients with type I diabetes with impaired awareness of hypoglycemia. Diabetes Care 17: 697–703. Hagstrom-Toft E, Bolinder J, Ungerstedt U, Arner P (1997). A circadian rhythm in lipid mobilization which is altered in IDDM. Diabetologia 40: 1070–8. Hoi-Hansen T, Pedersen-Bjergaard U, Thorsteinsson B (2005). Reproducibility and reliability of hypoglycaemic episodes recorded with continuous glucose monitoring system (CGMS) in daily life. Diabetic Medicine 7: 858–62. Jones TW, Porter P, Sherwin RS, Davis EA, O’Leary P, Frazer F et al. (1998). Decreased epinephrine responses to hypoglycemia during sleep. New England Journal of Medicine 338: 1657–62. Kerr D, Macdonald IA, Heller SR, Tattersall RB (1990). Alcohol causes hypoglycaemic unawareness in healthy volunteers and patients with Type 1 (insulin-dependent) diabetes. Diabetologia 33: 216–21. Kerr D, Sherwin RS, Pavalkis F, Fayad PB, Sikorski L, Rife F et al. (1993). Effect of caffeine on the recognition of and responses to hypoglycemia in humans. Annals of Internal Medicine 119: 799–804. Kerr D, Cheyne EH, Thomas P, Sherwin RS (2007). Influence of acute alcohol ingestion on the hormonal responses to modest hypoglycaemia in patients with Type 1 diabetes. Diabetic Medicine 24: 312–16. Kolaczynski JW, Ylikahri R, Harkonen M, Koivisto VA (1988). The acute effect of ethanol on counterregulatory response and recovery from insulin-induced hypoglycemia. Journal of Clinical Endocrinology and Metabolism 67: 384–8. Koivisto VA, Tulokas S, Toivonen M, Haapa E, Pelkonen R (1993). Alcohol with a meal has no adverse effects on postprandial glucose homeostasis in diabetic patients. Diabetes Care 16: 1612–4. Landolt HP, Werth E, Borbely AA, Dijk DJ (1995). Caffeine intake (200 mg) in the morning affects human sleep and EEG power spectra at night. Brain Research 675: 67–74. Laurienti PJ, Field AS, Burdette JH, Maldjian JA, Yen YF, Moody DM (2003). Relationship between caffeine-induced changes in resting cerebral perfusion and blood oxygenation level-dependent signal. American Journal of Neuroradiology 24: 1607–11. MacCuish AC, Munro JF, Duncan LJP (1970). Treatment of hypoglycaemic coma with glucagon, intravenous dextrose, and mannitol infusion in a hundred diabetics. Lancet ii: 946–9. MacCuish AC (1993). Treatment of hypoglycaemia. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M, eds. Edward Arnold, London: 212–21.
REFERENCES
119
MacDonald MJ (1987). Postexercise late-onset hypoglycemia in insulin-dependent diabetic patients. Diabetes Care 10: 584–8. McGowan K, Thomas W, Moran A (2002). Spurious reporting of nocturnal hypoglycemia by CGMS in patients with tightly controlled type 1 diabetes. Diabetes Care 25: 1499–503. Melki V, Ayon F, Fernandez M, Hanaire-Broutin H (2006). Value and limitations of the Continuous Glucose Monitoring System in the management of type 1 diabetes. Diabetes and Metabolism 32: 123–9. Moriarty KT, Maggs DG, Macdonald IA, Tattersall RB (1993). Does ethanol cause hypoglycaemia in overnight fasted patients with type 1 diabetes? Diabetic Medicine 10: 61–5. Muhlhauser I, Koch J, Berger M (1985a). Pharmacokinetics and bioavailability of injected glucagon: differences between intramuscular, subcutaneous, and intravenous administration. Diabetes Care 8: 39–42. Muhlhauser I, Berger M, Sonnenberg G, Koch J, Jorgens V, Schernthaner G et al. (1985b). Incidence and management of severe hypoglycemia in 434 adults with insulin-dependent diabetes mellitus. Diabetes Care 8: 268–73. Muhlhauser I, Toth G, Sawicki PT, Berger M (1991). Severe hypoglycemia in type 1 diabetic patients with impaired kidney function. Diabetes Care 14: 344–46. Nilsson A, Tideholm B, Kalen J, Katzman P (1988). Incidence of severe hypoglycaemia and its causes in insulin-treated diabetics. Acta Medica Scandinavica 224: 257–2. Pereira MA, Parker ED, Folsom AR (2006). Coffee consumption and risk of type 2 diabetes mellitus: an 11-year prospective study of 28 812 postmenopausal women. Archives of Internal Medicine 26: 1311–6. Polinsky RJ, Kopin IJ, Ebert MH, Weise V (1980). The adrenal medullary response to hypoglycemia in patients with orthostatic hypotension. Journal of Clinical Endocrinology and Metabolism 51: 1401–6. Potter J, Clarke P, Gale EA, Dave SH, Tattersall RB (1982). Insulin-induced hypoglycaemia in an Accident and Emergency Department: the tip of an iceberg? British Medical Journal 285: 1180–2. Pramming S, Thorsteinsson B, Bendtson I, Binder C (1990). The relationship between symptomatic and biochemical hypoglycaemia in insulin-dependent diabetic patients. Journal of Internal Medicine 228: 641–6. Pukakainen I, Koivisto VA, Yki-Jarvinen H (1991). No reduction in total hepatic glucose output by inhibition of gluconeogenesis with ethanol in NIDDM patients. Diabetes 40: 1319–27. Rasmussen BM, Orskov L, Schmitz O, Hermansen K (2001). Alcohol and glucose counterregulation during acute insulin-induced hypoglycemia in type 2 diabetic subjects. Metabolism 50: 451–7. Richardson T, Rozkovec A, Thomas P, Ryder J, Meckes C, Kerr D (2004). Influence of caffeine on heart rate variability in patients with long-standing type 1 diabetes. Diabetes Care 27: 1127–31. Richardson T, Thomas P, Ryder J, Kerr D (2005a). Influence of caffeine on frequency of hypoglycemia detected by continuous interstitial glucose monitoring system in patients with long-standing type 1 diabetes. Diabetes Care 28: 1316–20. Richardson T, Weiss M, Thomas P, Kerr D (2005b). Day after the night before: influence of evening alcohol on risk of hypoglycemia in patients with type 1 diabetes. Diabetes Care 28: 1801–2. Rivier C, Bruhn T, Vale W (1984). Effect of ethanol on the hypothalamic-pituitary-adrenal axis in the rat: role of corticotropin-releasing factor (CRF). Journal of Pharmacology and Experimental Therapy 229: 127–31. Rosen B, Miodovnik M, Holcberg G, Khoury J, Siddiqi T (1995). Hypoglycaemia: the price of intensive insulin therapy for pregnant women with insulin dependent diabetes mellitus. Obstetrics and Gynaecology 85: 417–22.
120
MODERATORS, MONITORING AND MANAGEMENT
Ryder RE, Owens DR, Hayes TM, Ghatei MA, Bloom SR (1990). Unawareness of hypoglycaemia and inadequate hypoglycaemic counterregulation: no causal relation with diabetic autonomic neuropathy. British Medical Journal 301: 783–7. The DCCT Research Group (1991). Epidemiology of severe hypoglycemia in the Diabetes Control and Complications Trial. American Journal of Medicine 90: 450–9. The Diabetes Control and Complications Trial Research Group (1995). Adverse events and their association with treatment regimens in the Diabetes Control and Complications Trial. Diabetes Care 18: 1415–27. The Diabetes Control and Complications Trial Research Group (1997). Hypoglycemia in the Diabetes Control and Complications Trial. The Diabetes Control and Complications Trial Research Group. Diabetes 46: 271–86. Ting JW, Lautt WW (2006). The effect of acute, chronic, and prenatal ethanol exposure on insulin sensitivity. Pharmacology and Therapeutics 111: 346–73. Trojan N, Pavan P, Iori E, Vettore M, Marescotti MC, Macdonald IA et al. (1999). Effect of different times of administration of a single ethanol dose on insulin action, insulin secretion and redox state. Diabetic Medicine 16: 400–7. Turner BC, Jenkins E, Kerr D, Sherwin RS, Cavan DA (2001). The effect of evening alcohol consumption on next-morning glucose control in type 1 diabetes. Diabetes Care 24: 1888–93. UK Hypoglycaemia Group (2007). Risk of hypoglycaemia in types 1 and 2 diabetes: effects of treatment modalities and their duration. Diabetologia 50: 1140–7. Veneman T, Mitrakou A, Mokan M, Cryer P, Gerich J (1993). Induction of hypoglycemia unawareness by asymptomatic nocturnal hypoglycemia. Diabetes 42: 1233–7. Vinik AI, Maser RE, Mitchell BD, Freeman R (2003). Diabetic autonomic neuropathy. Diabetes Care 26: 1553–79. Ward CM, Stewart AW, Cutfield RG (1990). Hypoglycaemia in insulin dependent diabetic patients attending an outpatients’ clinic. New Zealand Medical Journal 25: 339–41. Watson JM, Jenkins EJ, Hamilton P, Lunt MJ, Kerr D (2000). Influence of caffeine on the frequency and perception of hypoglycemia in free-living patients with type 1 diabetes. Diabetes Care 23: 455–9. Wehrenberg WB, Janowski BA, Piering AW, Culler F, Jones KL (1990). Glucocorticoids: potent inhibitors and stimulators of growth hormone secretion. Endocrinology 126: 3200–3. Yki-Jarvinen H, Nikkila EA (1985). Ethanol decreases glucose utilisation in healthy man. Journal of Clinical Endocrinology and Metabolism 61: 941–5.
6 Counterregulatory Deficiencies in Diabetes David Kerr and Tristan Richardson
INTRODUCTION Modern intensive education programmes for people with insulin-treated diabetes have failed to eliminate hypoglycaemia, and this important side-effect of insulin treatment may add to the psychological, as well as the physical, burden associated with this condition. Attempts to improve blood glucose control can increase the risk of severe hypoglycaemia. For example, within the Diabetes Control and Complications Trial (DCCT), improvements in HbA1c levels with intensive insulin therapy were associated with a three-fold increase in the risk of severe hypoglycaemia compared to individuals treated conventionally (The Diabetes Control and Complications Trial Research Group,1993). Despite the development of strategies to improve glycaemic control, while simultaneously trying to reduce the risk of recurrent severe hypoglycaemia, less than half of people with type 1 diabetes consistently achieve HbA1c levels < 75% (Jacqueminet et al., 2005), and for some individuals, fear of hypoglycaemia is the main barrier to achieving optimal glycaemic control (Cox et al., 1987) (see Chapter 14). The use of continuous glucose monitoring has indicated that the frequency and duration of hypoglycaemic events in patients with type 1 diabetes have probably been under-recognised (Figure 6.1) (Cheyne and Kerr, 2002), especially in children (Weintrob et al., 2004). In the UK in recent years, a number of structured education programmes have been developed, most of which are based on a German model (Muhlhauser et al., 1987), and have been focused on empowering patients to alter insulin doses more accurately according to the carbohydrate content of meals, the level of planned exercise, work demands and so on (DAFNE Study Group, 2002). Disappointingly, despite increasing numbers of patients having access to these educational programmes, along with the increased availability of analogue insulins, and with the evidence that these measures are reducing the number of patients who have microvascular complications (Figure 6.2), the rates of severe hypoglycaemia have not altered significantly (Figure 6.3) (Bulsara et al., 2004). Why is hypoglycaemia still such an important problem for people with type 1 diabetes despite these significant improvements in technology and the delivery of care? The answer frequently relates to the problem of defective glucose counterregulation and the often associated difficulty of recognising a fall in blood glucose (impaired awareness of hypoglycaemia – see Chapter 7) at a time when appropriate self-corrective action can be taken.
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
COUNTERREGULATORY DEFICIENCIES IN DIABETES
122
Glucose (mmol/L)
20.0 15.0 10.0 5.0 0.0 –5.0 12:00 AM
4:00 AM
8:00 AM
12:00 PM
4:00 PM
8:00 PM
12:00 AM
Figure 6.1 Interstitial glucose values over three days measured by continuous glucose monitoring and showing persistent nocturnal hypoglycaemia
50% Retinopathy Kidney disease
40% 30% 20% 10% 0% 25 years
Figure 6.2 Reduction in prevalence of microvascular complications over the past 25 years from specialist diabetes centres. Adapted from Rossing (2005), with kind permission from Springer Science and Business Media
NORMAL GLUCOSE COUNTERREGULATION Under normal circumstances, the brain uses glucose as its predominant fuel and is almost completely dependent upon a continuous supply of glucose from the peripheral circulation to maintain normal function. Consequently, to protect the delivery of glucose to the brain, a hierarchy of (counterregulatory) responses (Figure 6.4) are activated as peripheral blood glucose falls below normal (Mitrakou et al., 1991). The main components of this system of normal (i.e. non-diabetic) glucose counterregulation, which prevents or quickly corrects hypoglycaemia, are as follows: • A reduction in pancreatic -cell insulin secretion. • An increase in pancreatic -cell glucagon secretion. If hypoglycaemia is prolonged a number of other hormones are also released. These include epinephrine (adrenaline), growth hormone and cortisol.
NORMAL GLUCOSE COUNTERREGULATION
123
A 22 20 Rate / 100 Patient years
18 16 14 12 10 8 6
Incidence rate
4 2 1992
1994
1996
1998
2000
2002
Calender Year
Figure 6.3 Rates of severe hypoglycaemia in 1335 children between 1992 and 2002. Hypoglycaemia has remained problematical despite a fall in HbA1c of 0.2% per year and increased use of multiple daily injections, insulin analogues and pump therapy. Adapted from Bulsara et al. (2004) and reproduced courtesy of The American Diabetes Association
• Activation of the autonomic nervous system with the development of characteristic warning symptoms. • During more profound hypoglycaemia (blood glucose < 20 mmol/l), glucose delivery to the brain is enhanced as a consequence of increasing cerebral blood flow (Thomas et al., 1997). • Hepatic autoregulation, whereby the liver is able to respond to hypoglycaemia by directly increasing glucose production in the absence of detectable hormonal stimulation. Hepatic autoregulation only appears to have an influential role in glucose counterregulation during prolonged and severe hypoglycaemia (Tappy et al., 1999). The principal counterregulatory hormones, glucagon and epinephrine, directly increase the production of glucose by the liver as a consequence of the breakdown of hepatic glycogen stores (glycogenolysis) and the manufacture of glucose by gluconeogenesis. Epinephrine also promotes muscle glycogenolysis, proteolysis and lipolysis to provide substrates (lactate, alanine and glycerol) for further gluconeogenesis (Figure 6.5). As blood glucose falls below normal, these hormonal responses are not secreted on an ‘all or nothing’ basis. Individual hormones have specific blood glucose thresholds at which levels begin to rise above their baseline levels (Figure 6.6). For example, the glycaemic thresholds for the release of glucagon and epinephrine are well above the thresholds for the generation of warning symptoms and impairment of higher brain (cognitive) function (Mitrakou et al., 1991). These thresholds are not fixed but can be altered upwards or downwards according to
COUNTERREGULATORY DEFICIENCIES IN DIABETES
124
Hypothalamus
Pituitary gland Growth hormone
u vo er ic N nom Auto
Pancreas
ACTH
s s Sy
te m
Adrenal gland
Cortisol Glucagon Norepinephrine
Epinephrine
Figure 6.4 The brain is the key regulatory organ involved in glucose counterregulation
the prevailing quality of glycaemic control. They do not appear, however, to be influenced by the rate of fall of blood glucose within the hyper- or euglycaemic range. The key organ in coordinating the hormonal and other responses to hypoglycaemia is the brain. Although numerous neural areas have been proposed as the control centres for counterregulation, it is likely that neurones located within the ventro-medial nuclei of the hypothalamus are essential for integrating the hormonal responses to a fall in peripheral blood glucose (Borg et al., 1994), possibly via ATP-sensitive K+ channels (McCrimmon et al., 2005), although other glucoreceptors outside the brain are involved in initiating the counterregulatory responses, most notably within the liver (Smith et al., 2002). Although the brain was once thought to be insensitive to insulin, there is clinical evidence to suggest that insulin can act on the central nervous system (CNS) to influence the physiological responses to hypoglycaemia (Kerr et al., 1991). Recently, in studies using brain/neuronal insulin receptor knockout mice (i.e., mice with absent insulin receptor proteins in the brain), the induction of hypoglycaemia was associated with an attenuated epinephrine and an almost completely absent norepinephrine response, although glucagon release was unaffected when compared to control animals. Therefore it appears that insulin has a role in protecting the CNS against hypoglycaemia (Fisher et al., 2005).
NORMAL GLUCOSE COUNTERREGULATION
125
Figure 6.5 Principal metabolic effects of counterregulation in response to hypoglycaemia
Glucose (mmol / l) 4.0
Release of hormones
Warning symptoms 3.0
Cognitive impairment
Increased brain blood flow 2.0
Figure 6.6 Blood glucose thresholds for release of counterregulatory hormones, onset of warning symptoms of hypoglycaemia and cognitive impairment as blood glucose falls below normal
126
COUNTERREGULATORY DEFICIENCIES IN DIABETES
In type 1 diabetes there may be multiple defects in normal glucose counterregulation which significantly increase an individual’s risk of hypoglycaemia. Defects in hormonal counterregulation can result as a consequence of the following: • secretion of inadequate amounts of counterregulatory hormones; • alteration of the blood glucose threshold at which the hormones are released; i.e., a more profound hypoglycaemic stimulus is required before hormonal secretion increases above baseline; • diminished tissue sensitivity to a given plasma concentration of hormone. In type 1 diabetes, the most common scenario for increasing the risk of recurrent hypoglycaemia includes a reduced (but not absent) epinephrine response to falling blood glucose levels, sustained peripheral hyperinsulinaemia and an absent glucagon response – this constellation of defects is associated with a 25-fold increased risk of severe hypoglycaemia in people who are on intensive insulin therapy. If impaired hypoglycaemia awareness is present, the risk is increased six fold (see Chapter 7).
DEFECTIVE HORMONAL GLUCOSE COUNTERREGULATION The most important defects in glucose counterregulation in type 1 diabetes are: • failure of circulating plasma insulin levels to decline (i.e., as a consequence of exogenous insulin administration); • failure of glucagon secretion from pancreatic -cells; • attenuated epinephrine response to hypoglycaemia. Several factors are known to increase the risk of counterregulatory failure, including: • long duration of diabetes; • extremes of age; • improving glycaemic control; • sleep; • exercise; • recurrent hypoglycaemia – related to the degree rather than duration of antecedent hypoglycaemia (Davis et al., 2000a). The defects in glucose counterregulation are not ‘all or nothing’ and are influenced by a number of factors. Some of the defects may be reversible.
DEFECTIVE HORMONAL GLUCOSE COUNTERREGULATION
127
Glucagon In non-diabetic individuals, glucagon is secreted by pancreatic -cells within 30 minutes of the blood glucose falling below normal. It is released directly as a consequence of local tissue glucopenia and indirectly by sympathetic neural inputs to the pancreas. It is unclear whether local (pancreatic) tissue glucopenia or autonomic activation is the key process involved in stimulating the release of glucagon during hypoglycaemia. In people with type 1 diabetes, the glucagon secretory response to hypoglycaemia is initially diminished and is subsequently lost within a few years of the onset of diabetes (Gerich et al., 1973), although it can be released in response to other stimuli, such as exercise or an intravenous infusion of arginine (Gerich and Bolli, 1993). This indicates that the failure of the glucagon response is most probably a signalling rather than a structural defect. Loss of the glucagon response to hypoglycaemia often coexists with clinical evidence of autonomic neuropathy but the latter is not invariably present (Bolli et al. 1983). Although recent studies have suggested that an early sympathetic neuropathy limited to the pancreatic islets may be a potential mechanism, patients with pancreatic transplants (i.e., denervated islets) can still produce glucagon in response to hypoglycaemia (Diem et al., 1990). The cause of the defect in glucose counterregulation is not known but may include: • reduction in pancreatic -cell mass; • autonomic neuropathy; • local effect of insulin on normal -cell function; • generalised hyperinsulinaemia; • chronic hyperglycaemia; • increased pancreatic production of somatostatin; • accumulation of amylin within the islets; • local effect of insulin-like growth factor-1. An alternative hypothesis to explain the glucagon-counterregulatory defect suggests that in healthy individuals a decrease in intra-islet insulin in association with a decrease in -cell glucose is the usual signal for release of glucagon as peripheral blood glucose levels fall (Samols et al., 1972). Support for this comes from the observation that infusion of the -cell secretagogue, tolbutamide, prevents the glucagon response to hypoglycaemia in non-diabetic individuals (Banarer and Cryer, 2003). Similarly, supraphysiological levels of insulin have been shown to impair glucagon release in response to moderate hypoglycaemia (Kerr et al., 1991). Blunting of the normal glucagon response to hypoglycaemia can also be achieved in healthy volunteers by infusion of insulin-like growth factor-1 (IGF-1), the putative mediator of the somatotrophic action of growth hormone, at least in non-diabetic humans (Kerr et al., 1993) (Figure 6.7). Whether IGF-1 is involved in the pathogenesis of glucagon counterregulatory failure in patients with type 1 diabetes is unclear. Alternatively, it is possible that chronic hyperglycaemia may directly impair pancreatic -cell function through the mechanism of glucose toxicity similar to the effect that this has on -cell function. Nevertheless, improvements in glycaemic control, following introduction of
128
COUNTERREGULATORY DEFICIENCIES IN DIABETES
Figure 6.7 Infusion of insulin-like growth factor 1 abolishes the expected rise in glucagon when blood glucose was lowered to, and maintained at, 2.8 mmol/l in healthy volunteers. Reproduced from Kerr et al. (1993) by permission of The Journal of Clinical Investigation
intensive insulin therapy, invariably fail to restore the glucagon responses to hypoglycaemia (Amiel, 1991). The early appearance of an impaired response of glucagon, together with a temporal dissociation from deficiencies in other counterregulatory hormones, argues against this abnormality being mediated from within the central nervous system. Impairment of the glucagon response to hypoglycaemia results in greater suppression of hepatic glucose production by insulin. Since the ability of insulin to increase glucose utilisation is unaltered, glucose uptake by muscle and other tissues will exceed hepatic glucose production for much longer, resulting in more profound and prolonged hypoglycaemia. Whether or not the lack of a glucagon response is by itself sufficient to increase the frequency of severe hypoglycaemia is unclear; most patients with recurrent severe hypoglycaemia have impairment both of glucagon and epinephrine release.
Catecholamines An attenuated epinephrine response (from the adrenal medulla) to falling blood glucose levels together with an absent glucagon response markedly increases the risk of severe hypoglycaemia in individuals with type 1 diabetes. It is unknown how common epinephrine counterregulatory failure is in type 1 diabetes, but it is related to the duration of diabetes and the prevailing quality of glycaemic control. Estimates suggest this defect may exist in up to 45% of people with type 1 diabetes of long duration (Gerich and Bolli, 1993). The epinephrine response to hypoglycaemia can be subnormal in patients with type 1 diabetes who have no clinical evidence of autonomic neuropathy and also in individuals with secondary (pancreatic) diabetes. Like glucagon, the epinephrine secretory deficiency is stimulus-specific to hypoglycaemia, remaining intact in response to exercise. The isolated
DEFECTIVE HORMONAL GLUCOSE COUNTERREGULATION
129
failure of plasma epinephrine concentrations to rise in response to hypoglycaemia does not appreciably impair glucose counterregulation – unless it is combined with glucagon deficiency (as occurs in patients with long-standing type 1 diabetes), when the risk of severe and prolonged hypoglycaemia is significantly increased. As mentioned above, deficient sympathoadrenal responses to hypoglycaemia are major components of the clinical problem of defective glucose counterregulation and impaired awareness of hypoglycaemia in type 1 diabetes (Cryer, 2004; 2005). This may occur as a result of the following: • reduced release of catecholamines and acetyl choline; • possibly a reduction in tissue sensitivity to the actions of these substances, although this is controversial. Of clinical importance is the observation that a hypoglycaemic episode per se, adversely alters the subsequent sympathoadrenal responses to hypoglycaemia (Heller and Cryer, 1991). Cryer has termed this Hypoglycaemia Associated Autonomic Failure (HAAF) (Figure 6.8). In other words, a recent episode of hypoglycaemia, known as ‘antecedent hypoglycaemia’, (including an asymptomatic event) diminishes the sympathoadrenal, symptomatic and cognitive responses to subsequent hypoglycaemia. As a corollary, avoidance of hypoglycaemia (for as little as two weeks) markedly improves the responses (Cranston et al., 1994). Sympathoadrenal responses to hypoglycaemia are also affected by recent exercise, whether the fall in blood glucose level occurs when an individual is awake or asleep (Jones et al., 1998; Banarer and Cryer, 2003) (Figure 6.9) and by the time of day or night (Merl et al., 2004).
Insulin Deficient Diabetes (Imperfect Insulin Replacement) (No Insulin, No Glucagon)
Antecedent Hypoglycaemia Sleep Reduced Sympathoadrenal Responses to Hypoglycaemia
Antecedent Exercise
Reduced Epinephrine Responses
Reduced Sympathetic Neural Responses
Defective Glucose Counterregulation
Hypoglycaemia Unawareness
Recurrent Hypoglycaemia
Figure 6.8 Hypoglycaemia Associated Autonomic Failure (HAAF) in type 1 diabetes. Adapted from Cryer (2005) and reproduced courtesy of The American Diabetes Association
COUNTERREGULATORY DEFICIENCIES IN DIABETES
85
75
65
55
45
85 3000
500
2500
400
2000
300
1500
200
1000
100
pmol / l
Epinephrine (pg/ml)
600
500
0 0
Nominal glucose (mg / dl)
(b)
60 120 180 240 300 Time (min)
0
75
65
55
45
600
3000
500
2500
400
Morning awake
300 200
2000
pmol / l
Nominal glucose (mg / dl)
(a)
Epinephrine (pg / ml)
130
1500 Night awake
100
1000 500 Asleep
0 0
60 120 180 240 300 Time (min)
0
Figure 6.9 Plasma epinephrine (adrenaline) in non-diabetic subjects (a) and in patients with type 1 diabetes (b) studied in the morning while awake and during the night while awake and asleep. Adapted from Banarer and Cryer (2003) and reproduced courtesy of The American Diabetes Association
( morning awake; • night awake; asleep)
Cortisol and Growth Hormone Growth hormone (GH) and cortisol are thought to become important glucose-raising hormones only after hypoglycaemia has been prolonged for more than one hour. However, defects in cortisol and GH release can cause profound and prolonged hypoglycaemia because of a reduction in hepatic glucose production and, to a lesser extent, by exaggeration of insulin-stimulated glucose uptake by muscle. Abnormalities in growth hormone and cortisol secretion in response to hypoglycaemia are characteristic of long-standing type 1 diabetes, affecting up to a quarter of patients who have had diabetes for more than ten years. In rare cases, coexistent endocrine failure such as Addison’s disease or hypopituitarism also predisposes patients to severe hypoglycaemia. Pituitary failure, although uncommonly associated with type 1 diabetes, occasionally develops in young women as a consequence of ante-partum pituitary infarction. As an intact hypothalamic–pituitary–adrenal axis is important for adequate counterregulation, this axis should be formally assessed in any individual with brittle diabetes presenting with unexplained, recurrent hypoglycaemia (Hardy et al., 1994; Flanagan and Kerr, 1996). More commonly, ingestion of even modest amounts of alcohol can significantly attenuate normal growth hormone secretion and increase the risk of hypoglycaemia especially the following morning (see Chapter 5).
MECHANISMS OF COUNTERREGULATORY FAILURE At the onset of type 1 diabetes, hormonal counterregulation is usually normal but within five years of diagnosis, glucagon responses to hypoglycaemia become markedly impaired or even absent, although a glucagon response can occur if the hypoglycaemic stimulus is sufficiently profound (Frier et al. 1988; Hvidberg et al. 1998). After ten years of diabetes, patients usually have a sub-optimal epinephrine response to compound the absent glucagon response to a fall
MECHANISMS OF COUNTERREGULATORY FAILURE
131
in blood glucose (White et al., 1985) (Figure 6.10). Thus, patients with type 1 diabetes of long duration are at risk of severe and prolonged neuroglycopenia during hypoglycaemia as a direct consequence of inadequate glucose counterregulation. Although attenuated growth hormone and cortisol responses are less common, they are late manifestations in terms of diabetes duration. As mentioned previously, these defects in glucose counterregulation are not ‘all or nothing’ changes but can be influenced by the prevailing standard of glycaemic control and by the frequency of hypoglycaemic episodes. Various theories relate to the clinical observation that blood glucose thresholds for the release of counterregulatory hormone levels can change after periods of recurrent hypoglycaemia (Cryer, 2005). These may relate to changes at the level of the CNS, which co-ordinates the usual responses to low blood glucose levels. At present there is little evidence to suggest that the alterations associated with recurrent hypoglycaemia occur at glucose sensors outside the CNS, for example, within the portal vein.
(a)
(b)
type 1 diabetes 1–5 years
type 1 diabetes 14–31 years Insulin
5
Glucagon (pg / ml) Epinephrine (pg / ml) Glucose (mmol / I)
Glucagon (pg / ml) Epinephrine (pg / ml) Glucose (mmol / I)
Insulin
4 3 2 300 200 100 0 200
150
100
0
30
60
90 120 150
Time (min)
5 4 3 2 300 200 100 0 200
150
100
0
30
60
90 120 150
Time (min)
Figure 6.10 Influence of duration of diabetes on glucagon and epinephrine responses to hypoglycaemia in patients with type 1 diabetes (•) after (a) 1–5 years (glucagon response is blunted whereas epinephrine release is preserved); and (b) with long-standing diabetes, both responses become severely impaired. = non-diabetic controls. Reproduced from Textbook of Diabetes, 2nd edition (1997) Pickup J. and William G. (eds) by permission of Blackwell Science Ltd. Data sourced from Bolli et al. (1983). Copyright © 1983 American Diabetes Association. Reprinted with permission from The American Diabetes Association
132
COUNTERREGULATORY DEFICIENCIES IN DIABETES
Systemic Mediator The systemic mediator theory suggests that a substance is released in response to hypoglycaemia which attenuates subsequent sympathoadrenal responses to further episodes of hypoglycaemia. The initial candidate for this was cortisol, based on two observations: first, the attenuating effect of antecedent hypoglycaemia on later sympathoadrenal responses is absent in patients with primary adrenocortical failure; and second, in healthy volunteers, following infusions of cortisol (to supraphysiological levels) during euglycaemia, adrenomedullary epinephrine secretion and muscle sympathetic neural activity were reduced during subsequent hypoglycaemia (Davis et al., 1996; 1997). However, this effect of cortisol is lost if the prevailing cortisol levels are lowered towards those seen during hypoglycaemia or if recurrent hypoglycaemia is induced in animals that are genetically modified to have absent adrenocortical responses (Raju et al., 2003; McGuiness et al., 2005).
Brain Fuel Transport This mechanism is based on the hypothesis that following antecedent hypoglycaemia, glucose transport from blood into brain tissue is increased – in animals by increasing GLUT-1 transport across the brain microvasculature (McCall et al., 1986; Kumagai et al., 1995). In patients with type 1 diabetes whose treatment resulted in near normal glucose levels, impaired awareness of hypoglycaemia can develop – such patients are at increased risk of seizures and coma. Boyle et al. (1995) tested the hypothesis that during hypoglycaemia, these patients would have normal glucose uptake in the brain and consequently that sympathoadrenal activation would not occur, resulting in impaired awareness of hypoglycaemia. They found that there was no significant change in the uptake of glucose in the brain among the patients with type 1 diabetes who had the lowest HbA1c levels. Conversely, glucose uptake in the brain fell in patients with less wellcontrolled type 1 diabetes. The responses of plasma epinephrine and pancreatic polypeptide and the frequency of symptoms of hypoglycaemia were also lowest in the group with the lowest HbA1c values. They concluded that during hypoglycaemia, patients with nearly normal HbA1c values have normal glucose uptake in the brain, preserving cerebral metabolism, reducing the responses of counterregulatory hormones, and causing impaired awareness of hypoglycaemia (Boyle et al., 1995). However, these findings occurred after days of prolonged hypoglycaemia which is in contrast to the clinical observation that attenuated sympathoadrenal responses occur within hours of a hypoglycaemic event. More recently, studies using positron emission tomography have found no change in bloodto-brain glucose transport 24 hours after an episode of hypoglycaemia and no differences between individuals with and without hypoglycaemia awareness (Segel et al., 2001; Bingham et al., 2005). It remains possible, however, that there are changes in the transport of alternative cerebral fuels following antecedent hypoglycaemia.
Brain Metabolism It has been hypothesised that brain metabolism per se is altered following an episode of hypoglycaemia. Most research in this area has focused on the ventromedial nucleus of the hypothalamus. Glucose deprivation in the VMH (by administration of 2-deoxyglucose) activates the sympathoadrenal system and increases glucagon secretion whereas local perfusion
AGE, OBESITY AND GLUCOSE COUNTERREGULATION
133
Glucose
Utilisation Delivery
Adenosine
Caffeine Figure 6.11 Caffeine may act by uncoupling brain glucose demand (increased) and substrate delivery (decreased) through its actions on adenosine receptors. Reproduced from Brain Research Reviews, 17, Nehlig et al., 139–169, Copyright (1992), with permission from Elsevier
of the area with glucose suppresses these responses during systemic hypoglycaemia (Borg et al., 1995; Borg et al., 1997). The mechanisms involved are unknown but may be a consequence of increased glucokinase activity to enhance glucose metabolism in neurones in the region (Gabriely and Shamoon, 2005). However, it is likely that brain metabolism in areas and other signalling mechanisms within the CNS are influenced by recurrent antecedent hypoglycaemia (Cryer, 2005). The brain glycogen supercompensation hypothesis suggests that after a single episode of hypoglycaemia, there is a rebound increase in glycogen formation in brain astrocytes to provide additional substrates (e.g. lactate) for brain metabolism (Choi et al., 2003). Alterations of substrate delivery to the brain do appear to influence the magnitude of the hormonal counterregulatory response to hypoglycaemia in healthy volunteers and in patients with type 1 diabetes. Infusions of acetazolamide, a potent cerebral vasodilator, markedly attenuates these responses (Thomas et al., 1997) whereas ingestion of modest amounts of caffeine (to reduce substrate delivery) augments the responses (Debrah et al., 1996). The mechanisms of the latter is unknown but may involve antagonism of central adenosine receptors with uncoupling of brain blood flow (i.e., substrate delivery) and brain glucose metabolism (i.e., brain glucose demand) resulting in relative cerebral neuroglycopenia (Figure 6.11). This is discussed in more detail in Chapter 5.
AGE, OBESITY AND GLUCOSE COUNTERREGULATION In children with type 1 diabetes, the glucagon response to hypoglycaemia is markedly attenuated compared to non-diabetic individuals but compensated for by vigorous secretion of other counterregulatory hormones, particularly epinephrine, with the peak epinephrine responses being almost two-fold higher than in adults (Amiel et al., 1987). The total sympathoadrenal responses to hypoglycaemia are also influenced by pubertal stage (Ross et al., 2005).
134
COUNTERREGULATORY DEFICIENCIES IN DIABETES
Furthermore, it appears that the glycaemic thresholds for the secretion of epinephrine and growth hormone are set at a higher blood glucose level in non-diabetic children compared to adults (Jones et al., 1991). In children with type 1 diabetes, the secretion of epinephrine in response to hypoglycaemia commences at an even higher level. In children who have markedly elevated HbA1c values, there is a further shift of the blood glucose threshold to a higher level for the release of counterregulatory hormones. Advanced age, in otherwise healthy people, does not appear to diminish or delay counterregulatory responses to hypoglycaemia (Brierley et al., 1995), although the magnitude of responses of epinephrine and glucagon is lower at milder hypoglycaemic levels (around 3.4 mmol/l ) compared to younger non-diabetic subjects, but is much more comparable with a more profound hypoglycaemic stimulus (2.8 mmol/l) (Ortiz-Alonso et al., 1994) (See Chapter 11). The magnitude of counterregulatory responses to low blood glucose levels following preceding hypoglycaemia also appears to depend on the gender of experimental subjects, with men having blunted responses compared to women (Davis et al., 2000b). Both the autonomic nervous system and the hypothalamic–pituitary–adrenal axis are activated in excess in the morbidly obese. Before and after bariatric surgery (average weight loss 40 kg over 12 months), severely obese non-diabetic subjects, underwent a hyperinsulinaemic hypoglycaemic clamp (blood glucose 3.4 mmol/l). Before weight reduction, patients demonstrated brisk peak responses in glucagon, epinephrine, pancreatic polypeptide, and norepinephrine. After surgery and during hypoglycaemia, all these responses were attenuated and most markedly so for glucagon, which was totally abolished in association with a marked improvement in insulin sensitivity. In contrast, the growth hormone response was increased after weight reduction (Guldstrand et al., 2003).
HUMAN INSULIN AND COUNTERREGULATION At present there is no consistent evidence that the species of insulin is an important determinant of the counterregulatory response to hypoglycaemia. Over 25 clinical laboratory studies have examined the effect of insulin species on the counterregulatory response to hypoglycaemia induced by an intravenous bolus, intravenous infusion, or subcutaneous injection of insulin (Fisher and Frier, 1993; Jorgensson et al., 1994). Most of the studies showed no significant differences between the hormonal responses. Two studies showed a reduction in the epinephrine response to hypoglycaemia, and both of these studies also reported diminished autonomic symptoms to hypoglycaemia after human insulin (Schluter et al., 1982; Heine et al., 1989). A meta-analysis comparison of the effects of human and animal insulin as well as of the adverse reaction profiles did not show clinically relevant differences between species especially in terms of risk and responses to hypoglycaemia (Richter and Neises, 2005).
TREATMENT OF COUNTERREGULATORY FAILURE At present no treatment is available that will reverse the glucagon deficit that develops within a few years of the onset of type 1 diabetes (Figure 6.12). However, there are strategies
TREATMENT OF COUNTERREGULATORY FAILURE
135
Glucose
Glucagon
Glucagon
Time
Glucose
Epinephrine
Epinephrine
Time
Glucose
Glucagon and Epinephrine
Glucagon and Epinephrine
Time
Figure 6.12 Schematic representation of the consequences of defective glucagon, epinephrine or a combined defect of glucagon and epinephrine release during recovery from hypoglycaemia
to reduce the risk of hypoglycaemia from causing further hypoglycaemia and promoting Hypoglycaemia Associated Autonomic Failure: • relaxation of glycaemic targets; • use of multiple daily injections of insulin, utilising rapid-acting and basal analogues (Bolli, 2006); • consideration of switching to continuous subcutaneous insulin infusion therapy (Chase et al., 2006);
136
COUNTERREGULATORY DEFICIENCIES IN DIABETES
• intensive education including carbohydrate counting and appropriate blood glucose monitoring; • use of novel technologies to aid diagnosis, e.g. continuous glucose monitoring (Cheyne and Kerr, 2002); • discussion of patient factors, e.g. lipohypertrophy and other injection site problems, alcohol, caffeine consumption. Common psychological problems known to affect diabetes management adversely, such as anxiety, depression and eating disorders, have been extensively reported (Jacqueminet et al., 2005). Of particular relevance is the recognition of the role played by high levels of anxiety. Evidence-based treatment interventions are available for treating anxiety in the non-diabetic population; however a systematic review and meta-analysis of randomised controlled trials of psychological interventions for adults with diabetes has yet to be conducted. In clinical practice it continues to be recognised that there is a group of patients whose lives are completely disrupted by recurrent episodes of hyper and hypoglycaemia – the so-called ‘brittle diabetic’. The outlook for such patients is usually poor (Tattersall et al., 1991).
CONCLUSIONS • In non-diabetic individuals, clinically significant hypoglycaemia is an extremely rare event because of effective glucose counterregulation. This includes suppression of endogenous pancreatic insulin secretion, and release of glucagon, catecholamines, cortisol and growth hormone. • The brain is the critical organ for co-ordination of the physiological responses to low blood glucose levels. • People with diabetes almost inevitably lose their ability to release glucagon in response to a fall in blood glucose, within five years of diagnosis. After ten years, a significant proportion of patients also has deficient epinephrine responses and is at increased risk of more protracted hypoglycaemia and neuroglycopenia. • Modern treatment of type 1 diabetes, including intensive education and treatment regimens utilising new technologies, has failed to eradicate completely the problem of recurrent hypoglycaemia.
REFERENCES Amiel SA, Simonson DC, Sherwin RS, Lauriano AA, Tamborlane WV (1987). Exaggerated epinephrine responses to hypoglycemia in normal and insulin-dependent diabetic children. Journal of Pediatrics 110: 832–7. Amiel S (1991). Glucose counter-regulation in health and disease: current concepts in hypoglycaemia recognition and response. Quarterly Journal of Medicine 293: 707–27. Banarer S, Cryer PE (2003). Sleep-related hypoglycemia-associated autonomic failure in type 1 diabetes: reduced awakening from sleep during hypoglycemia. Diabetes 52: 1195–203.
REFERENCES
137
Bingham EM, Dunn JT, Smith D, Sutcliffe-Goulden J, Reed LJ, Mansden PK, Amiel SA (2005). Differential changes in brain glucose metabolism during hypoglycaemia accompany loss of hypoglycaemia awareness in men with type 1 diabetes mellitus. An [11C]-3-O-methyl-D-glucose PET study. Diabetologia 48: 2080–9. Bolli G, De Feo P, Compagnucci P, Cartechini MG, Angeletti G, Santeusanio F et al. (1983). Abnormal glucose counterregulation in insulin-dependent diabetes mellitus. Interaction of antiinsulin antibodies and impaired glucagon and epinephrine secretion. Diabetes 32: 134–41. Bolli GM (2006). Insulin treatment in type 1 diabetes. Endocrine Practice 12 Suppl 1: 105–9. Borg WP, During MJ, Sherwin RS, Borg MA, Brines ML, Shulman GI (1994). Ventromedial hypothalamic lesions in rats suppress counter regulatory responses to hypoglycemia. Journal of Clinical Investigation 93: 1677–82. Borg WP, Sherwin RS, During MJ, Borg MA, Shulman GI (1995). Local ventromedial hypothalamus glucopenia triggers counterregulatory hormone release. Diabetes 44: 180–4. Borg MA, Sherwin RS, Borg WP, Tamborlane WV, Shulman GI (1997). Local ventromedial hypothalamus glucose perfusion blocks counterregulation during systemic hypoglycemia in awake rats. Journal of Clinical Investigation 99: 361–5. Boyle PJ, Kempers SF, O’Connor AM, Nagy RJ (1995). Brain glucose uptake and unawareness of hypoglycemia in patients with insulin dependent diabetes mellitus. New England Journal of Medicine 333: 1726–31. Brierley EJ, Broughton DL, James OFW, Alberti KGMM (1995). Reduced awareness of hypoglycaemia in the elderly despite an intact counter-regulatory response. Quarterly Journal of Medicine 88: 439–45. Bulsara M, Holman CD, Davies EA, Jones TW (2004). The impact of a decade of changing treatment on rates of severe hypoglycemia in a population-based cohort of children with type 1 diabetes. Diabetes Care 27: 2293–8. Chase HP, Horner BP, McFann K, Yetzer H, Gaston J, Banion C et al. (2006). The use of insulin pumps with meal bolus alarms in children with type 1 diabetes to improve glycemic control. Diabetes Care 29: 1012–5. Cheyne E, Kerr D (2002) Making ‘sense’ of diabetes: using a continuous glucose sensor in clinical practice. Diabetes Metabolism Research Reviews 18 Suppl 1: S43–8. Choi I Y, Seaquist ER, Gruetter R (2003). Effect of hypoglycemia on brain glycogen metabolism in vivo. Journal of Neuroscience Research 72: 25–32. Cox DJ, Irving GA, Gonder-Frederick L, Nowacek G, Butterfield J (1987). Fear of hypoglycemia: quantification, validation and utilization. Diabetes Care 10: 617–21. Cranston I, Lomas J, Maran A, Macdonald IA, Amiel SA (1994). Restoration of hypoglycaemia awareness in patients with long-duration insulin-dependent diabetes. Lancet 344: 283–7. Cryer PE (2004). Diverse causes of hypoglycemia-associated autonomic failure in diabetes. New England Journal of Medicine 350: 2272–9. Cryer PE (2005). Mechanisms of hypoglycemia-associated autonomic failure and its component syndromes in diabetes. Diabetes 54: 3592–601. DAFNE Study Group (2002). Training in flexible, intensive insulin management to enable dietary freedom in people with type 1 diabetes: dose adjustment for normal eating (DAFNE) randomised controlled trial. British Medical Journal 325: 746. Davis SN, Shavers C, Costa F, Mosqueda-Garcia R (1996). Role of cortisol in the pathogenesis of deficient counterregulation after antecedent hypoglycemia in normal humans. Journal of Clinical Investigation 98: 680–91. Davis SN, Shavers C, Davis B, Costa F (1997). Prevention of an increase in plasma cortisol during hypoglycemia preserves subsequent counterregulatory responses. Journal of Clinical Investigation 100: 429–38. Davis SN, Mann S, Galassetti P, Neil R, Tate D, Ertl AC, Costa F (2000a). Effects of differing durations of antecedent hypoglycemia on counterregulatory responses to subsequent hypoglycemia in normal humans. Diabetes 49: 1897–903.
138
COUNTERREGULATORY DEFICIENCIES IN DIABETES
Davis SN, Fowler F, Costa S (2000b). Hypoglycemic counterregulatory responses differ between men and women with type 1 diabetes. Diabetes 49: 65–72. Debrah K, Sherwin RS, Murphy J, Kerr D (1996). Effect of caffeine on recognition of and physiological responses to hypoglycaemia in insulin-dependent diabetes. Lancet 347: 19–24 Diem P, Redmon JB, Abid M, Moran A, Sutherland DE, Halter JB, Robertson RP (1990). Glucagon, catecholamine and pancreatic polypeptide secretion in type I diabetic recipients of pancreas allografts. Journal of Clinical Investigation 86: 2008–13. Fisher M, Frier BM (1993). Hypoglycaemia and human insulin. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M, eds. Edward Arnold, London: 314–27. Fisher S, Brunning J, Lannon S, Kahn CR (2005). Insulin signaling in the central nervous system is critical for the normal sympathoadrenal response to hypoglycemia. Diabetes 54: 1447–51. Flanagan D, Kerr D (1996). Recurrent hypoglycemia and long-standing brittle diabetes. Endocrinology and Metabolism 3: 67–69. Frier BM, Fisher M, Gray CE, Beastall GH (1988). Counterregulatory hormonal responses to hypoglycaemia in type 1 (insulin-dependent) diabetes: evidence for diminished hypothalamic-pituitary hormonal secretion. Diabetologia 31: 421–9. Gabriely I, Shamoon H (2005). Fructose normalizes specific counterregulatory responses to hypoglycemia in patients with type 1 diabetes. Diabetes 54: 609–16. Gerich JE, Langlois M, Noacco C, Karam JH, Forsham PH (1973). Lack of glucagon response to hypoglycemia in diabetes: evidence for an intrinsic pancreatic alpha cell defect. Science 182: 171–3. Gerich JE, Bolli GB (1993). Counterregulatory failure. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M, eds. Edward Arnold, London: 253–67. Guldstrand M, Ahren B, Wredling R, Backman L, Lins P, Adamson U (2003). Alteration of the counterregulatory responses to insulin-induced hypoglycemia and of cognitive function after massive weight reduction in severely obese subjects. Metabolism 52: 900–7. Hardy KJ, Burge MR, Boyle P, Scarpello JHB (1994). A treatable cause of recurrent severe hypoglycemia. Diabetes Care 17: 722–4. Heine RJ, Van der Heyden EAP, Van der Veen EA (1989). Responses to human and porcine insulin in healthy subjects. Lancet 334: 946–9. Heller SR, Cryer PE (1991). Reduced neuroendocrine and symptomatic responses to subsequent hypoglycemia after 1 episode of hypoglycemia in non-diabetic humans. Diabetes 40: 223–6. Hvidberg A, Juel Christensen N, Hilsted J (1998). Counterregulatory hormones in insulin-treated diabetic patients admitted to an Accident and Emergency Department with hypoglycaemia. Diabetic Medicine 15: 199–204. Jacqueminet S, Masseboeuff N, Rolland M, Grimaldi A, Sachon C (2005). Limitations of the so-called ‘intensified’ insulin therapy in type 1 diabetes mellitus. Diabetes Metabolism 31(4 Pt 2): 4S45–50. Jones TW, Boulware SD, Kraemer DT, Caprio S, Sherwin RS, Tamborlane WV (1991). Independent effects of youth and poor diabetes control on responses to hypoglycemia in children. Diabetes 40: 358–63. Jones TW, Porter P, Sherwin RS, Davis EA, O’Leary P, Frazer F et al. (1998). Decreased epinephrine responses to hypoglycemia during sleep. New England Journal of Medicine 338: 1657–62. Jorgensen L, Dejgaard A, Pramming SK (1994). Human insulin and hypoglycaemia: a literature survey. Diabetic Medicine 11: 925–34. Kerr D, Reza M, Smith N, Leatherdale BA (1991). Importance of insulin in subjective, cognitive and hormonal responses to hypoglycemia in patients with IDDM. Diabetes 40: 1057–62. Kerr D, Tamborlane WV, Rife F, Sherwin RS (1993). Effect of insulin-like growth factor-1 on the responses to and recognition of hypoglycemia in humans. Journal of Clinical Investigation 91: 141–7. Kumagai AK, Kang YS, Boado RJ, Pardridge WM (1995). Upregulation of blood-brain barrier GLUT1 glucose transporter protein and mRNA in experimental chronic hypoglycemia. Diabetes 44: 1399–404.
REFERENCES
139
McCall AL, Fixman LB, Fleming N, Tornheim K, Chick W, Ruderman NB (1986). Chronic hypoglycemia increases brain glucose transport. American Journal of Endocrinology 251: E442–7. McCrimmon RJ, Evans ML, Fan X, McNay EC, Chan O, Ding Y et al. (2005). Activation of ATP-sensitive K+ channels in the ventromedial hypothalamus amplifies counterregulatory hormone responses to hypoglycemia in normal and recurrently hypoglycemic rats. Diabetes 54: 3169–74. McGuiness O, Ansari T, Potts J, Jacobson L (2005). Glucocorticoid insufficiency in corticotrophin releasing hormone knockout mice does not preserve autonomic responses during repeated hypoglycemia clamps. Diabetes 54 (Suppl): A26 (abstract). Merl V, Kern W, Peters A, Oltmanns KM, Gais S, Born J et al. (2004). Differences between nighttime and daytime hypoglycemia counterregulation in healthy humans. Metabolism 53: 894–8. Mitrakou A, Ryan C, Veneman T, Mokan M, Jenssen T, Kiss I et al. (1991). Hierarchy of glycemic thresholds for counterregulatory hormonal secretion, symptoms, and cerebral dysfunction. American Journal of Endocrinology 260: E67–74. Muhlhauser I, Bruckner I, Berger M, Cheta D, Jorgens V, Lonescu-Tirgoviste C et al. (1987). Evaluation of an intensified insulin treatment and teaching programme as routine management of type 1 (insulin-dependent) diabetes. The Bucharest–Dusseldorf Study. Diabetologia 30: 681–90. Nehlig A, Daval JL, Debry G (1992). Caffeine and the central nervous system: mechanisms of action, biochemical, metabolic and psychostimulant effects. Brain Research Reviews 17: 139–69. Ortiz-Alonso FJ, Galecki A, Herman WH, Smith MJ, Jacquez JA, Halter JB (1994). Hypoglycemia counterregulation in elderly humans: relationship to glucose levels. American Journal of Endocrinology 267: E497–506. Raju B, McGregor VP, Cryer PE (2003). Cortisol elevations comparable to those that occur during hypoglycemia do not cause hypoglycemia-associated autonomic failure. Diabetes 52: 2083–89. Richter B, Neises G (2005). ‘Human’ insulin versus animal insulin in people with diabetes mellitus. Cochrane Database Systematic Reviews Jan 25(1): CD003816. Ross LA, Warren RE, Kelnar CJ, Frier BM. (2005). Pubertal stage and hypoglycaemia counterregulation in type 1 diabetes. Archives of Disease in Childhood 90: 190–4. Rossing P. (2005) The changing epidemiology of diabetic microangiopathy in type 1 diabetes. Diabetologia 2005; 48: 1439–44. Samols E, Tyler J, Marks V (1972): Glucagon-insulin interrelationships. In: Glucagon: Molecular Physiology, Clinical and Therapeutic Implications. Lefebvre P, Unger RH, eds. Pergamon, Elmsford NY: 151–174. Schluter KJ, Petersen KG, Sontheimer K, Enzmann F, Kerp L (1982). Different counterregulatory responses to human insulin (recombinant DNA) and purified pork insulin. Diabetes Care 5 (suppl 2): 78–81. Segel SA, Fanelli CG, Dence CS, Markham J, Videen TO, Paramore DS et al. (2001). Blood-to-brain glucose transport, cerebral glucose metabolism and cerebral blood flow are not increased following hypoglycemia. Diabetes 50: 1911–17. Smith D, Pernet D, Reid H, Bingham E, Rosenthal JM, Macdonald IA et al. (2002). The role of hepatic portal glucose sensing in modulating responses to hypoglycaemia in man. Diabetologia 45: 1416–24. Tappy L, Chiorelo R, Berger M (1999). Autoregulation of glucose production in health and disease. Current Opinions in Clinical Nutrition and Metabolic Care 2: 161–4. Tattersall RB, Gregory R, Selby C, Kerr D, Heller S (1991). Course of brittle diabetes: 12 year follow up. British Medical Journal 302: 1240–3. The Diabetes Control and Complications Trial Research Group (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New England Journal of Medicine 329: 977–86. Thomas M, Sherwin R, Murphy J, Kerr D (1997). Importance of cerebral blood flow on recognition of and physiological responses to hypoglycemia. Diabetes 46: 829–33.
140
COUNTERREGULATORY DEFICIENCIES IN DIABETES
Weintrob N, Schechter A, Benzaquen H, Shalitin S, Lilos P, Galatzer A, Phillip M (2004). Glycemic patterns detected by continuous subcutaneous glucose sensing in children and adolescents with type 1 diabetes mellitus treated by multiple daily injections versus continuous subcutaneous insulin infusion. Archives of Pediatric and Adolescent Medicine 158: 677–84. White NH, Gingerich RL, Levandoski LA, Cryer PE, Santiago JV (1985). Plasma pancreatic polypeptide response to insulin-induced hypoglycemia as a marker for defective glucose counterregulation in insulin-dependent diabetes. Diabetes 34: 870–5.
7 Impaired Awareness of Hypoglycaemia Brian M. Frier Dangerous hypoglycaemia may occur without warning symptoms – E.P. Joslin et al. (1922)
INTRODUCTION The generation of symptoms in response to hypoglycaemia provides a fundamental defence for the brain, by alerting the affected individual to the imminent development of neuroglycopenia (Chapter 2). This should provoke an appropriate response – obtaining and ingesting some form of carbohydrate to reverse the low blood glucose. If these warning symptoms fail to occur, or they are delayed until the blood glucose has fallen to a level that causes disabling neuroglycopenia, serious consequences may ensue. When the normal warning mechanisms are deficient or are ignored and no avoiding action is taken, severe hypoglycaemia may occur, with progression to confusion, altered consciousness and eventual coma. An inadequate symptomatic warning often occurs in people with insulin-treated diabetes, in various circumstances and with differing causes, and is described as impaired awareness of hypoglycaemia or hypoglycaemia unawareness. This is an acquired abnormality that is effectively a complication of insulin therapy, and should be ranked alongside the microvascular complications of diabetes such as retinopathy, neuropathy or nephropathy, because its morbidity can be just as serious and disabling.
NORMAL RESPONSES TO HYPOGLYCAEMIA Acute hypoglycaemia induces a series of changes – hormonal, neurophysiological, symptomatic and cognitive – which occur at different and defined blood glucose concentrations (Figure 7.1). The thresholds at which these changes are triggered have been described in non-diabetic humans, most occurring within a relatively narrow range of blood glucose concentrations. In diabetic individuals these glycaemic thresholds are not static and permanent, but are dynamic and display plasticity, altering in response to external influences such as changes in glycaemic control and exposure to extremes of blood glucose. Thus the blood glucose level at which symptoms are activated can be modified through the ability of the brain to adapt to environmental change, that is, its exposure to prevailing blood glucose concentrations.
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
142
Arterialised venous blood glucose concentration (mmol/L)
5.0
4.6 mmol/L Inhibition of endogenous insulin secretion 3.8 mmol/L
4.0 Counterregulatory hormone release • Glucagon • Epinephrine 3.0
3.2–2.8 mmol/L Onset of symptoms • Autonomic • Neuroglycopenic
3.0–2.4 mmol/L Neurophysiological dysfunction • Evoked responses
2.0
3.0 mmol/L Widespread EEG changes
2.8 mmol/L Cognitive dysfunction • Inability to perform complex tasks < 1.5 mmol/L
1.0
Severe neuroglycopenia • Reduced conscious level • Convulsions • Coma
0
Figure 7.1 Hierarchy of endocrine, symptomatic and neurological responses to acute hypoglycaemia in non-diabetic subjects. Glycaemic thresholds are based on glucose concentrations in arterialised venous blood. Modified from Textbook of Diabetes, 2nd edition (1997) (eds J. Pickup and G. Williams), by permission of Blackwell Science Ltd
Depriving the brain of glucose causes it to malfunction, and cognitive impairment quickly becomes evident as an overt manifestation of neuroglycopenia. Some of these features are relatively subtle, and may not be detected immediately by the patient. A fall in blood glucose triggers activation of the peripheral autonomic nervous system via central hypothalamic autonomic centres within the brain, and stimulates the sympathoadrenal system. This promotes typical physiological responses including sweating, an increase in rate and contractility of the heart (sensed as a pounding heart), and tremor, these being some of the classical features of the autonomic reaction (Figure 7.2). Epinephrine (adrenaline) is secreted in large quantities from the adrenal medullae and contributes to some of the symptoms mainly by heightening the magnitude of the response. The early literature on hypoglycaemia and diabetes provides accurate descriptions of the autonomic features of acute hypoglycaemia, and patients and physicians alike commonly discussed hypoglycaemic ‘reactions’, a term that regrettably is now seldom used. It emphatically describes the sudden, and often florid, onset of the autonomic features of hypoglycaemia, which drive the individual to seek assistance or obtain a supply of glucose to relieve these unpleasant symptoms.
‘Awareness’ of Hypoglycaemia The generation of typical physiological responses to hypoglycaemia is perceived through sensory feedback to the brain, and after central processing, an appropriate motor response is made. Much has been made by some commentators of the predominant importance of autonomic symptoms in the detection of the onset of hypoglycaemia. This premise is
NORMAL RESPONSES TO HYPOGLYCAEMIA
143
We do not have rights to reproduce this figure electronically
Figure 7.2 Generation of neuroglycopenic and autonomic symptoms in response to hypoglycaemia. Autonomic activation and the involvement of the sympatho-adrenal system in the stimulation of representative end-organs associated with common autonomic symptoms of hypoglycaemia. Reproduced from Hypoglycaemia and Diabetes (eds B.M. Frier and M. Fisher), © 1993 Edward Arnold, by permission of Edward Arnold (Publishers) Ltd
based partly on the laboratory-based observation of non-diabetic subjects that autonomic symptoms commence at a higher blood glucose concentration (around 0.5 mmol/l) than neuroglycopenic symptoms (Mitrakou et al., 1991). In everyday experience reported by people with insulin-treated diabetes, a rapid decline in blood glucose does not permit a subjective distinction to be made between these different thresholds for the development of autonomic and neuroglycopenic symptoms, and people treated with insulin identify both types with equal frequency as their initial warning symptoms (Hepburn et al., 1992). It has been assumed that because neuroglycopenia may interfere with cognitive function, this will affect the individual’s ability to perceive and interpret neuroglycopenic cues such as the inability to concentrate, drowsiness or difficulty with mentation. This may be true when a falling blood glucose is not treated and is allowed to drop to a level associated with severe
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
144
neuroglycopenia, but most patients detect (and often rely upon) neuroglycopenic symptoms during early hypoglycaemia, and rate these as important as autonomic symptoms in providing a warning. It is the initial perception of any symptom of hypoglycaemia, irrespective of whether this is autonomic, neuroglycopenic or simply a vague sensation of apprehension or loss of well-being (a common early feature described by many) which constitutes ‘awareness’ of hypoglycaemia. Only the initial warning symptoms are important in this respect, and not the total spectrum or absolute number of symptoms, some of which occur too late to have any value in alerting the patient to the impending risk of a falling blood glucose. A major difference between the autonomic and the neuroglycopenic symptomatic response is that, once triggered, the autonomic response quickly reaches a maximum intensity which then gradually declines with time, whereas the neuroglycopenic response becomes more profound the further the blood glucose falls. This qualitative difference in response becomes important if early cues are ignored or are not detected, as progressive neuroglycopenia will eventually interfere with the individual’s ability to identify and self-treat the low blood glucose. When a person is fully awake, alert and on guard against possible hypoglycaemia, this symptomatic warning system generally works very effectively (Chapter 2). However, there are many times in everyday life when the symptoms may be either diminished or disregarded. This is particularly so during sleep when symptoms are seldom detected, or they may be ignored if a person is distracted by other activities, such as watching an interesting programme on television, participating in sport or concentrating on a task. Circumstances can modify the value of specific warning symptoms, making them difficult to interpret as features of hypoglycaemia. Examples include sweating on a hot day, shivering when the weather is cold or feeling drowsy during a boring meeting! All of these may represent early hypoglycaemia but are attributed to other causes by the affected person. A list of the factors that influence normal awareness of hypoglycaemia is shown in Box 7.1. The intensity of symptoms can vary and the value of individual symptoms as warning features may not be constant in any single individual. This is often not appreciated in the assessment of research findings, and it is difficult to extrapolate the careful measurement of symptomatic responses to hypoglycaemia in studies performed in a laboratory setting, to the hurly-burly of everyday life.
Box 7.1
Factors influencing normal awareness of hypoglycaemia
Internal Physiological Recent glycaemic control Degree of neuroglycopenia Symptom intensity/sensitivity
External Drugs Beta-adrenoceptor blockers (non-selective) Hypnotics, tranquillisers Alcohol
Psychological Focused attention Congruence; denial Competing explanations
Environmental Posture Distraction
Education Knowledge Symptom belief
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
145
Warning symptoms provide internal cues, but most people with insulin-treated diabetes also rely on external cues based on their experience of the timing of insulin administration in relation to food, the effect of delaying meals or the amount of food ingested, the effect of exercise on blood glucose and many other factors that can influence short-term glycaemic control. These cues are supplemented by blood glucose monitoring, which gives an exact and objective measure of prevailing glycaemia. Further useful feedback may be obtained from observers such as relatives or friends, many of whom become adept at noticing early neuroglycopenia before the onset of the patient’s subjective warning symptoms. ‘Awareness’ of hypoglycaemia is therefore distilled from a combination of resources, and has to be learned by people with newly diagnosed diabetes commencing insulin therapy. They have no previous experience of symptoms of hypoglycaemia and must receive appropriate education on the potential range of symptoms. Through experience they will recognise the cluster of symptoms peculiar to themselves, because symptoms are idiosyncratic. Awareness of hypoglycaemia therefore assists in protecting the individual from the risk of an unexpected fall in blood glucose. When awareness of hypoglycaemia becomes impaired or is absent while a person is awake, the individual becomes progressively vulnerable to the development of severe hypoglycaemia.
IMPAIRED AWARENESS OF HYPOGLYCAEMIA Definition No satisfactory or comprehensive definition of impaired hypoglycaemia awareness has been suggested to date. Many laboratory-based studies of experimental hypoglycaemia have used arbitrary definitions based on witnessed observations of subjects who fail to develop classical features of hypoglycaemia, or the failure of physiological or hormonal responses to exceed twice the standard deviation from mean basal levels. These are statistical devices, which take no account of subjective reality, require the application of sophisticated and unphysiological glucose clamp procedures, and have little direct application to clinical management. Asymptomatic biochemical hypoglycaemia occurs more frequently during routine blood glucose monitoring in diabetic patients who report impaired awareness of hypoglycaemia (Gold et al., 1994; Clarke et al., 1995) and such a record may alert the clinician to the possibility that an individual is developing this problem. A much higher rate of undetected hypoglycaemia in people with impaired awareness has been demonstrated during waking hours using continuous blood glucose monitoring (Kubiak et al., 2004). However, in clinical practice a careful history is essential in determining whether reduced warning symptoms of hypoglycaemia are a significant problem, and if this is occurring consistently. Patients who assert that they have a problem with perceiving the onset of symptoms of hypoglycaemia are generally correct in this belief (Clarke et al., 1995), so that the identification of impaired awareness of hypoglycaemia should be based principally on clinical history. Validated scoring systems to assess awareness of hypoglycaemia have been described by Gold et al. (1994) and Clarke et al. (1995), and supportive information can be derived from simultaneous inspection of the individual’s blood glucose results. Detailed questioning of a patient about his or her ability to detect the onset of hypoglycaemic symptoms may need to be supplemented by questioning close relatives, who often report a much higher rate of severe hypoglycaemia
146
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
(Heller et al., 1995; Jorgensen et al., 2003). This will provide a witnessed description of how hypoglycaemia develops in a patient, with information on its true frequency and severity. Patients often underestimate the frequency of severe hypoglycaemia, partly because of post-hypoglycaemia amnesia.
Classification In one study, Hepburn et al. (1990) subdivided hypoglycaemia awareness into three categories: normal, partial and absent awareness. These were defined as follows: • Normal awareness: the individual is always aware of the onset of hypoglycaemia. • Partial awareness: the symptom profile has changed with a reduction either in the intensity or in the number of symptoms and, in addition, the individual may be aware of some episodes of some episodes of hypoglycaemia but not of others. • Absent awareness: the individual is no longer aware of any episode of hypoglycaemia. Although the subdivision into partial and absent awareness is artificial, it reflects the natural history of this clinical problem, illustrating the gradual progression of this disability, and emphasising that in some patients the abnormality is severe (absent awareness) although total absence of clinical manifestations of hypoglycaemia (particularly the neuroglycopenic features) is exceptionally rare (Gold et al., 1994, Clarke et al., 1995). The problem may not be simply an absence of symptoms, but rather that the time during which warning symptoms can be detected is extremely short, allowing the affected individual a very limited opportunity to take avoiding action. Some patients describe how the onset of hypoglycaemia appears to have become much more rapid compared with their previous experience and progresses quickly to severe neuroglycopenia. However, impaired awareness may not necessarily evolve into total unawareness of hypoglycaemia, and may vary over time, presumably because of major influences of environmental factors on the generation and perception of symptoms. The above classification of awareness of hypoglycaemia is far from comprehensive. In addition, the state of hypoglycaemia awareness can be ascertained only when the individual is in a physical state in which recognition of the onset of hypoglycaemia is possible. Therefore, if the person is asleep, intoxicated, inebriated, anaesthetised or sedated, so that their conscious level is reduced, they are not able to perceive (as subjective symptoms) the normal physiological manifestations of hypoglycaemia. An individual’s awareness of hypoglycaemia can be evaluated only if hypoglycaemia occurs while the individual is awake. A further prerequisite is that the person must have had previous experience of hypoglycaemia at some time during treatment with insulin. In assessing the present state of hypoglycaemia awareness, it is desirable that the patient should have experienced one or more episodes of hypoglycaemia (confirmed biochemically) within a recent time interval such as the preceding year, so that a comparison of the symptoms can be made with earlier episodes of hypoglycaemia. A diagnosis of impaired hypoglycaemia awareness cannot be entertained or surmised if a patient has either never been exposed previously to acute hypoglycaemia or has only started to experience hypoglycaemic events very recently. Because hypoglycaemia awareness and its impairment is a continuum ranging from normality to
PREVALENCE OF IMPAIRED AWARENESS OF HYPOGLYCAEMIA
147
complete inability to detect the onset of hypoglycaemia, a classification of this condition will need to consider alterations in symptom intensity as well as detection of hypoglycaemia by any means and the ability of the patient to self-treat low blood glucose.
PREVALENCE OF IMPAIRED AWARENESS OF HYPOGLYCAEMIA Impaired awareness of hypoglycaemia is common in people treated with insulin. Although the chronic form of this acquired condition mainly affects those with type 1 diabetes, it appears that a similar problem does eventually emerge in patients with type 2 diabetes who have been treated with insulin for several years (Hepburn et al., 1993a). A prevalence of 8% was observed in a cohort of 215 patients in Edinburgh (Henderson et al., 2003). Because few patients with type 2 diabetes who require insulin therapy survive for a sufficiently long period to permit this complication to develop, impaired awareness is principally a problem associated with type 1 diabetes. It is not known whether impaired awareness of hypoglycaemia occurs in diabetic patients treated with oral antidiabetic agents. Impaired awareness of hypoglycaemia has been shown to be associated with strict glycaemic control (see Chapter 8), but significant modification of the symptomatic response to hypoglycaemia does not occur unless the glycated haemoglobin concentration is within the non-diabetic range (Boyle et al., 1995; Kinsley et al., 1995; Pampanelli et al., 1996). Only a small proportion of people with insulin-treated diabetes can sustain this degree of super-optimal glycaemic control indefinitely. In the Diabetes Control and Complications Trial (DCCT), with its extensive resources devoted to maintaining intensive insulin therapy, more than 40% of the patients in the group with strict glycaemic control achieved a HbA1c of 6.05% or less (the upper limit of the non-diabetic range) at some time during the study, but only 5% were able to maintain this level of glycaemic control continuously (The Diabetes Control and Complications Trial Research Group, 1993). The proportion of any insulintreated diabetic population that can achieve this therapeutic goal will depend on local policies regarding insulin therapy, the expertise of local diabetes specialist teams, available resources and the enthusiasm of individual patients. With the exception of a few highly motivated patients, most people treated with insulin are unable to maintain strict glycaemic control for protracted periods. In clinical practice this ‘acute’ form of hypoglycaemia unawareness is probably relatively uncommon. Nonetheless, the influence of strict glycaemic control on symptomatic and counterregulatory responses to hypoglycaemia has been studied extensively, and has provided insights into the potential pathogenetic mechanisms underlying impaired awareness of hypoglycaemia. Reduced warning symptoms of hypoglycaemia (of varying severity) occur in approximately one quarter of all insulin-treated patients. Cross-sectional population surveys in different European and North American populations of insulin-treated diabetic patients, using similar methods of assessment, have given remarkably consistent estimates (Table 7.1). Impaired awareness of hypoglycaemia becomes more common with increasing duration of insulin-treated diabetes (Hepburn et al., 1990), and almost 50% of patients experience hypoglycaemia without warning symptoms after 25 years or more of treatment (Pramming et al., 1991) (Figure 7.3). It appears therefore to be an acquired abnormality associated with insulin therapy.
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
148
Table 7.1 Prevalence of Hypoglycaemia Unawareness in population studies of insulin-treated diabetes
Country
Number of patients
Impaired awareness of hypoglycaemia (%)
Scotland Germany Denmark USA
302 523 411 628
23 25 27 20
Reference Hepburn et al. (1990) Muhlhauser et al. (1991) Pramming et al. (1991) Orchard et al. (1991)
Figure 7.3 Comparisons between the duration of diabetes and the percentage of 411 type 1 diabetic patients reporting (a) changes in symptoms of hypoglycaemia, (b) sweating and/or tremor as one of the two cardinal autonomic symptoms of hypoglycaemia, and (c) severe hypoglycaemic episodes without warning symptoms. Values are medians; shaded areas show 95% confidence limits. Reproduced from Pramming et al. (1991) by permission of John Wiley & Sons, Ltd
PATHOGENESIS OF IMPAIRED AWARENESS OF HYPOGLYCAEMIA
149
Frequency of Associated Severe Hypoglycaemia
Incidence (Events per subject per year)
Annual Prevalence (% Subjects)
It is apparent that impaired awareness of hypoglycaemia is a major risk factor for severe hypoglycaemia. In the DCCT, 36% of all episodes of severe hypoglycaemia occurred with no warning symptoms in patients while they were awake (The DCCT Research Group, 1991). In a population study in Edinburgh, retrospective assessment of the frequency of severe hypoglycaemia revealed that 90% of patients with impaired awareness of symptoms experienced severe hypoglycaemia in the preceding year, compared to 18% in a comparable group who had retained normal awareness (Hepburn et al., 1990). Prospective studies have confirmed the increase in frequency of mild and severe hypoglycaemia associated with impaired awareness of hypoglycaemia (Gold et al., 1994; Clarke et al., 1995), with a sixfold higher frequency of severe hypoglycaemia being documented in people with impaired awareness (Gold et al., 1994) (Figure 7.4).
Figure 7.4 Proportion of patients affected and event rates for severe hypoglycaemia in patients with type 1 diabetes with normal (, n = 31) or impaired (, n = 29) awareness of hypoglycaemia. Reproduced from Cryer and Frier (2004) by permission of John Wiley & Sons, Ltd. Data derived from Gold et al. (1994)
PATHOGENESIS OF IMPAIRED AWARENESS OF HYPOGLYCAEMIA The mechanisms underlying impaired awareness of hypoglycaemia are not known and may be multifactorial. Possible mechanisms are listed in Box 7.2.
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
150
Box 7.2
Impaired awareness of hypoglycaemia: possible mechanisms
CNS adaptation Chronic exposure to low blood glucose • glucose clamp (2.9 mmol/l) for 56 hours in non-diabetic subjects • insulinoma in non-diabetic patients • strict glycaemic control in diabetic patients Recurrent transient exposure to low blood glucose • antecedent hypoglycaemia CNS glucoregulatory failure • counterregulatory deficiency (hypothalamic defect?) • hypoglycaemia associated (central) autonomic failure (HAAF) Peripheral nervous system dysfunction • peripheral autonomic neuropathy • reduced peripheral adrenoceptor sensitivity
Altered Glycaemic Threshold for Initiation of Symptoms Symptoms of hypoglycaemia commence when the blood glucose reaches a specific level, and although this threshold may differ between individuals, it is usually constant and reproducible in the non-diabetic state (Vea et al., 1992). This blood glucose threshold for symptoms can be modified by protracted hypoglycaemia (Boyle et al., 1994) and is not fixed in people with diabetes who are treated with insulin, with its dynamic nature being demonstrated in various situations. In clinical practice, it has long been recognised that insulin-treated diabetic patients who have poor glycaemic control experience symptoms of hypoglycaemia when their blood glucose declines within a hyperglycaemic range (Maddock and Krall, 1953) and this has been shown to be associated with the onset of hypoglycaemic symptoms at a significantly higher blood glucose (4.3 mmol/l) compared to non-diabetic subjects (2.9 mmol/l) (Boyle et al., 1988). Conversely, strict glycaemic control modifies the glycaemic threshold for the onset of symptoms, which do not commence until blood glucose has declined to a lower level than that required in less well controlled patients to initiate a symptomatic response (see Chapter 8). The terminology that is used in relation to a change in the glycaemic threshold is potentially confusing. When a lower blood glucose is required to initiate a response, whether symptomatic, physiological or counterregulatory, the glycaemic threshold is said to be raised
PATHOGENESIS OF IMPAIRED AWARENESS OF HYPOGLYCAEMIA
151
or elevated; that is, a more profound hypoglycaemic stimulus is necessary to trigger the relevant response. Thus, strict glycaemic control raises the glycaemic threshold for the onset of symptoms, which do not occur until blood glucose has declined to a much lower concentration than would be observed in non-diabetic subjects. For many years, clinicians have recognised that the glycaemic threshold for the onset of hypoglycaemic symptoms is higher in patients with a long duration of type 1 diabetes who require a much lower blood glucose to provoke a symptomatic response. Lawrence (1941) wrote that ‘as years of insulin life go on, sometimes only after 5–10 years, I find it almost the rule that the type of insulin reactions change, the premonitory autonomic symptoms are missed out and the patient proceeds directly to the more serious manifestations affecting the central nervous system’. He astutely suggested that ‘the tissues may become attuned to a lower sugar concentration’. Recent studies in animals and humans have shown that the brain does adapt to chronic exposure to low blood glucose (see below) but this may not be beneficial to the individual with diabetes who is treated with insulin, i.e., it is a maladaptive response. An early study by Sussman et al. (1963) – revisited and extended by Hepburn et al. (1991) – showed that diabetic patients who had self-reported unawareness of hypoglycaemia did mount a sympatho-adrenal response to acute hypoglycaemia, but that this occurred at a lower blood glucose concentration than comparable diabetic subjects who had normal symptomatic awareness (Figure 7.5). However, the autonomic response was preceded by the
We do not have rights to reproduce this figure electronically
Figure 7.5 Venous blood glucose concentrations for the onset of the autonomic reaction in response to insulin-induced hypoglycaemia in individual non-diabetic control subjects, and in type 1 diabetic patients with normal and impaired awareness of hypoglycaemia and with autonomic neuropathy. Mean + SEM is shown for each group. Data derived from Hepburn et al. (1991) and reproduced from Hypoglycaemia and Diabetes (eds B.M. Frier and M. Fisher), © 1993 Edward Arnold, by permission of Edward Arnold (Publishers) Ltd
152
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
development of overt neuroglycopenia, which interfered with perception of the autonomic warning symptoms when they did eventually occur. This sequence of responses disrupts the ability of the individual subject to take appropriate action to self-treat low blood glucose. Similar findings have been reported by others (Grimaldi et al., 1990; Mokan et al., 1994; Bacatselos et al., 1995). In these studies, the counterregulatory hormonal responses to hypoglycaemia were delayed (Grimaldi et al., 1990; Hepburn et al. 1991) and their glycaemic thresholds had also shifted to occur at lower blood glucose concentrations (Mokan et al., 1994). This is consistent with the reported observation that impaired awareness of hypoglycaemia co-segregates with counterregulatory hormonal deficiency in people with longstanding type 1 diabetes (Ryder et al., 1990). In addition, Mokan et al. (1994) reported that cognitive dysfunction and neuroglycopenic symptoms in people with impaired awareness occurred at lower blood glucose levels than in people with type 1 diabetes who had normal awareness. This suggests that people with impaired awareness can function effectively with very low blood glucose concentrations, at which symptoms and cognitive impairment would normally occur in nondiabetic and aware diabetic subjects. The potential risk of this situation is apparent: it is akin to walking along the edge of a cliff on a dark night. With such a narrow glycaemic warning zone the propensity to rapidly develop severe neuroglycopenia is high and the margin for error is dangerously narrow. The results of these laboratory-based experimental studies of diabetic patients who have established hypoglycaemia unawareness are consistent with clinical observations of people with this acquired problem. At one moment they appear to be cerebrating normally (despite their blood glucose being low) then they rapidly become confused or drowsy, often with a vacant or dazed appearance and an inertia to seek some form of carbohydrate to reverse the neuroglycopenia. They may have to rely on relatives, friends or colleagues to identify the hypoglycaemia and provide treatment. This becomes a serious emergency if the patient is alone or if the insidious, but often rapid, development of neuroglycopenia goes unobserved. This explains the increased risk of progression to severe hypoglycaemia, and the higher rates reported in people with hypoglycaemia unawareness. Studies examining the effects of strict glycaemic control on symptomatic and counterregulatory responses to hypoglycaemia have also demonstrated a similar shift in glycaemic threshold for autonomic symptoms and an acute sympatho-adrenal response. However, the effect on glycaemic thresholds for neuroglycopenic symptoms and cognitive dysfunction remains controversial (see Chapter 8).
Peripheral Autonomic Neuropathy For many years, peripheral autonomic neuropathy was considered to be the principal cause of impaired awareness of hypoglycaemia (Hoeldtke et al., 1982). This was based on the assumption that the diminished secretion of epinephrine in response to hypoglycaemia (Hilsted et al., 1981; Bottini et al., 1997) would either prevent the generation of autonomic symptoms (such as sweating or a pounding heart) or reduce their intensity, resulting in an inability to perceive the onset of hypoglycaemia. Thus, autonomic neuropathy would interfere with the normal physiological responses stimulated by autonomic activation.
PATHOGENESIS OF IMPAIRED AWARENESS OF HYPOGLYCAEMIA
153
There are various reasons why this hypothesis is unlikely: • Although epinephrine can augment the intensity of a few autonomic symptoms of hypoglycaemia, it has a very limited role in their generation, which is modulated by sympathetic neural activation; the reduced secretory response of epinephrine in autonomic neuropathy is compensated by an increase in sensitivity of peripheral beta-adrenoceptors (Hilsted et al., 1987). • Diabetic subjects with autonomic neuropathy have normal physiological responses and experience typical autonomic symptoms during hypoglycaemia (Hilsted et al., 1981; Hepburn et al., 1993b), and no relationship has been found between autonomic dysfunction and hypoglycaemic symptoms (Berlin et al., 1987). • Impaired awareness of hypoglycaemia co-segregates with deficient counterregulatory hormonal responses and not with autonomic neuropathy (Ryder et al., 1990). • The prevalence of autonomic neuropathy is similar in patients with type 1 diabetes of long duration (more than 15 years), whether or not they have impaired awareness of hypoglycaemia (Hepburn et al., 1990). • Although impaired awareness of hypoglycaemia is a major risk factor for severe hypoglycaemia, the latter is either no more common in type 1 diabetic patients with autonomic neuropathy (Bjork et al., 1990; The DCCT Research Group, 1991), or is only modestly increased (Stephenson et al., 1996). • Autonomic neuropathy is not a determinant of whether glycaemic thresholds for autonomic (including symptomatic) responses to hypoglycaemia are affected by antecedent hypoglycaemia (Dagogo-Jack et al., 1993). Both impaired awareness of hypoglycaemia and peripheral autonomic neuropathy are common in people with type 1 diabetes of long duration, and frequently coexist. This does not prove a causal relationship, and it would appear that peripheral autonomic dysfunction does not have a prominent role in the pathogenesis of this syndrome. However, reduced sensitivity of cardiac beta-adrenoceptors to catecholamines has been observed in patients with type 1 diabetes who have impaired awareness of hypoglycaemia (Berlin et al., 1987). Hypoglycaemia per se reduces beta-adrenergic sensitivity in type 1 diabetes (Fritsche et al., 1998), and this sensitivity is increased after avoidance of hypoglycaemia for four months in people who have impaired awareness (Fritsche et al., 2001). The improved beta-adrenergic sensitivity correlated with a rise in autonomic symptom scores. Maladaptation of tissue sensitivity to catecholamines may therefore contribute to the development of hypoglycaemia unawareness even though autonomic neuropathy is not present.
Hypoglycaemia Associated Autonomic Failure The co-segregation of impaired hypoglycaemia awareness with counterregulatory deficiency suggests that they share a common underlying pathogenetic mechanism. These acquired abnormalities associated with hypoglycaemia in type 1 diabetes (Box 7.3) are characterised by a high frequency of severe hypoglycaemia and a common pathophysiological
154
Box 7.3
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
Acquired syndromes associated with hypoglycaemia in type 1 diabetes
• Counterregulatory deficiency • Impaired hypoglycaemia awareness • Altered glycaemic thresholds for counterregulatory and symptomatic responses
Figure 7.6 Schematic diagram of the concept of hypoglycaemia associated autonomic failure (HAAF), based on Cryer (1992)
feature, namely the elevated glycaemic thresholds (or lower blood glucose concentrations) that are required to trigger symptomatic and hormonal secretory responses. In other words, more profound hypoglycaemia is necessary to produce the usual symptomatic and counterregulatory responses to acute hypoglycaemia. Cryer (1992) has designated this group of abnormalities as a form of ‘hypoglycaemia associated autonomic failure’ (HAAF), and has speculated that recurrent severe hypoglycaemia may be the primary problem which establishes a vicious circle (Figure 7.6). It seems likely that this defect resides within the central nervous system. The possible mechanisms underlying HAAF and related hypoglycaemia syndromes in diabetes have been reviewed in detail (Cryer, 2005).
PATHOGENESIS OF IMPAIRED AWARENESS OF HYPOGLYCAEMIA
155
Central Nervous System Adaptation to Hypoglycaemia Some people with insulin-treated diabetes remain lucid, with no evidence of impaired cognitive function, when their blood glucose is low (often well below 3.5 mmol/l). Biochemical hypoglycaemia that is asymptomatic is commonly recorded by patients who have impaired awareness of hypoglycaemia, and they appear to have developed a neurological adaptation to chronic neuroglycopenia. The altered glycaemic threshold prevents the onset of warning symptoms and cognitive dysfunction until the blood glucose falls to a dangerously low level, which is extremely undesirable when striving for safe clinical management of insulin-treated diabetes. Although the human brain is dependent on a continuous supply of glucose for normal function, it can adapt to prolonged exposure to hypoglycaemia. This adaptation process takes at least several hours and possibly a few days to occur. Short-term exposure to acute hypoglycaemia (blood glucose 2.5 mmol/l) for 60 minutes in non-diabetic subjects showed no improvement in cognitive function and no reduction in symptom scores during this brief time interval (Gold et al., 1995a). However, when non-diabetic subjects were subjected to chronic hypoglycaemia (blood glucose 2.9 mmol/l) for 56 hours, using a glucose clamp, significant cerebral adaptation did occur (Boyle et al., 1994). The responses to acute hypoglycaemia (blood glucose 2.5 mmol/l) were compared before and after the period of chronic hypoglycaemia. Brain glucose uptake was initially reduced when blood glucose was below 3.6 mmol/l, but after a period of chronic hypoglycaemia uptake was preserved and cerebral function was maintained (Figure 7.7), demonstrating an effect of cerebral adaptation to chronic neuroglycopenia. The glycaemic thresholds for the onset of symptoms, counterregulatory hormonal secretion and cognitive dysfunction, were all modified and occurred at much lower blood glucose concentrations. A similar phenomenon has been observed in non-diabetic patients who had an insulinoma causing chronic hypoglycaemia; symptomatic responses to acute hypoglycaemia were blunted and counterregulatory hormonal responses were impaired but cognitive function was unaffected (Mitrakou et al., 1993). Surgical removal of the insulin-secreting tumour reversed these abnormalities indicating that they had resulted from cerebral adaptation to chronic hypoglycaemia. Strict glycaemic control in people with insulin-treated diabetes also alters the glycaemic thresholds for the development of counterregulatory hormones and symptoms (Chapter 8), so that a lower blood glucose concentration is required to trigger these responses. The observation that this requires a reduction in HbA1c to within the non-diabetic range (Kinsley et al., 1995) suggests that the median daily blood glucose in these individuals is relatively low, and the frequency of biochemical (and symptomatic) hypoglycaemia will be greater than in insulin-treated diabetic patients who are not as well controlled (Thorsteinsson et al., 1986). Boyle et al. (1995) have shown that those patients who had near normal HbA1c values, maintained normal uptake of glucose by the brain during hypoglycaemia, so preserving cerebral metabolism, reducing the counterregulatory responses to hypoglycaemia and diminishing symptomatic awareness. Although this capacity to maintain, and even increase, cerebral blood glucose uptake during hypoglycaemia is a protective response for the brain in these patients with strict glycaemic control, it is considered to be maladaptive, because it suppresses the normal symptomatic warning of responses and so risks the development of much more profound neuroglycopenia.
156
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
Figure 7.7 Rates of brain glucose uptake, epinephrine concentration and total symptoms of hypoglycaemia in non-diabetic subjects before and after prolonged hypoglycaemia. Initial day of investigation (hatched); after 56 hours of hypoglycaemia (solid). ∗ Significant difference from baseline for each of the two days. Reproduced from Boyle (1997), Diabetologia, 40, S69–S74. With kind permission of Springer Science and Business Media
During hypoglycaemia, glucose transport into the brain becomes rate-limiting, and brain energy metabolism deteriorates. The adaptive response results from an increased utilisation of glucose by the brain. In rodents, the transport of glucose across the blood–brain barrier is increased after several days of chronic hypoglycaemia (McCall et al., 1986). Further studies in rats of glucose transport activity across the blood–brain barrier have shown that when blood glucose was kept below 2.0 mmol/l for several days, changes in expression of the glucose transporter, GLUT-1, in brain microvasculature occurred in response to the chronic hypoglycaemia (Kumagai et al., 1995). This increase in GLUT-1 activity was responsible for the compensatory increase in glucose transport across the blood–brain barrier.
PATHOGENESIS OF IMPAIRED AWARENESS OF HYPOGLYCAEMIA
157
Antecedent (Episodic) Hypoglycaemia It has been recognised for many years that severe hypoglycaemia is associated with further episodes of severe hypoglycaemia, and one episode may influence the clinical manifestations of another occurring soon afterwards (Severinghaus, 1926). In recent years, several studies have shown that the symptomatic and counterregulatory responses to an episode of acute hypoglycaemia are diminished if a preceding (or antecedent) episode of hypoglycaemia has occurred within the previous 24 hours. Several studies have been performed in nondiabetic subjects (Table 7.2) and in people with insulin-treated diabetes (Table 7.3). Although these studies differ considerably in design and methods of inducing hypoglycaemia, in general it appears that antecedent hypoglycaemia of between one and two hours duration has a significant influence on the magnitude of the symptomatic and counterregulatory responses to subsequent hypoglycaemia occurring within the following 24 to 48 hours (Figure 7.8). The glycaemic thresholds for symptomatic and counterregulatory hormonal responses are altered by antecedent hypoglycaemia, and the degree to which subsequent responses are blunted are determined by the duration and depth of antecedent hypoglycaemia (Davis et al., 1997). Some of the physiological responses (e.g. sweating) may be blunted for longer than other responses following antecedent hypoglycaemia (George et al., 1995). Recurrent, shortlived (15 minutes) episodes of hypoglycaemia on four consecutive days, had no effect on counterregulatory and symptomatic responses in non-diabetic subjects (Peters et al., 1995), and so transient reductions in blood glucose may not produce this effect. Davis et al. (2000) have observed that a short duration of antecedent hypoglycaemia (20 minutes to lower and raise blood glucose from 3.9 to 2.9 mmol/l with the blood glucose being maintained for five minutes at 2.9 mmol/l) did not affect symptomatic awareness of hypoglycaemia, but did blunt the counterregulatory hormonal responses. Antecedent hypoglycaemia also reduces the counterregulatory responses to exercise on the following day, both in nondiabetic (Davis et al., 2000b) and type 1 diabetic subjects (Galassetti et al., 2003) and influences the metabolic responses, particularly diminishing endogenous glucose production in response to exercise. This may promote exercise-induced hypoglycaemia in type 1 diabetes. Some studies have examined the effect of antecedent hypoglycaemia on cognitive function, but in many the methods of assessment were inadequate and insufficient to provide definitive evidence of a change in cognitive response. Although some maintain that the glycaemic threshold for cognitive dysfunction is not altered by hypoglycaemia (see Chapter 8), an increasing number of studies have suggested that this does shift to a lower blood glucose concentration in the same manner as the thresholds for autonomic and counterregulatory responses (Veneman et al., 1993; Ovalle et al., 1998; Fanelli et al., 1998). Consistent with these observations, a study from Germany in non-diabetic men has shown that after a single episode of antecedent hypoglycaemia, subsequent hypoglycaemia had less effect on auditory-evoked brain potentials and short-term memory (Fruehwald-Schultes et al., 2000), demonstrating cerebral adaptation that preserves cognitive function. Nocturnal (episodic) hypoglycaemia, which is frequently not identified by patients, has been proposed as a mechanism for the induction of hypoglycaemia unawareness in people who give no history of recurrent hypoglycaemia (Veneman et al., 1993). The possible mechanisms of cerebral adaptation causing impaired awareness of hypoglycaemia are summarised in Box 7.4.
iv. infusion (2.2–2.5) 2 h
10
9
10
10
8
8
31
16
30
Mellman et al. (1994)
Robinson et al. (1995)
Peters et al. (1995)
George et al. (1995)
Davis et al. (1997)
Davis et al. (2000a)
Davis et al. (2000b)
Fruehwald-Schultes et al. (2000)
24 h
Clamp (2.9) 2 h (n = 15); 30 min (n = 16); 5 min (n = 10) Clamp (2.9) 2 h
BG = blood glucose; AH = antecedent hypoglycaemia; CR = counterregulatory (hormonal); iv = intravenous
Stepped clamp (2.6) 2.5 h 18–24 h
24 h
< 24 h
2 and 5 days
4 consecutive daily hypos
24 h and 6 days
1.5 h
4 consecutive daily hypos 7h
18 h
Interval before test hypoglycaemia
Stepped clamp (2.9) 2 h
iv. bolus injection of insulin (< 28) < 15 min Clamp (2.9) 2 h
Clamp (3.0) 2 h
Clamp (3.2) 2 h
Clamp (2.2–2.8) 1 h
Clamp (3.0) 2 h
10
9
No. of subjects
Heller and Cryer (1991) Widom and Simonson (1992) Veneman et al. (1993)
References
Method of induction, nadir BG (mmol/l) and duration of AH
Table 7.2 Studies of antecedent hypoglycaemia (AH) in non-diabetic humans
Clamp (3.1)
Exercise (90 min)
Clamp (2.9)
Clamp (2.9)
Clamp (2.5)
iv. bolus (< 28) X4
Stepped clamp (2.8) Clamp (2.5)
Stepped clamp (2.2) Stepped clamp (2.3)
Clamp (2.8)
Physiological responses still reduced after 1 week Magnitude of reduced CR response related to depth of AH Reduced CR responses in all studies. Symptom scores unaffected by short duration hypoglycaemia Blunting of CR responses to exercise. Attenuated metabolic responses (reduced endogenous glucose production) Auditory-evoked brain potentials and short-term memory less affected
Elevated BG thresholds for symptoms and CR responses Nocturnal hypo: BG thresholds for symptoms and CR responses elevated Elevated BG thresholds for symptoms and CR responses Reduced adrenaline and sweating at 6 days No effect on symptoms or CR responses
Reduced symptoms and CR responses
Test hypo: method and BG nadir Effect of AH on subsequent responses (mmol/l) to hypoglycaemia
Clamp (2.9) 2 h
24 h
5–10 h
48 h
Exercise for 90 min
Stepped clamp (2.5)
Stepped clamp (2.2) Stepped clamp (2.5)
Stepped clamp (1.7)
3 days
2 days
Stepped clamp (2.6)
Clamp (3.0)
Test hypo: method and BG nadir (mmol/l)
15 h
60 min
Interval before test hypoglycaemia
AH = antecedent hypoglycaemia; AN = autonomic neuropathy; iv = intravenous; CR = counterregulatory (hormonal)
16
Galassetti et al. (2003)
Clamp (2.6) 3.5 h, nocturnal
6
15
Clamp (2.8) 2 h, twice weekly for 1 month
8
George et al. (1997) Ovalle et al. (1998)
Fanelli et al. (1998)
iv. bolus injection of insulin (< 2.2) mins X 3 Clamp (2.8) 2 h
26 (± AN) 12 (non-diabetic)
18
Clamp (2.7) 2 h
13
Davis et al. (1992) Dagogo-Jack et al. (1993)
Lingenfelser et al. (1993)
Clamp (3.0) 2 h
No. of subjects
Reference
Method of induction, nadir BG (mmol/l) and duration of AH
Table 7.3 Studies of antecedent hypoglycaemia in people with insulin-treated diabetes
Physiological responses diminished for 2 days Reduced symptoms, CR and cognitive dysfunction. Fewer symptomatic episodes of clinical hypoglycaemia. Less deterioration in cognitive function after nocturnal hypoglycaemia (versus euglycaemia). Elevated BG threshold for cognitive dysfunction Blunting of CR responses to exercise. Three fold increase in glucose need
Reduced CR responses and hepatic glucose output Elevated BG thresholds for symptoms and autonomic responses Reduced symptoms, CR and neurophysiological responses
Effect of AH on responses to hypoglycaemia
160
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
Figure 7.8 Schematic representation of the effect of antecedent hypoglycaemia on the neuroendocrine and symptomatic responses to subsequent hypoglycaemia
Box 7.4 Mechanisms of cerebral adaptation causing impaired awareness of hypoglycaemia Symptomatic and Neuroendocrine responses to hypoglycaemia in insulin-treated diabetes are diminished in association with: • strict glycaemic control (HbA1c in non-diabetic range) • antecedent (episodic) hypoglycaemia • chronic (protracted) hypoglycaemia They may be restored by: • relaxation of glycaemic control • scrupulous avoidance of hypoglycaemia Although antecedent hypoglycaemia may induce transient impairment of awareness of hypoglycaemia it is unclear how this mechanism would induce chronic or prolonged loss of symptomatic perception. Although frequent, recurrent hypoglycaemia may have a contributory effect to inducing hypoglycaemia unawareness, presumably the hypoglycaemia has to be relatively protracted to induce prolonged cerebral adaptation, and the phenomenon is not limited to patients who have strict glycaemic control. The problem remains of explaining the induction of protracted or chronic hypoglycaemia unawareness, which often appears to be a permanent defect. Presumably repetitive hypoglycaemic insults to the brain (which are not necessarily severe) eventually ‘downregulate’ the central mechanisms that sense a low blood glucose and activate the glucoregulatory responses within the hypothalamus. There is evidence of a permanent redistribution of regional cerebral blood flow in diabetic patients with a history of recurrent severe hypoglycaemia (MacLeod et al., 1994a) with, in particular, a relative increase in blood flow to the frontal lobes. This may represent a chronic adaptive
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
161
response to protect vulnerable areas of the brain from recurrent, severe neuroglycopenia. However, a further study showed that the changes in regional cerebral blood flow in response to controlled hypoglycaemia in patients with type 1 diabetes occurred independently of the state of awareness of hypoglycaemia (MacLeod et al., 1996). The EEG changes associated with modest hypoglycaemia are more pronounced in patients with type 1 diabetes who have impaired awareness of hypoglycaemia (Tribl et al. 1996). Most studies suggest that a diffuse functional abnormality is present in the anterior part of the brain in diabetes, and this may be implicated in the impaired perception of hypoglycaemia. The pre-frontal areas of the cortex are closely connected to sub-cortical areas, and localised dysfunction could theoretically reduce the ability of the brain to perceive symptomatic hypoglycaemia. Further clues may be provided by neuroimaging studies. A study using positron emission tomography (PET) compared changes in global and regional brain glucose metabolism during euglycaemia and hypoglycaemia in 12 men with type 1 diabetes, six of whom had impaired awareness of hypoglycaemia (Bingham et al., 2005). Brain glucose content was reduced by hypoglycaemia in both groups with a relative increase in tracer uptake on the prefrontal cortical regions. Differences between the groups were observed in glucose handling in regions of the brain, and whereas the cerebral metabolic rate for glucose showed a relative rise in the aware subjects, it fell in the unaware subjects. Global neuronal activation was observed with hypoglycaemia in the aware patients, but was absent in the unaware, suggesting that cortical activation is a necessary correlate of hypoglycaemia awareness. Further studies using PET scans or other forms of neuroimaging may identify regional differences in the response to hypoglycaemia within the brains of people with impaired awareness, which will help to elucidate the functional abnormalities associated with this syndrome.
IMPAIRED AWARENESS OF HYPOGLYCAEMIA AND LONG-TERM EFFECT ON COGNITIVE FUNCTION Impaired awareness of hypoglycaemia is a major risk factor for severe hypoglycaemia, and patients with the chronic form of this condition have a six-fold higher frequency (Gold et al., 1994). It is possible therefore that impaired hypoglycaemia awareness may be associated with evidence of a decline in cognitive function. Hepburn et al. (1991) noted that diabetic patients with a history of impaired awareness of hypoglycaemia performed less well than those with normal awareness of hypoglycaemia on limited cognitive function testing, both at a normal blood glucose and during hypoglycaemia. This suggested that an acquired cognitive impairment may have been superimposed upon an increased susceptibility to neuroglycopenia. A modest, but insignificant, decline in intellectual function was noted with progressive loss of hypoglycaemia awareness in a population study (MacLeod et al., 1994b). Formal measurement of cognitive function during controlled hypoglycaemia (blood glucose 2.5 mmol/l) showed that patients with type 1 diabetes who had impaired hypoglycaemia awareness exhibited more profound cognitive dysfunction during acute hypoglycaemia than patients with normal awareness, and that this persisted for longer following recovery of blood glucose (Gold et al., 1995b). By contrast, a more recent study of people with type 1 diabetes, which examined the rate of recovery of differing domains of cognitive function after hypoglycaemia, showed that during hypoglycaemia (blood glucose 2.5 mmol/l) cognitive function did not deteriorate in those with impaired awareness, suggesting that
162
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
cerebral adaptation to hypoglycaemia had occurred in these patients (Zammitt et al., 2005). These apparently discrepant results leave this issue unresolved.
HUMAN INSULIN For 60 years after its discovery, insulin for therapeutic use was obtained from the pancreata of cattle and pigs. With the development of recombinant DNA technology it was possible to ‘genetically engineer’ molecules and insulin was the first protein to be made in this way, becoming available for the treatment of humans in the 1980s. Several of the existing animal insulin formulations were withdrawn, principally for commercial reasons, and human insulin rapidly became the most commonly prescribed form of insulin. The structure of human insulin differs from porcine insulin by a single amino acid and from bovine insulin by three amino acids. In initial trials it was not expected that human insulin would differ substantially in potency from animal insulin, but because human insulin was slightly purer than some of the animal insulins, patients were advised to reduce the dose by around 10% when converting from animal to human insulin. Detailed pharmacokinetic studies comparing human and animal insulins did not demonstrate any major differences, but human insulin has a slightly faster onset of action, a slightly shorter duration of action, and is less immunogenic than equivalent animal insulins. Most clinical studies, conducted on a worldwide scale, showed no significant differences between human and animal insulins in their clinical application. In Switzerland, however, one group of clinicians reported encountering serious clinical problems with the use of human insulin in patients with type 1 diabetes (Teuscher and Berger, 1987). In particular, they claimed that patients experienced more frequent hypoglycaemia with human insulin, and that warning symptoms were modified by human insulin, as a result of which many patients were unable to detect the onset of hypoglycaemia. A pathologist in the United Kingdom then claimed that the number of patients dying from severe hypoglycaemia had increased since the introduction of human insulin (see Chapter 12). The evidence for this irresponsible statement did not withstand scrutiny, but in the UK anecdotal reports emerged of problems experienced by patients with human insulin, and solicitors acting on behalf of over 400 patients tried to bring a legal action against the insulin manufacturers, alleging that human insulin gave less warning of hypoglycaemia. Additional claims included allegations that human insulin may have caused personality changes in individuals and even other disease states such as multiple sclerosis. This group action was abandoned in 1993 because of the lack of robust scientific evidence for these claims. However, this issue generated much controversy and heated debate and stimulated several studies comparing human with animal insulins, which are not reviewed here. A systematic review of the extensive literature on this topic examined whether published evidence supported a difference in the frequency and awareness of hypoglycaemia induced by human and animal insulins (Airey et al., 2000). A total of 52 randomized controlled trials were identified, 37 of which were of double-blind design, whereas others reported hypoglycemic outcomes as a secondary or incidental outcome during comparative investigations of efficacy or immunogenicity. Seven of the double-blind studies reported differences in frequency of hypoglycaemia or of symptomatic awareness, and four of the unblinded trials reported differences in hypoglycaemia. None of the four population time trend studies found any relationship between the increasing use of human insulin and hospital admission for hypoglycaemia
TREATMENT STRATEGIES
163
or unexplained death among people with insulin-treated diabetes. The authors observed that the least rigorous scientific studies gave the greatest support to the premise that treatment with human insulin influences the frequency, severity or symptoms of hypoglycaemia. The report concluded that the published evidence did not support the contention that human insulin is responsible either for the alleged problems with impaired awareness of hypoglycaemia or for a higher risk of severe hypoglycaemia, but were unable to decide whether some of the phenomena associated with the use of human insulin resulted from stricter glycaemic control (Airey et al., 2000). In clinical practice there are undoubtedly a small number of people with insulin-treated diabetes in whom the use of human insulin has not been satisfactory, being associated with frequent and unpredictable hypoglycaemia and a diminished sense of well-being. Whether this is related to the different pharmacokinetics of human insulin or represents an idiosyncratic response in individuals is unclear, but such patients clearly wish to retain the option to use animal species of insulin, and it is to be hoped that the availability of the animal species of insulins will be maintained. No problem with symptomatic awareness of hypoglycaemia has been reported with the use of the newer insulin analogues.
TREATMENT STRATEGIES When impaired awareness of hypoglycaemia is therapy-related, that is, resulting from strict glycaemic control, the approach to management is relatively simple. The total insulin dose should be reduced, attention paid to the appropriateness of the insulin regimen, and overall glycaemic control should be relaxed. Liu et al. (1996) reported an improvement in symptomatic and counterregulatory hormonal responses to hypoglycaemia after three months of less strict glycaemic control in a small group of insulin-treated patients, in whom the mean HbA1c rose from 6.9% to 8.0%. It has been claimed that impaired awareness of hypoglycaemia (and to some extent counterregulatory hormonal deficiency) can be reversed by scrupulous avoidance of hypoglycaemia through meticulous attention to diabetic management (Cranston et al., 1994; Dagogo-Jack et al., 1994; Fanelli et al., 1994). The effect that this had on glycaemic thresholds for cognitive dysfunction and the recovery of counterregulatory hormonal secretion to hypoglycaemia differed between these studies, but all demonstrated an improved symptomatic response following avoidance of hypoglycaemia for periods varying from three weeks to one year. However, the studies can be criticised for the following reasons: • Only a small number of patients were studied. • The definition of hypoglycaemia unawareness was based on an increased frequency of asymptomatic biochemical hypoglycaemia, and with the exception of the study by DagogoJack et al. (1994), was not based on having a history of hypoglycaemia unawareness. • In all studies there was a small but definite rise in glycated haemoglobin, suggesting that the improved symptomatic awareness was related primarily to relaxation of glycaemic control. Although the scrupulous avoidance of hypoglycaemia is clearly desirable, and may be beneficial to reducing the severity of hypoglycaemia unawareness, it is very difficult to
164
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
Figure 7.9 Augmentation of the normal secretory response of epinephrine (adrenaline) to, and awareness of, acute hypoglycaemia (blood glucose 2.8 mmol/l) by the prior ingestion of caffeine in insulin-treated diabetic patients. Derived from data in Debrah et al. (1996)
achieve as it is extremely time-consuming and labour-intensive both for patients and health professionals. The use of continuous subcutaneous insulin infusion overnight instead of isophane (NPH) insulin at bedtime has been shown to be beneficial in diabetic patients with impaired awareness of hypoglycaemia, improving warning symptoms and counterregulatory responses to hypoglycaemia, presumably by reducing the frequency of nocturnal hypoglycaemia (Kanc et al., 1998). The ingestion of caffeine uncouples the relationship between cerebral blood flow and glucose utilisation via antagonism of adenosine receptors, causing relative neuroglycopenia and earlier release of counterregulatory hormones during moderate hypoglycaemia. The prior consumption of caffeine augments the symptomatic and counterregulatory hormonal responses to a modest reduction of blood glucose in non-diabetic subjects (Kerr et al., 1993), and a similar phenomenon occurs in people with type 1 diabetes following the ingestion of a dose of caffeine equivalent to two or three cups of coffee (Debrah et al., 1996). The reduction in cerebral blood flow is sustained, the counterregulatory response is augmented (Figure 7.9) and greater awareness of hypoglycaemia occurs (see Chapter 5). This raises the prospect of identifying some form of therapeutic intervention, which utilises a similar mechanism to heighten the residual symptomatic response in people with type 1 diabetes who have impaired awareness of hypoglycaemia. The adenosine-receptor antagonist, theophylline, stimulates the secretion of catecholamines and reduces cerebral blood flow, and a single intravenous dose has been shown to enhance counterregulatory hormone responses to hypoglycaemia and partially restore perception of hypoglycaemic symptoms in patients with type 1 diabetes with impaired awareness of hypoglycaemia (de Galan et al., 2002). Glycaemic thresholds for haemodynamic and symptomatic responses were restored to normal. It is not known whether oral theophylline would be as effective, and whether the effects can be sustained, as the development of tolerance to these drugs is common. It is clearly desirable to avoid severe hypoglycaemia at all costs, and treatment strategies should be adopted to achieve this aim (Box 7.5). Frequent blood glucose monitoring is essential in affected patients, and may require occasional nocturnal measurements to detect
CONCLUSIONS
165
Box 7.5 Treatment strategies for patients with impaired awareness of hypoglycaemia • Frequent blood glucose monitoring (including nocturnal measurements). • Avoid blood glucose values < 40 mmol/l. • Set target range of blood glucose higher than for ‘aware’ patients (e.g. preprandial between 6.0–12.0 mmol/l; bedtime > 80 mmol/l) • Avoid HbA1c in non-diabetic range. • Use predominantly short-acting insulins (e.g. basal-bolus regimen; insulin analogues). • Regular snacks between meals and at bedtime, containing unrefined carbohydrate. • Appropriate additional carbohydrate consumption and/or insulin dose adjustment for premeditated exercise. • Learn to identify subtle neuroglycopenic cues to low blood glucose.
low blood glucose during the night. Blood glucose awareness training has been developed in the USA, with re-education of affected patients to recognise neuroglycopenic cues (Cox et al., 1995), but this also requires facilities and resources that are not available in most centres. Intensive insulin therapy is contraindicated in patients who have impaired awareness of hypoglycaemia and treatment goals have to be considered individually. The avoidance of severe hypoglycaemia is paramount as this may exacerbate the problem, and the use of mostly short-acting insulin (and possibly insulin analogues) in basal-bolus regimens may be particularly useful in avoiding biochemical and symptomatic hypoglycaemia without compromising overall glycaemic control.
CONCLUSIONS • An inadequate symptomatic warning to hypoglycaemia is common in people with insulin-treated diabetes and is described as impaired awareness of hypoglycaemia or hypoglycaemia unawareness. It increases in prevalence with duration of insulin-treated diabetes. • In people who report impaired awareness of hypoglycaemia, asymptomatic hypoglycaemia occurs more frequently during routine blood glucose monitoring. This may alert the clinician to the possibility that an individual is developing this problem. • Impaired awareness of hypoglycaemia may be associated with strict glycaemic control; significant modification of the symptomatic response occurs when the HbA1c concentration is within the non-diabetic range.
166
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
• The mechanisms underlying impaired awareness of hypoglycaemia may be multifactorial. Possible mechanisms include chronic exposure to a low blood glucose, antecedent hypoglycaemia, and central autonomic and glucoregulatory failure. • Antecedent hypoglycaemia has a significant influence on the magnitude of the symptomatic and counterregulatory responses to subsequent hypoglycaemia occurring within the following 48 hours. • When impaired awareness of hypoglycaemia results from strict glycaemic control, the total insulin dose should be reduced, attention paid to the suitability of the insulin regimen, and overall glycaemic control should be relaxed. • Impaired awareness of hypoglycaemia, and to some extent counterregulatory hormonal deficiency, can probably be reversed by scrupulous avoidance of hypoglycaemia through meticulous attention to diabetic management. • Intensive insulin therapy is contraindicated in patients who have impaired awareness of hypoglycaemia. The use of mostly short-acting insulins (including insulin analogues) in basal-bolus regimens may be particularly useful in avoiding biochemical and symptomatic hypoglycaemia without compromising overall glycaemic control.
REFERENCES Airey CM, Williams DRR, Martin PG, Bennett, CMT, Spoor PA (2000). Hypoglycaemia induced by exogenous insulin – ‘human’ and animal insulin compared. Diabetic Medicine 17: 416–32. Bacatselos SO, Karamitsos DT, Kourtoglou GI, Zambulis CX, Yovos JG, Vyzantiadis AT (1995). Hypoglycaemia unawareness in type 1 diabetic patients under conventional insulin treatment. Diabetes, Nutrition and Metabolism 8: 267–75. Berlin I, Grimaldi A, Payan C, Sachon C, Bosquet F, Thervet F, Puech AJ (1987). Hypoglycemic symptoms and decreased B-adrenergic sensitivity in insulin-dependent diabetic patients. Diabetes Care 10: 742–7. Bingham EM, Dunn JT, Smith D, Sutcliffe-Goulden J, Reed LJ, Marsden PK, Amiel SA (2005). Differential changes in brain glucose metabolism during hypoglycaemia accompany loss of hypoglycaemia awareness in men with type 1 diabetes mellitus. An (11 C)-3-O-methyl-D-glucose PET study. Diabetologia 48: 2080–9. Bjork E, Palmer M, Schvarcz E, Berne C (1990). Incidence of severe hypoglycaemia in an unselected population of patients with insulin-treated diabetes mellitus, with special reference to autonomic neuropathy. Diabetes, Nutrition and Metabolism 4: 303–9. Bottini P, Boschetti E, Pampanelli S, Ciofetta M, Del Sindaco P, Scionti L et al. (1997). Contribution of autonomic neuropathy to reduced plasma adrenaline responses to hypoglycemia in IDDM. Evidence for a nonselective defect. Diabetes 46: 814–23. Boyle PJ, Schwartz NS, Shah SD, Clutter WE, Cryer PE (1988). Plasma glucose concentrations at the onset of hypoglycemic symptoms in patients with poorly controlled diabetes and in non-diabetics. New England Journal of Medicine 318: 1487–92. Boyle PJ, Nagy RJ, O’Connor AM, Kempers SF, Yeo RA, Qualls C (1994). Adaptation in brain glucose uptake following recurrent hypoglycemia. Proceedings of National Academy of Science, USA 91: 9352–6. Boyle PJ, Kempers SF, O’Connor AM, Nagy RJ (1995). Brain glucose uptake and unawareness of hypoglycemia in patients with insulin-dependent diabetes mellitus. New England Journal of Medicine 333: 1726–31.
REFERENCES
167
Boyle PJ (1997). Alteration in brain glucose metabolism induced by hypoglycaemia in man, Diabetologia, 40, Suppl. S69–S74. Clarke WL, Cox DJ, Gonder-Frederick LA, Julian D, Schlundt D, Polonsky W (1995). Reduced awareness of hypoglycemia in adults with IDDM. A prospective study of hypoglycemic frequency and associated symptoms. Diabetes Care 18: 517–22. Cox D, Gonder-Frederick L, Polonsky W, Schlundt D, Julian D, Clarke W (1995). A multicenter evaluation of blood glucose awareness training – II. Diabetes Care 18: 523–8. Cranston I, Lomas J, Maran A, Macdonald I, Amiel SA (1994). Restoration of hypoglycaemia awareness in patients with long-duration insulin-dependent diabetes. Lancet 344: 283–7. Cryer PE (1992). Iatrogenic hypoglycemia as a cause of hypoglycemia-associated autonomic failure in IDDM. A vicious circle. Diabetes 41: 255–60. Cryer P and Frier BM (2004). Hypoglycaemia, International Textbook of Diabetes, 3rd edn, Ferrannini E., Zimmet P., De Fronzo R. and Keen H. (eds), 1079–100. Cryer PE (2005). Mechanisms of hypoglycemia-associated autonomic failure and its component syndromes in diabetes. Diabetes 54: 3592–601. Dagogo-Jack SE, Craft S, Cryer PE (1993). Hypoglycemia-associated autonomic failure in insulindependent diabetes mellitus. Journal of Clinical Investigation 91: 819–28. Dagogo-Jack S, Rattarasarn C, Cryer PE (1994). Reversal of hypoglycemia unawareness, but not defective glucose counterregulation, in IDDM. Diabetes 43: 1426–34. Davis MR, Mellman M, Shamoon H (1992). Further defects in counterregulatory responses induced by recurrent hypoglycemia in IDDM. Diabetes 41: 1335–40. Davis SN, Shevers C, Mosqueda-Garcia R, Costa F (1997). Effects of differing antecedent hypoglycemia on subsequent counterregulation in normal humans. Diabetes 46: 1328–35. Davis SN, Mann S, Galassetti P, Neill RA, Tate D, Ertl AC, Costa F (2000a). Effects of differing durations of antecedent hypoglycemia on counterregulatory responses to subsequent hypoglycemia in normal humans. Diabetes 49: 1897–903. Davis SN, Galassetti P, Wasserman DH, Tate D (2000b). Effects of antecedent hypoglycemia on subsequent counterregulatory responses to exercise. Diabetes 49: 73–81. de Galan BE, Tack CJ, Lenders JW, Pasman JW, Elving LD, Russel FG et al. (2002). Theophylline improves hypoglycemia unawareness in type 1 diabetes. Diabetes 51: 790–6. Debrah K, Sherwin RS, Murphy J, Kerr D (1996). Effect of caffeine on recognition of and physiological responses to hypoglycaemia in insulin dependent diabetes. Lancet 347: 19–24. Fanelli C, Pampanelli S, Epifano L, Rambotti AM, Di Vincenzo A, Modarelli F et al. (1994). Long-term recovery from unawareness, deficient counterregulation and lack of cognitive dysfunction during hypoglycaemia, following institution of rational, intensive insulin therapy in IDDM. Diabetologia 37: 1265–76. Fanelli CG, Paramore DS, Hershey T, Terkamp C, Ovalle F, Craft S, Cryer PE (1998). Impact of nocturnal hypoglycemia on hypoglycemic cognitive dysfunction in type 1 diabetes. Diabetes 47: 1920–7. Fritsche A, Stumvoll M, Grub M, Sieslack S, Renn W, Schmulling RM et al. (1998). Effect of hypoglycemia on beta-adrenergic sensitivity in normal and type 1 diabetic subjects. Diabetes Care 21: 1505–10. Fritsche A, Stefan N, Haring H, Gerich J, Stumvoll M (2001). Avoidance of hypoglycemia awareness by increasing beta-adrenergic sensitivity in type 1 diabetes. Annals of Internal Medicine 134: 729–36. Fruehwald-Schultes B, Born J, Kern W, Peters A, Fehm HL (2000). Adaptation of cognitive function to hypoglycemia in healthy men. Diabetes Care 23: 1059–66. Galassetti P, Tate D, Neill RA, Morrey S, Wasserman DH, Davis SN (2003). Effect of antecedent hypoglycemia on counterregulatory responses to subsequent euglycemic exercise in type 1 diabetes. Diabetes 52: 1761–9. George E, Harris N, Bedford C, Macdonald IA, Hardisty CA, Heller SR (1995). Prolonged but partial impairment of the hypoglycaemic physiological response following short-term hypoglycaemia in normal subjects. Diabetologia 38: 1183–90.
168
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
George E, Marques JL, Harris ND, Macdonald IA, Hardisty CA, Heller SR (1997). Preservation of physiological responses to hypoglycemia 2 days after antecedent hypoglycemia in patients with IDDM. Diabetes Care 20: 1293–8. Gold AE, MacLeod KM, Frier BM (1994). Frequency of severe hypoglycemia in patients with type 1 diabetes with impaired awareness of hypoglycemia. Diabetes Care 17: 697–703. Gold AE, Deary IJ, MacLeod KM, Thomson KJ, Frier BM (1995a). Cognitive function during insulininduced hypoglycemia in humans: short-term cerebral adaptation does not occur. Psychopharmacology 119: 325–33. Gold AE, MacLeod KM, Deary IJ, Frier BM (1995b). Hypoglycemia-induced cognitive dysfunction in diabetes mellitus: effect of hypoglycemia unawareness. Physiology and Behavior 58: 501–11. Grimaldi A, Bosquet F, Davidoff P, Digy JP, Sachon C, Landault C et al. (1990). Unawareness of hypoglycemia by insulin-dependent diabetics. Hormone and Metabolic Research 22: 90–5. Heller SR, Cryer PE (1991). Reduced neuroendocrine and symptomatic responses to subsequent hypoglycemia after 1 episode of hypoglycemia in non-diabetic humans. Diabetes 40: 223–6. Heller, SR, Chapman J, McCloud J, Ward J (1995). Unreliability of reports of hypoglycaemia by diabetic patients. British Diabetic Journal 310: 440. Henderson JN, Allen KV, Deary IJ, Frier BM (2003). Hypoglycaemia in insulin-treated type 2 diabetes: frequency, symptoms and impaired awareness. Diabetic Medicine 20: 1016–21. Hepburn DA, Patrick AW, Eadington DW, Ewing DJ, Frier BM (1990). Unawareness of hypoglycaemia in insulin-treated diabetic patients: prevalence and relationship to autonomic neuropathy. Diabetic Medicine 7: 711–7. Hepburn DA, Patrick AW, Brash HM. Thomson I, Frier BM (1991). Hypoglycaemia unawareness in type 1 diabetes: a lower plasma glucose is required to stimulate sympatho-adrenal activation. Diabetic Medicine 8: 934–45. Hepburn DA, Deary IJ, Frier BM (1992). Classification of symptoms of hypoglycaemia in insulintreated diabetic patients using factor analysis: relationship to hypoglycaemia unawareness. Diabetic Medicine 9: 70–5. Hepburn DA, MacLeod KM, Pell ACH, Scougal IJ, Frier BM (1993a). Frequency and symptoms of hypoglycaemia experienced by patients with type 2 diabetes treated with insulin. Diabetic Medicine 10: 231–7. Hepburn DA, MacLeod KM, Frier BM (1993b). Physiological, symptomatic and hormonal responses to acute hypoglycaemia in type 1 diabetic patients with autonomic neuropathy. Diabetic Medicine 10: 940–9. Hilsted J, Madsbad S, Krarup T, Sestoft L, Christensen NJ, Tronier B, Galbo H (1981). Hormonal, metabolic, and cardiovascular responses to hypoglycemia in diabetic autonomic neuropathy. Diabetes 30: 626–33. Hilsted J, Richter E, Madsbad S, Tronier B, Christensen NJ, Hildebrandt P et al. (1987). Metabolic and cardiovascular responses to epinephrine in diabetic autonomic neuropathy. New England Journal of Medicine 317: 421–6. Hoeldtke RD, Boden G, Shuman CR, Owen OE (1982). Reduced epinephrine secretion and hypoglycemia unawareness in diabetic autonomic neuropathy. Annals of Internal Medicine 96: 459–62. Jorgensen HV, Pedersen-Bjergaard U, Rasmussen AK, Borch-Johnsen K (2003). The impact of severe hypoglycemia and impaired awareness of hypoglycemia on relatives of patients with type 1 diabetes. Diabetes Care 26: 1106–9. Joslin EP, Gray H, Root HF. (1922). Insulin in hospital and home. Journal of Metabolic Research 2: 651–99. Kanc K, Janssen MMJ, Keulen ETP, Jacobs MAJM, Popp-Snijders C, Snoek FJ, Heine RJ (1998). Substitution of night-time continuous subcutaneous insulin infusion therapy for bedtime NPH insulin in a multiple injection regimen improves counterregulatory hormonal responses and warning symptoms of hypoglycaemia in IDDM. Diabetologia 41: 322–9.
REFERENCES
169
Kerr D, Sherwin RS, Pavalkis F, Fayad P, Sikorski L, Rife F et al. (1993). Effect of caffeine on the recognition of and responses to hypoglycemia in humans. Annals of Internal Medicine 119: 799–804. Kinsley BT, Widom B, Simonson DC (1995). Differential regulation of counterregulatory hormone secretion and symptoms during hypoglycemia in IDDM. Effect of glycemic control. Diabetes Care 18: 17–26. Kubiak T, Hermanns HJ, Schreckling HJ, Kulzer B, Haak T (2004). Assessment of hypoglycaemia awareness using continuous glucose monitoring. Diabetic Medicine 21: 487–90. Kumagai AK, Kang Y-S, Boado RJ, Pardridge WM (1995). Upregulation of blood–brain barrier GLUT 1 glucose transporter protein and mRNA in experimental chronic hypoglycemia. Diabetes 44: 1399–404. Lawrence RD (1941). Insulin hypoglycaemia. Changes in nervous manifestations. Lancet ii: 602. Lingenfelser T, Renn W, Sommerwerck U, Jung MF, Buettner UW, Zaiser-Kaschel H et al. (1993). Compromised hormonal counterregulation, symptom awareness, and neurophysiological function after recurrent short-term episodes of insulin-induced hypoglycemia in IDDM patients. Diabetes 42: 610–8. Liu D, McManus RM, Ryan EA (1996). Improved counter-regulatory hormonal and symptomatic responses to hypoglycemia in patients with insulin-dependent diabetes mellitus after 3 months of less strict glycemic control. Clinical and Investigative Medicine 19: 71–82. MacLeod KM, Hepburn DA, Deary IJ, Goodwin GM, Dougall N, Ebmeier KP, Frier BM (1994a). Regional cerebral blood flow in IDDM patients: effects of diabetes and of recurrent severe hypoglycaemia. Diabetologia 37: 257–63. MacLeod KM, Deary IJ, Graham KS, Hepburn DA, Frier BM (1994b). Hypoglycaemia unawareness in adult patients with type 1 diabetes: relationship to severe hypoglycaemia and cognitive impairment. Diabetes, Nutrition and Metabolism 7: 205–12. MacLeod KM, Gold AE, Ebmeier KP, Hepburn DA, Deary, IJ, Goodwin GM, Frier BM (1996). The effects of acute hypoglycemia on relative cerebral blood flow distribution in patients with type 1 (insulin-dependent) diabetes and impaired hypoglycemia awareness. Metabolism 45:974–80. McCall AL, Fixman LB, Fleming N, Tornheim K, Chick W, Ruderman NB (1986). Chronic hypoglycaemia increases brain glucose transport. American Journal of Endocrinology 251: E442–7. Maddock RK, Krall LP (1953). Insulin reactions. Manifestations and need for recognition of longacting insulin reactions. Archives of Internal Medicine 91: 695–703. Mellman MJ, Davis MR, Brisman M, Shamoon H (1994). Effect of antecedent hypoglycemia on cognitive function and on glycemic thresholds for counterregulatory hormone secretion in healthy humans. Diabetes Care 17: 183–8. Mitrakou A, Ryan C, Veneman T, Mokan M, Jenssen T, Kiss I et al. (1991). Hierarchy of glycemic thresholds for counterregulatory hormone secretion, symptoms and cerebral dysfunction. American Journal of Endocrinology 260: E67–74. Mitrakou A, Fanelli C, Veneman T, Perriello G, Calderone S, Platanisiotis D et al. (1993). Reversibility of unawareness of hypoglycemia in patients with insulinomas. New England Journal of Medicine 329: 834–9. Mokan M, Mitrakou A, Veneman T, Ryan C, Korytkowski M, Cryer P, Gerich J (1994). Hypoglycemia unawareness in IDDM. Diabetes Care 17: 1397–403. Muhlhauser I, Heirnemann L, Fritsche E, von Lennep K, Berger M (1991). Hypoglycemic symptoms and frequency of severe hypoglycemia in patients treated with human and animal insulin preparations. Diabetes Care 14: 745–9. Orchard TJ, Maser RE, Becker DJ, Dorman JS, Drash AL (1991). Human insulin use and hypoglycaemia: insights from the Pittsburgh Epidemiology of Diabetes Complications Study. Diabetic Medicine 8: 469–74. Ovalle F, Fanelli CG, Poramore DS, Hersley T, Craft S, Cryer PE (1998). Brief twice-weekly episodes of hypoglycemia reduce detection of clinical hypoglycemia in type 1 diabetes mellitus. Diabetes 47: 1472–9.
170
IMPAIRED AWARENESS OF HYPOGLYCAEMIA
Pampanelli S, Fanelli C, Lalli C, Ciofetta M, Del Sindorco P, Lepore M et al. (1996). Long-term intensive insulin therapy in IDDM: effects on HbA1c , risk for severe and mild hypoglycaemia, status of counterregulation and awareness of hypoglycaemia. Diabetologia 39: 677–86. Peters A, Rohloff F, Kerner W (1995). Preserved counterregulatory hormone release and symptoms after short term hypoglycemic episodes in normal men. Journal of Clinical Endocrinology and Metabolism 80: 2894–8. Pramming S, Thorsteinsson B, Bendtson I, Binder C (1991). Symptomatic hypoglycaemia in 411 type 1 diabetic patients. Diabetic Medicine 8: 217–22. Robinson AM, Parkin HM, Macdonald IA, Tattersall RB (1995). Antecedent hypoglycaemia in nondiabetic subjects reduces the adrenaline response for 6 days but does not affect the catecholamine response to other stimuli. Clinical Science 89: 359–66. Ryder REJ, Owens DR, Hayes TM, Ghatei MA, Bloom SR (1990). Unawareness of hypoglycaemia and inadequate hypoglycaemic counterregulation: no causal relation with diabetic autonomic neuropathy. British Medical Journal 301: 783–7. Severinghaus EL (1926). Hypoglycemic coma due to repeated insulin overdosage. American Journal of Medical Sciences 172: 573–80. Stephenson JM, Kempler P, Cavallo Peria P, Fuller JH, EURODIAB IDDM Complications Study Group (1996). Is autonomic neuropathy a risk factor for severe hypoglycaemia? The EURODIAB IDDM Complications Study. Diabetologia 39: 1372–6. Sussman KE, Crout JR, Marble A (1963). Failure of warning in insulin-induced hypoglycemic reactions. Diabetes 12: 38–45. Teuscher A, Berger WG (1987). Hypoglycaemia unawareness in diabetics transferred from beef/porcine insulin to human insulin. Lancet ii: 382–5. The DCCT Research Group (1991). Epidemiology of severe hypoglycemia in the Diabetes Control and Complications Trial. American Journal of Medicine 90: 450–9. The Diabetes Control and Complications Trial Research Group (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. The New England Journal of Medicine 329: 977–86. Thorsteinsson B, Pramming S, Lauritzen T, Binder C (1986). Frequency of daytime biochemical hypoglycaemia in insulin-treated diabetic patients: relation to daily median blood glucose concentrations. Diabetic Medicine 3: 147–51. Tribl G, Howorka K, Heger G, Anderer P, Thoma H, Zeitlhofer J (1996). EEG topography during insulin-induced hypoglycemia in patients with insulin-dependent diabetes mellitus. European Neurology 36: 303–9. Vea H, Jorde R, Sager G, Vaaler S, Sundsfjord J (1992). Reproducibility of glycaemic thresholds for activation of counterregulatory hormones and hypoglycaemic symptoms in healthy subjects. Diabetologia 35: 958–61. Veneman T, Mitrakou A, Mokan M, Cryer P, Gerich J (1993). Induction of hypoglycemia unawareness by asymptomatic nocturnal hypoglycemia. Diabetes 42: 1233–7. Widom B, Simonson DC (1992). Intermittent hypoglycemia impairs glucose counterregulation. Diabetes 41: 1335–40. Zammitt NN, Warren RE, Deary IJ, Frier BM (2005). Recovery of cognitive function after insulininduced hypoglycaemia in people with type 1 diabetes with either normal or impaired awareness of hypoglycaemia. Diabetologia 48 (Suppl 1): A293 (abstract).
8 Risks of Strict Glycaemic Control Stephanie A. Amiel INTRODUCTION The benefit of strict glycaemic control in diminishing the risks of the development of longterm complications of diabetes is beyond doubt, but the negative aspects of such therapies need to be considered, and their risks identified, understood and minimised. Modern intensified insulin management need not necessarily increase the risk of iatrogenic problems and can deliver better glycaemic control more safely than in the past, although substantial scope remains for improvement. New drugs for type 2 diabetes may offer greater opportunities to achieve near-normoglycaemia but may also bring new risks. These risks need to be explained carefully to every patient, who can then make an individual, informed choice about the management of their diabetes. The risks of intensified insulin therapy, the focus of this chapter, are those of insulin itself – intensified. Thus the major side-effects are weight gain (The Diabetes Control and Complications Trial Research Group, 1988) and hypoglycaemia (The Diabetes Control and Complications Trial Research Group, 1993; 1995a; 1997). Both of these problems may appear to be minimised with modern strategies for patient self-management, at least in published studies (Jorgens et al., 1993; DAFNE Study Group, 2002; Plank 2004 et al.; Samann et al. 2005), yet they remain serious issues for large numbers of people. Weight gain, attributed primarily to the resolution of caloric loss in glycosuria (Carlson and Campbell, 2003), is theoretically responsive to dietary strategies, but insulin and peripheral insulin sensitizers do cause lipogenesis and fluid retention, both of which contribute to a rise in weight that may be unacceptable to patients. Evidence is accumulating about the potential effects of insulin and other anti-diabetic agents on appetite control and satiety that may make the control of weight more difficult. Although the long-term diminution of risk of vascular complications is now established beyond doubt, the sudden institution of strict glycaemic control after a prolonged period of hyperglycaemia, can transiently, but sometimes seriously, destabilise microvascular disease (Agardh et al., 1992; The Diabetes Control and Complications Trial Research Group, 1998). In type 1 diabetes, the long-term follow up of the DCCT cohorts unequivocally has extended the evidence to include slowed progression of macrovascular, as well as microvascular, disease (Nathan et al., 2003; Writing team for the Diabetes Control and Complications Trial/Epidemiology of Diabetes Intervention and Complications Research Group, 2002), and so the risks of intensified therapy need to be balanced against the potentially large gains. A new risk, of unknown magnitude, is the increasing use of novel insulins, which have different properties from endogenous human insulin and thus, at least in theory, may have different side-effects.
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
172
RISKS OF STRICT GLYCAEMIC CONTROL
When cohorts of patients are studied rather than individuals, other potential risks of the demands of intensified insulin therapies, and in particular the inherent psychosocial strains, do not emerge as a problem. Concerns have been expressed that greater use of home blood glucose monitoring may increase anxiety, particularly in type 2 patients who are not taking insulin, and who have limited means at their disposal of responding to high blood glucose values (Franciosi et al., 2001). In contrast, in type 1 diabetes, research suggests that patients may prefer intensified treatment regimens. In the Diabetes Control and Complications Trial, patients in the intensive treatment arm of the study had an overall improvement in their subjective feelings of control and well-being, although it was offset by a greater fear of hypoglycaemia (The Diabetes Control and Complications Trial Research Group, 1996a). However, this balance may be particularly positive in people who actively choose to use intensified therapies. In the DAFNE Trial, all participants had selected the intensive programme of treatment, and significant and apparently lasting benefits in quality of life measures were demonstrated using the intensified management strategy (DAFNE Research Group, 2002). However, it must be acknowledged that when individual patients are exhorted to achieve perfection in glycaemic control, they may experience difficulties and frustration with the impossible task of trying to eliminate blood glucose readings that lie outside the normal range, especially if they are not equipped to act upon such readings. In general, the main risk of intensified diabetes therapy remains hypoglycaemia. This chapter examines the problem of hypoglycaemia that is specifically associated with strict glycaemic control, an area that has aroused much concern and controversy. Most comments relate to patients with type 1 diabetes. The risks of severe hypoglycaemia associated with strict control in insulin-treated type 2 diabetes are likely to be similar, but occur much later in the natural history of the disease, when insulin deficiency is profound (see Chapter 11).
DEFINITION OF HYPOGLYCAEMIA It is difficult to determine a frequency of hypoglycaemia without first defining what is meant by ‘hypoglycaemia’. In many studies, hypoglycaemia is documented by self-reporting, which may be very unreliable (Heller et al., 1995). Retrospective analyses suffer from problems of recall, and accurate documentation of hypoglycaemia is obtained only in prospective research studies that require biochemical verification of low blood glucose concentrations (see Chapter 3). Hypoglycaemia can be categorised by its symptomatology and its severity, but no real consensus exists. ‘Mild’ hypoglycaemia is usually defined as an episode that a person recognises and treats themselves and does not significantly disrupt daily living; ‘severe’ hypoglycaemia is an episode in which blood glucose has fallen to a level where the patient has become so disabled that assistance is required from another person (The Diabetes Control and Complications Trial Research Group, 1991). Alternatively, ‘severe’ hypoglycaemia may be defined by the requirement for parenteral treatment (intramuscular glucagon or intravenous dextrose), with or without hospital admission, or by the development of coma (The Diabetes Control and Complications Trial Research Group, 1987). A category of ‘moderate’ hypoglycaemia, in which an individual requires external assistance but which falls short of requiring parenteral therapy or developing a coma, or the division of hypoglycaemia into grades of severity has also been used (Limbert et al., 1993). Obviously the definition used will affect the estimate of incidence, and if severe hypoglycaemia is defined solely as coma, rates will be lower than if all episodes requiring assistance are included.
CONTRIBUTORS TO INCREASED RISK OF SEVERE HYPOGLYCAEMIA
173
Various levels of blood glucose concentration have been used to define mild and moderate hypoglycaemia biochemically, but there is now recognition that the blood glucose concentrations which have been used arbitrarily to define pathological ‘spontaneous’ hypoglycaemia (such as < 22 mmol/l) are unsuitable for defining hypoglycaemia in people with diabetes. An arterialised plasma glucose concentration of around 3.6 mmol/l may be sufficient to cause physiological autonomic responses in healthy volunteers (Chapter 1). Subtle changes in cognitive function can initially be detected by formal testing below 4.0 mmol/l, although clinically relevant cognitive impairment does not occur until the arterialised plasma glucose concentration has fallen to approximately 3.0 mmol/l (Chapter 2). There is evidence that if plasma glucose falls below 3.0 mmol/l for a period of time, this can reduce the symptomatic responses to a further episode of hypoglycaemia occurring within the following 24 hours (Heller and Cryer, 1991), so-called antecedent hypoglycaemia (Chapter 7). The continuous avoidance of low blood glucose concentrations (below 3.0 mmol/l) can restore normal hormonal and symptomatic responses to hypoglycaemia (Fanelli et al., 1993; Cranston et al., 1994), and it follows that values below 3.0 mmol/l can be considered to be hypoglycaemic. Diabetes UK coined the phrase ‘make four the floor’ to protect against potentially dangerous hypoglycaemia and, most recently, a panel convened by the American Diabetes Association concluded in favour of a concentration of 3.9 mmol/l (Working Group on Hypoglycemia, American Diabetes Association, 2005). This was based on research data, showing evidence of counterregulatory responses at blood glucose concentrations (often arterial or arterialised) of 3.9 mmol/l and the demonstration of a small reduction in the glucagon response to hypoglycaemia induced immediately afterwards. However, this defines ‘hypoglycaemia’ at blood glucose concentrations that commonly occur in healthy non-diabetic people and may therefore encourage over-treatment. It is probably more appropriate to differentiate between targets for adjusting therapy, which may properly be above 4.0 mmol/l, and ‘hypoglycaemia’, which implies a pathological cause and requires intervention.
CONTRIBUTORS TO INCREASED RISK OF SEVERE HYPOGLYCAEMIA IN PATIENTS UNDERTAKING INTENSIFIED INSULIN THERAPY Factors Predisposing Patients to Severe Hypoglycaemia in Intensified Insulin Therapy The relationship between impaired symptomatic awareness of hypoglycaemia and an increased rate of severe hypoglycaemia is well established (Hepburn et al., 1990; Gold et al., 1994; Clarke et al., 1995), although affected patients in these studies were not subject to strict glycaemic control. The association between counterregulatory failure and increased risk of severe hypoglycaemia is also well recognised (Ryder et al., 1990). Indeed, counterregulatory failure was proposed as a predictor of risk of severe hypoglycaemia in the subsequent application of intensified therapy (White et al., 1983), and it was not until later that the ability of intensified therapy to cause counterregulatory failure was suggested (Simonson et al., 1985a). It is indeed very important to appreciate that neither asymptomatic nor severe hypoglycaemia are restricted to people using intensified insulin therapy.
174
RISKS OF STRICT GLYCAEMIC CONTROL
Apart from a previous history of severe hypoglycaemia, the greatest risk may be the degree of insulin deficiency, as reflected by the absence of C-peptide (Muhlhauser et al., 1998), as well as the glycaemic control prior to embarking upon intensified therapy and the determination to reach the glycaemic targets (Bott et al., 1994; Muhlhauser et al., 1998). Preservation of endogenous insulin is not affected by intensification of insulin therapy, although there is evidence to suggest that if strict glycaemic control is imposed when diabetes is diagnosed, this may result in more prolonged preservation of endogenous insulin secretion (Shah et al., 1989, The Diabetes Control and Complications Trial Research Group, 1998b). Other factors, related to the patient rather than to treatment, may increase the risk of severe hypoglycaemia, including social class (Muhlhauser et al., 1998) and possibly genetics. In a study from Denmark, much of the risk of severe hypoglycaemia was attributed to ACE genotype (PedersenBjergaard et al., 2003), although this has not been confirmed and has aroused controversy; also, the absence of traditional risk factors in the Danish study is a cause for concern.
The Effects of Intensified Insulin Therapy Upon Risk of Severe Hypoglycaemia In the DCCT, a clear link was demonstrated between intensified insulin therapy and the frequency of severe hypoglycaemia. In that trial, a three-fold higher rate of severe hypoglycaemia was recorded by the patients in the intensive treatment arm when compared with those on conventional therapy (The Diabetes Control and Complications Trial Research Group, 1991; 1993; 1997). This persisted throughout the entire study, although absolute rates declined gradually in both groups. Furthermore, the risk of severe hypoglycaemia was higher for any given HbA1c , for the people receiving intensive treatment. This phenomenon has not been adequately explained. It is now known that exposure to hypoglycaemia per se can induce defects in counterregulation and loss of subjective awareness of hypoglycaemia (Heller and Cryer, 1991; George et al., 1995; 1997; Davis et al., 1997). It has been assumed that intensive therapy exposes the patient to a greater frequency of mild hypoglycaemia that is sufficient to induce such defects and thereby increase the risk of severe hypoglycaemia by that mechanism. However, methods for delivering intensified diabetes therapy have subsequently improved. Modern methods that focus on transferring skills of insulin adjustment to the patients themselves are reported to achieve improvements in HbA1c with multiple daily injection therapy regimens, without causing more episodes of severe hypoglycaemia, and in their most successful forms achieve a parallel reduction of hypoglycaemia (Jorgens et al., 1993; DAFNE Study Group, 2002; Plank et al., 2004; Samann et al., 2005). The judicious use of insulin analogues in intensified regimens may be associated with slightly less risk of hypoglycaemia (Ashwell et al., 2006), whereas the use of continuous subcutaneous insulin infusion (CSII) with pumps is associated with a much lower frequency of severe hypoglycaemia, and has been used successfully as treatment for patients with problematical hypoglycaemia (Bode et al., 1996, Rodrigues et al., 2005) and in the context of clinical trials (Hoogma et al., 2006).
The Link Between Intensified Insulin Therapy and Risk of Severe Hypoglycaemia Patients describe symptoms of hypoglycaemia at a wide range of blood glucose concentrations. In an individual patient, the main determinant of the blood glucose concentration
CONTRIBUTORS TO INCREASED RISK OF SEVERE HYPOGLYCAEMIA
175
at which protective responses commence is probably the recent prevailing range of blood glucose concentration to which the patient has been exposed. For example, when patients with poorly-controlled type 2 diabetes were studied with a controlled hypoglycaemic challenge after blood glucose had been normalised overnight, their epinephrine responses to hypoglycaemia were triggered at higher blood glucose values than in well-controlled patients (Korzon-Burakowska et al., 1998). As mentioned earlier, the first indication that strict glycaemic control might cause abnormal responses to hypoglycaemia was observed when controlled hypoglycaemia was induced in a small group of patients with type 1 diabetes before, and after, they had been treated with intensified insulin therapy (Simonson et al., 1985a). Following the improvement in glycaemic control, the magnitude of the counterregulatory hormonal response to an abrupt lowering of blood glucose to 2.8 mmol/l was significantly less than observed previously. This study had been planned to investigate the potential of better glycaemic control to restore some of the defects of normal counterregulation that develop in people with type 1 diabetes (see Chapter 6), so these results were unexpected. The importance of these preliminary observations was underlined by a subsequent study in which patients with type 1 diabetes receiving intensified insulin treatment were found to have impaired glucose counterregulation (Amiel et al., 1987). During an intravenous infusion of insulin, most patients were unable to maintain arterialised plasma glucose above 3.0 mmol/l, in contrast with conventionallytreated diabetic patients whose glycaemic control was not as good (as demonstrated by higher glycated haemoglobin concentrations) or non-diabetic volunteers. The intensively-treated diabetic patients were less symptomatic, and although the rise in their plasma epinephrine was of similar magnitude to the other groups, this occurred only when the hypoglycaemia was more profound. Further studies of hypoglycaemia, using a stepped glucose clamp to produce a controlled reduction of blood glucose, confirmed that the symptomatic and hormonal responses started at lower blood glucose concentrations in patients with strict glycaemic control, and were delayed in onset and diminished in magnitude for any given blood glucose concentration (Amiel et al., 1988) (Figure 8.1). The delayed onset and diminished vigour of symptomatic and hormonal responses to hypoglycaemia in strictly-controlled diabetic subjects offers a partial explanation for the increased
Figure 8.1 The effect of intensified diabetes therapy (IRx) on epinephrine responses to a slow reduction in plasma glucose over four hours. Copyright © 1988 American Diabetes Association. From Amiel et al., 1988. Reprinted with permission from The American Diabetes Association
176
RISKS OF STRICT GLYCAEMIC CONTROL
occurrence of asymptomatic biochemical hypoglycaemia. The risk may be particularly manifested when the glycated haemoglobin concentration is reduced to within, or just above, the non-diabetic range (Box 8.1). This was shown in a study of 34 subjects with type 1 diabetes who had a wide range of total HbA1 values (Kinsley et al., 1995). They were subjected to a stepped glucose clamp to lower arterialised blood glucose to 2.3 mmol/l and the responses were compared with a non-diabetic control group. Symptomatic responses (particularly autonomic) and some counterregulatory hormonal responses were diminished in the seven diabetic subjects who had a total HbA1 of 7.85% or less, i.e., glycaemic control that was within their local nondiabetic range of total HbA1 . A very similar study by Pampanelli et al. (1996) produced identical observations in 10 of 33 subjects, whose HbA1c was within the local non-diabetic range, and in whom it was also noted that the onset of some aspects of cognitive dysfunction was delayed. Current evidence would suggest that it is the increased exposure to episodic hypoglycaemia, associated with the treatment strategy that is promoting the problem. Most importantly, a series of studies has shown that hypoglycaemia awareness and counterregulatory hormone responses can be restored in well-controlled diabetic subjects by avoidance of blood glucose concentrations below 3.0 mmol/l in daily life, confirming the circular link between hypoglycaemia exposure and impaired awareness of hypoglycaemia (Fanelli et al., 1993; Cranston et al., 1994). Thus, although impaired awareness of hypoglycaemia is a major problem in clinical practice, it is by no means exclusively confined to intensified therapy. Although the risk remains greater with lower mean glucose and glycated haemoglobin concentrations, impaired awareness is reversible, at least in the setting of carefully controlled research studies, by scrupulous avoidance of even modest hypoglycaemia in daily life (Fanelli et al., 1993; Cranston et al., 1994). Although this may result in a deterioration of glycaemic control as the problem was reversed, with a rise in mean HbA1c from 6.9% to 8.0% in one small study of seven patients with impaired awareness of hypoglycaemia (Liu et al., 1996), this is not inevitable (Cranston et al., 1994). It is possible for avoidance of hypoglycaemia to result in an improvement of glycated haemoglobin, as post-hypoglycaemia hyperglycaemia is eradicated. Much less work has been done in type 2 diabetes, although there is increasing evidence that in patients with insulin-treated type 2 diabetes of long duration, the prevalence of severe hypoglycaemia is not greatly different from people with type 1 diabetes (see Chapter 11). Modern trends of starting insulin earlier in type 2 diabetes, when insulin deficiency is not
Box 8.1
Effects of strict glycaemic control in type 1 diabetes
• Reduction in microvascular and macrovascular complications. • Potential increase in risk of severe hypoglycaemia. • Diminished counterregulatory and symptomatic responses to hypoglycaemia. • Altered glycaemic thresholds for activation of responses (i.e., lower blood glucose required). • Promotion of increased frequency of exposure to hypoglycaemia which exacerbates impaired awareness of hypoglycaemia. • Tendency to weight gain.
CEREBRAL ADAPTATION
177
severe, are likely to reduce the overall risk of hypoglycaemia in patients with insulin-treated type 2 diabetes. Recent studies using bedtime basal insulin as the first line of intensifying diabetes treatment for type 2 patients who are not achieving glycaemic targets, have reported a low risk of severe hypoglycaemia, even when using conventional insulins (Yki-Jarvinen et al., 2006). However, caution is indicated when patients require conversion to full insulin therapy. In a small study of poorly-controlled patients treated with oral medication, in which responses to hypoglycaemia were measured before and after improving glycaemic control with insulin, counterregulatory responses and the blood glucose thresholds at which these were initiated were modified, as occurs in type 1 diabetes (Korson-Burakowska et al., 1998).
CEREBRAL ADAPTATION When hypoglycaemia occurs, the stimulus for counterregulation appears to be a fall in the cerebral metabolic rate of glucose. Boyle et al. (1994) measured arteriovenous differences in glucose concentration in the human brain during hypoglycaemia to show that the rate of uptake of glucose (and by implication of metabolism) falls before most of the counterregulatory responses and cognitive changes occur. They also demonstrated that this fall in metabolic rate of the brain was reduced in healthy volunteers who were made acutely hypoglycaemic following a period of 56 hours of protracted moderate hypoglycaemia, suggesting that the metabolism of the human brain can adapt to prolonged exposure to low blood glucose. This enables the brain to maintain its metabolism and continue to function in response to subsequent hypoglycaemia. A further study in diabetic patients showed that diabetic patients with strict glycaemic control and impaired awareness of hypoglycaemia were able to maintain the rate of cerebral uptake of glucose during experimental hypoglycaemia, while others with normal symptomatic awareness exhibited a marked fall in cerebral uptake of glucose, associated with symptomatic and counterregulatory hormonal responses (Figure 8.2) (Boyle et al., 1995). These data led to the hypothesis that impaired awareness of hypoglycaemia and defective glucose counterregulation may result from an adaptation in the sensitivity of the cerebral glucose sensor, which allows it to sustain its metabolic rate (and so not trigger counterregulation) during subsequent hypoglycaemia (see Chapter 7). However, the expectation that patients with impaired awareness of hypoglycaemia will show an increase in brain glucose metabolic rate at any given blood glucose concentration has not been supported by neuroimaging studies. In studies utilising positron emission tomography that used various tracers for glucose to measure either the metabolic rate of brain glucose or glucose tracer uptake in humans, several investigators have failed to find differences during euglycaemia or hypoglycaemia that could be in accord with prevailing glycaemic control (Cranston et al., 2001; Segal et al., 2001). One study found evidence during hypoglycaemia of a difference in the change in uptake of the glucose tracer, de-oxyglucose, in the brain region around the hypothalamus in intensively-treated diabetic subjects who had impaired awareness of hypoglycaemia (Cranston et al., 2001), which is of interest because animal studies have implicated this region (among others) in sensing hypoglycaemia. In more recent studies, the difference in brain glucose metabolism in subjects with impaired awareness of hypoglycaemia was a failure of increase of cerebral metabolic rate during hypoglycaemia, associated with a failure to generate or perceive symptoms (Bingham et al., 2005). These data are compatible with the concept that cortical activation is important for perception of symptoms and that this fails in people who develop a loss of awareness of
178
RISKS OF STRICT GLYCAEMIC CONTROL
Figure 8.2 Changes from baseline (mean ± SD) in (a) glucose uptake in the brain and (b) hypoglycaemia symptom scores, and plasma concentrations of (c) epinephrine and (d) pancreatic polypeptide during hypoglycaemia in patients with type 1 diabetes with differing degrees of glycaemic control (black bars), and in non-diabetic subjects (grey bars). Reproduced from Boyle et al. (1995) with permission. Copyright © 1995 Massachusetts Medical Society
CEREBRAL ADAPTATION
179
hypoglycaemia. It is becoming evident that changes in symptomatic responses and cortical function in hypoglycaemia are driven by complex mechanisms associated with, but not exclusively controlled directly by, the changes in the glucose metabolic rate of neurones. Some cognitive functions are better preserved than others during hypoglycaemia in subjects who have previous experience of hypoglycaemia than in hypoglycaemia-naive subjects who have normal counterregulation (Fanelli et al., 1993; Boyle et al., 1995). This does not entirely fit the clinical picture of patients becoming significantly confused during hypoglycaemia while remaining asymptomatic. One measure of cognitive function, the choice reaction time, does not appear to adapt, and when hypoglycaemia is induced slowly, it deteriorates at similar levels of blood glucose in most subjects, irrespective of their previous glycaemic experience and their state of hypoglycaemia awareness (Maran et al., 1995). Other measures of cognitive function also deteriorate at similar levels of blood glucose in diabetic subjects who have had very disparate experiences of preceding glycaemia (Widom and Simonson, 1990; Amiel et al., 1991; Hvidberg et al., 1996). The ability of the brain to adapt its metabolic and functional capacity according to previous glycaemic experience varies across different regions of the brain. Regions of the brain that detect hypoglycaemia, and some parts of the cerebral cortex, may be able to adapt more effectively to antecedent hypoglycaemia than other areas, to sustain glucose metabolism during subsequent exposure. As blood glucose falls this would effectively destroy the normal protective hierarchy of corrective and symptomatic responses that precede cognitive impairment, replacing it with the dangerous situation whereby cognitive impairment is the initial response to hypoglycaemia, with autonomic responses not occurring until the blood glucose declines to a much lower level. In this situation the patient becomes too confused and unable to recognise the warning symptoms and so take appropriate corrective action (Figure 8.3).
Figure 8.3 The change in hierarchy of responses to hypoglycaemia (a) before and (b) after intensified insulin therapy in type 1 diabetes mellitus
180
RISKS OF STRICT GLYCAEMIC CONTROL
The magnitude of the change in glycaemic thresholds for various functions of the brain in response to strict control of diabetes is variable. Where glucose thresholds for cognitive dysfunction do alter in people with impaired awareness of hypoglycaemia, the differences between the blood glucose thresholds for the symptomatic and autonomic responses and those for the onset of cognitive impairment are much smaller. As a result, the window of opportunity for the patient to recognise that hypoglycaemia is developing is much narrower, giving less time for corrective action to be taken. As described above, the molecular mechanisms controlling the thresholds for activation of the various components of the counterregulatory responses remain the subject of intense research.
OTHER RISKS OF INTENSIFIED INSULIN THERAPY Diabetic Ketoacidosis and Hyperinsulinaemia Although severe hypoglycaemia was indisputably the major metabolic side-effect of intensive insulin therapy in the DCCT, concerns have been expressed that some intensive treatment regimens may also increase the risk of developing ketosis. This was primarily related to the use of CSII (with insulin pump therapy) and was thought to relate to the absence of any intermediate-acting or background insulin in the event of pump failure. In insulin pump therapy, soluble or fast-acting analogue insulin is delivered steadily by a slow infusion of very low doses throughout the day. The insulin delivery is accelerated before meals to deliver boluses, akin to giving intermittent subcutaneous injections of short-acting insulin. Because the basal insulin is delivered in a very low volume and there is no depot of intermediateacting insulin in the subcutaneous tissues to act as a reservoir, an interruption in the delivery of insulin can rapidly lead to hyperglycaemia and even ketosis, especially if the patient’s blood glucose is already elevated (Castilloa et al., 1996). This may occur as a result of disconnection of the pump, air in the delivery system, blockage in the tubing or more rarely, mechanical failure of the pump. The apparently high risk of diabetic ketoacidosis (DKA) with insulin pump therapy was first described when pumps left the experimental centres where they had been invented and were taken up for more general use (Knight et al., 1985), although the same centre that reported a problem with DKA also reported satisfactory experience overall with pump therapy (Knight et al., 1986). In 1997, a metaanalysis of trials of CSII has indicated that the rate of DKA was significantly higher (Egger et al., 1997) and the rate of development of DKA was also slightly greater in intensivelytreated patients in the DCCT, although many of those patients used multiple injections of insulin to improve their glycaemic control (The Diabetes Control and Complications Trial Research Group, 1995b). However, a more recent meta-analysis, including studies using more modern equipment and up-dated algorithms for using pump therapy, is more reassuring (Pickup et al., 2002). Current intensive treatment regimens focus more on transferring skills of flexible insulin dose adjustment more effectively to the patients and in this setting, DKA rates are not higher. Indeed, experienced centres have utilised insulin pump therapy to help people avoid recurrent DKA (Rodrigues et al., 2005). However, the risk is worth reiterating, as it can return when new technologies are applied in inexperienced settings. Intensive insulin therapy often leads to a redistribution of the times of administration of insulin rather than a straightforward increase in dosage, and concerns have been raised
THERAPEUTIC MANIPULATION
181
Figure 8.4 Severe hypoglycaemia (events per 100 patient-years) at baseline with multiple daily injections (MDI) and by year on continuous subcutaneous insulin infusion (CSII). Copyright © 1996 American Diabetes Association. From Bode et al., 1996. Reprinted with permission from The American Diabetes Association
that continuous peripheral hyperinsulinaemia may be deleterious. This anxiety may be more theoretical than real, as improved glycaemic control in type 1 diabetes improves insulin sensitivity (Simonson et al., 1985b), which ultimately should prevent hyperinsulinaemia. However, achieving adequate plasma concentrations of insulin in the hepatic circulation is always likely to be at the cost of promoting hyperinsulinaemia in the systemic circulation, as insulin has to be delivered by subcutaneous injection. This potential over-insulinisation may contribute to the risk of hypoglycaemia, and when insulin is delivered into the portal system, as with intraperitoneal infusion systems, hypoglycaemia is less frequent at any given blood glucose level (Lassmann-Vague et al., 1996; Dunn et al., 1997). However, one study (Figure 8.4) demonstrated a pronounced and sustained reduction in the frequency of severe hypoglycaemia following the transfer of patients from multiple injections of insulin to CSII (Bode et al., 1996), and so appropriate temporal distribution of the action of insulin may, be the critical factor in preventing hypoglycaemia.
THERAPEUTIC MANIPULATION Avoidance of Hypoglycaemia It is important to stress that management of the potentially devastating syndrome of impaired awareness of hypoglycaemia and deficient counterregulation, with its high risk of severe hypoglycaemia, should not be an excuse for encouraging poor glycaemic control. However, some patients with these acquired syndromes are not suitable for intensive insulin therapy and
182
RISKS OF STRICT GLYCAEMIC CONTROL
very strict glycaemic control, and blood glucose targets may have to be higher when faced with these problems. Frequent blood glucose monitoring is essential to identify biochemical hypoglycaemia, and the use of basal-bolus insulin regimens (which predominantly use shortacting insulins) may be beneficial in avoiding recurrent hypoglycaemia. However, frequent blood glucose monitoring alone can sometimes exacerbate the problem, unless the patient is instructed in how to use the data to adjust insulin regimens prospectively to avoid problems, rather than to react by immediate treatment of the inevitable occasional high reading. There have been few detailed behavioural studies in people with impaired awareness of hypoglycaemia but it may be a particular problem where the links between high blood glucose and risk of vascular complications have been very well accepted by the patient, who may need convincing that transient hyperglycaemia is not a problem. Clinical experience suggests that many patients with such problems ‘glucose chase’, taking corrective action whenever they identify a high blood glucose concentration. Correcting this behaviour, in particular avoiding postprandial glucose correction, and re-training patients to use glucose monitoring to seek patterns for future dose adjustment, can be very beneficial. Programmes that teach patients to test blood glucose before eating and to use the information to adjust the dose of insulin required for the immediate meal, allow people to live with greater flexibility. By recording the blood glucose results, recurrent patterns in changes can be sought against which prospective adjustments to insulin regimens can be made. These measures allow HbA1c to be improved while lowering the risk of severe hypoglycaemia. It is important to recognise that the preservation of physiological defences to hypoglycaemia is dependent upon the avoidance of hypoglycaemia in daily life, and not on tolerance of chronic hyperglycaemia and an elevated glycated haemoglobin. As discussed above, the studies that have attempted to restore awareness of hypoglycaemia by avoidance of hypoglycaemia did not cause any major deterioration of glycaemic control, although a modest increase in HbA1c of around 0.5–1.0% occurred in two studies (Fanelli et al., 1993; Dagogo-Jack et al., 1994). Anecdotally, average blood glucose concentrations and HbA1c may even improve with strategies for avoiding hypoglycaemia, by preventing wide fluctuations in blood glucose. Strategies to avoid hypoglycaemia can be very time-consuming and labour-intensive, for the patient as well as for the physician, and require several supportive measures, such as having to maintain daily telephone contact between the patient and the medical and nursing staff (Fanelli et al., 1993). It took Cranston and colleagues (1994) up to 12 months for the subjects taking part in their study to achieve three consecutive weeks when the home blood glucose readings did not fall below 3.0 mmol/l. Two of three studies demonstrated partial recovery of the counterregulatory responses to hypoglycaemia (Fanelli et al., 1993; Cranston et al., 1994), but one did not (Dagogo-Jack et al., 1994), although the symptoms of hypoglycaemia were restored. The number of subjects was small in all of these studies. However, the studies were often done in patients with an established problem of hypoglycaemia and newer tools are at hand to help prevent or reverse the problems. Educational strategies cannot be emphasised too strongly. They remain to be tested specifically in people who have major problems with recurrent severe hypoglycaemia but their ability to achieve better glycaemic control with less frequent hypoglycaemia is a major improvement over the outcomes of the DCCT. For patients with type 1 diabetes, the introduction of fast-acting insulin analogues for insulin replacement at meal-times has reduced hypoglycaemia, particularly at night, because of the much shorter duration of action of fast-acting analogues (Brunelle et al., 1998; Home et al., 2000). Likewise, the longeracting insulin analogues have been associated with a lower risk of nocturnal hypoglycaemia
PATIENTS UNSUITABLE FOR STRICT CONTROL
183
(Pieber et al., 2000; Vague et al., 2003) and regimens that combine short and longer-acting insulin analogues claim to have a lower risk of hypoglycaemia in patients with type 1 diabetes (Hermansen et al., 2004; Ashwell et al., 2005), although the methods of measurement of hypoglycaemia in many of these trials was much less robust than was discussed earlier. The potential of CSII to reduce hypoglycaemia has also been highlighted. The use of faster and less painful glucose monitoring devices will facilitate home monitoring, although it remains critical that the data obtained are utilised through appropriate patient education. The advent of real-time glucose monitors may allow patients to avoid hypoglycaemia, as they can take action to interrupt a steady decline in blood glucose concentration that they would not previously have had the opportunity to observe. The value of these new technologies remains to be proven in routine clinical use, but they do hold promise. The main defence against recurrent hypoglycaemia with its consequent blunting of subjective symptomatic awareness remains the establishment of therapeutic goals that are realistic for individual patients. The physician’s tendency to concentrate on eliminating hyperglycaemia has led to subnormal blood glucose values being ignored, a practice worsened by the belief of some physicians and patients alike that, because an episode of biochemical hypoglycaemia is asymptomatic, it is not important. There is no doubt that a clinically detectable deterioration in performance of some aspects of cognitive function occurs in human subjects at arterialised blood glucose concentrations of 3.0 mmol/l (see Chapter 2), and an absence of symptoms at that level should ring alarm bells with the patient’s physician. Given that healthy non-diabetic subjects do not commonly exhibit fasting blood glucose concentrations below 4.0 mmol/l, it seems wholly unnecessary to encourage or even permit such subnormality in patients with diabetes (one exception to this maxim being pregnancy where healthy non-diabetic women do exhibit lower blood glucose levels as discussed in Chapter 10). With intensive insulin therapy the therapeutic targets should be near-normal blood glucose levels (before meals 4.0–7.0 mmol/l, after meals 4.0–9.0 mmol/l, depending on time of testing), with a slightly higher than normal glucose at bedtime (7.0–9.0 mmol/l) to reduce the risk of hypoglycaemia occurring during the night. Blood glucose measured during the night may be a little lower (≥ 36 mmol/l), but in view of the evidence presented above, patients should avoid allowing it to fall any lower than this.
PATIENTS UNSUITABLE FOR STRICT CONTROL The DCCT demonstrated that any reduction in glycated haemoglobin is associated with a reduced risk of microvascular complications over time, and the benefits are greater with higher glycated haemoglobin concentrations (The Diabetes Control and Complications Trial Research Group, 1996b). A cross-sectional study which suggested that the risk reduction for nephropathy is near-maximal at a glycated haemoglobin of 8% (Krolewski et al., 1995) cannot be extrapolated to other microvascular complications, because in the DCCT no glycaemic threshold (estimated by glycated haemoglobin) for the development of retinopathy was demonstrated in a patient group whose average HbA1c was 7%. The long-term followup of the DCCT cohort has confirmed that intensive therapy benefits macrovascular as well as microvascular risk and that its effects are sustained (Writing team for the Diabetes Control and Complications Trial/ Epidemiology of Diabetes Interventions and Complications Research Group, 2002). Thus, unless an individual already has normal glycated haemoglobin with no problematical hypoglycaemia, no patient who has diabetes is unsuitable for attempts
RISKS OF STRICT GLYCAEMIC CONTROL
184
to improve their glycaemic control, especially with modern techniques that are able to deliver improved control without increasing hypoglycaemia risk. In practice, however, there are patients in whom attempts to achieve a near normal glycated haemoglobin are not appropriate (Box 8.2). Patients with advanced complications, especially retinopathy, have not been shown to benefit and a sudden improvement in glycaemic control may cause acceleration in severity of pre-proliferative or early proliferative retinopathy (Hanssen et al., 1986). Although some authorities claim that this should not be a contraindication to improving glycaemic control (Chantelau and Kohner, 1997), as yet there is no real evidence for benefit in advanced cases and the retinopathy should be treated appropriately before glycaemic control is intensified. Similarly, in patients with established renal impairment and severe macrovascular disease, attempts to treat elevated blood pressure and plasma lipids and to encourage patients to stop smoking may be more beneficial than targeting glycaemic control alone. As intensive insulin therapy is aimed at achieving benefit over a period of five to ten years or more, patients with a reduced life expectancy should not be exposed to the risks and rigours associated with this treatment regimen. This applies also to elderly patients, who may be frail and physically inactive. Very young patients may not be good candidates for very strict glycaemic control. Poor control should not be encouraged in children, as growth may be jeopardised, and there is some evidence that pre-pubertal glycaemic control may influence the later risk of complications (Donaghue et al., 1997; Holl et al., 1998). However, very small children, who are very insulin-sensitive, may be at risk of intellectual damage if exposed to recurrent severe hypoglycaemia (see Chapters 9 and 13).
Box 8.2
Application of strict glycaemic control in type 1 diabetes
Caution required: • Long duration of insulin-treated diabetes (counterregulatory deficiences). • Previous history of severe hypoglycaemia. • Established impaired awareness of hypoglycaemia. • History of epilepsy. • Patient unwilling to do home blood glucose monitoring. Contraindicated: • Extremes of age. • Ischaemic heart disease. • Unstable diabetic retinopathy (can be instituted after treatment). • Advanced diabetic complications. • Limited life expectancy (e.g. serious coexisting disease).
REFERENCES
185
It is the patient who determines the degree of glycaemic control that they feel is worth the effort. Patients with very erratic life styles, and those who are not prepared to commit themselves to regular self-monitoring of blood glucose, with frequent attention to the timing of injection and adjustment of dosage of insulin, cannot safely undertake measures to achieve near-normoglycaemia. A compromise must be reached after a full discussion of the risks. Patients currently experiencing problematical hypoglycaemia may not wish to aim for glycaemic targets near the normal range, although the regimens of intensive insulin therapy may still be appropriate for them if they can eliminate hypoglycaemia from their daily lives. This may also be true of people undertaking dangerous or physically demanding jobs, who may deliberately set higher blood glucose targets to protect against hypoglycaemia, but who should be encouraged to practice regular self-monitoring and adjustment of insulin doses. It is the informed patient who must determine their own therapeutic aims at any given time. The doctor’s role is to try to ensure that the patient has the knowledge to make appropriate decisions and to provide the tools to achieve these aims.
CONCLUSIONS • The principal risks of intensive insulin therapy are hypoglycaemia and weight gain. • In earlier studies such as the DCCT, patients using intensive insulin regimens were three times more likely to have an episode of severe hypoglycaemia than those on conventional insulin regimens, but newer techniques for training patients to use insulin flexibly can deliver improved glycaemic control without any increase in severe hypoglycaemia, and sometimes with reduced hypoglycaemia occurrence. • It is likely that exposure to hypoglycaemia during intensive therapy impairs the counterregulatory responses to hypoglycaemia and symptomatic awareness and this may be seen particularly if glycated haemoglobin is within the non-diabetic range. • Total avoidance of hypoglycaemia can restore the symptomatic response to hypoglycaemia, but achieving this once problematic hypoglycaemia has been established may be demanding and time-consuming, both for patients and healthcare professionals. Nevertheless, the long-term benefits of good diabetic control in type 1 diabetes are unequivocal and current technologies should help more patients achieve it. • It is essential that all definitions of good glycaemic control include an absence of severe hypoglycaemia as well as near-normal glycated haemoglobin. Intensified treatment regimens should be adjusted to incorporate both.
REFERENCES Agardh CD, Eckert B, Agardh E (1992). Irreversible progression of severe retinopathy in young type-1 insulin-dependent diabetes mellitus patients after improved metabolic control. Journal of Diabetic Complications 6: 96–100. Amiel SA, Tamborlane WV, Simonson DC, Sherwin RS (1987). Defective glucose counterregulation after strict glycemic control of insulin-dependent diabetes mellitus. New England Journal of Medicine 316: 1376–83.
186
RISKS OF STRICT GLYCAEMIC CONTROL
Amiel SA, Sherwin RS, Simonson DC, Tamborlane WV (1988). Effect of intensive insulin therapy on glycemic thresholds for counterregulatory hormone release. Diabetes 37: 901–7. Amiel SA, Pottinger RC, Archibald HR, Chusney G, Cunnah DTF, Prior PF, Gale EAM (1991). Effect of antecedent glucose control on cerebral function during hypoglycemia. Diabetes Care 14: 109–18. Ashwell SG, Amiel SA, Bilous RW, Dashora U, Heller SR, Hepburn DA et al. (2006). Improved glycaemic control with insulin glargine plus insulin lispro: a multicentre, randomized, cross-over trial in people with type 1 diabetes. Diabetic Medicine 23: 285–92. Bingham EM, Dunn JT, Smith D, Sutcliffe-Goulden J, Reed LJ, Marsden PK, Amiel SA (2005). Differential changes in brain glucose metabolism during hypoglycaemia accompany loss of hypoglycaemia awareness in men with type 1 diabetes mellitus. An [11C]-3-O-methyl-D-glucose PET study. Diabetologia 48: 2080–9. Bode BW, Steed RD, Davidson PC (1996). Reduction in severe hypoglycemia with long-term continuous subcutaneous insulin infusion in type 1 diabetes. Diabetes Care 19: 324–7. Bott U, Jorgens V, Grusser M, Bender R, Muhlhauser I, Berger M (1994). Predictors of glycaemic control in type 1 diabetic patients after participation in an intensified treatment and teaching programme. Diabetic Medicine 11: 362–71. Boyle PJ, Nagy RJ, O’Connor AM, Kempers SF, Yeo RA, Qualls C (1994). Adaptation in brain glucose uptake following recurrent hypoglycemia. Proceedings of the National Academy of Sciences of the United States of America 91: 9352–6. Boyle PJ, Kempers SF, O’Connor AM, Nagy RJ (1995). Brain glucose uptake and unawareness of hypoglycemia in patients with insulin-dependent diabetes mellitus. New England Journal of Medicine 333: 1726–31. Brunelle BL, Llewelyn J, Anderson JH Jr, Gale EA, Koivisto VA (1998). Meta-analysis of the effect of insulin lispro on severe hypoglycemia in patients with type 1 diabetes. Diabetes Care 21: 1726–31. Carlson MG, Campbell PJ (1993). Intensive insulin therapy and weight gain in IDDM. Diabetes 42: 1700–7. Castilloa MJ, Sheen AJ, Lefebvre PJ (1996). The degree/rapidity of the metabolic deterioration following interruption of a continuous subcutaneous insulin infusion is influenced by the prevailing blood glucose level. Journal of Clinical Endocrinology and Metabolism 81: 1975–8. Chantelau E, Kohner EM (1997). Why some cases of retinopathy worsen when diabetic control improves. British Medical Journal 315: 1105–6. Clarke WL, Cox DJ, Gonder-Frederick LA, Julian D, Schlundt D, Polonsky W (1995). Reduced awareness of hypoglycemia in adults with IDDM. A prospective study of hypoglycemic frequency and associated symptoms. Diabetes Care 18: 517–22. Cranston I, Lomas J, Maran A, Macdonald IA, Amiel SA (1994). Restoration of hypoglycaemia awareness in patients with long-duration insulin-dependent diabetes. Lancet 344: 283–7. Cranston I, Reed LJ, Marsden PK, Amiel SA (2001). Changes in regional brain (18)Ffluorodeoxyglucose uptake at hypoglycemia in type 1 diabetic men associated with hypoglycemia unawareness and counter-regulatory failure. Diabetes 50: 2329–36. DAFNE Study Group (2002). Training in flexible, intensive insulin management to enable dietary freedom in people with type 1 diabetes: Dose Adjustment For Normal Eating (DAFNE) randomised controlled trial. British Medical Journal 325: 746. Dagogo-Jack S, Rattarasarn C, Cryer PE (1994). Reversal of hypoglycemia unawareness, but not defective glucose counterregulation, in IDDM. Diabetes 43: 1426–34. Davis SN, Shavers C, Mosqueda-Garcia R, Costa F (1997). Effects of differing antecedent hypoglycemia on subsequent counterregulation in normal humans. Diabetes 46: 1328–35. Donaghue KC, Fung AT, Hing S, Fairchild J, King J, Chan A et al. (1997). The effect of prepubertal diabetes duration on diabetes microvascular complications in early and late adolescence. Diabetes Care 20: 77–80. Dunn FL, Nathan DM, Scavini M, Selam JL, Wingrove TG (1997). Long term therapy of IDDM with an implantable insulin pump. The Plantable Insulin Pump Trial Study Group. Diabetes Care 20: 59–63.
REFERENCES
187
Egger M, Davey Smith G, Stettler C, Diem P (1997). Risk of adverse effects of intensified treatment in insulin-dependent diabetes mellitus: a meta-analysis. Diabetic Medicine 14: 919–28. Fanelli CG, Epifano L, Rambotti AM, Pampanelli S, Di Vincenzo A, Modanelli F et al. (1993). Meticulous prevention of hypoglycemia normalizes the glycemic thresholds and magnitude of most of neuroendocrine responses to, symptoms of, and cognitive function during hypoglycemia in intensively treated patients with short-term IDDM. Diabetes 42: 1683–9. Franciosi M, Pellegrini F, De Berardis G, Belfiglio M, Cavaliere D, Di Nardo B et al., QuED Study Group (2001). Diabetic patients: an urgent need for better educational strategies. Diabetes Care 24: 1870–7. George E, Harris N, Bedford C, Macdonald IA, Hardisty CA, Heller SR (1995). Prolonged but partial impairment of the hypoglycaemic physiological response following short-term hypoglycaemia in normal subjects. Diabetologia 38: 1183–90. George E, Marques JL, Harris ND, Macdonald IA, Hardisty CA, Heller SR (1997). Preservation of physiological responses to hypoglycemia 2 days after antecedent hypoglycemia in patients with IDDM. Diabetes Care 20: 1293–8. Gold AE, MacLeod KM, Frier BM (1994). Frequency of severe hypoglycemia in patients with type 1 diabetes with impaired awareness of hypoglycemia. Diabetes Care 17: 697–703. Hanssen KF, Dahl-Jorgensen K, Lauritzen T, Feldt-Rasmussen B, Brinchmann-Hansen O, Deckert T (1986). Diabetic control and microvascular complications: the near-normoglycaemic experience. Diabetologia 29: 677–84. Heller SR, Cryer PE (1991). Reduced neuroendocrine and symptomatic responses to subsequent hypoglycemia after 1 episode of hypoglycemia in non-diabetic humans. Diabetes 40: 223–6. Heller S, Chapman J, McLoud J, Ward J (1995). Unreliability of reports of hypoglycaemia by diabetic patients. British Medical Journal 310: 440. Hepburn DA, Patrick AW, Eadington DW, Ewing DJ, Frier BM (1990). Unawareness of hypoglycaemia in insulin-treated diabetic patients: prevalence and relationship to autonomic neuropathy. Diabetic Medicine 7: 711–7. Hermansen K, Fontaine P, Kukolja KK, Peterkova Y, Leth G, Gall MA (2004). Insulin analogues (insulin detemir and insulin aspart) versus traditional human insulins (NPH insulin and regular human insulin) in basal-bolus therapy for patients with type 1 diabetes. Diabetologia 47: 622–9. Holl RW, Lang GE, Grabert M, Heinze E, Lang GK, Debatin KM (1998). Diabetic retinopathy in pediatric patients with type 1 diabetes: effect of diabetes duration, prepubertal and pubertal onset of diabetes, and metabolic control. Journal of Pediatrics 132: 790–4. Home PD, Lindholm A, Riis A, European Insulin Aspart Study Group (2000). Insulin aspart versus human insulin in the management of long-term blood glucose control in type 1 diabetes mellitus: a randomized controlled trial. Diabetic Medicine 17: 762–70. Hoogma RP, Hammond PJ, Gomis R, Kerr D, Bruttomesso D, Bouter KP et al. 5 Nations Study Group (2006). Comparison of the effects of continuous subcutaneous insulin infusion (CSII) and NPH-based multiple daily insulin injections (MDI) on glycaemic control and quality of life: results of the 5-nations trial. Diabetic Medicine 23: 141–7. Hvidberg A, Fanelli CG, Hershey T, Terkamp C, Craft S, Cryer PE (1996). Impact of recent antecedent hypoglycemia on hypoglycemic cognitive dysfunction in non-diabetic humans. Diabetes 45: 1030–6. Jorgens V, Grusser M, Bott U, Mulhauser I, Berger M (1993). Effective and safe translation of intensified insulin therapy to general internal medicine departments. Diabetologia 36: 99–105. Kinsley BT, Widom B, Simonson DC (1995). Differential regulation of counterregulatory hormone secretion and symptoms during hypoglycemia in IDDM. Effect of glycemic control. Diabetes Care 18: 17–26. Knight G, Jennings AM, Boulton AJ, Tomlinson S, Ward JD (1985). Severe hyperkalaemia and ketoacidosis during routine treatment with an insulin pump. British Medical Journal 291: 371–2. Knight G, Boulton AJ, Drury J, Ward JD (1986). Long term glycaemic control by alternative regimens in a feasibility study of continuous subcutaneous insulin infusion. Diabetes Research 3: 335–8.
188
RISKS OF STRICT GLYCAEMIC CONTROL
Korzon-Burakowska A, Hopkins D, Matyka K, Lomas J, Pernet A, Macdonald I, Amiel S (1998). Effects of glycemic control on protective responses against hypoglycemia in type 2 diabetes. Diabetes Care 21: 283–90. Krolewski AS, Laffel LM, Krolewski M, Quinn M, Warram JH (1995). Glycosylated hemoglobin and the risk of microalbuminuria in patients with insulin-dependent diabetes mellitus. New England Journal of Medicine 332: 1251–5. Lassmann-Vague V, Belicar P, Alessis C, Raccah D, Vialettes B, Vague P (1996). Insulin kinetics in type I diabetic patients treated by continuous intraperitoneal insulin infusion: influence of anti-insulin antibodies. Diabetic Medicine 13: 1051–5. Limbert C, Schwingshandl J, Haas J, Roth R, Borkenstein M (1993). Severe hypoglycemia in children and adolescents with IDDM: frequency and associated factors. Journal of Diabetes Complications 7: 216–20. Liu D, McManus RM, Ryan EA (1996). Improved counter-regulatory hormonal and symptomatic responses to hypoglycemia in patients with insulin-dependent diabetes mellitus after 3 months of less strict glycemic control. Clinical and Investigative Medicine 19: 71–82. Maran A, Lomas J, Macdonald IA, Amiel SA (1995). Lack of preservation of higher brain function during hypoglycaemia in patients with intensively-treated IDDM. Diabetologia 38: 1412–18. Muhlhauser I, Overmann H, Bender R, Bott U, Berger M (1998). Risk factors of severe hypoglycaemia in adult patients with type I diabetes – a prospective population based study. Diabetologia 41: 1274–82. Nathan DM, Lachin J, Cleary P, Orchard T, Brillon DJ, Backlund JY et al. (2003). Intensive diabetes therapy and carotid intima-media thickness in type 1 diabetes mellitus. New England Journal of Medicine 348: 2294–303. Pampanelli S, Fanelli C, Lalli C, Gofetta M, Del Sindaco P, Lepore M et al. (1996). Long-term intensive therapy in IDDM: effects on HbA1c , risk for severe and mild hypoglycaemia, status of counterregulation and awareness of hypoglycaemia. Diabetologia 39: 677–86. Pedersen-Bjergaard U, Agerholm-Larsen B, Pramming S, Hougaard P, Thorsteinsson B (2003). Prediction of severe hypoglycaemia by angiotensin-converting enzyme activity and genotype in type 1 diabetes. Diabetologia 46: 89–96. Pickup J, Mattock M, Kerry S (2002). Glycaemic control with continuous subcutaneous insulin infusion compared with intensive insulin injections in patients with type 1 diabetes: meta-analysis of randomised controlled trials. British Medical Journal 324: 705–8. Pieber TR, Eugene-Jolchine I, Derobert E (2000). Efficacy and safety of HOE 901 versus NPH insulin in patients with type 1 diabetes. The European Study Group of HOE 901 in type 1 diabetes. Diabetes Care 23: 157–62. Plank J, Kohler G, Rakovac I, Semlitsch BM, Horvath K, Bock G et al. (2004). Long-term evaluation of a structured outpatient education programme for intensified insulin therapy in patients with type 1 diabetes: a 12-year follow-up. Diabetologia 47: 1370–75. Rodrigues I, Reid HA, Ismail K, Amiel SA (2005). Indications and efficacy of continuous subcutaneous insulin infusion (CSII) therapy in type 1 diabetes mellitus: a clinical audit in a specialist service. Diabetic Medicine: 22: 842–9. Ryder RE, Owens DR, Hayes TM, Ghatei MA, Bloom SR (1990). Unawareness of hypoglycaemia and inadequate hypoglycaemic counterregulation: no causal relation with diabetic autonomic neuropathy. British Medical Journal 301: 783–7. Samann A, Muhlhauser I, Bender R, Kloos C, Muller UA (2005). Glycaemic control and severe hypoglycaemia following training in flexible, intensive insulin therapy to enable dietary freedom in people with type 1 diabetes: a prospective implementation study. Diabetologia 48: 1965–70. Segal SA, Fanelli CG, Dence CS, Markham J, Videen TO, Paramore DS et al. (2001). Blood-to-brain glucose transport, cerebral glucose metabolism, and cerebral blood flow are not increased after hypoglycemia. Diabetes 50: 1911–7. Shah SC, Malone JL, Simpson NE (1989). A randomized trial of intensive insulin therapy in newly diagnosed insulin-dependent diabetes mellitus. New England Journal of Medicine 320: 550–4.
REFERENCES
189
Simonson DC, Tamborlane WV, DeFronzo RA, Sherwin RS (1985a). Intensive insulin therapy reduces counterregulatory hormone responses to hypoglycemia in patients with type 1 diabetes. Annals of Internal Medicine 103: 184–90. Simonson DC, Tamborlane WV, Sherwin RS, Smith JD, DeFronzo RA (1985b). Improved insulin sensitivity in patients with type 1 diabetes mellitus after CSII. Diabetes 34 (suppl 3): 80–6. The Diabetes Control and Complications Trial Research Group (1987). The Diabetes Control and Complications Trial (DCCT). Results of a feasibility study. Diabetes Care 10: 1–19. The Diabetes Control and Complications Trial Research Group (1988). Weight gain associated with intensive therapy in the Diabetes Control and Complications Trial. Diabetes Care 11: 567–73. The Diabetes Control and Complications Trial Research Group (1991). Epidemiology of severe hypoglycemia in the Diabetes Control and Complications Trial. American Journal of Medicine 90: 450–9. The Diabetes Control and Complications Trial Research Group (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New England Journal of Medicine 329: 977–86. The Diabetes Control and Complications Trial Research Group (1995a). Adverse events and their association with treatment regimens in the Diabetes Control and Complications Trial. Diabetes Care 18: 1415–27. The Diabetes Control and Complications Trial Research Group (1995b). Implementation of treatment protocols in the Diabetes Control and Complications Trial. Diabetes Care 18: 361–76. The Diabetes Control and Complications Trial Research Group (1996a). Influence of intensive diabetes treatment on quality-of-life outcomes in the Diabetes Control and Complications Trial. Diabetes Care 19: 195–203. The Diabetes Control and Complications Trial Research Group (1996b). The absence of a glycemic threshold for the development of long-term complications: the perspective of the Diabetes Control and Complications Trial. Diabetes 45: 1289–98. The Diabetes Control and Complications Trial Research Group (1997). Hypoglycemia in the Diabetes Control and Complications Trial. Diabetes 46: 271–86. The Diabetes Control and Complications Trial Research Group (1998). Effect of intensive therapy on residual beta-cell function in patients with type 1 diabetes in the Diabetes Control and Complications Trial. A randomized, controlled trial. Annals of Internal Medicine 128: 517–23. Vague P, Selam JL, Skeie S, De Leeuw I, Elte JW, Haahr H et al. (2003). Insulin detemir is associated with more predictable glycemic control and reduced risk of hypoglycemia than NPH insulin in patients with type 1 diabetes on a basal-bolus regimen with premeal insulin aspart. Diabetes Care 26: 590–6. White NH, Skor DA, Cryer PE, Levandoski LA, Bier DM, Santiago JV (1983). Identification of type 1 diabetic patients at increased risk for hypoglycemia during intensive therapy. New England Journal of Medicine 308: 485–91. Widom B, Simonson DC (1990). Glycemic control and neuropsychologic function during hypoglycemia in patients with insulin-dependent diabetes mellitus. Annals of Internal Medicine 112: 904–12. Working Group on Hypoglycemia, American Diabetes Association (2005). Defining and reporting hypoglycemia in diabetes: a report from the American Diabetes Association Workgroup on Hypoglycemia. Diabetes Care 28: 1245–9. Writing team for the Diabetes Control and Complications Trial/ Epidemiology of Diabetes Interventions and Complications Research Group (2002). Effect of intensive therapy on the microvascular complications of type 1 diabetes mellitus. Journal of the American Medical Association 287: 2563–9. Yki-Jarvinen H, Kauppinen-Makelin R, Tiikkainen M, Vahatalo M, Virtamo H, Nikkila K et al. (2006). Insulin glargine or NPH combined with metformin in type 2 diabetes: the LANMET study. Diabetologia 49: 442–51.
9 Hypoglycaemia in Children with Diabetes Krystyna A. Matyka
INTRODUCTION Sub-optimal care of children with type 1 diabetes mellitus carries devastating consequences. Young children, previously thought to be protected from the early development of microvascular complications, have been found to be at significant risk of these complications that can present in adolescence (Danne et al., 1994; Solders et al., 1997; Schultz et al., 1999). Yet they are also at risk of detrimental neuropsychological sequelae, which are thought to be related to recurrent episodes of hypoglycaemia that may accompany intensification of insulin therapy aimed at decreasing the risk of these microvascular complications. The exaggerated metabolic demands of the growing child, combined with a lifestyle that is unpredictable even on a day to day basis, make children very vulnerable to both repeated and severe episodes of hypoglycaemia (Allen et al., 2001). This chapter examines the aetiology, physiology, consequences and management of episodes of hypoglycaemia during this dynamic time of life.
DEFINITION OF HYPOGLYCAEMIA IN CHILDHOOD The definition of hypoglycaemia in childhood has been extremely controversial. It has been suggested that children can tolerate lower levels of blood glucose, especially as the developing brain can use alternative substrates for cerebral metabolism. This has been supported by the clinical finding that some children with diabetes appear to be ‘normal’ when their blood glucose concentrations are low as demonstrated by home blood glucose monitoring. However, difficulties arise as young children are not expected to perform complex psychomotor tasks and clinical detection of mild changes in performance can be difficult. This subject is not easy to study because of the ethical difficulties of performing studies of normal glucose homeostasis during fasting in young children, which could be used to define the limits of normality of blood glucose concentrations. Fasting glucose requirements will, in part, be determined by brain glucose uptake and the central nervous system has a pivotal role in carbohydrate metabolism throughout life. The brain of infants and children can use glucose at a rate of 3–5 mg/kg/min, equivalent to almost all endogenous production as defined by stable isotope studies of cerebral glucose metabolism, and a linear correlation exists between glucose production and estimated brain
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
192
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
size, from premature infants to adult life (Bier et al., 1977). There is a marked change in the correlation between body weight and hepatic glucose production corresponding to mid-puberty (approximately 40 kg) indicating that brain growth is virtually complete (Bier et al., 1977). However, the developing brain also has the ability to use alternative substrates for cerebral metabolism, and during fasting young children have higher concentrations of ketones and lactate when compared to adults (Haymond et al., 1982). A number of studies have examined metabolic parameters, including blood glucose concentrations, following fasts of varying duration in children. On the whole these studies have been used to provide normative data for use in the clinical evaluation of children with potential disorders of metabolism; as a result the fasts have been prolonged and sampling infrequent. There is little detailed information on metabolic variables during a short fast as is experienced by children with diabetes, for example when they go to sleep. One study of nocturnal glucose homeostasis in 39 children, subsequently found to have a constitutional short stature, demonstrated a cyclical variation in blood glucose concentrations during the night with a periodicity of 80–120 minutes (Stirling et al., 1991). At some stage of the night blood glucose levels fell transiently to below 3 mmol/l in 5% of the children, but it is difficult to know if these children are representative of those who grow normally. Other studies have examined glucose and metabolite profiles intermittently during prolonged fasts of 24–36 hour duration (Chaussain, 1973; Chaussain et al., 1974; Chaussain et al., 1977; Saudubray et al., 1981; Haymond et al., 1982; Kerr et al., 1983; Lamers et al., 1985a; Lamers et al., 1985b). Fasting glucose concentrations varied depending on the duration of the fast and age of child studied. After 24 hours, blood glucose was found to range from 3.0–3.5 mmol/l. Some of these studies suggested a positive correlation between fasting blood glucose concentrations and age (Chaussain et al., 1977; Saudubray et al., 1981; Lamers et al., 1985a).
Definition of Hypoglycaemia in Childhood Diabetes Although there has been great controversy regarding the biochemical definition of hypoglycaemia for the diagnosis of pathological states in the paediatric population (Koh et al., 1988), these arguments should not apply to type 1 diabetes. The management of type 1 diabetes involves striving towards the maintenance of glucose concentrations well within the physiological range rather than merely outside of the pathological range. The only study from those discussed above to examine glucose concentrations after a duration of fasting that would represent that occurring as part of normal daily living (14 hours) found a fasting glucose concentration of 43 ± 01 mmol/l in children aged 3–15 years old (Lamers et al., 1985b). Guidelines now suggest that diabetes management should aim to keep plasma glucose above 4 mmol/l (‘four should be the floor’ was recommended in a Diabetes UK report in 1996), and this is probably a reasonable recommendation for children. Both in research studies and in clinical care, hypoglycaemia is often subdivided into degrees of severity based on the intervention required. An example of such a classification that was suggested by the International Society for Paediatric and Adolescent Diabetologists (ISPAD) is shown in Table 9.1. It should be noted that the usual adult definition of ‘mild’ (i.e., self-treated) hypoglycaemia cannot be applied in young children who rely on their adult carers for their diabetes management, because they are unlikely to be able to treat the episodes themselves.
SIGNS AND SYMPTOMS OF HYPOGLYCAEMIA
193
Table 9.1 Example of classification of hypoglycaemia (data sourced from ISPAD Consensus Guidelines, 2000) Grade Mild (Grade 1) Moderate (Grade 2) Severe (Grade 3)
Description Aware of, responds to and self-treats the hypoglycaemia Cannot respond to hypoglycaemia and requires help from someone else, but oral treatment is successful Semi-conscious or unconscious or in coma ± convulsions and may require parenteral therapy (glucagon or IV glucose)
PREVALENCE OF HYPOGLYCAEMIA As the definition of hypoglycaemia has been controversial, studies of prevalence have often used variable definitions, both for daytime hypoglycaemia and for that occurring during sleep. Hypoglycaemia is notoriously under-reported, as episodes, particularly mild ones, are quickly forgotten. Even severe episodes may be overlooked. The individual themselves may have amnesia for the event and, if it occurred away from their normal environment, nobody may document what took place. A number of studies have examined the prevalence of severe hypoglycaemia in the paediatric population (Table 9.2). Some studies included only episodes of coma/convulsion whereas others also included those events in which neurological impairment was severe enough to require intervention. Hypoglycaemia has been shown to be a significant problem but, given the methodological complexities, prevalence rates of severe hypoglycaemia have varied greatly from 3.1 to 85.7 episodes per 100 patient-years. It is likely that less severe episodes are much more common and under-reported.
Nocturnal Hypoglycaemia Studies have also examined the prevalence of nocturnal hypoglycaemia. In these studies a biochemical rather than a symptomatic definition of hypoglycaemia was used and has been very variable, ranging from 3.0 to 3.8 mmol/l. The prevalence of hypoglycaemia has varied from 10 to 55% but it is noticeable that the more recent studies have detected a higher prevalence (Beregszaszi et al., 1997; Lopez et al., 1997; Porter et al., 1997; Matyka et al., 1999a). The majority of these episodes do not wake the patient and, in everyday life, a great number of episodes of nocturnal hypoglycaemia will be completely undetected.
SIGNS AND SYMPTOMS OF HYPOGLYCAEMIA Daytime Hypoglycaemia The classical symptoms of hypoglycaemia, as described in adults (Chapter 2), are classified into three distinct groups: autonomic, neuroglycopenic and non-specific (Deary et al., 1993). In adults the sympathoadrenal response during hypoglycaemia is primarily responsible for the classical autonomic symptoms which alert the individual to the falling glucose
195
2897 146
13–17
3.2–25.5
0–18
1–18
0–18
0–18
1–19
1–24
1–19
0–34
DCCT (1994)
Bognetti et al. (1997)
Mortensen et al. (1997)
Nordfelt and Ludvigsson (1997)
Davis et al. (1997)
Davis et al. (1998)
Rosilio et al. (1998)
Tupola et al. (1998b)
Thomsett et al. (1999)
Allen et al. (2001) 415
268
376
2579
709
657
187
Number of subjects
Age (years)
Study
prospective prospective cross-sectional
prospective retrospective
coma ± seizures; requiring assistance coma ± seizures; requiring assistance coma ± seizures or glucagon injection
coma ± seizures or glucagon injection coma ± seizures
prospective
prospective prospective
coma ± seizures; requiring assistance
coma
cross-sectional
retrospective
prospective
Study type
coma ± seizures
coma, seizure or requiring assistance
Definition
7% of patients 4 yrs after diagnosis; 4% after 6.5 years
5
3.1
45
7.8 15.4
4.8 13.1
15–19 10.1–12.6
22
14.9
85.7
No. episodes/ 100 patient years
lower HbA1c ; older age
younger age; lower HbA1c ; number of clinic visits
lower HbA1c ; higher insulin dose
lower HbA1c ; more exercise; more daily blood glucose measurements
younger age lower HbA1c
younger age; lower HbA1c
lower age at onset; longer disease duration; lower HbA1c
younger age; lower HbA1c
younger age
lower HbA1c
Correlations
Table 9.2 Summary of studies examining prevalence of severe hypoglycaemia since the Diabetes Control and Complications trial in 1993
7–16
0–19
1.2–15.8
11–18
1.6–17.1
0–9
Levine et al. (2001)
Rewers et al. (2002)
Craig et al. (2002)
Holl et al. (2003)
Vanelli et al. (2005)
Wagner et al. (2005)
6309
3560
872
1190
1243
300 prospective
prospective
cross-sectional cross-sectional
coma ± seizures
coma ± seizures
coma ± seizures coma ± seizures ; episodes requiring parenteral therapy; episodes requiring outside assistance episodes requiring outside assistance prospective
prospective
episodes requiring outside assistance
22.6
17.6
17.9
36
19
62
younger age; longer diabetes duration; higher insulin dose; insulin regimen; centre experience
none apparent
not evaluated
younger age; males; > 3 insulin injections/day; longer diabetes duration
young age; increased diabetes duration; lower HbA1c ; underinsurance
lower HbA1c ; younger age
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
196
concentration so that corrective action can be taken. The classical autonomic symptoms occur in adults between 3.0 and 3.6 mmol/l and include: sweating, palpitations, hunger and shaking. If glucose concentrations fall further, neuroglycopenia will develop, usually at a glucose concentration around 2.8 mmol/l, and if it falls further the individual may not be able to correct the hypoglycaemia themselves. The most common neuroglycopenic symptoms are: confusion, drowsiness, odd behaviour, speech difficulty and incoordination. If blood glucose continues to fall, coma or convulsion could ensue although the glycaemic threshold at which this occurs is not certain. The symptoms of hypoglycaemia in childhood differ from adults. In one study, children and parents were asked which symptoms and signs alerted them to an episode of hypoglycaemia (McCrimmon et al., 1995). The authors reported that symptoms did not separate into distinct autonomic and neuroglycopenic categories and behavioural symptoms were prominent. Another study examined the frequency of hypoglycaemia in a group of children and adolescents using a three-month diary (Tupola et al., 1998a). Episodes of hypoglycaemia (defined as a blood glucose ≤ 3 mmol/l) were documented along with symptoms. Of the patients (aged 2.5–21 years), 52% had a total of 287 episodes of hypoglycaemia, the majority of which (77%) were mild. The most common presenting symptoms were weakness, tremor, hunger and drowsiness; 39% of symptoms were classified as ‘adrenergic’ (autonomic) and 61% as neuroglycopenic or non-specific behavioural (Table 9.3). The dominant symptoms were different in different age groups of children. Those children less than six years of age had fewer autonomic symptoms than adolescents, with the commonest symptom being drowsiness in young children and tremor in the older children (Tupola et al., 1998a).
Nocturnal Hypoglycaemia The majority of episodes of nocturnal hypoglycaemia do not awaken the child from sleep and thus go undetected, even in those who have normal awareness of hypoglycaemia during waking hours (Gale and Tattersall, 1979; Bendtson et al., 1993; Porter et al., 1996; Table 9.3 Reported symptoms during 221 episodes of mild hypoglycaemia (data sourced from Tupola et al., 1998b) Symptom
Prevalence (%)
autonomic
tremor hunger sweating
22 14 4
neuroglycopenic
drowsiness irritability/aggressiveness dizziness poor concentration blurred vision
34 4 1 1 1
non-specific
weakness nausea abdominal pain headache tearfulness
7 7 2 2 1
RISK FACTORS FOR HYPOGLYCAEMIA
197
Beregszaszi et al., 1997; Lopez et al., 1997; Matyka et al., 1999a). The reasons for this lack of awareness of nocturnal episodes of hypoglycaemia are unclear, but are discussed in Chapter 4. As symptom generation depends on intact autonomic and counterregulatory defence mechanisms, it has been postulated that diminished counterregulation overnight may result in lack of arousal when hypoglycaemia occurs during sleep (Bendtson et al., 1993; Jones et al., 1998).
RISK FACTORS FOR HYPOGLYCAEMIA The discussion of risk factors naturally overlaps with that of prevention of hypoglycaemia, and a fuller review of specific interventions is provided in the section entitled ‘Prevention’ on page 207. Table 9.3 presents the results of studies that have been performed to examine the prevalence of severe hypoglycaemia in childhood since the Diabetes Control and Complications Trial (DCCT) was published (The Diabetes Control and Complications Trial Research Group, 1993). Although a number of studies have suggested correlations between younger age and strict glycaemic control, it is important to note that many individual episodes of hypoglycaemia may be explained by missed meals or unplanned exercise, but this would not have been addressed in epidemiological studies.
Glycaemic Control Insulin requirements vary with age and are approximately 0.5–1 U/kg/day before puberty and 1.5–2 U/kg/day during adolescence, reflecting the insulin resistance that is present during this period of rapid growth and development (Dunger, 1992). Despite numerous developments in terms of novel insulin preparations and modes of delivery, people with type 1 diabetes still experience varying states of insulin deficiency or excess that are difficult to control and predict. This is probably most evident in adolescents with type 1 diabetes in whom peripheral hyperinsulinaemia is achieved in an attempt to replace adequate levels of insulin in the portal circulation during the pubertal growth spurt (Dunger, 1992). The DCCT highlighted the dilemma faced by all patients with type 1 diabetes (The Diabetes Control and Complications Trial Research Group, 1993). Attempts at improving glycaemic control, by intensifying diabetes management, in an effort to decrease the likelihood of the long-term microvascular complications of diabetes led to a significant increase in the risk of severe hypoglycaemia (SH). In the DCCT, subjects in the intensified treatment group had a three-fold higher risk of SH (The Diabetes Control and Complications Trial Research Group, 1993). A group of 195 adolescents, aged between 13 and 17 years, took part in this trial (The Diabetes Control and Complications Trial Research Group, 1994). Although the benefits of improved glycaemic control in terms of microvascular complications were still significant, the adolescents found it more difficult to achieve the low HbA1c concentration than adults (806 ± 013 versus 712 ± 003%; p < 0001). Despite this, adolescents had a greater tendency towards experiencing severe hypoglycaemia: 85.7 events per 100 patient-years versus 56.9 events in the adult cohort (The Diabetes Control and Complications Trial Research Group, 1994). However, in a European-wide clinical audit, which was designed to look at metabolic control in children and adolescents, it was found that severe hypoglycaemia was as common in those centres where metabolic control was poor, as in
198
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
We do not have rights to reproduce this figure electronically
Figure 9.1 Rates of (a) Severe hypoglycaemia, and (b) average HbA1c by calender year. Reproduced from Bulsara et al. (2004) with permission from The American Diabetes Association
hypoglycaemia was as common in those centres where metabolic control was poor, as in those centres that achieved better control as judged by HbA1c , suggesting that research does not always reflect clinical experience (Mortensen et al., 1997). Since 1993, ample opportunity has been present to refine the approaches to intensive insulin therapy and to improve education both for patients and physicians. Longitudinal studies of the incidence of hypoglycaemia are unusual but one audit study from a large paediatric clinic in Western Australia demonstrated an interesting trend in incidence of SH over a period of ten years (Bulsara et al., 2004). Over the first five years of the study, the incidence increased by 29% in conjunction with a decline in the average HbA1c of about 0.2% per year. Despite a continued improvement in glycaemic control, the incidence of SH appeared to plateau at this clinical centre suggesting that improved diabetes management, from more effective insulin regimens or better education, can improve blood glucose concentrations without a concomitant increase in incidence of hypoglycaemia (Figure 9.1).
Nocturnal Insulin Requirements The mismatch of insulin delivery and insulin requirements on standard insulin regimens is particularly evident during the night and most episodes of severe hypoglycaemia occur during sleep (Edge et al., 1990a; The Diabetes Control and Complications Trial Research Group, 1997). Current insulin replacement regimens tend to result in hyperinsulinaemia in the early part of the night, although physiological insulin requirement is at its nadir between 24:00–03:00 hours, and so exacerbates the risk of hypoglycaemia at this time (Matyka et al., 1999a; Mohn et al., 1999; Ford-Adams et al., 2003). Insulin requirements then peak between 04:00–08:00 hours and a ‘dawn phenomenon’ occurs which can lead to fasting hyperglycaemia (Bolli and Gerich, 1984; Edge et al., 1990b) and is thought to result from GH secretion during the later part of the night (De Feo et al., 1990; Edge et al., 1990b)
RISK FACTORS FOR HYPOGLYCAEMIA
199
exacerbated further by the delayed effects of daytime physical activity on muscle glucose metabolism and the prolonged period of fasting that occurs overnight, especially in young children. This suggests that the overnight period is the time of greatest hypoglycaemia risk (The Diabetes Control and Complications Trial Research Group, 1997).
Intensive Insulin Regimens Few studies have examined the impact of insulin regimen on the risk of hypoglycaemia in children. The DCCT did find a significantly higher risk of hypoglycaemia among the 195 adolescents who participated in the study although this was a comparison of overall glycaemic control and not of specific regimens (The Diabetes Control and Complications Trial Research Group, 1994). A number of studies examining prevalence of hypoglycaemia have found an inverse correlation between hypoglycaemia risk and glycated haemoglobin (Table 9.2). The Hvidore Study Group has formed a collaboration between 21 international paediatric centres from 18 countries (Holl et al., 2003). The group surveyed paediatric diabetes management of 2873 children aged up to 18 years in 1995 and restudied 872 of these children in 1998. Although the use of multiple injection regimens increased from 42% to 71% this did not lead to an improvement in glycaemic control as judged by glycated haemoglobin concentrations. Although there was a tendency towards an increase in the frequency of severe hypoglycaemia in the group of children/adolescents who had had an intensification of insulin regimen, this did not reach statistical significance, perhaps because of the low number of events recorded (Holl et al., 2003). Another study of more than 6000 children has suggested that injection regimen and centre experience, as judged by the size of the clinic, may be significant risk factors for severe hypoglycaemia (Wagner et al., 2005). In this study of children aged up to nine years, an increased risk of hypoglycaemia was observed in those children taking four insulin injections daily or on insulin pump therapy, compared to those children taking one to three injections daily. It is important to note that even the more recent studies do not include data acquired since the introduction of the insulin analogues or the use of more physiological and intensive insulin regimens. Small studies of a few children in which insulin analogues have been compared with human insulins suggest that the risk of hypoglycaemia may be lower with analogues without compromising glycaemic control (see later section on hypoglycaemia prevention on page 209).
Diet Children with type 1 diabetes have the same nutritional requirements as children without diabetes. However, meals and snacks containing a high proportion of carbohydrate, have to be regularly distributed throughout the day to avoid the extremes of hypo- and hyperglycaemia (Magrath et al., 1993). This can be an issue for children who do not want to be different from their peers and do not want to eat at times when there friends are playing. Toddlers can present a special problem as many do not eat regular meals but graze during the day. Surprisingly, there has been little systematic study of the role of both quantity and quality of dietary components on the risk of hypoglycaemia. However, some studies of the
200
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
prevalence of hypoglycaemia have found that a number of episodes can be attributed to missed meals (Daneman et al., 1989; Davis et al., 1997; Tupola et al., 1998b). The impact of dietary interventions in the avoidance of hypoglycaemia, mainly during sleep, have been examined, and this is discussed later in the section on hypoglycaemia prevention.
Physical Activity As early as 1926, it was found that exercise could potentiate the hypoglycaemic effect of insulin in patients with type 1 diabetes (Lawrence, 1926). During the first 5–10 minutes of exercising, muscle glycogen is used as the primary source of energy (Price et al., 1994). Subsequently, fuel is provided increasingly by circulating glucose, through gluconeogenesis and free fatty acids (FFAs), the release of which are under hormonal control and dependent predominantly on the portal glucagon : insulin ratio (Ahlborg et al., 1974). The acute effects of exercise are followed by restoration of the metabolic milieu. Muscle glucose uptake remains elevated as glycogen stores are replenished and although insulin sensitivity is enhanced in the period after exercise, increased glucose uptake by skeletal muscle can occur even in the absence of insulin (Cartee and Holloszy, 1990). The time taken to restore muscle glycogen to pre-exercise levels will depend on the intensity and duration of the exercise performed, and the timing and amount of dietary carbohydrate intake, but it can take several hours – typically 6–20 hours (Ivy and Holloszy, 1981; Richter, 1996). Until recently there has been little systematic study of the impact of exercise in childhood on glucose homeostasis. One study used continuous glucose monitoring to study a standardised exercise protocol in a group of children who were using continuous subcutaneous insulin infusion (CSII). Glucose profiles were examined both during and after exercise on a cycle ergometer with the infusion pump either switched on or off (Admon et al., 2005). Hypoglycaemia was more common after exercise than during it, and this was true whether CSII was on or off. All subjects had one to three episodes of symptomatic hypoglycaemia within 2.5 to 12 hours after exercise and four subjects had asymptomatic hypoglycaemia during exercise, only one of whom had consumed extra carbohydrate because their preexercise blood glucose had been below 5.5 mmol/l. Another study examined the impact of daytime exercise on overnight blood glucose profiles in 50 subjects aged 10–18 years on intensive insulin regimens (Tsalikian et al., 2005). On one occasion they were studied during a day of afternoon exercise, involving four periods of 15 minutes each on a treadmill at a heart rate estimated to be 55% of maximum effort for this age group, and on a separate occasion during a rest day. In this study, 22% of subjects developed hypoglycaemia during exercise; overnight hypoglycaemia was more common during the night after the afternoon exercise than during a night after the rest day (Tsalikian et al., 2005).
Age Studies of prevalence of hypoglycaemia have consistently found that younger children, especially those under the age of five years, are at increased risk of hypoglycaemia. This may be a consequence of increased insulin sensitivity, irregular eating patterns or impaired symptomatic awareness.
COUNTERREGULATION IN CHILDHOOD
201
Genetics Recent studies suggest that molecular markers may influence hypoglycaemia risk. A polymorphism in the gene encoding for angiotensin converting enzyme (ACE) has been described indicating the presence (insertion, I) or absence (deletion, D) of a 287 base pair sequence within intron 16 resulting in three genotypes: II, ID and DD. These genotypes are strongly related to serum ACE concentration with the highest values in DD and the lowest values in II genotypes (Rigat et al., 1990). Danish studies of adults with type 1 diabetes observed that patients with an II (insertion) genotype, who had a low serum ACE activity, had a significantly lower frequency of severe hypoglycaemia (Pedersen-Bjergaard et al., 2001; Pedersen-Bjergaard et al., 2003). A Swedish study of children with type 1 diabetes has reported a six-fold lower frequency of severe hypoglycaemia in those patients who had low serum ACE activity (Nordfelt and Samuelsson, 2003). However, a study of 585 children and adolescents in an Australian centre has not confirmed this association (Bulsara et al., 2007), so this remains controversial.
Clinic Experience Recent therapeutic developments, with the availability of novel insulin analogues and the greater use of intensive insulin regimens, have required re-education not only of patients but also the multidisciplinary team. Few studies have examined the impact of the clinic structure on risk of hypoglycaemia. One large multicentre study in Germany did show a link between small clinic size (< 50 children) and an increased risk of severe hypoglycaemia (Wagner et al., 2005). Another longitudinal study of prevalence of severe hypoglycaemia in Australia showed an increase in hypoglycaemia risk as the mean glycated haemoglobin in the clinic declined. However, over the final five years of the study the risk of hypoglycaemia plateaued while the HbA1c continued to decrease, suggesting the possible benefit of familiarity among patients, healthcare professionals or both (Bulsara et al., 2004) (Figure 9.1).
COUNTERREGULATION IN CHILDHOOD The physiology of counterregulation is the subject of Chapters 1 and 6, but a brief overview of studies of counterregulation in childhood will be presented here. Despite the ethical and practical problems of inducing hypoglycaemia in children for research purposes, a number of studies have been performed (Amiel et al., 1987; Brambilla et al., 1987; Singer-Granick et al., 1988; Hoffman et al., 1991; Jones et al., 1991; Bjorgaas et al., 1997a; Ross et al., 2005). These studies have all examined experimentally-induced hypoglycaemia, either using an insulin infusion method or a hyperinsulinaemic, hypoglycaemic glucose clamp technique. The results of individual hormonal responses are discussed briefly.
Glucagon As in adults the glucagon response during hypoglycaemia is lost in children with diabetes (Amiel et al., 1987; Jones et al., 1991; Ross et al., 2005). This is also the case in toddlers,
202
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
aged 18–57 months old, who have a very short duration of diabetes (Brambilla et al., 1987). This means that individuals with diabetes are more reliant on adequate epinephrine responses to correct hypoglycaemia.
Epinephrine The majority of studies suggest that children have exaggerated epinephrine responses to hypoglycaemia, with peak values that are two-fold higher than those found in adults (Amiel et al., 1987). Data from prepubertal children have been analysed separately from pubertal children and no significant differences were found in epinephrine responses (Amiel et al., 1987; Ross et al., 2005). One study, using an insulin infusion as opposed to a hyperinsulinaemic clamp, did suggest that epinephrine responses were blunted in a group of poorly-controlled adolescents (Bjorgaas et al., 1997a). The reason for the discrepancy in results is not clear. Glucose thresholds for counterregulatory responses have received very little attention. In a study of poorly-controlled adolescents (average total HbA1 : 15%) glucose thresholds for epinephrine secretion were significantly higher, with the poorly-controlled adolescents releasing epinephrine at a glucose concentration of 4.7 mmol/l compared to 3.9 mmol/l in healthy adolescents (Jones et al., 1991).
Growth Hormone None of the reported studies have identified defects in GH release during hypoglycaemia. Generally GH has been found to increase in response to hypoglycaemia both in children with diabetes and in non-diabetic controls (Amiel et al., 1987; Brambilla et al., 1987; Jones et al., 1991).
Cortisol Cortisol, like growth hormone, becomes more important as hypoglycaemia becomes prolonged. Studies of hypoglycaemia in children have shown variable results. Brambilla found no increase in cortisol in either the diabetic or control group of toddlers studied (Brambilla et al., 1987), whereas others have documented an increase in cortisol both in children with and without diabetes (Amiel et al., 1987; Jones et al., 1991).
Effect of Sleep Stage on Counterregulation One study of nocturnal hypoglycaemia in prepubertal children on conventional insulin regimens, found that the median glucose nadir during episodes of nocturnal hypoglycaemia was 1.9 mmol/l (range: 1.1–3.3 mmol/l) and the median duration was 270 minutes (range: 30–630 minutes) (Matyka et al., 1999a). This is similar to adults in whom the average duration of hypoglycaemia (glucose below 2 mmol/l) during the night was found to be three hours in a group of adults with poorly-controlled diabetes (Gale and Tattersall, 1979). Conventional wisdom would argue that prolonged episodes of hypoglycaemia are unusual as hypoglycaemia is promptly corrected by counterregulatory defence mechanisms. However,
CONSEQUENCES OF HYPOGLYCAEMIA
203
studies of nocturnal hypoglycaemia suggest that counterregulatory responses may be blunted during sleep (see Chapter 4). A study of spontaneous nocturnal hypoglycaemia in prepubertal children with diabetes demonstrated blunted and delayed counterregulatory hormone responses during sleep (Matyka et al., 1999b). Peak epinephrine response was only 0.9 nmol/l and there was a marked delay between the glucose falling below 3.5 mmol/l and a significant rise in epinephrine; the mean delay was 170 minutes. Another study examined counterregulatory hormone responses during hypoglycaemia that was experimentally-induced during the time of night when slow wave sleep predominates, and these responses were compared with those to hypoglycaemia induced during the day with the subjects awake and then again when they were awake during the night (Jones et al., 1998). Adolescents with diabetes and healthy controls participated in the study. Epinephrine responses during hypoglycaemia were blunted when hypoglycaemia was induced during slow wave sleep compared to when hypoglycaemia was induced when subjects were awake during the day or during the night (Jones et al., 1998). Studies of the physiology of sleep have demonstrated both variations in autonomic tone and cerebral glucose metabolism going from slow wave sleep through to rapid eye movement sleep which may influence the counterregulatory response during sleep (Maquet et al., 1990; Parmeggiani and Morrison, 1990).
CONSEQUENCES OF HYPOGLYCAEMIA Cognitive Impairment Severe hypoglycaemia can cause catastrophic cerebral damage when it is profound and prolonged, and in very young children this may be a risk associated with a variety of causes (Lucas et al., 1988). Glucose is critical not only as the major fuel for cerebral metabolism but also as a precursor of essential substrates which are essential for normal brain development (Glazer and Weber, 1971). Concerns have been raised that recurrent hypoglycaemia could affect long-term academic achievement in children with type 1 diabetes, from the effects of hypoglycaemia disrupting school performance to the possibility of damage accumulating over time.
Acute effects Few studies have studied the impact of acute hypoglycaemia on cognitive performance in children or adolescents. One study of the effects of experimentally-induced mild hypoglycaemia (3.1–3.6 mmol/l) found decrements in tests of mental flexibility and on measures that required planning and decision making and a rapid response, although the results were variable within subjects (Ryan et al., 1990). The learning ability of children, who spend much of their day at school assimilating information, could be seriously compromised if they experience frequent episodes of even mild hypoglycaemia during their time in class. Children can also be affected if they miss lessons because of severe hypoglycaemic events and could be compromised by asymptomatic episodes of hypoglycaemia that go undetected and therefore untreated. There is some evidence to suggest that children may be particularly susceptible to mild episodes of hypoglycaemia – studies examining P300 evoked potentials and EEG changes in response to
204
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
hypoglycaemia have found that abnormalities commence at a higher blood glucose level in children than in adults (Jones et al., 1995; Bjorgaas et al., 1998). The effect of nocturnal hypoglycaemia on cognitive function has received little attention. Cognitive performance could be directly affected by hypoglycaemia, but also by sleep disturbance. The few studies of nocturnal hypoglycaemia in children and adolescents with type 1 diabetes have shown no deleterious effect on overall sleep physiology (Bendtson et al., 1992; Porter et al., 1996; Matyka et al. 2000, Pillar et al., 2003) or on cognitive function the following morning.
Long-term effects Early studies of children with diabetes suggested that they were ‘mentally superior’ (West et al., 1934). However, in recent years concern has been expressed about the potential impact of recurrent episodes of hypoglycaemia on intellectual performance. It is beyond the scope of this chapter to provide a comprehensive review on this topic and only more recent studies are reviewed. The developing brain is extremely vulnerable to all types of cerebral trauma. Studies of children who have experienced closed head injuries suggest that the consequences may be delayed, with subtle cerebral damage becoming evident with time as normal developmental milestones are delayed. A large number of studies have examined the impact of recurrent hypoglycaemia in childhood (Ryan et al., 1985; Golden et al., 1989; Bjorgaas et al., 1997b; Hershey et al., 1999; Rovet and Ehrlich, 1999; Northam et al., 2001; Wysocki et al., 2003). Almost without exception, the results have shown a possible link between severe hypoglycaemia and decrements in cognitive performance and that those children most at risk of cognitive impairment have been those diagnosed early in life – usually less than five years of age (Ryan et al., 1985; Golden et al., 1989; Bjorgaas et al., 1997b; Northam et al., 2001). Deficiencies have been found in several cognitive domains but are more likely in those originating in the frontal lobe. One of the most impressive longitudinal studies has been that of Northam and colleagues. Cognitive performance was assessed in a large group of children (123 at baseline) with newly-diagnosed diabetes, aged 3–14 years, who were compared to healthy controls at three months, two years and six years following diagnosis (Northam et al., 2001). At six years, children with diabetes performed more poorly in measures of intelligence, attention, processing speed, long-term memory and executive skills. Attention, processing speed and executive skills were especially affected in those children who had developed diabetes when less than four years of age (Northam et al., 2001). Severe hypoglycaemia was associated with lower verbal and full-scale intelligence quotient (IQ) scores. The authors concluded that recurrent hypoglycaemia was a potential explanation for these cognitive deficits but could not exclude an effect of chronic hyperglycaemia. Another study has performed a cognitive test battery and structural neuroimaging using Magnetic Resonance Imaging in a group of young adults with type 1 diabetes and compared the findings between those with either early onset disease, defined as less than seven years, and late onset disease (Ferguson et al., 2005). Physiological risk variables such as diabetes duration, evidence of microvascular disease and retrospective reporting of preceding exposure to severe hypoglycaemia were also assessed. The patients with early onset diabetes were found to have deficits in non-verbal intelligence, information processing ability and psychomotor speed. The authors also found higher ventricular volumes and higher frequency
CONSEQUENCES OF HYPOGLYCAEMIA
205
of mild ventricular atrophy in those with early onset diabetes. None of the findings were related to the presence of microvascular disease or diabetes duration, suggesting that the cumulative effects of hyperglycaemia were unlikely to be causative. However, no definite link was confirmed between exposure to severe hypoglycaemia and any of these defects, although the study was limited by retrospective reporting of the history of preceding hypoglycaemia over a long period of time (Ferguson et al., 2005). Another study did not find a difference in tests of cognitive function over a period of 18 months in 142 patients taking part in a study examining the impact of intensive therapy versus usual care (Wysocki et al., 2003). The study group had a wide age range from 6–15 years and data from younger children were not analysed separately. Although there are methodological problems in designing studies to assess long-term cognitive function in children who have an ongoing chronic disorder, these studies do raise anxieties. It is felt that until hypoglycaemia can be reliably avoided, glycaemic control should be less intensively managed in younger children to avoid the risk of severe hypoglycaemia. This would put younger children at greater risk of developing the long-term microvascular complications of diabetes in an attempt to avoid the possible cognitive defects that may be associated with recurrent hypoglycaemia.
Hypoglycaemic Hemiplegia This is a rare complication of acute hypoglycaemia in which the patient recovers from the hypoglycaemia with a transient hemiparesis lasting no more than 24 hours. When neuroimaging is performed it is rare to find an abnormality. There is no evidence of any severe sequelae to this neurological manifestation of severe neuroglycopenia (Pocecco and Ronfani, 1998).
Fear of Hypoglycaemia Both the children with diabetes and their parents worry about the prospect of a severe episode of hypoglycaemia (Gold et al., 1997; Clarke et al., 1998; Gonder-Frederick et al., 1997; Marrero et al., 1997; Nordfelt and Ludvigson, 2005) (see Chapter 14). In one study severe hypoglycaemia caused more fear than the prospect of an episode of diabetic ketoacidosis (Nordfelt and Ludvigson, 2005). Although there is little evidence that this modifies behaviour to attempt hypoglycaemia avoidance, such as relaxing glycaemic control or eating more snacks, the adverse effects on quality of life should not be underestimated (Gold et al., 1997).
Prediction of Nocturnal Hypoglycaemia Overnight glucose profiles have been shown to be extremely variable. As a result studies of overnight hypoglycaemia have been unable to provide a ‘safe’ glucose value with which to go to bed. What is more useful is the fasting blood glucose concentration on the following morning. One study showed that the median fasting blood glucose at 07:00 hours was significantly lower following hypoglycaemia than a night with no hypoglycaemia (3.7 [1.4–10.6] versus 8.5 [3.8–19.2] mmol/l, p = 000001) (Matyka et al., 1999a) (Figure 9.2).
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
206 20
Glucose (mmol / l)
15
10
5
0
hypoglycaemia
no hypoglycaemia
Figure 9.2 Glucose values at 07.00 hours following a night of hypoglycaemia () and a night of no hypoglycaemia (). Reproduced by permission of K. Matyka, PhD thesis ‘Nocturnal hypoglycaemia in prepubertal children with type 1 diabetes mellitus’, University of London
The occurrence of a ‘Somogyi phenomenon’, whereby overnight hypoglycaemia promotes glucose counterregulation and causes fasting hyperglycaemia, has not been demonstrated in studies of nocturnal hypoglycaemia in children (Porter et al., 1996; Beregszaszi et al., 1997; Matyka et al., 1999a).
MANAGEMENT OF HYPOGLYCAEMIA The management of acute hypoglycaemia will depend on the severity of the episode. The International Society for Paediatric and Adolescent Diabetologists has provided guidelines for the management of acute episodes of hypoglycaemia based on severity (ISPAD, 2000). Blood glucose measurement is the only way to confirm hypoglycaemia if the diagnosis is uncertain and also confirm the return of the blood glucose to normal after treatment. Figure 9.3 shows the flow diagram for the management of hypoglycaemia.
Prevention When faced in clinic with a child who is having recurrent episodes of hypoglycaemia, a detailed history should be obtained regarding the timing of hypoglycaemia, insulin regimen, dietary intake and the relation to periods of physical activity. This will enable an assessment to be made of possible risk factors and inform how these may be avoided. If no obvious cause is found then other pathology should be sought, such as coincidental coeliac disease or the possibility of Addison’s disease, although these are relatively rare causes of recurrent hypoglycaemia (see Chapter 3). When contemplating preventive management the following aspects should be considered.
MANAGEMENT OF HYPOGLYCAEMIA
207
Check blood glucose – Glucose meter test and formal laboratory glucose if possible Hypoglycaemia
Severe
Mild
CONSCIOUS i.e. gag reflex intact
Oral glucose e.g. 5–15 grams of glucose or 100 mls sweet drink
Glucogel One ampoule orally
Success
UNCONSCIOUS
I.M. Glucagon <5 yrs 0.5 mg >5 yrs 1.0 mg
No success If no response in 15 mins give 1–2 mls/kg of 10% dextrose I.V. Repeat until there is a clinical response.
As symptoms improve or normoglycaemia is restored add oral complex carbohydrate e.g. biscuit, bread and so on. If unable to tolerate oral carbohydrate may need a glucose infusion e.g. 5–10% glucose at maintenance rate
Figure 9.3 Management of hypoglycaemia
Education Great importance is placed on education in the management of type 1 diabetes and structured education programmes are now an essential part of any diabetes service provision. Despite this, little validation has been made of the use of structured educational programmes in children. One study from Scandinavia has shown the benefits of a focused education package (Nordfelt et al., 2003). In this study families were given both written and video information on diabetes. One group were given general information on diabetes management and the other was given material to educate about the importance and means of hypoglycaemia avoidance. Although no differences in glycated haemoglobin were found between the two
208
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
groups, those who had the targeted intervention had a significantly reduced rate of severe hypoglycaemia (Nordfelt et al., 2003).
Insulin As noted earlier, the DCCT suggested that attempts to intensify insulin regimens may increase the risk of severe hypoglycaemia. Recent introductions of analogue insulins, however, indicate that improved glycaemic control does not always lead to hypoglycaemia, although few studies have been designed specifically to examine the benefits of different insulin regimens on the risk of hypoglycaemia. One study that compared insulin lispro with soluble (short-acting) insulin as part of a basal bolus regimen, showed small but statistically significant reductions in the prevalence of hypoglycaemia over a 30 day period when insulin lispro was being used (Holcombe et al., 2002). Other studies have found benefits of insulin analogue regimens on nocturnal hypoglycaemia. One randomised cross-over study in adolescents compared insulin lispro and glargine, as part of a daily multiple injection regimen, to human soluble and isophane insulins. Nocturnal hypoglycaemia was 43% lower with the analogue regimen, although no difference was observed in self-reported symptomatic hypoglycaemia (Murphy et al., 2003). Another study in prepubertal children examined the benefits of a thrice daily insulin regimen, where the evening dose of mixed insulin was replaced by a rapid-acting insulin analogue with the evening meal and isophane insulin before bed (Ford-Adams et al., 2003). Although there was no difference in glycated haemoglobin between the two treatment arms, the prevalence of hypoglycaemia was lower in the early part of the night (22.00–04.00 hours) when the analogue was used. Although not every patient is suited to using insulin pump therapy, clinic-based studies of CSII therapy have shown that more stable blood glucose control can be achieved without an increased risk of hypoglycaemia. In an American study describing the experience of using insulin pumps in a paediatric clinic, it was found that 50 adolescents on multiple daily injections experienced 134 episodes of severe hypoglycaemia per 100 patient-years compared to 76 episodes per 100 patient-years in the 25 adolescents who opted for pump therapy (Boland et al., 1999).
Diet Few studies have examined the impact of dietary interventions on hypoglycaemia risk except for nocturnal hypoglycaemia. The major dietary modification has been that of the introduction of a larger proportion of starch, as a form of long-acting carbohydrate, as part of the evening snack (Ververs et al., 1993; Kaufman et al., 1995; Detlofson et al., 1999; Matyka et al., 1999a). In these studies the benefits of starch have been inconsistent. One study found a lower frequency of nocturnal hypoglycaemia, although capillary sampling was performed only intermittently during the night and some episodes of hypoglycaemia may have been undetected (Kaufman et al., 1995). Others found no beneficial effect of cornstarch on the prevention of nocturnal hypoglycaemia although blood glucose concentrations fell more slowly, but in one study this occurred at the expense of promoting hyperglycaemia (Ververs et al., 1993; Matyka et al., 1999a). Although not designed to examine the impact of diet on hypoglycaemia, a study of a low glycaemic index diet has been shown to improve glycaemic
CONCLUSIONS
209
control without an increase in rate of hypoglycaemia, and appeared to have enhanced quality of life (Gilbertson et al., 2001)
Exercise When adequate plasma insulin concentrations are available, exercise can lead to acute hypoglycaemia. However if exercise is performed at a time of relative insulin deficiency, hyperglycaemia with ketosis can occur. In addition, delayed hypoglycaemia may occur as muscle glycogen stores recover mainly overnight (Admon et al., 2005; Tsalikian et al., 2005). Although few data are currently available regarding the most appropriate management of planned periods of physical activity, a number of guidelines have been proposed. The International Society for Paediatric and Adolescent Diabetologists has published guidelines on the Internet (www.ispad.org). These recommend that careful monitoring of blood glucose is essential to match food and insulin to the intensity of exercise and that a reduction of insulin should be considered. Additional slowly absorbed carbohydrate will be necessary, especially at bedtime, if exercise has been performed in the afternoon or early evening. From the data available so far (Admon et al., 2005; Tsalikian et al., 2005), these guidelines do seem a reasonable approach to the avoidance of both exercise related hypo- and hyperglycaemia. It is important, however, to work with the child and family to provide individually-tailored recommendations that are tried and tested for the child. The management of unpredictable episodes of physical activity are likely to remain a problem until a cure for diabetes is found.
CONCLUSIONS • Hypoglycaemia is a common problem for children with type 1 diabetes, especially young children, and for their families. • Behavioural symptoms are more common in childhood than more typical autonomic symptoms. The majority of episodes of nocturnal hypoglycaemia are totally asymptomatic. • Episodes of hypoglycaemia remain a significant barrier when striving for a degree of glycaemic control that will delay or prevent the development of the microvascular complications of diabetes. • Glucagon responses during hypoglycaemia are lost early in the course of type 1 diabetes. Epinephrine responses during overnight hypoglycaemia are blunted in both healthy children and those with type 1 diabetes and may contribute to the lack of symptoms of many episodes of hypoglycaemia overnight. • Concerns remain about the possible long-term implications of hypoglycaemia in terms of cognitive dysfunction although hyperglycaemia may also be important. • More recent data suggest that novel insulin analogues and regimens may enable improvements in glycaemic control to be achieved without a concomitant increase in the risk of hypoglycaemia.
210
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
• It is important to consider both the child and their family when assessing the causes of recurrent episodes of hypoglycaemia and when suggesting possible avoidance strategies. • All children with diabetes should have a personalised action plan for hypoglycaemia avoidance.
REFERENCES Admon G, Weinstein Y, Falk B, Weintrob N, Benzaquen H, Ofan R et al. (2005). Exercise with and without an insulin pump among children and adolescents with type 1 diabetes mellitus. Pediatrics 116: 348–55. Ahlborg G, Felig P, Hagenfeldt L, Hemdler R, Wahren J (1974). Substrate turnover during prolonged exercise in man. Journal of Clinical Investigation 53: 1080–90. Allen C, LeCaire T, Palta M, Daniles K, Meredith M, D’Alessio DJ, for the Wisconsin Diabetes Registry Project (2001) Risk factors for frequent and severe hypoglycemia in type 1 diabetes. Diabetes Care 24: 1878–81. Amiel SA, Simonson DC, Sherwin RS, Lauritano AA, Tamborlane WV (1987). Exaggerated epinephrine responses to hypoglycemia in normal and insulin-dependent diabetic children. Journal of Pediatrics 110: 832–7. Bendtson I, Gade J, Thomsen CE, Rosenfalck A, Wildschiodtz G (1992). Sleep disturbances in IDDM patients with nocturnal hypoglycemia. Sleep 15: 74–81 Bendtson I, Rosenfalck AM, Binder C (1993). Nocturnal versus diurnal counterregulation to hypoglycemia in type 1 (insulin-dependent) diabetic patients. Acta Endocrinologica 128: 109–15. Beregszaszi M, Tubiana-Rufi N, Benali K, Noel M, Bloch J, Czernichow P (1997). Nocturnal hypoglycemia in children and adolescents with insulin-dependent diabetes mellitus: Prevalence and risk factors. Journal of Pediatrics 131: 27–33. Bier DM, Leake RD, Haymond MW, Arnold KJ, Gruenke LD, Sperling MA, Kipnis DM (1977). Measurement of ‘true’ glucose production rates in infancy and childhood with 6,6-dideuteroglucose. Diabetes 26: 1016–23. Bjorgaas M, Vik T, Sand T, Birkeland K, Sager G, Vea H, Jorde R (1997a). Counterregulatory and symptom responses to hypoglycaemia in diabetic children. Diabetic Medicine 14: 433–41. Bjorgaas M, Gimse R, Vik T, Sand T (1997b). Cognitive function in type 1 diabetic children with and without episodes of severe hypoglycaemia. Acta Paediatrica 86: 148–53. Bjorgaas M, Sand T, Vik T, Jorde R (1998). Quantitative EEG during controlled hypoglycaemia in diabetic and non-diabetic children. Diabetic Medicine 15: 30–7. Bognetti F, Brunelli A, Meschi F, Viscardi M, Bonfanti R, Chiumello G (1997). Frequency and correlates of severe hypoglycaemia in children and adolescents with diabetes mellitus. European Journal of Paediatrics 156: 589–91. Boland EA, Grey M, Oesterle A, Fredrickson L, Tamborlane WV (1999). Continuous subcutaneous insulin infusion. A new way to lower risk of severe hypoglycemia, improve metabolic control, and enhance coping in adolescents with type 1 diabetes. Diabetes Care 22: 1779–84. Bolli G, Gerich J (1984). The dawn phenomenon – a common occurrence in both non-insulin dependent and insulin dependent diabetes mellitus. New England Journal of Medicine 310: 746–50. Brambilla P, Bougneres PF, Santiago JV, Chaussin JL, Pouplard A, Castano L (1987). Glucose counterregulation in pre-school-age children with recurrent hypoglycemia during conventional treatment. Diabetes 36: 300–4. Bulsara MK, Holman CDJ, Davis EA, Jones TW (2004). The impact of a decade of changing treatment on rates of severe hypoglycemia in a population-based cohort of children with type 1 diabetes. Diabetes Care 27: 2293–8.
REFERENCES
211
Bulsara MK, Holman CDJ, van Bockxmeer FM, Davis EA, Gallego PH, Beilby JP et al. (2007). The relationship between ACE genotype and risk of severe hypoglycaemia in a large population-based cohort of children and adolescents with type 1 diabetes. Diabetologia 50: 965–71. Cartee GD, Holloszy JO (1990). Exercise increases susceptibility of muscle glucose transport to activation by various stimuli. American Journal of Endocrinology 258: E390–3. Chaussain J-L (1973). Glycemic response to 24-hour fast in normal children and children with ketotic hypoglycemia. Journal of Pediatrics 82: 438–43. Chaussain J-L, Georges P, Olive G, Job J-C (1974). Glycemic response to 24-hour fast in normal children and children with ketotic hypoglycemia: II Hormonal and metabolic changes. Journal of Pediatrics 85: 776–81. Chaussain J-L, Georges P, Calzada L, Job J-C (1977). Glycemic response to 24-hour fast in normal children: III Influence of age. Journal of Pediatrics 91: 711–4. Clarke WL, Gonder-Frederick A, Snyder AL, Cox DJ (1998). Maternal fear of hypoglycemia in their children with insulin dependent diabetes mellitus. Journal of Pediatric Endocrinology and Metabolism 11: 189–94. Craig ME, Handelsman P, Donaghue KC, Chan A, Blades B, Laina R et al., NSW/ACT HbA(1c) Study Group (2002). Predictors of glycaemic control and hypoglycaemia in children and adolescents with type 1 diabetes from NSW and ACT. Medical Journal of Australia 177: 235–8. Daneman D, Frank M, Perlman K, Tamm J, Ehrlich R (1989). Severe hypoglycemia in children with insulin-dependent diabetes mellitus: frequency and predisposing factors. Journal of Pediatrics 5: 681–5. Danne T, Weber B, Hartmann R, Enders I, Burger W, Hovener G (1994). Long-term glycemic control has a nonlinear association to the frequency of background retinopathy in adolescents with diabetes. Follow-up of the Berlin Retinopathy Study. Diabetes Care 17: 1390–6. Davis EA, Keating B, Byrne GC, Russell M, Jones TW (1997). Hypoglycemia: incidence and clinical predictors in a large population based sample of children and adolescents with IDDM. Diabetes Care 20: 20–5. Davis EA, Keating B, Byrne GC, Russell, Jones TW (1998). Impact of improved glycaemic control on rates of hypoglycaemia in IDDM. Archives of Disease in Childhood 78: 111–5. Deary IJ, Hepburn DA, MacLeod KM, Frier BM (1993). Partitioning the symptoms of hypoglycaemia using multi-sample confirmatory factor analysis. Diabetologia 36: 771–7. De Feo P, Torlone E, Fanelli C, Santeusanio F, Brunetti P, Bolli GB (1990). Nocturnal spikes of growth hormone secretion cause the dawn phenomenon in type 1 (insulin dependent) diabetes mellitus by decreasing hepatic (and extrahepatic) sensitivity to insulin in the absence of insulin waning. Diabetologia 33: 52–9. Detlofson I, Kroon M, Aman J (1999). Oral bedtime cornstarch supplementation reduces the risk for nocturnal hypoglycaemia in young children with type 1 diabetes. Acta Paediatrica 88: 595–7. Dunger DB (1992). Diabetes in puberty. Archives of Disease in Childhood 67: 569–70. Edge JA, Matthews DR, Dunger DB (1990a). Failure of current insulin regimes to meet the overnight requirements of adolescents with insulin-dependent diabetes. Diabetes Research 15: 109–12. Edge JA, Matthews DR, Dunger DB (1990b). The dawn phenomenon is related to overnight growth hormone release in adolescent diabetics. Clinical Endocrinology 33: 729–37. Ferguson SC, Blane A, Wardlaw J, Frier BM, Perros P, McCrimmon RJ, Deary IJ (2005). Influence of an early-onset age of type 1 diabetes on cerebral structure and cognitive function. Diabetes Care 28: 1431–7. Ford-Adams ME, Murphy NP, Moore EJ, Edge JA, Ong KL, Watts AP et al. (2003). Insulin lispro: a potential role in preventing nocturnal hypoglycaemia in young children with diabetes mellitus. Diabetic Medicine 20: 656–60. Gale EAM, Tattersall RB (1979). Unrecognised nocturnal hypoglycaemia in insulin-treated diabetics. Lancet i: 1049–52.
212
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
Gilbertson HR, Brand-Miller JC, Thorburn AW, Evans S, Chondros P, Werther GA (2001). The effect of flexible low glycemic index dietary advice versus measured carbohydrate exchange diets on glycemic control in children with type 1 diabetes. Diabetes Care 24: 1137–43. Glazer RI, Weber G (1971). Incorporation of [6-3 H] glucose into lipid, protein, RNA and DNA slices of differentiating rat cerebral cortex. Journal of Neurochemistry 18: 1569–76. Gold AE, Deary IJ, Frier BM (1997). Hypoglycaemia and non-cognitive aspects of psychological function in insulin-dependent (type 1) diabetes mellitus (IDDM). Diabetic Medicine 14: 111–8. Golden MP, Ingersoll GM, Brack CJ, Russell BA, Wright JC, Huberty TJ (1989). Longitudinal relationship of asymptomatic hypoglycemia to cognitive function in IDDM. Diabetes Care 12: 89–93. Gonder-Frederick L, Cox D, Kovatchev B, Julian D, Clarke W (1997). The psychosocial impact of severe hypoglycemic episodes on spouses of patients with IDDM. Diabetes Care. 20: 1543–6. Haymond MW, Karl IE, Clarke WL, Pagliara AS, Santiago JV (1982). Differences in circulating gluconeogenic substrates during short-term fasting in men, women and children. Metabolism 31: 33–42. Hershey T, Bhargava N, Sadler M, White NH, Craft S (1999). Conventional versus intensive diabetes therapy in children with type 1 diabetes – effects on memory and motor speed. Diabetes Care 22: 1318–24. Hoffman RP, Singer-Granick C, Drash AL, Becker DJ (1991). Plasma catecholamine responses to hypoglycemia in children and adolescents with IDDM. Diabetes Care 14: 81–8. Holcombe JH, Zalani S, Arora VK, Mast CJ (2002). Lispro in Adolescents Study Group. Comparison of insulin lispro with regular human insulin for the treatment of type 1 diabetes in adolescents. Clinical Therapeutics 24: 629–38. Holl RW, Swift PG, Mortensen HB, Lynggaard H, Hougaard P, Aanstoot HJ et al., on behalf of the Hvidore Study Group (2003). Insulin injection regimens and metabolic control in an international survey of adolescents with type 1 diabetes over 3 years: results from the Hvidore Study Group. European Journal of Paediatrics 162: 22–9. ISPAD Consensus Guidelines for the Management of type 1 Diabetes Mellitus in Children and Adolescents. Swift P, ed. Medical Forum International, The Netherlands. 2000. Ivy JL, Holloszy JO (1981). Persistent increase in glucose uptake by rat skeletal muscle following exercise. American Journal of Physiology 241: C200–3. Jones TW, Boulware SD, Kraemer DT, Caprio S, Sherwin RS, Tamborlane WV (1991). Independent effects of youth and poor diabetes control on responses to hypoglycemia in children. Diabetes 40: 358–63. Jones TW, Borg WP, Boulware SD, McCarthy G, Sherwin RS, Tamborlane WV (1995). Enhanced adrenomedullary response and increased susceptibility to neuroglycopenia: Mechanisms underlying the adverse effects of sugar ingestion in healthy children. Journal of Pediatrics 126: 171–7. Jones TW, Porter P, Sherwin RS, Davis EA, O’Leary P, Frazer F et al. (1998). Decreased epinephrine responses to hypoglycemia during sleep. New England Journal of Medicine 338: 1657–62. Kaufman FR, Halvorson M, Kaufman ND (1995). A randomized, blinded trial of uncooked cornstarch to diminish nocturnal hypoglycemia at diabetes camp. Diabetes Research and Clinical Practice. 30: 205–9. Kerr DS, Hansen IL, Levy MM (1983). Metabolic and hormonal responses of children and adolescents to fasting and 2-deoxyglucose. Metabolism 22: 951–9. Koh THHG, Eyre JA, Aynsley-Green A (1988). Neonatal hypoglycaemia – the controversy regarding definition. Archives of Disease in Childhood 63: 1386–8. Lamers KJB, Doesburg WH, Gabreels FJM, Lemmens WAJG, Romson AC, Wevers RA, Renier WO (1985a). The concentration of blood components related to fuel metabolism during prolonged fasting in children. Clinica Chimica Acta 152: 155–63. Lamers KJB, Doesburg WH, Gabreels FJM, Romson AC, Renier WO, Wevers RA, Lemmens WAJG (1985b). Reference values of blood components related to fuel metabolism in children after an overnight fast. Clinica Chimica Acta 145: 17–26.
REFERENCES
213
Lawrence RD (1926). The effects of exercise on insulin action in diabetes. British Medical Journal 1: 648–52. Levine B, Anderson BJ, Butler DA, Antisdel JE, Brackett J, Laffel LMB (2001). Predictors of glycemic control and short-term adverse outcomes in youths with type 1 diabetes. Journal of Pediatrics 139: 197–203. Lopez MJ, Oyarzabal M, Barrio R, Hermoso F, Lopez JP, Rodriguez M et al. (1997). Nocturnal hypoglycaemia in IDDM patients younger than 18 years. Diabetic Medicine 14: 772–7. Lucas A, Morley R, Cole TJ (1988). Adverse neurodevelopmental outcome of moderate neonatal hypoglycaemia. British Medical Journal 297: 1304–8. McCrimmon RJ, Gold AE, Deary IJ, Kelnar CJ, Frier BM (1995). Symptoms of hypoglycemia in children with IDDM. Diabetes Care 18: 858–61. Magrath G, Hartland BV, the Nutrition Subcommittee of the British Diabetic Association’s Professional Advisory Committee (1993). Dietary recommendations for children with diabetes: an implementation paper. Journal of Human Nutrition and Dietetics 6: 491–507. Maquet P, Dive D, Salmon E, Sadzot B, Franco G, Poirrier R et al. (1990). Cerebral glucose utilization during the sleep-wake cycle in man determined by positron emission tomography and 18F-2-fluoro2-deoxy-D-glucose method. Brain Research 513: 136–43. Marrero DG, Guare JC, Vandagriff JL, Fineberg NS (1997). Fear of hypoglycemia in the parents of children and adolescents with diabetes: maladaptive or healthy response? Diabetes Educator 23: 281–6. Matyka KA, Wiggs L, Pramming S, Stores G, Dunger DB (1999a). Cognitive function and mood following profound nocturnal hypoglycaemia during conventional therapy of diabetes. Archives of Disease in Childhood 81: 138–42. Matyka KA, Crowne EC, Havel PJ, Macdonald IA, Matthews D, Dunger DB (1999b). Counterregulation during spontaneous nocturnal hypoglycemia in prepubertal children with insulin dependent diabetes mellitus. Diabetes Care 22: 1144–50. Matyka KA, Crawford C, Wiggs L, Dunger DB, Stores G (2000). Alterations in sleep physiology in young children with insulin dependent diabetes mellitus: relationship to nocturnal hypoglycemia. Journal of Pediatrics 137: 233–8. Mohn A, Matyka KA, Harris DA, Ross KM, Edge JA, Dunger DB (1999). Lispro or regular insulin for multiple injection therapy in adolescence. Differences in free insulin and glucose levels overnight. Diabetes Care 22: 27–32. Mortensen HB, Hougaard P, for the Hvidore Study Group on Childhood Diabetes (1997). Comparison of metabolic control in a cross-sectional study of 2,873 children and adolescents with IDDM from 18 countries. Diabetes Care 20: 714–20. Murphy NP, Keane SM, Ong KK, Ford-Adams M, Edge JA, Acerini C, Dunger DB (2003). Randomized cross-over trial of insulin glargine plus lispro or NPH insulin plus regular human insulin in adolescents with type 1 diabetes on intensive insulin regimens. Diabetes Care 26: 799–804. Nordfelt S, Ludvigsson J (1997). Severe hypoglycemia in children with IDDM: a prospective population study 1992–1994. Diabetes Care 20: 497–503. Nordfelt S, Samuelsson U (2003). Serum ACE predicts severe hypoglycemia in children and adolescents with type 1 diabetes. Diabetes Care 26: 274–8. Nordfelt S, Johansson C, Carlsson E, Hammersjo J-A (2003). Prevention of severe hypoglycaemia in type 1 diabetes: a randomised controlled population study. Archives of Disease in Childhood 88: 240–5. Nordfelt S, Ludvigsson J (2005). Fear and other disturbances of severe hypoglycaemia in children and adolescents with type 1 diabetes. Journal of Pediatric Endocrinology 18: 83–91. Northam EA, Anderson PJ, Jacobs R, Hughes M, Warne GL, Werther GA (2001). Neuropsychological profiles of children with type 1 diabetes 6 years after disease onset. Diabetes Care 24: 1541–6.
214
HYPOGLYCAEMIA IN CHILDREN WITH DIABETES
Parmeggiani PL, Morrison AR (1990). Alterations in autonomic functions during sleep. In: Central Regulation of Autonomic Functions. Loewy AD and Spyer KM, eds. Oxford University Press, Oxford: 367–86. Pedersen-Bjergaard U, Agerholm-Larsen, Pramming S, Hougaard P, Thorsteinsson B (2001). Activity of angiotensin converting enzyme and risk of severe hypoglycaemia in type 1 diabetes mellitus. Lancet 357: 1248–53. Pedersen-Bjergaard U, Agerholm-Larsen, Pramming S, Hougaard P, Thorsteinsson B (2003). Prediction of severe hypoglycaemia by angiotensin converting enzyme activity and genotype in type 1 diabetes. Diabetologia 46: 89–96. Pillar G, Schuscheim G, Weiss R, Malhotra A, McCowen KC, Shlitner A et al. (2003). Interactions between hypoglycemia and sleep architecture in children with type 1 diabetes mellitus. Journal of Pediatrics 142: 163–8. Pocecco M, Ronfani L (1998). Transient focal neurologic deficits associated with hypoglycaemia in children with insulin-dependent diabetes mellitus. Italian Collaborative Paediatric Diabetologic Group. Acta Paediatrica 87: 542–4. Porter P, Byrne G, Stick S, Jones TW (1996). Nocturnal hypoglycaemia and sleep disturbances in young teenagers with insulin dependent diabetes mellitus. Archives of Disease in Childhood 75: 120–3. Porter PA, Keating B, Byrne G, Jones TW (1997). Incidence and predictive criteria of nocturnal hypoglycemia in young children with insulin-dependent diabetes mellitus. Journal of Pediatrics 130: 366–72. Price TB, Rothman DL, Taylor R, Avison MJ, Shulman GI, Shulman RG (1994). Human muscle glycogen resynthesis after exercise: insulin-dependent and independent phases. Journal of Applied Physiology 76: 104–11. Rewers A, Chase HP, Mackenzie T, Walravens P, Roback M, Rewers M et al. (2002). Predictors of acute complications in children with type 1 diabetes. Journal of the American Medical Association 287: 2511–8. Richter EA (1996). Glucose utilization. In: Exercise: Regulation and Integration of Multiple Systems. Rowell LB and JT Shepherd JT, eds. Oxford University Press, New York: 912–51. Rigat B, Hubert C, Alhenc-Gelas F, Cambien F, Corvol P, Soubrier F (1990). An insertion/deletion polymorphism in the angiotensin-1 converting enzyme gene accounting for half the variance of serum enzyme levels. Journal of Clinical Investigation 86: 1343–6. Rosilio MR, Cotton JB, Wieliczko MC, Genrault B, Carel JC, Couvaras O et al., on behalf of the French Pediatric Diabetes Group (1998). Factors associated with glycemic control. Diabetes Care 21: 1146–53. Ross LA, Warren RE, Kelnar CJH, Frier BM (2005). Pubertal stage and hypoglycaemia counterregulation in type 1 diabetes. Archives of Disease in Childhood 90: 190–4. Rovet JF, Ehrlich RM (1999). The effect of hypoglycemic seizures on cognitive function in children with diabetes: a seven year prospective study. Journal of Pediatrics 134: 503–6. Ryan C, Vega A, Drash A (1985). Cognitive deficits in adolescents who developed diabetes early in life. Pediatrics 75: 921–7. Ryan CM, Atchison J, Puczynski S, Puczynski M, Arslanian S, Becker D (1990). Mild hypoglycemia associated with deterioration of mental efficiency in children with insulin-dependent diabetes mellitus. Journal of Pediatrics 117: 32–8. Saudubray JM, Marsac C, Limal JM, Dumurgier E, Charpentier C, Ogier H, Coude FX (1981). Variation in plasma ketone bodies during a 24-hour fast in normal and hypoglycemic children: Relationship to age. Journal of Pediatrics 98: 904–8. Schultz CJ, Konopelska-Bahu T, Dalton RN, Carroll TA, Stratton I, Gale EA et al. (1999). Microalbuminuria prevalence varies with age, sex, and puberty in children with type 1 diabetes followed from diagnosis in a longitudinal study. Oxford Regional Prospective Study Group. Diabetes Care 22: 495–502.
REFERENCES
215
Singer-Granick C, Hoffman RP, Kerensky K, Drash AL, Becker DJ (1988). Glucagon responses to hypoglycemia in children and adolescents with IDDM. Diabetes Care 11: 643–9. Solders G, Thalme B, Aguirre-Aquino M, Brandt L, Berg U, Persson A (1997). Nerve conduction and autonomic nerve function in diabetic children. A 10 year follow-up study. Acta Paediatrica 86: 361–6. Stirling HF, Darling JAB, Kelnar CJH (1991). Nocturnal glucose homeostasis in normal children. Hormone Research 35: 54 (abstract). The Diabetes Control and Complications Trial Research Group (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New England Journal of Medicine 329: 977–86. The Diabetes Control and Complications Trial Research Group (1994). Effect of intensive diabetes treatment on the development and progression of long-term complications in adolescents with insulin-dependent diabetes mellitus: Diabetes Control and Complications Trial. Journal of Pediatrics 125: 177–88. The Diabetes Control and Complications Trial Research Group (1997). Hypoglycemia in the Diabetes Control and Complications Trial. Diabetes 46: 271–86. Thomsett M, Shield G, Batch J, Cotterill A (1999). How well are we doing? Metabolic control in patients with diabetes. Journal of Paediatrics and Child Health 35: 479–82. Trumper BG, Reschke K, Molling J (1995). Circadian variation of insulin requirement in IDDM: the relationship between circadian change in insulin demand and diurnal patterns of growth hormone, cortisol and glucagon during euglycaemia. Hormone and Metabolic Research 27: 141–7. Tsalikian E, Mauras N, Beck RW, Tamborlane WV, Janz KF, Chase HP et al., Diabetes Research In Children Network Direcnet Study Group (2005). Impact of exercise on overnight glycemic control in children with type 1 diabetes. Journal of Pediatrics 147: 528–34. Tupola S, Rajantie J (1998a). Documented symptomatic hypoglycaemia in children and adolescents using multiple daily injection therapy. Diabetic Medicine 15: 492–6. Tupola S, Rajantie J, Maenpaa J (1998b). Severe hypoglycaemia in children and adolescents during multiple-dose insulin therapy. Diabetic Medicine 15: 695–9. Vanelli M, Cerutti F, Chiarelli F, Lorini R, Meschi F, MCDC – Italy Group (2005). Nationwide crosssectional survey of 3560 children and adolescents with diabetes in Italy. Journal of Endocrinological Investigation 28: 692–9. Ververs MT, Rouwe C, Smit GP (1993). Complex carbohydrates in the prevention of nocturnal hypoglycaemia in diabetic children. European Journal of Clinical Nutrition 47: 268–73. Wagner VM, Grabert M, Holl RW (2005). Severe hypoglycaemia, metabolic control and diabetes management in children with type 1 diabetes in the decade after the Diabetes Control and Complications Trial – a large-scale multicentre study. European Journal of Paediatrics 164: 73–9. West H, Richey A, Eyre MB (1934). Study of intelligence levels of juvenile diabetics. Psychological Bulletin 31: 598. Wysocki T, Harris MA, Mauras N, Fox L, Taylor A, Jackson SC, White NH (2003). Absence of adverse effects of severe hypoglycemia on cognitive function in school-aged children with diabetes over 18 months. Diabetes Care 26: 1100–5.
10 Hypoglycaemia in Pregnancy Ann E. Gold and Donald W.M. Pearson
INTRODUCTION Diabetes mellitus is one of the most common medical conditions affecting women during their reproductive years. A successful outcome of diabetic pregnancy can usually be anticipated with current management strategies, although an adverse outcome is still more common than in the non-diabetic population (Casson et al., 1997; Penney et al., 2003a; Evers et al., 2004; Jensen et al., 2004; Confidential Enquiry into Maternal and Child Health, 2005). Meticulous control of blood glucose before conception and throughout gestation is the cornerstone of management to reduce congenital anomalies, neonatal morbidity and mortality. However, striving for continuous normoglycaemia comes at a cost. Many women experience an increased frequency of hypoglycaemia, accompanied by impaired awareness of hypoglycaemia or modification of their hypoglycaemic symptoms. This chapter describes why hypoglycaemia is a recognised problem during pregnancy and how this influences the management of diabetic pregnancies. Population studies have shown that in many countries the average age of mothers with diabetes (type 1 and type 2 diabetes) during pregnancy is around 30 years (Penney et al., 2003a). At the time of their first pregnancy, women with type 1 diabetes have on average had diabetes for over ten years, whereas some will have been exposed to the long-term effects of chronic hyperglycaemia for much longer when they conceive. Since the microvascular complications of diabetes are associated with the duration of the condition, many women have established microangiopathy at the time of conception. Careful preparation for pregnancy and regular obstetrical and medical surveillance throughout pregnancy and delivery are mandatory (SIGN, 2001; American Diabetes Association, 2003), along with rapid access to specialist paediatric facilities for the neonate who may be heavy for dates and premature delivery, often by caesarean section.
METABOLIC CHANGES DURING PREGNANCY Fundamental changes occur in maternal metabolism and physiology during pregnancy. Over 280 days the mother’s weight increases on average by 12.5 kg. The main increase in weight occurs in the second half of pregnancy and is caused by the growth of the conceptus, the enlargement of maternal organs, maternal storage of fat and protein and an increase in maternal blood volume and interstitial fluid. An increase in the basal metabolic rate results in the need for increased energy intake. In addition throughout pregnancy maternal metabolism
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
218
HYPOGLYCAEMIA IN PREGNANCY
adapts to ensure an adequate supply of nutrients to the growing fetus and developing placenta. A normal pregnancy is characterised by major alterations of glucose homeostasis. Fasting glucose declines and, although the plasma glucose is elevated after an oral glucose tolerance test, the mean plasma glucose level is around 4 mmol/l during the third trimester of a normal non-diabetic pregnancy on a normal diet (Paretti et al., 2001). Development of the placenta in the uterus during the first trimester of pregnancy occurs in a low oxygen environment when maternal blood supply is restricted. During this time fetal metabolism is heavily anaerobic, which may serve to protect the developing embryo from oxygen free radical-mediated teratogenesis. At the start of the second trimester, when organogenesis is complete, the maternal circulation develops to support fetal growth. In the second and third trimester the development of insulin resistance leads to increased insulin secretion to avoid abnormal increases in glucose, free fatty acids and amino acids. In normal pregnancy insulin sensitivity is decreased by between 30 and 60%. Changing hormonal levels make a major contribution to insulin resistance. Human placental lactogen (HPL) has actions similar to growth hormone. It increases lipolysis with a rise in free fatty acids which are a steady source of energy for the mother and fetus during periods of starvation. Progesterone is also associated with insulin resistance. Maternal lipid stores increase during pregnancy and adipokines and cytokines may play a role in the development of increasing insulin resistance. The cytokine tumour necrosis factor-alpha (TNF-) rises as the fat mass increases and can be related to insulin resistance. In pregnant women a decrease of adiponectin has been shown to relate to increasing insulin resistance in the third trimester. In women with type 1 diabetes the physiological development of insulin resistance during pregnancy poses challenges to the expectant mother who is attempting to maintain normoglycaemia. Many other changes in physiology occur in pregnancy. The complex process of placental development is mostly complete by the end of the second trimester though the placenta continues to expand with the growing fetus. In the third trimester maternal metabolism switches from anabolism to catabolism, permitting an enhanced transfer of nutrients across the placenta to sustain rapid fetal growth. The placenta is an active organ in this process. In addition to synthesising various hormones the placenta regulates the transfer of maternal fuels to the fetus and facilitates maternal metabolic adaptation at different stages of pregnancy. Cells in contact with the maternal circulation and fetal circulation have a range of receptors, transporters and channels on both placental surfaces. At term the placenta of the mother with diabetes shows a number of differences from those in women who do not have diabetes. These include changes in morphology, blood flow, transport and metabolism. This is important since transplacental transport of glucose is a facilitated process and net transfer is strongly dependent on the concentration gradient of glucose between the maternal and fetal blood. However, the correlation between various indices of glucose control – e.g. HbA1c and fetal growth – is poor, suggesting that factors other than maternal hyperglycaemia contribute to accelerated fetal growth (Penney et al., 2003b). Up-regulation of placental glucose transporters in type 1 diabetes may contribute to increased placental glucose transfer and stimulate fetal growth even if the mother has excellent glycaemic control. Transport of amino acids across the human placenta is an active process resulting in amino acid concentrations in the fetal circulation that are substantially higher than those in the maternal circulation. Management strategies in women with type 1 diabetes need to take the metabolic adaptations of pregnancy into account. Although insulin resistance is the characteristic feature
FREQUENCY OF HYPOGLYCAEMIA IN DIABETIC PREGNANCY
219
of the later stage of pregnancy, in the first trimester a modest increase in insulin sensitivity occurs. The Diabetes In Early Pregnancy study (DIEP) reported declining insulin requirements in the middle of the first trimester of pregnancy in women with type 1 diabetes (Jovanovic et al., 2001). Over-insulinisation at this stage may be an issue since women will be striving for optimal glycaemic control during the crucial period of organogenesis. Hyperemesis gravidarum may also contribute to an increased risk of hypoglycaemia.
FREQUENCY OF HYPOGLYCAEMIA IN DIABETIC PREGNANCY Hypoglycaemia is a frequent problem experienced by women with diabetes during pregnancy. A number of studies have reported the frequency of hypoglycaemia during pregnancy in women with diabetes but comparison is made difficult by the variations used in the definition of hypoglycaemia and the methods used to collect the data. Table 10.1 summarises the frequency of hypoglycaemia in some studies of women with pre-gestational diabetes. In all of the studies, with the exception of that by Persson and Hanson (1993), severe hypoglycaemia was common during pregnancy in women with pre-gestational diabetes. In the study by Persson and Hanson (1993), a lower incidence of hypoglycaemia was reported using an intensive insulin regimen combined with very frequent self-monitoring, which may partly account for the difference in frequency as compared with the other studies. Most studies have demonstrated that the peak incidence of hypoglycaemia occurs during the first and second trimesters. Kimmerle et al. (1992) observed that 84% of hypoglycaemic episodes that resulted in impaired consciousness occurred before week 20. A peak incidence of hypoglycaemia was observed during weeks 10–15 in a study of women receiving intensive insulin therapy (Rosenn et al., 1995). In the Diabetes Control and Complications Trial (DCCT), a similar number of episodes of hypoglycaemia was recorded in the first and second trimesters and fewer episodes were reported during the third trimester. In a larger study of 323 women, severe hypoglycaemia was almost 2.5 times more frequent in the first trimester compared with the third trimester (Evers et al., 2002a; 2004). However, the reported incidence varies considerably between studies, which may represent differences in patient groups and management strategies as well as varying definitions of severe hypoglycaemia. Nocturnal hypoglycaemia is particularly common during pregnancy (Kimmerle et al., 1992). The advent of continuous blood glucose monitoring (CBGM) has confirmed this high incidence of nocturnal hypoglycaemia during pregnancy in mothers with type 1 diabetes (Yogev et al., 2003). In this study, 34 women were monitored for a 72-hour period between weeks 16 and 32 of pregnancy. During this short time period, nocturnal hypoglycaemic events were recorded in 26 (76%) women but only 17 of the patients experienced symptoms. In all of the affected patients an interval of 1–4 hours elapsed before any clinical manifestations of hypoglycaemia were apparent. Comparison of two national audits of diabetic pregnancy in Scotland has shown that significantly more women experienced severe hypoglycaemia in 1998–1999 than in 2003– 2004 (41 versus 30%). This difference may be explained by a number of factors, such as more women attending for pre-pregnancy care in 2003–2004 and the application of newer insulin regimens that employed short-acting analogues (Scottish Diabetes in Pregnancy Study Group, 2004).
HYPOGLYCAEMIA IN PREGNANCY
220
Table 10.1 Summary of studies of the incidence of ‘severe’ hypoglycaemia in pre-gestational diabetic pregnancy Reference
Type of diabetes
Kimmerle et al., 1992
Type 1
Persson et al., 1993 Rosenn et al., 1995
Type 1
DCCT, 1996
Type 1
Masson et al., 2003
Type 1 (all taking lispro insulin) Type 1 (all taking lispro insulin) Type 1
Garg et al., 2003 Evers et al. 2002a; 2004 Scottish Diabetes in Pregnancy Study Group, 2004
Type 1
Type 1
Definition of severe hypoglycaemia
Impaired conscious level responding to glucose/glucagon Requiring external help for recovery Subdivided into: a) requiring external help b) coma/seizure Seizure or loss of consciousness
n
Women experiencing severe hypoglycaemia during pregnancy 41% (77% episodes occurred during sleep) 4.4%
77 (85 pregnancies) 113
71% 34%
84
17% in intensive group 19.8% in conventional group 27%
180 (270 pregnancies)
76
Requiring external help for recovery
23% overall
62
Requiring external help for recovery Requiring external help for recovery
41% 1st trimester 17% 3rd trimester 30% overall 20% 1st trimester 17% 2nd trimester 10% 3rd trimester
278 (2002) 323 (2004) 155
Requiring external help for recovery
Hypoglycaemia also occurs in women with type 2 diabetes, who often require insulin in the pre-pregnancy period in order to optimise glycaemic control, and also in women who develop gestational diabetes (typically in the late second and third trimesters). However, the frequency of hypoglycaemia has not been documented with accuracy in these groups; it is the authors’ impression that hypoglycaemia occurs much less frequently than in type 1 diabetes.
Why are Women with Pre-gestational Diabetes at Greater Risk of Hypoglycaemia during Pregnancy? Great importance is placed upon maintaining good glycaemic control throughout diabetic pregnancy when women are highly motivated and often achieve levels of glycated
FREQUENCY OF HYPOGLYCAEMIA IN DIABETIC PREGNANCY
221
haemoglobin that are close to the non-diabetic range. In addition, because of the changes in insulin sensitivity in the first trimester, their risk of severe hypoglycaemia is increased. It has been shown that women with a previous history of hypoglycaemia are at increased risk of experiencing severe hypoglycaemia during pregnancy (Kimmerle et al., 1992; Evers et al., 2002a). Women who demonstrate a greater fluctuation in blood glucose during pregnancy are also at greater risk of experiencing severe hypoglycaemia (Rosenn et al., 1995) and women who are not using an ‘intensive’ insulin regimen have an enhanced risk. Evers et al. (2002a) observed that the risks of experiencing severe hypoglycaemia during the first trimester were significantly greater in women who had duration of diabetes greater than ten years, HbA1c less than 6.5% or an insulin dose of greater than 0.1 units of insulin per kg body weight. It is possible that impaired counterregulation could contribute to the increased risk of hypoglycaemia. A rat model demonstrated that the glucagon and epinephrine (but not norepinephrine) responses to hypoglycaemia (plasma glucose 3.4 mmol/l) were suppressed during pregnancy (Rossi et al., 1993); these data suggest that the counterregulatory responses to hypoglycaemia in rats may be impaired. A few studies have examined hormonal counterregulatory responses in pregnant women with type 1 diabetes during experimental hypoglycaemia during pregnancy. All studies utilised a hyperinsulinaemic hypoglycaemic clamp technique, except that by Nisell et al. (1994), which used an insulin bolus injection to induce hypoglycaemia. The results of these studies are shown in Table 10.2. The studies vary in terms of the time of gestation at the time of study and the glucose nadir reached and it is difficult to determine whether counterregulation is impaired during pregnancy as a result of pregnancy per se or because of differences in glycaemic control and the presence of diabetes. The study by Rosenn et al. (1996) probably had the most appropriate study design to answer this question. One criticism, however, is that the glucose nadir achieved was only 3.3 mmol/l, which may not have been sufficient to stimulate a counterregulatory response in the non-diabetic control group. The studies provide conflicting evidence as to whether the counterregulatory response to hypoglycaemia is deficient during pregnancy. No studies have examined counterregulatory effects during the first trimester at the time when the frequency of, and exposure to, hypoglycaemia are at a peak. However, to conduct such studies would raise major ethical concerns. Few studies have examined the development of impaired hypoglycaemia awareness during pregnancy although most clinicians would agree that this is particularly problematical during the first trimester. Evers et al. (2002a) observed that severe hypoglycaemia in the first trimester was more likely to occur in women with reduced symptomatic awareness of hypoglycaemia. In laboratory-induced hypoglycaemia Björklund et al. (1998a) did measure symptomatic responses to hypoglycaemia during the third trimester, and also postnatally, and found that symptoms such as ‘inability to concentrate’, ‘headache’ and ‘pounding heart’ were less prominent during pregnancy compared with during the postnatal period. However, it is difficult to ascertain whether this is a consequence of differences in glycaemic control or the incidence of hypoglycaemia during these two time periods. Fear of hypoglycaemia is also greater in women who have experienced severe hypoglycaemia (Evers et al., 2002a) and this is an important problem for the diabetic mother which should be addressed during antenatal care.
HYPOGLYCAEMIA IN PREGNANCY
222
Table 10.2 Summary of studies examining counterregulatory responses to hypoglycaemia in women with pre-gestational diabetes Reference
Gestation at time of study
Control group
Blood glucose nadir during study
Diamond et al., 1992
21–37 weeks Non-pregnant, 2.5 mmol/l non-diabetic age matched women
Nisell et al., 1994
Last trimester, and 8–12 weeks post-partum
Rosenn et al., 1996
Acted as own controls
3.2 mmol/l
3.3 mmol/l Non-pregnant Non-diabetic, age-matched. (1); Cases were 24–28 (2); also studied 32–34 (3) on 3 occasions i.e. acted as own controls
Björklund 30–34 weeks Acted as own controls et al., (1); 1998a 5–13 months postnatal (2)
2.3 mmol/l
Björklund 30–34 weeks No controls et al., 1998b
2.3 mmol/l
Results
No glucagon response in cases. Epinephrine release suppressed in cases; cf. controls. Lower blood glucose for epinephrine and growth hormone release in cases; cf. controls No glucagon or cortisol response at either time point.
n
9 type 1 cases; 7 controls
8 pregestational type 1 patients; 1 gestational
Similar increases in epinephrine and norepinephrine on both occasions 17 type 1 Reduced epinephrine cases; response in cases; cf. 10 controls controls at all study times. Reduced epinephrine response in cases during pregnancy; cf. pre-pregnancy. Growth hormone responses reduced in pregnancy in cases and controls; cf. pre-pregnancy Epinephrine response 10 similar in pregnancy and postnatal. Dehydroepiandrosterone increased more rapidly and less sustained during pregnancy. Growth hormone responses reduced in pregnancy Increase in placental growth hormone, but no changes in other placental hormones during hypoglycaemia
10
CLINICAL MANAGEMENT BEFORE AND DURING PREGNANCY
223
CLINICAL MANAGEMENT BEFORE AND DURING PREGNANCY Pre-conception Care The advantages of planned pregnancy should be regularly emphasised to women with diabetes during their reproductive years and effective contraceptive advice should be part of routine clinical care. After menarche, all girls with type 1 diabetes should be aware of the importance of pregnancy planning, because the infants of mothers who have attended for pre-pregnancy care have fewer major congenital malformations and require shorter periods in special care facilities than infants of mothers who do not attend for pre-conception care (Fuhrmann et al., 1983; Steel et al., 1989; 1990; Kitzmiller et al., 1991; Ray et al., 2001). Attendance for structured pre-pregnancy care is also associated with a reduction in the rate of spontaneous abortion. Pre-pregnancy counselling should address ways of minimising the risk of severe hypoglycaemia both before and during pregnancy. Ideally this should be discussed during the pre-pregnancy period. However, if the pregnancy has not been planned, women should be made aware of their increased risk of hypoglycaemia during pregnancy and how to avoid and manage potential episodes, particularly as the greatest risk is during the first trimester.
Organisation of Clinical Care In many centres clinical care is delivered by a multidisciplinary combined obstetric/diabetic team with very regular out-patient reviews to assess metabolic control and obstetric progress (Figure 10.1). Home blood glucose monitoring results are assessed and insulin regimen and dietary intake modified to optimise glycaemic control and HbA1c (Figure 10.2). Most women present for booking at around eight weeks gestation when an early scan will provide an accurate estimate of gestational age. This is important to allow the optimal time of delivery to be determined. Screening is performed routinely for Down’s syndrome and neural-tube defects. Although the prevalence of congenital anomaly has declined following the introduction of pre-conceptional counselling, the incidence of congenital malformation is still higher than in the non-diabetic population. A detailed ultrasound scan at around 20 weeks is performed to detect severe congenital anomalies, particularly to identify major malformation of the heart and the central nervous system. Frequent scanning is performed later in pregnancy to monitor fetal growth. In the third trimester regular cardiotocography, Doppler ultrasound and fetal movement charts are used to monitor fetal progress. Figure 10.3 demonstrates the measurement of abdominal circumference (AC). Sequential measurements of AC are used to monitor fetal growth. Good glycaemic control reduces stillbirth rate, neonatal hypoglycaemia and respiratory distress syndrome. Women should strive to maintain blood glucose levels as near to the nondiabetic range as possible without an excessive risk of hypoglycaemia. This usually means blood glucose target levels between 4 and 7 mmol/l. The diabetes team, but in particular the diabetes specialist nurses and specialist midwives, have an important role in educating women on the need for home blood glucose monitoring (usually four to six times a day) and in introducing intensive insulin regimens if the women are not already on these programmes. Maternal issues in pregnancy are shown in Table 10.3.
HYPOGLYCAEMIA IN PREGNANCY
224
Type 1 and type 2 diabetes in pregnancy; Gestational diabetes on insulin Blood tests Pre-pregnancy
HbA1c
Hb
U&E
TFT
<12 weeks
HbA1c
Hb
U&E
TFT
12 weeks
HbA1c
16 weeks
HbA1c
20 weeks
HbA1c
24 weeks
HbA1c
28 weeks
HbA1c
Other Tests Rubella
BTS
Eyes
MSU/MA
Eyes
MSU/MA
Scanning Miscellaneous
FVS
Folic acid
Dietician Midwife
Folic acid
Dietician Midwife
FVS
Dietician Midwife
AFP
Eyes
MSU/MA
Detailed
Midwife
Growth
Hb
U&E
TFT
BTS
Growth
Anti-D if Rh -ve
30 weeks
32 weeks
Dietician Midwife
HbA1c
Eyes
MSU/MA
Growth
34 weeks
36 weeks
Counselling
Anti-D if Rh -ve
HbA1c
Growth
Midwife
37 weeks
38 weeks
(Doppler)
39 weeks
HbA1c
Postnatal
HbA1c
(Doppler)
Hb
MSU/MA
Figure 10.1 Chart used in the combined diabetic antenatal clinic at Aberdeen Maternity Hospital, Scotland, indicating when investigations and scans should be undertaken
Figure 10.2 Example of home blood glucose monitoring chart used in the combined antenatal diabetic clinic at Aberdeen Maternity Hospital, Scotland
226
HYPOGLYCAEMIA IN PREGNANCY
Figure 10.3 Ultrasound scan of at 32 weeks gestation demonstrating measurement of abdominal circumference (AC)
Optimising Insulin Regimens Data is limited with regard to studies of different insulin regimens during pregnancy and many studies are small and observational in nature. The effects of conventional versus intensive insulin therapy were studied in the women participating in the DCCT (The Diabetes Control and Complications Trial Research Group, 1996). In women who had initially been
CLINICAL MANAGEMENT BEFORE AND DURING PREGNANCY
227
Table 10.3 Maternal issues in pregnancy Diabetes-related Hypoglycaemia Increased risk of ketoacidosis Retinopathy may deteriorate Nephropathy and fluid retention Pre-eclampsia/hypertension Increased caesarean section rate
Obstetric Preterm delivery Increased caesarean section rate Macrosomia and shoulder dystocia
assigned to conventional treatment, the occurrence of hypoglycaemia resulting in impaired conscious level was 4.0 per 100 pregnant-months compared with 2.8 in those receiving intensive treatment. Those women in the conventional group who had changed to intensive therapy before conception experienced significantly fewer episodes of hypoglycaemia during pregnancy than those who had not intensified their regimen until after conception (2.0 versus 5.0 episodes per 100 patient years). This suggests that it may be preferable, where possible, to intensify the insulin regimen during the pre-pregnancy period rather than making the change once conception has occurred. Another study of 118 women with pre-gestational diabetes compared twice daily free mixing of soluble and isophane insulins with a basal-bolus regimen of soluble insulin before meals and bedtime isophane insulin (Nachum et al., 1999). The authors reported no difference between the regimens in the incidence of hypoglycaemia, although those taking four injections of insulin per day achieved better glycaemic control. One interpretation of the data is that the greater the insulin dosage, the better the glycaemic control. However, the overall incidence of recorded hypoglycaemia was remarkably low (10%) and almost half the women had type 2 diabetes. Short-acting insulin analogues have received some attention and one randomised trial of insulin lispro in pregnancies complicated by type 1 diabetes has been reported (Persson et al., 2002). Thirty-three women were studied with 16 being randomised to lispro and 17 to soluble insulin at gestational week 15. Most of the women had very good glycaemic control at entry to the study (mean HbA1c 6.5%, range 4.5–8.6%). Only two patients experienced severe hypoglycaemia during the remainder of their pregnancies and both were on human soluble insulin, although it is important to remember that the study was undertaken in the latter half of pregnancy when the risk of hypoglycaemia is usually lower. However, a significantly higher rate of biochemical hypoglycaemia (< 30 mmol/l) was documented in the women using lispro (who were monitoring blood glucose six times daily). No differences were observed between the groups with regard to glycated haemoglobin, although women taking lispro had lower postprandial glucose excursions. In an observational study in which women received lispro before and during pregnancy, the prevalence of severe hypoglycaemia was 23%, with no greater incidence of adverse outcomes compared to other regimens (Garg et al., 2003). A similar prospective study that was designed to assess progression of retinopathy during pregnancy in women on lispro, demonstrated that the number of subjective hypoglycaemic events did not differ between those taking lispro and those taking soluble insulin, although the overall rates of hypoglycaemia were relatively low (Loukovaara et al., 2003).
228
HYPOGLYCAEMIA IN PREGNANCY
In gestational diabetes, when lispro was compared with human soluble (regular) insulin, a significant difference was demonstrated in the incidence of hypoglycaemia (Jovanovic et al., 1999). However, the recorded incidence of hypoglycaemia was again very low and it is not appropriate to extrapolate this data to pregnancy in type 1 diabetes. In a recent large study of diabetic pregnancy from the Netherlands (Evers et al., 2004), the rate of severe hypoglycaemia in a subgroup of 35 patients who were using lispro, did not differ from that observed in patients using other insulin regimens. The congenital malformation rates in those women using lispro have not differed from those on soluble insulin (Wyatt et al., 2004). A recently reported multi-centre study by Mathies on et al. (2007) has demonstrated the safe use of insulin aspart in Type 1 diabetic pregnancy, with a non-significantly lower incidence of severe hypoglycaemia compared to human soluble insulin at comparable levels of glycaemic control (Hb A1c mostly <6.5%). In summary, although short-acting insulin analogues are often used because they may be associated with a lower incidence of severe hypoglycaemia than soluble insulin, there is no firm evidence to support this premise in pregnant women with type 1 diabetes. It is possible that the results of studies are biased by the exclusion of women who have a history of significant severe hypoglycaemia. In addition, studies in pregnant women with type 1 diabetes may not be comparable with those in non-pregnant women with type 1 diabetes; pregnant women tend to be more highly motivated and monitor their glycaemic control very intensively. Many women have now completed pregnancy using the long-acting insulin analogue insulin glargine, and although this does appear to be a safe preparation, only anecdotal case reports are available with no larger studies to support its use. The data on the use of insulin pumps and the frequency of hypoglycaemia during pregnancy are very limited. One small study of 25 women did not observe any differences in glycaemic control or frequency of hypoglycaemia compared with conventional intensive insulin therapy. However, only episodes of hypoglycaemic coma were recorded and the rates in both groups were too low to draw any firm conclusions (Lapolla et al., 2003). In clinical practice the optimal insulin regimen for a patient will depend on a variety of factors including patient preference, lifestyle and previous problems with hypoglycaemia. Many women will use a basal-bolus regimen using isophane insulin at bedtime or a long-acting insulin analogue, once daily, to control fasting blood glucose and avoid nocturnal hypoglycaemia. Some women take a second dose of isophane before breakfast to achieve satisfactory glycaemic control between meals during the daytime hours. The mealtime ‘boluses’ of short-acting insulin are taken usually before or after breakfast, lunch and the evening meal. The dosages vary depending on the amount of carbohydrate in the meal and the anticipated expenditure of energy after the meal. Short-acting insulin analogues are increasingly used to achieve satisfactory postprandial glycaemia control. A basal-bolus regimen has the advantage of increased flexibility with regard to exercise, and the timing and size of the meal, and it is most likely to work effectively in the appropriately educated and motivated patient. Some women cannot achieve acceptable glycaemic control on currently available insulin regimens. Continuous subcutaneous insulin infusion (CSII) may provide an alternative for this group. As demonstrated by the DCCT data, it is sensible to institute any changes a few months prior to conception. This ensures that the women are well educated about the regimen and able to cope with varying situations. It is very important that the patient using CSII is fully educated and supported to ensure that the benefits from this type of regimen are achieved. Midwifery and obstetric staff may need help and guidance for this type of approach during
CLINICAL MANAGEMENT BEFORE AND DURING PREGNANCY
229
Table 10.4 Insulin treatment in Gestational Diabetes Mellitus (GDM) (SIGN 2001) • Intensive insulin regimens are occasionally recommended to achieve normal blood glucose levels and fetal growth • Women requiring insulin require a comprehensive education package regarding all aspects of insulin treatment Consider insulin treatment for GDM if: • The fasting blood glucose is over 6 mmol/l or the 2 hr postprandial glucose is over 7 mmol/l, despite appropriate lifestyle adjustments • The ultrasound scans show excessive growth with an abdominal circumference greater than the 95 percentile for gestation (some authorities recommend > 75 percentile)
pregnancy and labour. In some European countries, a significant percentage of women use this type of treatment during pregnancy but in many countries only a relatively small percentage use CSII. Although an appropriate insulin regimen is very important many other factors can influence insulin absorption and effectiveness. Pregnancy is a good time to review all factors that could contribute to suboptimal glycaemic control. All practical aspects of insulin administration should be reviewed and injection sites examined. Some women with gestational diabetes may require insulin during the last few weeks of pregnancy. The management of insulin treatment in gestational diabetes is summarised in Table 10.4.
Dietary and Lifestyle Management Dietary advice is also essential before, during and after pregnancy. Such advice will encourage the intake of foods with a high level of complex carbohydrates, soluble fibre and vitamins. Balancing of carbohydrate intake and insulin dose is important and many women may find it helpful to count carbohydrate portions and learn appropriate correction factors for high blood glucose levels. Dietetic review is therefore appropriate and a review of ‘sick-day rules’ should be undertaken. In the early weeks of pregnancy, where nausea and vomiting are common, this is particularly important. During the later weeks of pregnancy, women may prefer to take smaller but more frequent meals and their insulin regimen should be adjusted accordingly. In general, women are also advised to reduce their intake of saturated fat. Neural tube defects in high risk pregnancies are associated with low levels of folate. A large study of non-diabetic women has shown that the prescription of 4 mg of folate supplement, pre- and peri-conception, provides protection against neural tube defects in women at high risk. In Scotland, all women are advised to take 5 mg of folic acid pre-conception, and continue this preparation during pregnancy until around 12 weeks of gestation. Smoking should be discouraged. If women are used to taking regular physical activity this should be encouraged, along with advice about reducing insulin dosages before exercise and ensuring appropriate carbohydrate intake.
230
HYPOGLYCAEMIA IN PREGNANCY
Correction of Hypoglycaemia Advice concerning the correction of hypoglycaemia by appropriate dietary intake should be given and is similar to that for the non-pregnant patient with diabetes (see Chapter 5). Patients, their partners and other family members should be advised on the use of glucose gels, such as GlucoGel, for emergency use, and should keep a supply of this available. Instruction about the use of glucagon should also be given to partners or close family members. This should be done at the first review, since hypoglycaemia is more common in the first trimester.
Management of Delivery The infant of a diabetic mother is at increased risk during labour and delivery. Mothers with diabetes must be recognised to be a high risk obstetric patient and delivered in a unit that can provide experienced obstetric care and immediate access to neonatal intensive support.
Timing of Delivery There is no evidence to support the continuation of a type 1 diabetic pregnancy beyond 40 weeks and, provided glycaemic control has been satisfactory and fetal growth has not been excessive, induction of labour and delivery at around 39 weeks appears to provide a compromise between delivering the baby too early and avoiding late unexplained intrauterine death.
Management of Diabetes during Labour Management of diabetes in labour is summarised in Table 10.5. The aim is to achieve maternal normoglycaemia during labour using an intravenous dextrose/insulin infusion. The insulin infusion rate must be halved immediately after delivery to prevent postnatal hypoglycaemia, since there is a rapid increase in insulin sensitivity after separation of the placenta. After delivery, the insulin dose should be reduced to the dosages used before pregnancy. This will usually necessitate halving the doses required during pregnancy. If the mother decides on breast feeding, a further reduction of the insulin dosage may be required to avoid hypoglycaemia.
MATERNAL COMPLICATIONS OF DIABETES DURING PREGNANCY The complications experienced during pregnancy include those resulting from hypoglycaemia and from the microvascular complications of diabetes.
MATERNAL COMPLICATIONS OF DIABETES DURING PREGNANCY
231
Table 10.5 Management of diabetes during labour and delivery (clinical circumstances may lead to modifications in this regimen) • Fast the patient on the day of delivery for an elective section; otherwise fast from the onset of established labour. • Set up an intravenous insulin infusion using a syringe pump, e.g. 50 units short-acting analogue in 50 ml saline. Start at 1–2 units per hour. • Simultaneously commence a 10% dextrose +01% KCI intravenous infusion at a rate of 100 ml/hr. • Monitor bedside glucose hourly and set a target blood glucose of 4–6 mmol/l. • Use the algorithm (figure) to adjust the insulin infusion. • Halve the rate of the insulin infusion on separation and delivery of the placenta. Continue the dextrose infusion as before. • Revert to the pre-pregnancy dosages of subcutaneous insulin when a normal eating pattern is established. Insulin infusion scale Start insulin at 2 units per hour. Monitor capillary bedside glucose hourly. Capillary blood glucose (mmol/l) <4 Reduce insulin by 0.5 unit/hr
4–6 Continue current rate
>6 Increase by 0.5 unit/hr
The Risks of Maternal Hypoglycaemia to the Mother The morbidity and mortality of hypoglycaemia in diabetes is discussed in Chapter 12. During pregnancy the risks are similar but compounded by the potential risk to the fetus caused by injury. A maternal death was reported in one series (Rayburn et al., 1986). One particular concern in the pre-pregnancy and pregnancy clinics is the risk of hypoglycaemia while driving. One woman was involved in a motor vehicle accident after experiencing a hypoglycaemia-induced convulsion and fracturing her tibia and fibula. When planning pregnancy, advice on the avoidance of hypoglycaemia while driving should be reinforced for patients treated with insulin (see Chapter 14), particularly with respect to testing blood glucose before every journey and ensuring appropriate management of hypoglycaemia if it occurs while driving. They should also be warned that they will be advised to cease driving if their warning symptoms of hypoglycaemia are significantly reduced. Women who develop gestational diabetes and require insulin treatment should be given similar advice although the risk of developing severe hypoglycaemia is lower.
232
HYPOGLYCAEMIA IN PREGNANCY
Microvascular Complications of Pregnancy Retinopathy is common and may progress during pregnancy. Therefore retinal screening should be undertaken during each trimester (Figure 10.1). In the longer term, parous women with type 1 diabetes have significantly lower levels of all types of retinopathy compared with non-parous women. The associated significant differences in HbA1c suggest that the improved glycaemic control that is associated with pregnancy may be sustained over time with beneficial effects on long-term complications. Thus women should be reassured that strict glycaemic control during, and immediately after, pregnancy can effectively reduce the long-term risk of retinopathy. In pregnancy, diabetic nephropathy of any degree is less common than retinopathy, and requires specialist management.
COMPLICATIONS IN THE INFANT OF THE DIABETIC MOTHER The potential complications for the infant of a mother with diabetes are shown in Table 10.6. Newly born infants with diabetic mothers are at increased risk of developing hypoglycaemia. To avoid this risk, maternal blood glucose should be kept as close to normal as possible during pregnancy and particularly during labour and delivery. Early feeding of the newly born infant is essential and careful monitoring is mandatory.
The Risks of Maternal Hypoglycaemia to the Fetus/infant The potential risks of hypoglycaemia to the offspring can be considered in three main ways: the teratogenic effects of hypoglycaemia, other immediate effects on the fetus and delayed effects.
Table 10.6 Potential problems affecting the infant of a mother with diabetes Neonatal and perinatal • Hypoglycaemia • Stillbirth • Increased perinatal mortality • Congenital anomalies • Preterm delivery • Macrosomia • Birth trauma at delivery • Polycythaemia • Respiratory distress syndrome
COMPLICATIONS IN THE INFANT OF THE DIABETIC MOTHER
233
Data in animals suggest that hypoglycaemia in the first trimester could be teratogenic (Buchanan et al., 1986; Akazawa et al., 1987, 1989; Ellington, 1987; Smoak and Sadler, 1990). However, it is very difficult to extrapolate these results to human pregnancy as the exposure to hypoglycaemia in these animal studies would equate to sustained hypoglycaemia in humans of at least five hours. There also appears to be a differential effect between species making it inappropriate to extrapolate results from one species to another (Eriksson et al., 1986). In non-diabetic women in whom hypoglycaemic coma was induced as a treatment for psychiatric illness, the fetus was affected more frequently when hypoglycaemia was induced during the first trimester (Impastato et al., 1964). However many potential confounding variables were present, and the blood glucose profiles in women with type 1 diabetes are completely different from the non-diabetic woman. When viewing the outcomes of diabetic pregnancy in humans, the overwhelming evidence supports the achievement of strict glycaemic control in early pregnancy, which is usually associated with a higher risk of hypoglycaemia but greatly reduces the risks of congenital malformation (Kitzmiller et al., 1991; Kimmerle et al., 1992; Evers et al., 2004; Rosenn et al., 1995; The Diabetes Control and Complications Trial Research Group, 1996). This suggests that the risk of any teratogenic effect of hypoglycaemia is greatly outweighed by the benefits of good glycaemic control. A large follow-up study of 329 infants born to mothers with diabetes demonstrated a high perinatal fetal mortality rate of 21%, but no conclusive evidence could implicate hypoglycaemia or any other iatrogenic factor (Farquhar, 1969). In another study of women who had relatively good glycaemic control (Mills et al., 1988), all episodes of documented hypoglycaemia between gestational weeks 5 and 12 were recorded and no association with fetal malformations was observed. Similarly, in a study by Kitzmiller et al. (1991), all episodes of symptomatic hypoglycaemia were documented and no association with fetal malformation was found. In addition to the reduced risk of congenital malformations, other complications of pregnancy appear to be fewer in women who experience hypoglycaemia during pregnancy. The incidence of macrosomia has been shown to be significantly lower in women affected by severe hypoglycaemia during the third trimester (relative risk 0.66) (Evers et al., 2002b, Evers et al., 2004). With respect to the immediate effects of maternal hypoglycaemia on the fetus in utero, some data are available from studies that have induced experimental hypoglycaemia during pregnancy. Early studies suggested that fetal heart rate variability may be decreased during maternal hypoglycaemia (Stangenberg et al., 1983). However, later studies did not demonstrate changes in fetal movements, breathing or heart rate during maternal hypoglycaemia (Nisell et al., 1994, Reece et al., 1995). Björklund et al. (1996) also studied the effects of maternal blood glucose of 2.2 mmol/l on fetal heart rate. They observed an increase in frequency and amplitude of fetal heart rate accelerations, but no potentially harmful effects on fetal heart rate or on umbilical Doppler waveform analysis were seen. In a more recent study of 116 women who experienced ‘hypoglycaemia’ rather than hyperglycaemia while undergoing a glucose tolerance test to screen for gestational diabetes, no adverse effects on fetal growth or other perinatal outcomes were observed, although the level of blood glucose that was designated as ‘hypoglycaemia’ (4.9 mmol/l) may well be considered too high to represent this state (Calfee et al., 1999). The later developmental effects in the infant that has been subjected to maternal hypoglycaemia may not become apparent until later in childhood. However, there are many
234
HYPOGLYCAEMIA IN PREGNANCY
potential confounding factors when studying this aspect. For example, a woman who has been exposed to hypoglycaemia may also have experienced hyperglycaemia, which could also have a potentially deleterious effect on cognition and it is not possible to separate these effects. Early studies did demonstrate that the children of diabetic mothers developed cerebral dysfunction (Yssing, 1974) and one study reported an association between the IQ of offspring and the presence of acetonuria, but not hypoglycaemia, during pregnancy in the mother (Churchill et al., 1969). A Japanese study of 33 children born to mothers with diabetes demonstrated significantly lower intelligence scores than a control group when tested at three years of age (Yamashita et al., 1996). However, no correlation was observed between IQ and maternal HbA1c scores during pregnancy. An electrophysiological study of 60 infants (34 controls and 26 children from mothers with diabetes) performed at six months of age, also demonstrated deficits in recognition memory (suggesting hippocampal damage) in the offspring from mothers with diabetes that was unrelated to maternal glycaemic control, although the incidence of maternal hypoglycaemia was not measured (Nelson et al., 2000). In summary, there are no data that suggest that hypoglycaemia is teratogenic in humans or hypoglycaemia has an immediate adverse effect on the fetus. Some studies suggest that the offspring of diabetic mothers may have some differences in cerebral function, although these differences have not been attributed to maternal hypoglycaemia.
CONCLUSIONS • The outlook for the offspring of women with type 1 diabetes has improved dramatically over the last 30 years, and the benefit to the infant from optimal glycaemic control during pregnancy is evident. • The risk of maternal hypoglycaemia is considerable, particularly in the first and second trimesters, and severe hypoglycaemia is common. • Factors that may contribute to increased risk of hypoglycaemia include the requirement for strict glycaemic control, possible effects of pregnancy to cause counterregulatory deficiency and altered symptomatic awareness, differing insulin regimens and lifestyle/dietary factors. • Detailed education is essential to minimise the risk of hypoglycaemia. • There is no evidence to suggest that hypoglycaemia has an adverse effect on the human fetus or the infant of a diabetic mother, although significant maternal morbidity may occur. • Good organisation of pre-pregnancy and pregnancy services is required to reduce the risks of maternal and infant morbidity during diabetic pregnancy.
REFERENCES Akazawa M, Akazawa S, Hashimoto M, Yamaguchi Y, Kuriya N, Toyama K et al. (1987). Effects of hypoglycaemia on early embryogenesis in rat embryo organ culture. Diabetologia 30: 791–6.
REFERENCES
235
Akazawa M, Akazawa S, Hashimoto M, Akashi M, Yamazaki H, Tahara D et al. (1989). Effects of brief exposure to insulin-induced hypoglycemic serum during organogenesis in rat embryo culture. Diabetes 38: 1573–8. American Diabetes Association (2003). Preconception care of women with diabetes. Diabetes Care 26 (Suppl 1): S91–3. Björklund AO, Adamson UKC, Almström NHH, Enocksson EA, Gennser GM, Lins PE, Westgren LM (1996). Effects of hypoglycaemia on fetal heart activity and umbilical artery Doppler velocity waveforms in pregnant women with insulin-dependent diabetes mellitus. British Journal of Obstetrics and Gynaecology 103: 413–20. Björklund A, Adamson U, Andreasson K, Carlstrom K, Hennen G, Ogout A, Lins PE, Westgren M (1998a). Hormonal counterregulation and subjective symptoms during induced hypoglycemia in insulin-dependent diabetes mellitus patients during and after pregnancy. Acta Obstetrica Gynecologica Scandinavica 77: 625–34. Björklund AO, Adamson UKC, Carlström KAM, Hennen G, Igout A, Lins PE, Westgren LM (1998b). Placental hormones during induced hypoglycaemia in pregnant women with insulin-dependent diabetes mellitus: evidence of an active role for placenta in hormonal counter-regulation. British Journal of Obstetrics and Gynaecology 105: 649–55. Buchanan TA, Schemmer JK, Freinkel N (1986). Embryotoxic effects of brief maternal insulinhypoglycemia during organogenesis in the rat. Journal of Clinical Investigation 78: 643–9. Calfee EF, Rust OA, Bofill JA, Ross EL, Morrison JC (1999). Maternal hypoglycemia: is it associated with adverse perinatal outcome? Journal of Perinatology 19: 379–82. Casson IF, Clarke CA, Howard CV, McKendrick O, Pennycook S, Pharoah PO et al. (1997). Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study. British Medical Journal 315: 275–8. Churchill JA, Berendes HW, Nemore J (1969). Neuropsychological deficits in children of diabetic mothers. American Journal of Obstetrics and Gynecology 105: 257–68. Confidential Enquiry into Maternal and Child Health (2005). Pregnancy in women with type 1 and type 2 diabetes in 2002–2003. England, Wales and Northern Ireland. CEMACH, London. Diamond MP, Reece EA, Caprio S, Jones TW, Amiel SA, DeGennaro N, Laudano A et al. (1992). Impairment of counterregulatory hormone responses to hypoglycemia in pregnant women with insulin-dependent diabetes mellitus. American Journal of Obstetrics and Gynecology 166: 70–7. Ellington SK (1987). Development of rat embryos cultured in glucose-deficient media. Diabetes 36: 1372–8. Eriksson UT, Dahlström VE, Lithell HO (1986). Diabetes in pregnancy: influence of background and maternal diabetic state on the incidence of skeletal malformations in the fetal rat. Acta Endocrinologica (Copenhagen) 112 (Suppl 277): 66–73. Evers IM, ter Braak EWMT, de Valk HW, der Schout BV, Janssen N, Visser GA (2002a). Risk indicators predictive of severe hypoglycemia during the first trimester of type 1 diabetic pregnancy. Diabetes Care 25: 554–9. Evers IM, de Walk HW, Mol BWJ, ter Braak EWMT, Visser GHA (2002b). Macrosomia despite good glycaemic control in type 1 diabetic pregnancy; results of a nationwide study in The Netherlands. Diabetologia 45: 1484–9. Evers IM, de Valk HW, Visser GHA (2004). Risk of complications of pregnancy in women with type 1 diabetes: nationwide prospective study in the Netherlands. British Medical Journal 328: 915–8. Farquhar JW (1969). Prognosis for babies born to diabetic mothers in Edinburgh. Archives of Disease in Childhood 44: 36–47. Fuhrmann K, Reiker H, Semmler K, Glockner E (1983). Prevention of congenital malformation in infants of insulin dependent diabetic mothers. Diabetes Care 6: 219–23. Garg SK, Frias JP, Anil S, Gottlieb PA, MacKenzie T, Jackson WE. (2003). Insulin lispro therapy in pregnancies complicated by type 1 diabetes: glycemic control and maternal and fetal outcomes. Endocrine Practice 9: 187-93.
236
HYPOGLYCAEMIA IN PREGNANCY
Impastato DJ, Gabriel AR, Lardaro EH (1964). Electric and insulin shock therapy during pregnancy. Diseases of the Nervous System 25: 542–6. Jensen DM, Damm P, Moelsted-Pedersen L, Ovesen P, Westergaard JG, Moeller M, Beck-Nielsen H (2004). Outcomes in type 1 diabetic pregnancies: a nationwide, population-based study. Diabetes Care, 27: 2819–23. Jovanovic L, Ilic S, Pettitt DJ, Hugo K, Gutierrez M, Bowsher RR, Bastyr EJ (1999). Metabolic and immunologic effects of insulin lispro in gestational diabetes. Diabetes Care 22: 1422–7. Jovanovic L, Knopp RH, Brown, Conleg MR, Park E, Mills JL et al. (2001). Declining insulin requirement in the late first trimester of diabetic pregnancy. Diabetes Care 24: 1130–6. Kimmerle R, Heinemann L, Delecki A, Berger M (1992). Severe hypoglycemia incidence and predisposing factors in 85 pregnancies of type 1 diabetic women. Diabetes Care 15: 1034–7. Kitzmiller JL, Gavin LA, Gin GD, Jovanovic-Petersen L, Main EK, Zingran WD (1991). Preconception care of diabetes. Glycemic control prevents congenital abnormalities. Journal of the American Medical Association 265: 731–6. Lapolla A, Dalfrà MG, Masin M, Bruttomesso D, Piva I, Crepaldi C et al. (2003). Analysis of outcome of pregnancy in type 1 diabetics treated with insulin pump or conventional insulin therapy. Acta Diabetologica 40: 143–9. Loukovaara S, Immonen I, Teramo KA, Kaaja R (2003). Progression of retinopathy during pregnancy in type 1 diabetic women treated with insulin lispro. Diabetes Care 26: 1193–8. Masson EA, Patmore JE, Brash PD, Baxter M, Caldwell G, Gallen IW et al. (2003). Pregnancy outcome in type 1 diabetes mellitus treated with insulin lispro (Humalog). Diabetic Medicine 20: 46–50. Mathieson ER, Kinsley B, Amiel SA, Heller S, McCance D, Duran S, Bellaire S, Raben A on behalf of the Insulin Aspart Pregnancy Study Group (2007). Maternal glycemic control and hypoglycaemia in Type 1 diabetic pregnancy. Diabetes Care 30: 771–6. Mills JL, Knopp RH, Simpson JL, Jovanovic-Peterson L, Metzger BE, Holmes, LB et al. (1988). Lack of relation of increased malformation rates in infant of diabetic mothers to glycemic control during organogenesis. New England Journal of Medicine 318: 671–6. Nachum Z, Ben-Shlomo I, Weiner E, Shalev E (1999). Twice versus four times daily insulin dose regimens for diabetes in pregnancy: randomised controlled trial. British Medical Journal 319: 1223–7. Nelson CA, Wewerka S, Thomas KM, Tribby-Walbridge S, deRegnier R-A, Georgieff M (2000). Neurocognitive sequelae of infants of diabetic mothers. Behavioral Neuroscience 114: 950–6. Nisell H, Persson B, Hanson U, Lunell NO, Nylund L, Sarby B, Thronstrom S (1994). Hormonal, metabolic, and circulatory responses to insulin-induced hypoglycemia in pregnant and nonpregnant women with insulin-dependent diabetes. American Journal of Perinatology 11: 231–6. Paretti E, Mecacci I, Papini M, Cloni R, Carignani L, Mignosa M et al. (2001). Third trimester maternal glucose levels from diurnal profiles in non-diabetic pregnancies. Correlation with sonographic perameters of fetal growth. Diabetes Care 24: 1319–23. Penney GC, Mair G, Pearson DW (2003a). Outcomes of pregnancies in women with type 1 diabetes in Scotland: a national population-based study. British Journal of Obstetrics and Gynaecology 110: 315–18. Penney GC, Mair G, Pearson DW (2003b). The relationship between birth weight and maternal glycated haemoglobin (HbA1c ) concentration in pregnancies complicated by type 1 diabetes. Diabetic Medicine 20: 162–6. Persson B, Hansson U (1993). Hypoglycaemia in pregnancy. Bailliere’s Clinical Endocrinology and Metabolism 7: 731–9. Persson B, Swahn M-L, Hjertberg R, Hanson U, Nord E, Nordlander E, Hansson L-O (2002). Insulin lispro therapy in pregnancies complicated by type 1 diabetes mellitus. Diabetes Research and Clinical Practice 58: 115–21.
REFERENCES
237
Ray JG, O’Brien TE, Chan WS (2001). Preconception care and the risk of congenital anomalies in the offspring of women with diabetes mellitus: a meta-analysis. Quarterly Journal of Medicine 94: 435–44. Rayburn W, Piehl E, Jacober S, Schork A, Ploughman L (1986). Severe hypoglycemia during pregnancy: its frequency and predisposing factors in diabetic women. International Journal of Obstetrics and Gynecology 24: 263–8. Reece EA, Hagay Z, Roberts AB, DeGennaro N, Homko CJ, Connolly-Diamond M, Sherwin R et al. (1995). Fetal Doppler and behavioral responses during hypoglycemia induced with the insulin clamp technique in pregnant diabetic women. American Journal of Obstetrics and Gynecology 172: 151–5. Rosenn BM, Miodovnik M, Holcberg G, Khoury JC, Siddiqi TA (1995). Hypoglycemia: the price of intensive insulin therapy for pregnant women with insulin-dependent diabetes mellitus. Obstetrics and Gynecology 85: 417–22. Rosenn BM, Miodovnik M, Khoury JC, Siddiqi TA (1996). Counterregulatory hormonal responses to hypoglycemia during pregnancy. Obstetrics and Gynecology 87: 568–74. Rossi G, Lapaczewski P, Diamond MP, Jacob RJ, Shulman GI, Sherwin RS (1993). Inhibitory effect of pregnancy on counterregulatory hormone responses to hypoglycemia in awake rat. Diabetes 42: 1440–5. Scottish Diabetes in Pregnancy Study Group (2004). Audit of pregnancies in women with diabetes. Findings from a one year audit. Report to Scottish Programme for Clinical Effectiveness In Reproductive Health (SPCERH). SIGN Executive (2001). SIGN 55 Management of Diabetes. Scottish Intercollegiate Guidelines Network, Edinburgh: 34–8. Smoak IW, Sadler TW (1990). Embryopathic effects of short-term exposure to hypoglycemia in mouse embryos in vitro. American Journal of Obstetrics and Gynecology 63: 619–24. Stangenberg M, Persson B, Stange L, Carlstrom K (1983). Insulin-induced hypoglycemia in pregnant diabetics. Maternal and fetal cardiovascular reactions. Acta Obstetrica Gynecologica Scandinavica 62: 249–52. Steel JM, Johnstone FD, Smith AF (1989). Pre-pregnancy preparation. In: Carbohydrate Metabolism in Pregnancy and the Newborn: IV. Sutherland HW, Stowers JM and Pearson DWM, eds. SpringerVerlag, Berlin: 129–39. Steel JM, Johnstone FD, Hepburn D, Smith AF (1990). Can pre-pregnancy care of diabetic women reduce the risk of abnormal babies? British Medical Journal 301: 1070–3. The Diabetes Control and Complications Trial Research Group (1996). Pregnancy outcomes in the Diabetes Control and Complications Trial. American Journal of Obstetrics and Gynecology 174: 1343–53. Wyatt JW, Frias JL, Hoymet HE, Jovanovic L, Kaaja R, Brown F et al., the IONS study group (2004). Congenital anomaly rate in offspring of mothers with diabetes treated with insulin lispro during pregnancy. Diabetic Medicine 22: 803–7. Yamashita Y, Kawano Y, Kuriya N, Murakami Y, Matsuishi T, Yoshimatsu K, Kato H (1996). Intellectual development of offspring of diabetic mothers. Acta Paediatrica 85: 1192–6. Yssing M (1974). Oestriol excretion in pregnant diabetes related to long term prognosis of surviving children. Acta Endocrinologica 182: 95–104. Yogev Y, Chen R, Ben-Haroush A, Phillip M, Jovanovic L, Hod M (2003). Continuous glucose monitoring for the evaluation of gravid women with type 1 diabetes mellitus. Obstetrics and Gynecology 101: 633–8.
11 Hypoglycaemia in Type 2 Diabetes and in Elderly People Nicola N. Zammitt and Brian M. Frier
INTRODUCTION Hypoglycaemia has been thought to be related almost exclusively to the management of type 1 diabetes and has received limited attention as a problem associated with the treatment of type 2 diabetes. However, as the prevalence of type 2 diabetes escalates and life expectancy increases, it is inevitable that the number of older people with insulin-treated diabetes will increase steadily. Furthermore, the current emphasis on adherence to stricter glycaemic targets is encouraging the earlier use of insulin to treat type 2 diabetes, so influencing the prevalence and severity of hypoglycaemia in this group. Because hypoglycaemia is increasing as a clinical problem in people with type 2 diabetes, many of whom are elderly, it is pertinent to consider how age and type 2 diabetes per se affect the pathophysiology and frequency of hypoglycaemia.
PATHOPHYSIOLOGY OF HYPOGLYCAEMIA Most studies that have examined counterregulatory responses to hypoglycaemia have been conducted in young adults with, and without, type 1 diabetes. The pathophysiological differences that exist between type 1 and type 2 diabetes have the potential to modify counterregulatory mechanisms, as does the increasing prevalence of type 2 diabetes in middle-aged and elderly individuals. The increasing age of the population with diabetes will also influence counterregulation. The responses to hypoglycaemia can be considered in terms of generation of symptom responses and counterregulatory hormonal changes. The two are independent, although some hormones such as epinephrine (adrenaline) can contribute to symptom generation; both responses are mediated centrally through glucose sensors in the brain. It is convenient to examine symptoms and counterregulation separately.
The Effects of Ageing on the Responses to Hypoglycaemia Symptoms The symptoms of hypoglycaemia (Chapter 2) have been classified as autonomic, neuroglycopenic and non-specific (Deary et al., 1993). In elderly people who have diabetes,
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
240
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE Table 11.1 Common symptoms of hypoglycaemia in the elderly derived by Principal Components Analysis (adapted from Jaap et al. (1998) with permission from John Wiley & Sons, Ltd) Neuroglycopenic Poor concentration Lightheadedness Weakness Confusion Drowsiness Dizziness
Autonomic
Neurological
Sweating Shaking Anxiety Pounding heart
Unsteadiness Poor co-ordination Blurred vision Double vision Slurred speech
an additional group of neurological symptoms affecting vision and co-ordination has been identified (Jaap et al., 1998; McAulay et al., 2001) (Table 11.1) and non-specific symptoms of general malaise are less prominent than in young adults. These neurological symptoms (and signs) that are generated as manifestations of hypoglycaemia in elderly people may be confused with other conditions such as a transient ischaemic attack or a vaso-vagal episode. Many elderly people with type 2 diabetes have limited knowledge of the symptoms and treatment of hypoglycaemia (Pegg et al., 1991; Thomson et al., 1991) and their relatives and carers are equally ill-informed (Mutch and Dingwall-Fordyce, 1985). This magnifies the risk that early hypoglycaemia will neither be identified nor treated, and will therefore progress to the point where external help is required. Even in young adults, knowledge of diabetes and its treatment declines with time (Lawrence and Cheely, 1980), and in elderly people who are vulnerable to the effects of age-related cognitive decline, the need for regular educational reinforcement is greater, although seldom undertaken. Elderly people report a profile of hypoglycaemic symptoms that is different from that of young adults, and the intensity of their symptoms is less (Brierley et al., 1995; Matyka et al., 1997). A small British study that compared the responses to hypoglycaemia of young and elderly non-diabetic adults, showed that symptom scores were significantly lower in the older group, with autonomic and neuroglycopenic symptoms being affected equally (Brierley et al., 1995). Two Canadian studies have implicated diminished autonomic activation as the reason for the lower magnitude of symptomatic response in the elderly and have suggested that the attenuation in symptom intensity is a feature of increasing age, independent of any effects of diabetes (Meneilly et al., 1994a; Meneilly et al., 1994b). The glycaemic thresholds at which symptomatic responses to hypoglycaemia are generated (Chapter 7) alter with age. In young adults, symptoms are generated at a blood glucose level that is approximately 1.0 mmol/l higher than the level at which cognitive function becomes impaired (Schwartz et al., 1987). This allows sufficient time for remedial action to be taken before the development of severe neuroglycopenia (Mitrakou et al., 1991). However, in older subjects these thresholds are much closer together and occur almost simultaneously (around 30 ± 02 mmol/l) (Matyka et al., 1997) (Figure 11.1), which truncates the time available for self-treatment before the development of incapacitating neuroglycopenia. Thus, the altered symptom profile, lower symptom intensity and altered glycaemic thresholds in the elderly combine to increase the risk of progression to severe hypoglycaemia in this age group.
PATHOPHYSIOLOGY OF HYPOGLYCAEMIA
241
Arterialised blood glucose (mmol/l) 4.0
SYMPTOMS 3.5
Younger men (3.6 mmol/l)
SYMPTOMS 3.0
Older men (3.0 mmol/l) REACTION TIME
2.5
FOUR CHOICE REACTION TIME
Younger men (2.6 mmol/l)
2.0
Figure 11.1 Glycaemic thresholds for subjective symptomatic awareness of hypoglycaemia and for the onset of cognitive dysfunction in young and elderly non-diabetic males. Figure based on data derived from Matyka et al. (1997) and reproduced from McAulay and Frier (2001), with permission from John Wiley and Sons, Ltd
Counterregulation Early studies examining counterregulatory responses in elderly non-diabetic adults yielded conflicting results, partly because interpretation of the data was confounded by the presence of co-morbidities in many of the participants. However, subsequent studies suggest that there may be age-related alterations in counterregulation. One study used an intravenous infusion of insulin to compare the counterregulatory responses to hypoglycaemia in non-diabetic elderly and young adults (Marker et al., 1992). This study suggested that with advancing age the secretion of growth hormone and cortisol is diminished, with modest attenuation of blood glucose recovery in older non-diabetic adults, in whom the rise in plasma epinephrine is slower than in younger subjects (Marker et al., 1992). The secretion of glucagon and the rate of insulin clearance were also lower. A reduced rate of clearance of insulin has been noted in several studies (Minaker et al., 1982; Reaven et al., 1982; Fink et al., 1985). The changes observed by Marker and colleagues were unaffected by preceding physical training, suggesting that they are not simply a consequence of a more sedentary lifestyle associated with ageing (Marker et al., 1992). The glycaemic thresholds for the secretion of glucagon and epinephrine in response to hypoglycaemia also occur at a lower blood glucose level than in younger subjects. In young non-diabetic adults, these hormones are released at a blood glucose level of 3.3 mmol/l, compared to approximately 2.8 mmol/l in older adults (Meneilly et al., 1994a). With increasing age, the intensity of the hypoglycaemic stimulus (as demonstrated by blood glucose nadir) appears to influence the magnitude of the counterregulatory response.
242
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
In clamp studies comparing elderly and young non-diabetic subjects, the magnitude of the glucagon and epinephrine responses was lower in the elderly group during mild hypoglycaemia (blood glucose 3.3 mmol/l). However, both age groups achieved a similar magnitude of response at a lower blood glucose of 2.8 mmol/l, indicating that in the elderly, counterregulatory responses are preserved during more profound hypoglycaemia (Ortiz-Alonso et al., 1994). Two other similar studies in non-diabetic elderly subjects have not demonstrated any significant age-related impairment of the counterregulatory hormonal responses to hypoglycaemia (Brierley et al., 1995; Matyka et al., 1997). Furthermore, although the symptomatic and counterregulatory responses to hypoglycaemia may be modified by advancing age, it is not known at what age these changes become apparent or whether these effects are influenced by gender or the menopause in women. When designing studies and analysing results, few investigations of hypoglycaemia in type 2 diabetes have taken age or sex into account and few studies have included people aged over 70 years.
The Effects of Type 2 diabetes on the Responses to Hypoglycaemia Interpretation of the results of early studies of the counterregulatory responses to hypoglycaemia is constrained by their technical limitations and only a few of the more recent studies were designed to control for potentially confounding variables and embraced modern methodology. Symptoms Type 2 diabetes is a heterogeneous disorder and individuals with this condition may be treated with one or more treatment modalities. This raises the possibility that the agent used to treat the condition could influence the symptoms of hypoglycaemia. However, when the symptomatic responses to hypoglycaemia were examined in a group of non-diabetic subjects by inducing hypoglycaemia with insulin or a sulphonylurea (tolbutamide), neither the nature nor the intensity of symptoms differed with either agent (Peacey et al., 1996). Some studies have compared the symptoms associated with insulin-induced hypoglycaemia in people with type 1 and type 2 diabetes. In a retrospective population survey, people with insulin-treated type 2 diabetes reported a similar symptom profile to people with type 1 diabetes, who were matched for duration of insulin therapy, but not for age or duration of diabetes (Hepburn et al., 1993). Furthermore, glucose clamp studies have shown that after controlling for age, no significant differences exist in the symptoms of hypoglycaemia between people with type 1 and type 2 diabetes (Levy et al., 1998). The symptoms of hypoglycaemia in type 2 diabetes appear to be unaffected by the nature of the disorder or by the choice of treatment modality. Counterregulation Early studies of the counterregulatory responses to hypoglycaemia in type 2 diabetes were limited by factors such as differences in blood glucose nadir between the diabetic and control groups, poorly matched or absent control groups and the inconsistent methods used to induce
PATHOPHYSIOLOGY OF HYPOGLYCAEMIA
243
hypoglycaemia with techniques such as intravenous or subcutaneous bolus injections of insulin (De Galan and Hoekstra, 2001). Three studies that have examined counterregulatory responses to hypoglycaemia in people with type 2 diabetes, who were being treated with diet alone or with oral medication, have provided important observations. These studies demonstrated that the secretion of counterregulatory hormones occurs at higher blood glucose levels in individuals with type 2 diabetes than in non-diabetic subjects (Korzon-Burakowska et al., 1998; Spyer et al., 2000) and in people with type 1 diabetes (Levy et al., 1998). In one study, a stepped glucose clamp was used to examine how glycaemic control influenced the counterregulatory response to hypoglycaemia in 11 subjects with type 2 diabetes who were diet-treated or were taking sulphonylureas, and they were compared with 10 subjects with type 1 diabetes (Levy et al., 1998). The group with type 2 diabetes was older than the group with type 1 diabetes, so two non-diabetic control groups were matched for age and body weight to the diabetic subjects. In the subjects with type 2 diabetes, counterregulatory hormones were released at higher blood glucose levels than in those with type 1 diabetes. However, males were over-represented in the group with type 2 diabetes, and six of the subjects with type 2 diabetes required high insulin infusion rates, so introducing potentially confounding variables and limiting interpretation of the results of this study. There is evidence that hyperinsulinaemia per se can suppress the release of glucagon in response to hypoglycaemia (Diamond et al., 1991; Liu et al., 1991a; Liu et al., 1991b; Fanelli et al., 1994) while increasing catecholamine and cortisol release (Davis et al., 1993) in non-diabetic and type 1 diabetic subjects. Similarly, in non-diabetic and in type 1 diabetic subjects, women have attenuated counterregulatory responses to hypoglycaemia compared to men (Amiel et al., 1993; Davis et al., 2000; Sandoval et al., 2003). The effects of gender and hyperinsulinaemia on counterregulation have not yet been determined in type 2 diabetes. People with type 2 diabetes may have greater protection against hypoglycaemia because the counterregulatory responses commence at higher blood glucose levels than in nondiabetic people or in those with type 1 diabetes (Korzon-Burakowska et al., 1998; Levy et al., 1998; Spyer et al., 2000). However, when HbA1c is lowered with intensive therapy in type 1 diabetes, the thresholds for the counterregulatory responses are shifted to a lower glycaemic level (Cryer, 2002; Cryer et al., 2003) and the same phenomenon appears to occur in type 2 diabetes (Korzon-Burakowska et al., 1998; Levy et al., 1998). Counterregulatory deficiencies are associated with increasing duration of type 1 diabetes; it has been suggested that similar counterregulatory deficiencies may develop in type 2 diabetes as the disorder progresses in severity. The secretory response of glucagon to hypoglycaemia is lost early in the natural history of type 1 diabetes (Chapter 6) and, although the catecholamine response compensates for several years, it too declines with time (Cryer, 2002). There is disagreement over whether the glucagon response in type 2 diabetes is modestly diminished (Bolli et al., 1984; Meneilly et al., 1994b; Shamoon et al., 1994) or preserved (Boden et al., 1983; Heller et al., 1987; Korzon-Burakowska et al., 1998; Levy et al., 1998). It is probable that the ability of an individual to secrete glucagon in response to hypoglycaemia is related to their residual insulin secretory capacity and the ability to suppress endogenous insulin secretion when blood glucose falls (Israelian et al., 2005). People with type 2 diabetes constitute a heterogeneous group, with varying degrees of insulin deficiency. Most investigators who have reported an intact glucagon response have studied people with type 2 diabetes who were treated satisfactorily with oral anti-diabetic agents and were therefore unlikely to have been insulin-deficient (Boden et al., 1983; Heller et al.,
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
244
Growth hormone (µg/l)
150 125 100 75 = Non-diabetic subjects = T2DM Subjects
3.0
9.0 6.0 3.0 0 750
MEAN + SEM
Cortisol (nM)
Epinephrine (pM × 10–3)
Glucagon (pg/ml)
1987; Levy et al., 1998; Spyer et al., 2000). Furthermore, with one exception (Meneilly et al., 1994b) all of these studies examined middle-aged subjects in their fifth or sixth decade, despite the fact that most people with type 2 diabetes are older than this. The effects of ageing on counterregulation have not been addressed in these studies. To assess the role of insulin deficiency in promoting counterregulatory hormonal deficiencies, the counterregulatory responses to hypoglycaemia were examined in 15 non-diabetic controls and in 13 people with type 2 diabetes, six of whom were treated with insulin, and who were demonstrated to be insulin-deficient with low plasma C-peptide, while the remaining seven were being treated with oral anti-diabetic agents (Segel et al., 2002). The glucagon response to hypoglycaemia was preserved in the patients on oral agents and in the nondiabetic controls, but was virtually absent in the insulin-deficient patients, demonstrating an association between acquired counterregulatory abnormalities and insulin deficiency in type 2 diabetes. Deficient counterregulatory responses to hypoglycaemia were also observed in a group of patients with type 2 diabetes who had moderate beta cell failure (Israelian et al., 2006). In this study, the reduction in endogenous insulin secretion that normally occurs during hypoglycaemia was delayed and reduced, and responses of glucagon and growth hormone were impaired (Israelian et al., 2006) (Figure 11.2). Thus, multiple defects in the counterregulatory responses to hypoglycaemia emerge as beta cell function fails in type 2 diabetes. In type 1 diabetes, individuals who experience frequent hypoglycaemia can develop a condition that has been termed Hypoglycaemia Associated Autonomic Failure (HAAF) (Cryer, 1992; Cryer, 2004). The underlying premise is that recurrent hypoglycaemia leads
2.0 1.0
500 250 0
0.0 –30
0
30 60 Time (min)
90
120
–30
0
30 60 Time (min)
90
120
Figure 11.2 Hypoglycaemia-induced responses of glucagon, epinephrine, cortisol and growth hormone in non-diabetic subjects and in subjects with type 2 diabetes. In type 2 diabetes significantly lower responses to hypoglycaemia were observed for glucagon (45% reduction, p < 005) and growth hormone (50% reduction, p < 004). Reprinted from Israelian et al. (2006) Metabolism, with permission from Elsevier
PATHOPHYSIOLOGY OF HYPOGLYCAEMIA
245
to failure of the centrally mediated counterregulatory response to hypoglycaemia, resulting in impaired awareness of hypoglycaemia (Chapter 7). Impaired hypoglycaemia awareness has been associated primarily with type 1 diabetes but people with insulin-deficient type 2 diabetes are also at risk of developing HAAF. In the study by Segel and colleagues (Segel et al., 2002), subjects with type 2 diabetes underwent a hypoglycaemic clamp on the first day of the study followed by a subsequent period of hypoglycaemia later that same day. They were then exposed to a second hypoglycaemic clamp during the following day, at which time the plasma glucose levels required to activate the hormonal and symptomatic responses were significantly lower than during the first hypoglycaemic clamp. These findings implicate a role for antecedent hypoglycaemia in modifying the glycaemic thresholds at which responses to hypoglycaemia are triggered in type 2 diabetes; this suggests that individuals with insulintreated type 2 diabetes can develop HAAF. The preservation of the epinephrine response to hypoglycaemia, in conjunction with possible insulin resistance in some individuals, may provide further defensive mechanisms to protect people with type 2 diabetes against hypoglycaemia. The lipolytic effects of epinephrine can outweigh the anabolic effects of insulin on insulin-resistant adipose tissue, resulting in a rise in plasma free fatty acids in response to hypoglycaemia in type 2 diabetes (Bolli et al., 1984; Heller et al., 1987; Shamoon et al., 1994), but not in type 1 diabetes (Maggs et al., 1997). Epinephrine secretion during hypoglycaemia may therefore have a greater protective effect in insulin-resistant patients. Epinephrine also stimulates the mobilisation of glucose stored in the kidney and this may compensate in part for the impaired production of hepatic glucose observed in individuals who have a deficient glucagon response to hypoglycaemia (Woerle et al., 2003).
Moderators of Hypoglycaemia in type 2 diabetes Although factors such as sleep, consumption of alcohol and the timing of exercise are known to influence the risk of hypoglycaemia in type 1 diabetes (see Chapters 3 and 5), their effects are not known in type 2 diabetes. Prolonged treatment with insulin (greater than ten years) is a reliable predictor of increased risk of severe hypoglycaemia in type 2 diabetes (Donnelly et al., 2005) and when people with type 2 diabetes become insulin-deficient, their frequency of severe hypoglycaemia rises towards that of individuals with type 1 diabetes who are matched for duration of treatment with insulin (Hepburn et al., 1993). Impaired awareness of hypoglycaemia is a major significant risk factor for severe hypoglycaemia in people with type 1 diabetes (Gold et al., 1994), but is less common in type 2 diabetes (Hepburn et al., 1993; Henderson et al., 2003) with only 8% having impaired awareness in a retrospective survey of 215 individuals with insulin-treated type 2 diabetes. However, the incidence of hypoglycaemia was nine times greater in this small sub-group than in those with normal awareness (Henderson et al., 2003). Continuous glucose monitoring systems (CGMS) can detect asymptomatic hypoglycaemia. In a prospective study of 30 individuals with type 2 diabetes and 40 patients with type 1 diabetes, asymptomatic hypoglycaemia was detected in 14 (47%) of the group with type 2 diabetes and in 25 (63%) of the group with type 1 diabetes (Chico et al., 2003). An Australian study used CGMS to examine the frequency of hypoglycaemia over six days in 25 patients treated with sulphonylureas. In 14 (56%) of these subjects, readings of < 22 mmol/l lasting for at least 15 minutes were recorded. The participants had no symptomatic awareness of these episodes
246
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
(Hay et al., 2003). Although impaired awareness of hypoglycaemia is more common in type 1 diabetes, it may be more frequent in type 2 diabetes than is appreciated. Variability of blood glucose levels also appears to be an important moderator of the risk of hypoglycaemia. In a prospective observational trial, a predominantly male cohort of people with insulin-treated type 2 diabetes recorded blood glucose profiles for eight weeks and reported all episodes of hypoglycaemia over one year. The probability of all hypoglycaemia was highest in those who had a low mean blood glucose with a high standard deviation, implicating the variability of blood glucose values as a predictor of risk of hypoglycaemia in insulin-treated type 2 diabetes (Murata et al., 2004). Exercise and alcohol can also influence the risk of hypoglycaemia. The rate of skeletal muscle glucose uptake during physical exertion is normal in people with type 2 diabetes, but hepatic glucose output is impaired compared to non-diabetic individuals (Chipkin et al., 2001); this can result in exercise-related hypoglycaemia during physical exertion. Exercise reduces postprandial blood glucose concentrations (Minuk et al., 1981; Larsen et al., 1997) and improves insulin sensitivity for up to 16 hours after physical activity (Trovati et al., 1984; Devlin et al., 1987), exposing the individual to an increased risk of hypoglycaemia for a prolonged period. However, a combination of moderate exercise and ingestion of alcohol did not provoke acute hypoglycaemia in C-peptide-positive, middle-aged individuals with type 2 diabetes, regardless of whether they had eaten or fasted before exercise (Rasmussen et al., 1999). Although alcohol impairs the counterregulatory responses to hypoglycaemia in type 1 diabetes (Avogaro et al., 1993), it does not appear to affect recovery from hypoglycaemia in type 2 diabetes (Rasmussen et al., 2001).
FREQUENCY OF HYPOGLYCAEMIA IN TYPE 2 DIABETES Mild hypoglycaemia is defined by the ability to self-treat, while the need for external assistance denotes a severe episode. The frequency of hypoglycaemia in people with type 1 diabetes is described in Chapter 3. Mild hypoglycaemia occurs on average around twice weekly (Pramming et al., 1991; Pedersen-Bjergaard et al., 2004) and the estimated incidence of severe hypoglycaemia ranges from 1.0 to 1.7 episodes per patient per year (MacLeod et al., 1993; ter Braak et al., 2000; Pedersen-Bjergaard et al., 2004) with an annual prevalence of between 30% (MacLeod et al., 1993; The Diabetes Control and Complications Trial Research Group, 1993; Stephenson et al., 1994) and 40% (ter Braak et al., 2000). The frequency of hypoglycaemia in type 2 diabetes cannot be summarised in an equally succinct manner because of the heterogeneity of this disorder and the range of treatment modalities available. Furthermore, many people with type 2 diabetes are elderly and the frequency of hypoglycaemia is often underestimated in this group (McAulay et al., 2001). Most studies have been conducted retrospectively and both the definitions of hypoglycaemia and the nature of the treatment modalities that have been examined differ between studies, thus hindering comparison. Studies have demonstrated that individuals with type 1 diabetes can reliably remember the occurrence of severe hypoglycaemia after an interval of one year, but that recall of mild hypoglycaemia becomes unreliable after a one week period (Pramming et al., 1991; Pedersen-Bjergaard et al., 2003). In people with insulin-treated type 2 diabetes, recall of severe hypoglycaemia is similarly robust over a period of one year (Akram et al., 2003) but the reliability of their recall of mild hypoglycaemia has not been examined. Finally, the treatment regimens and patient inclusion criteria used in clinical
FREQUENCY OF HYPOGLYCAEMIA IN TYPE 2 DIABETES
247
trials may not be representative of the patient characteristics observed and the therapeutic interventions employed within an unselected diabetic out-patient population and this may limit the extent to which such data can be extrapolated. Table 11.2b summarises available epidemiological data on hypoglycaemia in type 2 diabetes.
Hypoglycaemia and Oral Antidiabetic Agents Hypoglycaemia caused by oral antidiabetic agents is primarily associated with the insulin secretagogues. It is seldom a side-effect of treatment with thiazolidinediones or alphaglucosidase inhibitors, and has been reported only infrequently in association with metformin, usually when food intake is restricted (UK Prospective Diabetes Study (UKPDS) Group 1998a, United Kingdom Prospective Diabetes Study Group 1998c). The frequency of hypoglycaemia secondary to sulphonylurea therapy is considerably lower than that attributed to insulin (UK Prospective Diabetes Study (UKPDS) Group 1998b; Miller et al., 2001; Leese et al., 2003) but is probably underestimated (Hartling et al., 1987). In a retrospective six-month study in England of 219 people with type 2 diabetes treated with sulphonylureas and metformin, 20% of those taking sulphonylureas (either alone or in combination with metformin) had experienced hypoglycaemic symptoms (Jennings et al., 1989). Sulphonylureas can also cause severe hypoglycaemia. Over a seven-year period, the Swedish Adverse Drug Reactions Advisory Committee reported 19 cases of glipizide-associated severe hypoglycaemia presenting with reduced consciousness, with two deaths (Asplund et al., 1991). Incidences of severe hypoglycaemia of 0.224 episodes per 100 person-years with longacting sulphonylureas versus 0.075 episodes per 100 patient-years were reported among 28 patients (median age 73 years) who were admitted to hospital with severe hypoglycaemia in Switzerland (Stahl and Berger, 1999). The frequency of severe hypoglycaemia may be underestimated in this study as patients receiving treatment in the community were excluded. A recent study, A Diabetes Outcome Progression Trial (ADOPT), in which newly diagnosed patients with type 2 diabetes were randomised to treatment with rosiglitazone, metformin and glibenclamide for a median of four years, prospectively recorded the hypoglycaemic events associated with each treatment (Kahn et al., 2006). One (0.1%) serious and 142 (9.8%) total events were recorded for rosiglitazone, one (0.1%) serious and 168 (11.6%) total events for metformin and eight (0.6%) serious and 557 (38.7%) total events for glibenclamide. The frequency of hypoglycaemia relates to the individual pharmacokinetic properties of each sulphonylurea (Table 11.3), with the long-acting agents such as chlorpropamide, glibenclamide and long-acting glipizide being associated with the greatest risk (Stahl and Berger, 1999; Del Prato et al., 2002; Rendell 2004). Glibenclamide is associated with a greater risk of severe hypoglycaemia than gliclazide (Tessier et al., 1994) because active metabolites prolong its hypoglycaemic effects for 24 hours (Jonsson et al., 2001; Rendell 2004). Glibenclamide also attentuates the glucagon response to hypoglycaemia in nondiabetic volunteers (ter Braak et al., 2002) and in people with type 2 diabetes (LandstedtHallin et al., 1999; Banarer et al., 2002). Several drugs may potentiate the hypoglycaemic effects of sulphonylureas (Table 11.4). Risk factors for severe hypoglycaemia associated with sulphonylurea therapy include age, a past history of cardiovascular disease or stroke, renal failure, reduced food intake, alcohol ingestion and interactions with other drugs (Hartling et al., 1987; Seltzer et al., 1989; Asplund et al., 1991; Campbell et al., 1994; Shorr et al., 1997; Ben-Ami et al., 1999; Burge et al., 1999; Harrigan, et al., 2001).
Retrospective, structured interview
T2DM, OHA only
219 (203: SU, 16: metformin) 6 months
Design
Subjects
Number
Hepburn, 1993∗
10 years
3935
T2DM, OHA and insulin
104 T1DM, 104 T2DM 1 year
Insulin-treated T1DM and T2DM
Prospective, Retrospective multicentre RCT questionnaire
UKPDS 33, 1998
NA
Insulin: SH
2.3%
36.5% 10%
82.7%
Miller, 2001
NA
56% conventional 93% intensive
NA
NA
18–35 months
153
All T2DM
0%
30%
0%
16%
7 months
1055
Type 2
Prospective, Retrospective multicentre RCT interview
Abraira, 1995
Leese, 2003
Donnelly, 2005
15%
64%
NA
NA
1 year
215
7.3%
NA
0.8%
NA
I year
160
3%
45%
NA
NA
94 T1DM 173 T2DM One month
Retrospective PopulationProspective questionnaire based dataset analysis Insulin-treated 69 T1DM; Insulin-treated T2DM T2DM SU T1DM & (22), insulin T2DM (66)
Henderson, 2003
16.5%
NA
NA
NA
1 year
401
Insulin-treated T2DM
Retrospective questionnaire
Akram, 2006
Abbreviations: T1DM = type 1 diabetes; T2DM = type 2 diabetes; NA = not applicable; SH = severe hypoglycaemia; SU = sulphonylurea; LOC = loss of consciousness; RCT = randomised clinical trial; OHA = oral hypoglycaemic agent; Hypos = hypoglycaemia; ∗ = only figures for type 2 DM given
NA
Metformin 0% NA ⎫ SUs ⎬ Glibenclamide 313% Glibenclamide 17.0% Chlorpropamide 136% Chlorpropramide 202% ⎭ 11.0% overall Gliclazide 131% Glibenclamide NA 0.6% Chlorpropramide 0.4%
Insulin: all hypos
OHA: SH
OHA: all hypos
PREVALENCE
Duration
Jennings, 1989
Study
Table 11.2a Prevalence of severe hypoglycaemia (hypos) in type 2 diabetes (adapted from Zammitt and Frier, 2005).
FREQUENCY OF HYPOGLYCAEMIA IN TYPE 2 DIABETES
249
Table 11.2b Incidence of severe hypoglycaemia (hypos) in type 2 diabetes. Figures are expressed as events per patient per year (adapted from Zammitt and Frier, 2005) Study
VA-CSDM 1995
Henderson 2003
Leese 2003
Donnelly 2005
Akram, 2006
INCIDENCE OHA: all hypos
NA
NA
NA
NA
NA
OHA: SH
NA
NA
NA
NA
Insulin: all hypos
1.5 (standard) 16.5 (intensive) 0.02
NA
SU: 0.009 Metformin: 0.0005 NA
16.37
NA
0.28
0.12 (T1 and T2DM)
0.35
0.44
Insulin: SH
Abbreviations: T1DM = type 1 diabetes; T2DM = type 2 diabetes; NA = not applicable; SH = severe hypoglycaemia; SU = sulphonylurea; LOC = loss of consciousness; RCT = randomised clinical trial; OHA = oral hypoglycaemic agent; Hypos = hypoglycaemia
Table 11.3 Pharmacokinetics of the sulphonylureas (data sourced from Kobayashi et al. (1984); Campbell et al. (1994); DeFronzo (1999); Harrower (2000); Harrigan et al. (2001); Schernthaner (2003))
Generation
Name
tmax (hours)
t1/2 (hours)
Duration of action (hours)
Renal excretion of active metabolite
First
Chlorpropamide Tolbutamide
2–7 3–4
36 3–28
60 6–12
Yes Insignificant
Second
Glipizide Glipizide GITS (Glucotrol XL) Glibenclamide Gliclazide Gliclazide MR
1–3 6–12
7 7
12–24 24
No No
2–6 2–3 4–6
10 12–14
12–24
Yes Minimal (5%) Minimal
Glimepiride
2–3
5–9
Third
24 16–24
Yes (?)
tmax = time to peak; t1/2 = half-life
Table 11.4 Drugs that potentiate the effects of sulphonylurea drugs used to treat type 2 diabetes (data sourced from Campbell et al. (1994); Harrigan et al. (2001))
Mechanism Drugs
Decreased hepatic metabolism Chloramphenicol H2 blockers Ciprofloxacin Warfarin Sulphonamides Monoamineoxidase inhibitors
Decreased renal excretion Allopurinol Probenecid Aspirin
Displacement of SU from albumin binding sites
Increases plasma concentration
Fibrates Trimethoprim Aspirin Warfarin Sulphonamides
Fluconazole, Miconazole
Inhibition of gluconeogenesis Alcohol
250
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
Some of the newer sulphonylureas such as glimepiride are associated with a lower risk of hypoglycaemia. In a population-based study in Germany, the incidence of hypoglycaemia was examined in people with type 2 diabetes who had attended a hospital emergency department over a four-year period (Holstein et al., 2001). A total of 45 episodes were recorded in individuals who had been taking a sulphonylurea, of which 38 were associated with glibenclamide. Glimepiride was implicated in only six episodes, despite more frequent prescribing of glimepiride. Glimepiride was compared with a modified release form of gliclazide in a multicentre, European trial (Schernthaner et al., 2004a). Glycated haemoglobin values improved by around 1.0% in both groups but modified release gliclazide was implicated in fewer cases of hypoglycaemia (3.7%) compared to glimepiride (8.9%). Severe hypoglycaemia did not occur in either group. The oral glucose prandial regulators, repaglinide and nateglinide, induce less frequent hypoglycaemia than the sulphonylureas because of their rapid onset of action and their selective stimulation of insulin secretion in the presence of carbohydrate but not in the fasting state (Strange et al., 1999; Nattrass and Lauritzen, 2000; Culy and Jarvis, 2001). In a randomised multi-centre trial comparing repaglinide with nateglinide, slightly lower HbA1c values were achieved after 16 weeks on repaglinide at the expense of mild hypoglycaemic episodes in 7% of patients compared to no episodes in the nateglinide group (Rosenstock et al., 2004).
Studies Comparing Hypoglycaemia Secondary to Insulin or Oral Antidiabetic Agents It is often difficult to compare the frequency of hypoglycaemia ascribed to insulin and oral antidiabetic agents because in many studies the participants were receiving treatment with both modalities. The United Kingdom Prospective Diabetes Study Group (1998a) reported the prevalence of hypoglycaemia in people with type 2 diabetes using different therapies. A higher frequency of hypoglycaemia was observed in association with intensive, compared to conventional, treatment whether using sulphonylureas or insulin. The prevalence of hypoglycaemia across the different treatment groups is summarised in Table 11.5. Of the oral agents used in this study, metformin had the lowest recorded rate of hypoglycaemia while glibenclamide had the highest. With intensive treatment, hypoglycaemia occurred most frequently in the insulin-treated patients, and the prevalence of hypoglycaemia was lower in the first decade of the study than in later years. The prevalence of hypoglycaemia was lower when the groups were analysed on an ‘intention to treat’ basis because several patients in the conventional-treatment groups required an increasing number of different therapies with escalating doses as their glycaemic control deteriorated. Although the patients were questioned about the occurrence of hypoglycaemia at every four-monthly review, only the most severe episode was recorded each time. This study cannot therefore provide an accurate estimate of the incidence of hypoglycaemia in type 2 diabetes and the overall prevalence of severe hypoglycaemia of 2% is deceptive as it does not indicate the rise in prevalence with increasing duration of type 2 diabetes (United Kingdom Prospective Diabetes Study Group, 1998a). A systematic review of randomised controlled trials comparing insulin monotherapy with combination therapy with insulin and oral antidiabetic agents confirmed the relative safety of the latter treatment regimen and 13 out of 14 studies did not show any significant difference
FREQUENCY OF HYPOGLYCAEMIA IN TYPE 2 DIABETES
251
Table 11.5 Hypoglycaemia episodes per year by intention to treat analysis and actual therapy for intensive and conventional treatment in the UKPDS: non-overweight patients (data sourced from UK Prospective Diabetes Study (UKPDS) Group (1998a; 1998b)) Mean proportion of patients per year with hypoglycaemia (%) Conventional Chlorpropamide Glibenclamide Insulin Metformin ≥ 1 episode MH in first 10 years
Normal BMI Overweight participants
12 09
110 121
177 175
365 340
– 42
≥ 1 episode SH in first 10 years
Normal BMI Overweight participants
01 07
04 06
06 25
23 03
– 0
≥ 1 episode MH in first 10 years (intention to treat)
Normal BMI Overweight participants
10 79
16 152
21 205
28 255
– 83
≥ 1 episode SH in first 10 years (intention to treat)
Normal BMI Overweight participants
07 0.7
10 1.2
14 1.0
18 2.0
– 0.6
MH = mild hypoglycaemia; SH = severe hypoglycaemia; BMI = body mass index
in hypoglycaemia rates between the two therapeutic strategies (Goudswaard et al., 2004). A randomised multicentre study compared the use of twice daily human Mixtard 30 (Novo Nordisk) with once daily insulin glargine in combination with metformin and glimepiride and observed that the combination regimen of insulin and oral agents achieved better glycaemic control with fewer confirmed episodes of hypoglycaemia (4.07 versus 9.97 episodes per patientyear, p < 00001) (Janka et al., 2005). In an observational study of 41 people with type 2 diabetes treated with oral antidiabetic drugs and bedtime isophane (NPH) insulin, 49% had experienced mild hypoglycaemia since commencing insulin with an incidence of four episodes per patient per year and no episodes of severe hypoglycaemia (Allen et al., 2004). Several studies have compared the frequency of hypoglycaemia secondary to oral antidiabetic agents compared to that associated with insulin treatment. A retrospective population-based study in Tennessee examined the frequency of ‘serious’ hypoglycaemia over a four-year period in 19932 Medicaid patients with type 2 diabetes aged 65 or older (Shorr et al., 1997). ‘Serious’ hypoglycaemia was defined as an episode resulting in a fatal outcome or requiring hospital treatment. The frequency of severe hypoglycaemia was therefore underestimated by excluding events other than those that culminated in hospitalisation or death of the patient. The reported incidence of ‘serious’ hypoglycaemia with sulphonylureas was 1.23 episodes per 100 person-years and 2.76 episodes per 100 person-years with insulin treatment. In Atlanta, a six-month, retrospective survey was conducted of 1055 patients with type 2 diabetes, treated with oral antidiabetic drugs or insulin and who were predominantly female
252
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
African-Americans (Miller et al., 2001). A quarter of the group had experienced at least one episode of hypoglycaemia during the study period. The prevalence of hypoglycaemia increased with more intensive treatment, the highest rate being associated with insulin. Severe hypoglycaemia occurred in 0.5% of patients, all of whom had been treated with insulin. However, this study is limited by its reliance on patient recall of hypoglycaemia and the ethnicity and gender of the study group, which makes it impossible to extrapolate the outcomes to European populations. A 12-month prospective multicentre British study tested the hypothesis that the risk of hypoglycaemia in individuals with insulin-treated type 2 diabetes of short duration is comparable to those taking sulphonylureas and is lower than in patients with newly diagnosed type 1 diabetes (UK Hypoglycaemia Study Group, 2007). Monthly questionnaires were used to record self-reported hypoglycaemia and a 72-hour period of continuous blood glucose monitoring was used at the beginning and end of the study period to detect asymptomatic hypoglycaemia. No difference was observed in the prevalences of severe hypoglycaemia in those with type 2 diabetes treated with sulphonylureas, and in individuals with type 2 diabetes treated with insulin for less than two years (7% in both groups). Symptomatic hypoglycaemia in patients with type 2 diabetes recently commenced on insulin therapy (< 2 years) was considerably lower than in those with type 1 diabetes of less than five years duration (median rate one versus 22 episodes per subject per year, p < 0001) (Figure 11.3), but was higher in patients with type 2 diabetes > 5 years, showing a significantly higher prevalence in those with a longer duration of insulin therapy.
Proportion reporting at least one servere hypo
1.0
0.8
0.6
0.4
0.2
0.0 type 2 treated with sulphonylureas
type 2 < 2 yrs insulin
type 2 > 5 yrs insulin
type 1 < 5 yrs insulin
type 1 > 15 yrs insulin
Figure 11.3 Proportion of patients with type 1 diabetes for < 5 years and > 15 years, and type 2 diabetes in different treatment groups (sulphonylureas, insulin < 2 years, insulin > 5 years), who experienced one or more episodes of self-reported severe hypoglycaemia during 9–12 months of follow-up in the UK Hypoglycaemia Group Study. Reproduced from UK Hypoglycaemia Study Group (2007), Diabetologia, with kind permission of Springer Science and Business Media
FREQUENCY OF HYPOGLYCAEMIA IN TYPE 2 DIABETES
253
Hypoglycaemia and Insulin In the USA, the Veterans Affairs Cooperative Study in type 2 Diabetes (VA CSDM) examined glycaemic control and complications and compared a simple (‘standard’) insulin regimen (administered once daily) with an intensive (‘stepped’) regimen (Abraira et al., 1995). The participants in this trial had diabetes of relatively short duration (mean ± SD 78 ± 4 years), were all insulin-treated males, and were followed up for only 18–35 months. The overall incidence of severe hypoglycaemia was 0.02 episodes per patient per year with no significant difference between the standard and stepped treatment groups. The frequency of mild hypoglycaemia was significantly higher in the intensively treated group (stepped versus standard: 16.5 versus 1.5 episodes per patient per year). However, blood glucose was monitored less frequently in the standard treatment group, which may have caused under-reporting of asymptomatic hypoglycaemia. The participants in the VA CSDM study had a relatively short duration of diabetes. A retrospective survey in Edinburgh of 215 people with insulin-treated type 2 diabetes observed that the frequency of hypoglycaemia increased with the duration of insulin therapy and the duration of type 2 diabetes (Henderson et al., 2003) (Figure 11.4) and was inversely proportional to HbA1c concentration. The annual prevalence of severe hypoglycaemia was 15% with an overall incidence of 0.28 episodes per patient per year. The relationship between duration of insulin treatment and prevalence of severe hypoglycaemia has been replicated in a more recent 12-month prospective multicentre British survey (UK Hypoglycaemia Study Group 2007) (Figure 11.3). A similar incidence was reported in a retrospective study in Denmark of 401 patients with insulin-treated type 2 diabetes, 66 (16.5%) of whom had experienced at least one episode of severe hypoglycaemia in the preceding year giving an overall incidence of 0.44 episodes per patient
40 n = 29
35
Prevalence (%)
30 25 20 15
n = 39 n = 147
10 5 0 1–5
6–10
>10
Duration of insulin therapy (years)
Figure 11.4 Prevalence of severe hypoglycaemia in relation to duration of insulin therapy in patients with type 2 diabetes. Reproduced from Henderson et al. (2003) by permission of Blackwell Publishing
254
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
per year, but no relationship to HbA1c was observed (Akram et al., 2006). A retrospective study of 600 unselected insulin-treated diabetic patients performed a decade earlier in Edinburgh had observed an incidence of severe hypoglycaemia of 0.73 episodes per patient per year in the 56 people with type 2 diabetes compared with 1.7 episodes per patient per year in the 544 with type 1 diabetes (MacLeod et al., 1993). An earlier survey in the same centre compared the frequency of severe hypoglycaemia in 86 people with insulin-treated type 2 diabetes with 86 people with type 1 diabetes, matched for duration of insulin treatment and insulin dose (Hepburn et al., 1993). The frequency of severe hypoglycaemia was similar in the two groups and a positive correlation was found between the frequency of severe hypoglycaemia and duration of treatment with insulin (r = 039, p < 0001). The introduction of insulin analogues has been claimed to lower the risk of hypoglycaemia and several studies have compared the risk of hypoglycaemia in people with type 2 diabetes treated with either conventional insulin or insulin analogues. The risk of hypoglycaemia has been reported to be lower with long-acting insulin glargine in some studies (Yki-Järvinen et al., 2000; Rosenstock et al., 2001; Rosenstock et al., 2003; Riddle et al., 2003) and with insulin detemir (Hermansen et al., 2006) compared with isophane (NPH) insulin. In one study (Riddle et al., 2003), symptomatic hypoglycaemia was significantly lower using combination treatment with metformin and glargine compared to metformin and isophane (NPH) insulin, but no difference in biochemically confirmed hypoglycaemia was observed between the two groups. Insulin glargine was also associated with a lower frequency of hypoglycaemia than premixed insulins (Janka et al., 2005; Raskin et al., 2005). Rapidacting insulin analogues, such as lispro and glulisine, also appeared to limit the frequency of hypoglycaemia in people with type 2 diabetes when compared to short-acting soluble (regular) insulins (Anderson et al., 1997; Bastyr et al., 2000; McAulay and Frier, 2003; Dailey et al., 2004). However, several studies have not observed a significantly lower incidence of hypoglycaemia when using insulin analogues in comparison with conventional insulin (Ross et al., 2001; Raslova et al., 2004; Schernthaner et al., 2004b; Haak et al., 2005). Continuous subcutaneous insulin infusion (CSII) is associated with a lower risk of severe hypoglycaemia in type 1 diabetes (Weissberg-Benchell et al., 2003). This method of insulin delivery is not routinely employed in people with type 2 diabetes, but a 12-month prospective randomised study of 107 adults with insulin-treated type 2 diabetes showed no significant difference in the rates of mild or severe hypoglycaemia between CSII and multiple insulin injections (Herman et al., 2005).
Hypoglycaemia and Newer Treatment Modalities for type 2 diabetes Inhaled insulin An alternative route of delivery of insulin is by inhalation into the lungs. Inhaled insulin has been compared with subcutaneous insulin in several studies and the frequency of hypoglycaemia is equivalent to that of subcutaneous administration of insulin (Hermansen et al., 2004; Hollander et al., 2004).
MORBIDITY OF HYPOGLYCAEMIA AND NEED FOR EMERGENCY TREATMENT
255
Incretin mimetics Glucagon-like peptide 1 (GLP-1) is an incretin hormone that promotes glucose-dependent insulin secretion and inhibition of glucagon production. GLP-1 is rapidly degraded in vivo by the ubiquitous enzyme Dipeptidyl Peptidase IV (DPP-IV) and no stable oral preparation is available. GLP-1 mimetics are associated with improvements in glycaemic control (Zander et al., 2002; Fineman et al., 2003; Koltermann et al., 2003; Degn et al., 2004a) without appearing to cause hypoglycaemia in people with type 2 diabetes (Knop et al., 2003; Madsbad et al., 2004). Exenatide, a synthetic GLP-1 receptor agonist, stimulates insulin release only in the presence of glucose (Degn et al., 2004b) and, although it suppresses glucagon production, the effect on glucagon suppression is abolished during hypoglycaemia and the counterregulatory response to insulin-induced hypoglycaemia is preserved (Nauck et al., 2002, Degn et al., 2004b). Although incretin mimetics may cause reactive hypoglycaemia in non-diabetic individuals (Meier and Nauck, 2005), they do not appear to cause hypoglycaemia in people with type 2 diabetes (Knop et al., 2003, Vilsbøll et al., 2001). Few studies to date have quantified the risk of hypoglycaemia associated with DPP-IV inhibitors. In a one-year placebo-controlled trial in which the DPP-IV inhibitor, vildagliptin was added to metformin, only four confirmed episodes of mild (self-treated) hypoglycaemia were recorded in the 51 patients in the treatment arm (with no episodes in the placebo arm) and severe hypoglycaemia did not occur (Ahren et al., 2004).
MORBIDITY OF HYPOGLYCAEMIA AND NEED FOR EMERGENCY TREATMENT Many people with type 1 diabetes regard severe hypoglycaemia with the same degree of trepidation as that reserved for the advanced complications of diabetes such as loss of sight or renal failure (Pramming et al., 1991). Hypoglycaemia is not simply extremely unpleasant for the individual concerned; it has the potential risk of severe morbidity and may precipitate major vascular events such as stroke, myocardial infarction, acute cardiac failure and ventricular arrhythmias (Landstedt-Hallin et al., 1999; McAulay and Frier, 2001; Desouza et al., 2003) (see Chapter 12). Healthcare professionals may not always recognise the causative role of hypoglycaemia when treating these secondary events, especially if they are unfamiliar with some of the age-related neurological manifestations of hypoglycaemia. The elderly are particularly at risk of hypoglycaemia-related physical injury and bone fractures as a result of their general frailty and the presence of co-morbidities, such as osteoporosis (McAulay and Frier 2001). In a seven-year review of 102 cases of hypoglycaemic coma secondary to either insulin or glibenclamide, 92 patients had type 2 diabetes; seven sustained physical injury, five died, two suffered myocardial ischaemia and one patient had a stroke (Ben-Ami et al., 1999). In type 1 diabetes, relatives often treat severe hypoglycaemia at home, but while people with insulin-treated type 2 diabetes experience severe hypoglycaemia less frequently, they appear to require the assistance of the emergency services with equal frequency. This might suggest that people with insulin-treated diabetes are at greater risk of morbidity and disability during hypoglycaemia, and they and their relatives may be less able to cope than younger people with type 1 diabetes. In addition, many people with insulin-treated type 2 diabetes live alone. A population survey in the region of Tayside in Scotland indicated that the annual
256
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
rate of severe hypoglycaemia requiring emergency medical intervention was similar in these groups (Leese et al., 2003). All episodes of severe hypoglycaemia requiring input from the emergency medical services in one year were identified. A total of 160 people with diabetes required treatment for 244 episodes of severe hypoglycaemia. Emergency treatment was required for 7.1% of those with type 1 diabetes, 7.3% of those with insulin-treated type 2 diabetes and 0.8% of people taking oral antidiabetic agents. In a different prospective survey in the same region of Tayside, the occurrence of hypoglycaemia was monitored over a period of one month in a cohort of 267 people with insulin-treated diabetes (both type 1 and type 2) (Donnelly et al., 2005). The prevalence of all forms of hypoglycaemia in the group with insulin-treated type 2 diabetes was 45% with an incidence of 16.4 episodes per patient per year, compared to an incidence of 42.9 episodes per patient per year in type 1 diabetes. The incidence of severe hypoglycaemia was 0.35 episodes per patient per year in the group with type 2 diabetes and 1.15 episodes per patient per year in those with type 1 diabetes. The figures for the incidences were extrapolated from prospective data collected over one month but these calculated rates for people with type 1 diabetes are consistent with those recorded in other European studies (Pramming et al., 1991; MacLeod et al., 1993; ter Braak et al., 2000; Pedersen-Bjergaard et al., 2004), suggesting that the data collected were representative of the annual event rate. In this study, only 10% of the group with type 1 diabetes experiencing severe hypoglycaemia required emergency service treatment compared to one in three of the group with type 2 diabetes. Thus the frequency of severe hypoglycaemia recorded in people with type 2 diabetes was higher than anticipated and their need to enlist the help of the emergency services was greater than in those with type 1 diabetes.
CONCLUSIONS • Ageing modifies the counterregulatory and symptomatic responses to hypoglycaemia. • In older people, effective self-treatment of hypoglycaemia may be compromised by the close proximity of the glycaemic thresholds for the onset of symptoms and cognitive dysfunction, which occur almost simultaneously. • In type 2 diabetes, counterregulatory responses to hypoglycaemia commence at higher blood glucose levels than those observed in non-diabetic adults or in people with type 1 diabetes, and this may have a protective effect. When insulin therapy is introduced and HbA1c is reduced, these thresholds are shifted to lower blood glucose levels. • With progressive insulin deficiency, people with type 2 diabetes develop counterregulatory hormonal deficiencies and impaired symptomatic awareness, similar to the acquired hypoglycaemia syndromes of type 1 diabetes. • Hypoglycaemia in type 2 diabetes occurs most frequently with insulin therapy, but sulphonylurea-induced hypoglycaemia is also a significant but under-estimated problem. • Although less common than in type 1 diabetes, the frequency of hypoglycaemia in insulintreated type 2 diabetes rises progressively with increasing duration of insulin treatment. • People with insulin-treated type 2 diabetes are more likely to require the assistance of emergency medical services to treat severe hypoglycaemia than those with type 1 diabetes.
REFERENCES
257
• The paucity of data from elderly people with diabetes is of concern, as the morbidity of hypoglycaemia is greater in this age group, their presenting features of hypoglycaemia may be misinterpreted and prompt treatment may not be provided.
REFERENCES Abraira C, Colwell JA, Nuttall FQ, Sawin CT, Nagel NJ, Comstock JP et al., the VA CSDM Group (1995). Veterans Affairs Cooperative Study on glycemic control and complications in Type II Diabetes (VA CSDM). Diabetes Care 18: 1113–23. Ahren B, Landin-Olsson M, Jansson PA, Svensson M, Holmes D, Schweizer A (2004). Inhibition of dipetidyl-peptidase-4 reduces glycemia, sustains insulin levels, and reduces glucagon levels in type 2 diabetes. Journal of Clinical Endocrinology and Metabolism 89: 2078–84. Akram K, Pedersen-Bjergaard U, Borch-Johnsen K, Thorsteinsson B (2003). Recall of severe hypoglycaemic episodes and course of hypoglycaemia awareness in insulin treated type 2 diabetes in one year follow-up. Diabetologia 46 (Suppl 2) A304 (Abstract). Akram K, Pedersen-Bjergaard U, Carstensen B, Borch-Johnsen K, Thorsteinsson B (2006). Frequency and risk factors of severe hypoglycaemia in insulin-treated type 2 diabetes: a cross-sectional survey. Diabetic Medicine 23: 750–6. Allen KV, McAulay V, Sommerfield AJ, Frier BM (2004). Hypoglycaemia is uncommon with a combination of antidiabetic drugs and bedtime NPH insulin for type 2 diabetes. Practical Diabetes International 21: 179–82. Amiel SA, Maran A, Powrie JK, Umpleby AM, Macdonald IA (1993). Gender differences in counterregulation to hypoglycaemia. Diabetologia 36: 460–4. Anderson JH Jr, Brunelle RL, Keohane P, Koivisto VA, Trautmann ME, Vignati L, DiMarchi R (1997). Mealtime treatment with insulin analog improves postprandial hyperglycemia and hypoglycemia in patients with non-insulin-dependent diabetes mellitus. Multicenter Insulin Lispro Study Group. Archives of Internal Medicine 157: 1249–55. Asplund K, Wilholm B-E, Lundman B (1991). Severe hypoglycaemia during treatment with glipizide. Diabetic Medicine 8: 726–31. Avogaro A, Beltramello P, Gnudi L, Maran A, Valerio A, Miola M et al. (1993). Alcohol intake impairs glucose counterregulation during acute insulin-induced hypoglycemia in IDDM patients. Evidence for a critical role of free fatty acids. Diabetes 42: 1626–34. Banarer S, McGregor V, Cryer PE (2002). Intraislet hyperinsulinemia prevents the glucagon response to hypoglycemia despite an intact autonomic response. Diabetes 51: 958–65. Bastyr EJ, Huang Y, Brunelle RL, Vignati L, Cox DJ, Kotsamos JG (2000). Factors associated with nocturnal hypoglycemia among patients with type 2 diabetes new to insulin therapy. Diabetes Obesity and Metabolism 2: 39–46. Ben-Ami H, Nagachandran P, Mendelson A, Edoute Y (1999). Drug-induced hypoglycemic coma in 102 diabetic patients. Arch Intern Med 159: 281–4. Boden G, Soriano M, Hoeldtke RD, Owen OE (1983). Counterregulatory hormone release and glucose recovery after hypoglycemia in non-insulin-dependent patients. Diabetes 32: 231–7. Bolli G, Tsalikian E, Haymond M, Cryer P, Gerich JE (1984). Defective glucose counterregulation after subcutaneous insulin in noninsulin dependent diabetes mellitus: paradoxical lack of compensatory increase in glucose production, roles of insulin resistance, abnormal neuroendocrine responses and islet paracrine interactions. Journal of Clinical Investigation 73: 1532–41. Brierley EJ, Broughton DL, James OFW, Alberti KGMM (1995). Reduced awareness of hypoglycaemia in the elderly despite an intact counterregulatory response. Quarterly Journal of Medicine 88: 439–45.
258
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
Burge MR, Zeise T-M, Sobhy TA, Rassam AG, Schade DS (1999). Low-dose ethanol predisposes elderly fasted patients with type 2 diabetes to sulfonylurea-induced low blood glucose. Diabetes Care 22: 2037–43. Campbell IW, Chalmers J, Herlihy OM (1994). Sulphonylurea-induced hypoglycaemia in elderly people with diabetes. Practical Diabetes 11: 102–3. Chico A, Vidal-Rios P, Subira M, Novials A (2003). The continuous glucose monitoring system is useful for detecting unrecognised hypoglycemias in patients with type 1 and type 2 diabetes but is not better than frequent capillary glucose measurements for improving metabolic control. Diabetes Care 26: 1153–7. Chipkin SR, Klugh SA, Chasan-Taber L (2001). Exercise and diabetes. Cardiology Clinics 19: 489–505. Cryer PE (1992). Iatrogenic hypoglycemia as a cause of hypoglycemia-associated autonomic failure in IDDM. A vicious cycle. Diabetes. 41: 255–60 Cryer PE (2002). Hypoglycaemia: The limiting factor in the glycaemic management of type I and type II diabetes. Diabetologia 45: 937–48. Cryer PE, Davis SN, Shamoon H (2003). Hypoglycemia in diabetes. Diabetes Care 26: 1902–12. Cryer PE (2004). Diverse causes of hypoglycemia-associated autonomic failure in diabetes. New England Journal of Medicine 350: 2272–9. Culy CR, Jarvis B (2001). Repaglinide: a review of its therapeutic use in type 2 diabetes mellitus. Drugs 61: 1625–60. Dailey G, Rosenstock J, Moses RG, Ways K (2004). Insulin glulisine provides improved glycemic control in patients with type 2 diabetes. Diabetes Care 27: 2363–8. Davis SN, Goldstein RE, Jacobs J, Price L, Wolfe R, Cherrington AD (1993). The effects of differing insulin levels on the hormonal and metabolic response to equivalent hypoglycemia in normal humans. Diabetes 42: 263–72. Davis SN, Fowler S, Costa F (2000). Hypoglycemic counterregulatory responses differ between men and women with type 1 diabetes. Diabetes 49: 56–72. De Galan BE, Hoekstra JBL (2001). Glucose counterregulation in type 2 diabetes mellitus. Diabetic Medicine 18: 519–27. Deary IJ, Hepburn DA, MacLeod KM, Frier BM (1993). Partitioning the symptoms of hypoglycaemia using multi-sample confirmatory factor analysis. Diabetologia 36: 771–7. DeFronzo RA (1999). Pharmacologic therapy for type 2 diabetes mellitus. Annals of Internal Medicine 131: 281–303. Degn KB, Juhl CB, Sturis J, Jakobsen G, Brock B, Chandramouli V et al. (2004a). One week’s treatment with the long-acting glucagon-like peptide 1 derivative liraglutide (NN2211) markedly improves 24h glycemia and - and -cell function and reduces endogenous glucose release in patients with type 2 diabetes. Diabetes 53: 1187–94. Degn KB, Brock B, Juhl CB, Djurhuus CB, Grubert J, Kim D et al. (2004b). Effect of intravenous infusion of exenatide (synthetic Exendin-4) on glucose-dependent insulin secretion and counterregulation during hypoglycemia. Diabetes 53: 2397–403. Del Prato, Aragona M, Coppelli A (2002). Sulfonylureas and hypoglycaemia. Diabetes Nutrition and Metabolism 15: 444–51. Desouza C, Salazar H, Cheong B, Murgo J, Fonseca V (2003). Association of hypoglycemia and cardiac ischemia. Diabetes Care 26: 1485–89. Devlin JT, Hirshman M, Horton ED, Horton ES (1987). Enhanced peripheral and splanchnic insulin sensitivity in NIDDM men after single bout of exercise. Diabetes 36: 434–39. Diamond MP, Hallerman L, Starick-Zych K, Jones TW, Connolly-Howard M, Tamborlane WV, Sherwin RS (1991). Suppression of counterregulatory hormone response to hypoglycemia by insulin per se. Journal of Clinical Endocrinology and Metabolism 72: 1388–90. Donnelly LA, Morris AD, Frier BM, Ellis JD, Donnan PT, Durrant R et al., for the DARTS/MEMO collaboration (2005). Frequency and predictors of hypoglycaemia in type 1 and insulin-treated type 2 diabetes: a population based study. Diabetic Medicine 22: 449–55.
REFERENCES
259
Fanelli C, Pampanelli S, Epifano L, Rambotti AM, Ciofetta M, Modarelli F et al. (1994). Relative roles of insulin and hypoglycaemia on induction of neuroendocrine responses to, symptoms of and deterioration of cognitive function in hypoglycaemia in male and female humans. Diabetologia 37: 797–807. Fineman MS, Bicsak TA, Shen LZ, Taylor K, Gaines E, Varns A et al. (2003). Effect on glycemic control of exenatide (synthetic exendin-4) additive to existing metformin and/or sulfonylurea treatment in patients with type 2 diabetes. Diabetes Care 26: 2370–7. Fink RI, Revers RR, Kolterman OG, Olefsky JM (1985). The metabolic clearance of insulin and the feedback inhibition of insulin secretion are altered with ageing. Diabetes 34: 275–80. Gold AE, MacLeod KM, Frier BM (1994). Frequency of severe hypoglycemia in patients with type 1 diabetes with impaired awareness of hypoglycemia. Diabetes Care 17: 697–703. Goudswaard A, Furlong NJ, Rutten G, Stolk R, Valk G (2004). Insulin monotherapy versus combinations of insulin with oral hypoglycaemic agents in patients with type 2 diabetes mellitus (Cochrane review). Cochrane Database of Systematic Reviews 4: CD003418. Haak T, Tiengo A, Draeger E, Suntum M, Waldhausl W (2005). Lower within-subject variability of fasting blood glucose and reduced weight gain with insulin detemir compared to NPH insulin in patients with type 2 diabetes. Diabetes Obesity and Metabolism 7: 56–64. Harrigan RA, Nathan MS, Beattie P (2001). Oral agents for the treatment of type 2 diabetes mellitus: pharmacology, toxicity and treatment. Annals of Emergency Medicine 38: 68–78. Harrower A (2000). Gliclazide modified release: from once-daily administration to 24-hour blood glucose control. Metabolism 49 (Suppl 2): 7–11. Hartling SG, Faber OK, Wegmann M-L, Wahlin-Boll E, Melander A (1987). Interaction of ethanol and glipizide in humans. Diabetes Care 10: 683–6. Hay LC, Wilmhurst EG, Fulcher G (2003). Unrecognized hypo- and hyperglycemia in well-controlled patients with type 2 diabetes mellitus: the results of continuous glucose monitoring. Diabetes Technology and Therapeutics 5: 19–26. Heller SR, Macdonald IA, Tattersall RB (1987). Counterregulation in type II (non-insulin-dependent) diabetes mellitus. Normal endocrine and glycaemic responses up to 10 years after diagnosis. Diabetologia 30: 924–9. Henderson JN, Allen KV, Deary IJ, Frier BM (2003). Hypoglycaemia in insulin-treated type 2 diabetes: frequency, symptoms and impaired awareness. Diabetic Medicine 20: 1016–21. Hepburn DA, MacLeod KM, Pell ACH, Scougal IJ, Frier BM (1993). Frequency and symptoms of hypoglycaemia experienced by patients with type 2 diabetes treated with insulin. Diabetic Medicine 10: 231–7. Herman WH, Ilag LL, Johnson SL, Martin CL, Sinding J, AL Harthi A et al. (2005). A clinical trial of continuous subcutaneous insulin infusion versus multiple daily injections in older adults with type 2 diabetes. Diabetes Care 28: 1568–73. Hermansen K, Ronnemaa T, Petersen AH, Bellaire S, Adamson U (2004). Intensive therapy with inhaled insulin via the AERx insulin diabetes management system: a 12-week proof-of-concept trial in patients with type 2 diabetes. Diabetes Care 27: 162–7. Hermansen K, Davies M, Derezinski T, Martinez Ravn G, Clauson P, Home P, on behalf of the Levemir Treat-To-Target Group (2006). A 26-week, randomised, parallel, treat-to-target trial comparing insulin detemir with NPH insulin as add-on therapy to oral glucose-lowering drugs in insulin-naïve people with type 2 diabetes. Diabetes Care 29: 1269–74. Hollander PA, Blonde L, Rowe R, Mehta AE, Milburn JL, Hershon KS et al., for the Exubera Phase III Study Group (2004). Efficacy and safety of inhaled insulin (Exubera) compared with subcutaneous insulin therapy in patients with type 2 diabetes. Diabetes Care 27: 2356–62. Holstein A, Plaschke A, Egberts E-H (2001). Lower incidence of severe hypoglycemia in patients with type 2 diabetes treated with glimepiride versus glibenclamide. Diabetes Metabolism Research and Reviews 17: 467–73.
260
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
Israelian Z, Gosmanov NR, Szoke E, Schorr M, Bokhari S, Cryer PE et al. (2005). Increasing the decrement in insulin secretion improves glucagon responses to hypoglycemia in advanced type 2 diabetes. Diabetes Care 28: 2691–6. Israelian Z, Szoke E, Woerle J, Bokhari S, Schorr M, Schwenke DC et al. (2006). Multiple defects in counterregulation of hypoglycemia in modestly advanced type 2 diabetes mellitus. Metabolism 55: 593–8. Jaap AJ, Jones GC, McCrimmon RJ, Deary IJ, Frier BM (1998). Perceived symptoms of hypoglycaemia in elderly type 2 diabetic patients treated with insulin. Diabetic Medicine 15: 398–401. Janka HU, Plewe G, Riddle MC, Klieb-Frisch C, Schweitzer MA, Yki-Järvinen H (2005). Comparison of basal insulin added to oral agents versus twice-daily premixed insulin as initial therapy for type 2 diabetes. Diabetes Care 28: 254–9. Jennings AM, Wilson RM, Ward JD (1989). Symptomatic hypoglycemia in NIDDM patients treated with oral hypoglycemic agents. Diabetes Care 12: 203–7. Jonsson A, Chan JC, Rydberg T, Vaaler S, Hallengren B, Cockram CS et al. (2001). Pharmacodynamics and pharmacokinetics of intravenous glibenclamide in Caucasian and Chinese patients with type-2 diabetes. European Journal of Clinical Pharmacology 55: 721–7. Kahn S, Haffner SM, Heise MA, Herman WH, Holman RR, Jones NP et al., on behalf of the ADOPT study group (2006). Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. New England Journal of Medicine 355: 2427–43. Knop FK, Vilsbøll T, Larsen S, Madsbad S, Holst J, Krarup T (2003). No reactive hypoglycemia after subcutaneous administration of glucagon-like peptide-1 in lean type 2 diabetic patients and in patients with diabetes secondary to chronic pancreatitis. Diabetes Care 26: 2581–7. Kobayashi K, Kimura M, Sakuguchi T, Hase A, Matsuoka A (1984). Pharmacokinetics of gliclazide in healthy and diabetic subjects. Journal of Pharmacological Science 73: 1684–7. Kolterman OG, Buse JB, Fineman MS, Gaines E, Heintz S, Bisack TA et al. (2003). Synthetic Exendin-4 (exenatide) significantly reduces postprandial and fasting plasma glucose in subjects with type 2 diabetes. Journal of Clinical Endocrinology and Metabolism 88: 3082–8. Korzon-Burakowska A, Hopkins D, Matyka K, Lomas J, Pernet A, Macdonald I, Amiel S (1998). Effects of glycemic control on protective responses against hypoglycemia in type 2 diabetes. Diabetes Care 21: 283–90. Landstedt-Hallin L, Adamson U, Lins P-E (1999). Oral glibenclamide suppresses glucagon secretion during insulin-induced hypoglycemia in patients with type 2 diabetes. Journal of Clinical Endocrinology and Metabolism 84: 3140–5. Larsen JJ, Dela F, Kjaer M, Galbo H (1997). The effect of moderate exercise on postprandial glucose homeostasis in NIDDM patients. Diabetologia 40: 447–53. Lawrence PA, Cheely J (1980). Deterioration of diabetic patients’ knowledge and management skills as determined during out-patient visits. Diabetes Care 3: 214–8. Leese GP, Wang J, Broomhall JKelly P, Marsden A, Morrison W et al., for the DARTS/MEMO Collaboration (2003). Frequency of severe hypoglycemia requiring emergency treatment in type 1 and type 2 diabetes: a population-based study of health service resource use. Diabetes Care 26: 1176–80. Levy CJ, Kinsley BT, Bajaj M, Simonson DC (1998). Effect of glycemic control on glucose counterregulation during hypoglycemia in NIDDM. Diabetes Care 21: 1330–8. Liu D, Moberg E, Kollind M, Lins PE, Adamson U (1991a). A high concentration of circulating insulin suppresses the glucagon response to hypoglycemia in normal man. Journal of Clinical Endocrinology and Metabolism 73: 1123–8. Liu DT, Adamson UC, Lins PE, Kollind ME, Moberg EA, Andreasson K (1991b). Inhibitory effect of circulating insulin on glucagon secretion during hypoglycemia in type I diabetic patients. Diabetes Care 15: 59–65. MacLeod KM, Hepburn DA, Frier BM (1993). Frequency and morbidity of severe hypoglycaemia in insulin-treated diabetic patients. Diabetic Medicine 10: 238–45.
REFERENCES
261
Madsbad S, Schmitz O, Ranstam J, Jakobsen G, Matthews DR, on behalf of the NN2211-1310 international study group (2004). Improved glycemic control with no weight increase in patients with type 2 diabetes after once-daily treatment with the long-acting Glucagon-Like Peptide 1 analog Liraglutide (NN2211): a 12-week, double-blind, randomized, controlled trial. Diabetes Care 27: 1335–42. Maggs DG, Jacob R, Rife F, Caprio S, Tamborlane WV, Sherwin RS (1997). Counterregulation in peripheral tissues: effect of systemic hypoglycemia on levels of substrates and catecholamines in human skeletal muscle and adipose tissue. Diabetes 46: 70–6. Marker JC, Cryer PE, Clutter WE (1992). Attenuated glucose recovery from hypoglycemia in the elderly. Diabetes 41: 671–8. Matyka K, Evans M, Lomas J, Cranston I, Macdonald I, Amiel SA (1997). Altered hierarchy of protective responses against severe hypoglycemia in normal aging in healthy men. Diabetes Care 20: 135–41. McAulay V, Deary IJ, Frier BM (2001). Symptoms of hypoglycaemia in people with diabetes. Diabetic Medicine 18: 690–705. McAulay V, Frier BM (2001) Hypoglycaemia. In: Diabetes in Old Age. Second edition. Sinclair AJ, Finucane P, eds. John Wiley and Sons, Ltd, Chichester: 133–52. McAulay V, Frier BM (2003). Insulin analogues and other developments in insulin therapy for diabetes. Expert Opinion in Pharmacotherapy 4: 1141–56. Meier JJ, Nauck MA (2005). Glucagon-like peptide 1 (GLP-1) in biology and pathology. Diabetes Metabolism Research and Reviews 21: 91–117. Meneilly GS, Cheung E, Tuokko H (1994a). Altered responses to hypoglycemia of healthy elderly people. Journal of Clinical Endocrinology and Metabolism 78: 1341–8. Meneilly GS, Cheung E, Tuokko H (1994b). Counterregulatory hormone responses to hypoglycemia in the elderly patient with diabetes. Diabetes 43: 403–10. Miller CD, Philips LS, Ziemer DC, Gallina DL, Cook CB, El-Kebbi IM (2001). Hypoglycemia in patients with type 2 diabetes mellitus. Archives of Internal Medicine 161: 1653–9. Minaker KL, Rowe JW, Torino R, Pallotta JA (1982). Influence of age on clearance of insulin in man. Diabetes 31: 851–5. Minuk HL, Vranic M, Marliss EB, Hanna AK, Albisser AM, Zinman B (1981). Glucoregulatory and metabolic response to exercise in obese noninsulin-dependent diabetes. American Journal of Endocrinology 240: E458–64. Mitrakou A, Ryan C, Veneman T, Mokan M, Jenssen T, Kiss I et al. (1991). Hierarchy of glycemic thresholds for counterregulatory hormone secretion, symptoms and cerebral dysfunction. American Journal of Endocrinology 266: E67–74. Murata GH, Hoffman RM, Shah JH, Wendel CS, Duckworth WC (2004). A probabilistic model for predicting hypoglycemia in type 2 diabetes mellitus. Archives of Internal Medicine 164: 1445–50. Mutch WJ, Dingwall-Fordyce I (1985). Is it a hypo? Knowledge of the symptoms of hypoglycaemia in elderly diabetic patients. Diabetic Medicine 2: 54–6. Nattrass M, Lauritzen T (2000). Review of prandial glucose regulation with repaglinide: a solution to the problem of hypoglycemia in the treatment of type 2 diabetes? International Journal of Obesity and Metabolic Diseases 24 (Suppl 3): S21–31. Nauck MA, Heimesaat MM, Behle K, Holst JJ, Nauck MS, Ritzel R et al. (2002). Effects of glucagonlike peptide 1 on counterregulatory hormone responses, cognitive functions and insulin secretion during hyperinsulinemic stepped hypoglycemic clamp experiments on healthy volunteers. Journal of Clinical Endocrinology and Metabolism 87: 1239–46. Ortiz-Alonso FJ, Galecki A, Herman WH, Smith MJ, Jacquez JA, Halter JB (1994). Hypoglycemia counterregulation in elderly humans: relationship to glucose levels. American Journal of Endocrinology 267: E497–506. Peacey SR, George E, Rostami-Hodjegan A, Bedford C, Harris N, Hardisty CA et al. (1996). Similar physiological and symptomatic responses to sulphonylurea and insulin induced hypoglycaemia in normal subjects. Diabetic Medicine 13: 634–41.
262
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
Pedersen-Bjergaard U, Pramming S, Thorsteinsson B (2003). Recall of severe hypoglycaemia and self-estimated state of awareness in type 1 diabetes. Diabetes Metabolism Research and Reviews 19: 232–40. Pedersen-Bjergaard U, Pramming S, Heller SR, Wallace TM, Rasmussen AK, Jorgensen HV et al. (2004). Severe hypoglycaemia in 1076 adult patients with type 1 diabetes: influence of risk markers and selection. Diabetes Metabolism Research and Reviews 20: 479–86. Pegg A, Fitzgerald D, Wise D, Singh BM, Wise PH (1991). A community-based study of diabetesrelated skills and knowledge in elderly people with insulin-requiring diabetes. Diabetic Medicine 8: 778–81. Pramming S, Thorsteinsson B, Bendtson I, Binder C (1991). Symptomatic hypoglycaemia in 411 type 1 diabetic patients. Diabetic Medicine 8: 217–22. Raskin P, Allen E, Hollander P, Gabbay RA, Hu P, Bode B, Garber A, for the INITIATE Study Group (2005). Initiating insulin therapy in type 2 diabetes: a comparison of biphasic and basal insulin analogs. Diabetes Care 26: 260–5. Raslova K, Bogoev M, Raz I, Leth G, Gall MA, Hancu N (2004). Insulin detemir and insulin aspart: a promising basal-bolus regimen for type 2 diabetes. Diabetes Research and Clinical Practice 66: 193–201. Rasmussen BM, Christiansen C, Rasmussen OW, Hansen C, Hermansen K (1999). Alcohol and postexercise metabolic responses in type 2 diabetes. Metabolism 48: 597–602. Rasmussen BM, Orskov L, Schmitz O, Hermansen K (2001). Alcohol and glucose counterregulation during acute insulin-induced hypoglycemia in type 2 diabetic subjects. Metabolism 50: 451–7. Reaven GM, Greenfield MS, Mondon CE, Rosenthal M, Wright D, Reaven EP (1982). Does insulin removal rate from plasma decline with age? Diabetes 31: 670–73. Rendell M (2004). The role of sulphonylureas in the management of type 2 diabetes mellitus. Drugs 64: 1339–58. Riddle MC, Rosenstock J, Gerich J, on behalf of the insulin glargine 4002 study investigators (2003). Randomized addition of glargine or human NPH insulin to oral therapy of type 2 diabetic patients. Diabetes Care 26: 3080–6. Rosenstock J, Schwartz SL, Clark CM, Park GD, Donley DW, Edwards MB (2001). Basal insulin therapy in type 2 diabetes. Diabetes Care 24: 631–6. Rosenstock J, on behalf of the HOE 901/204 Study Investigators Group (2003). Safety and efficacy of insulin glargine (HOE 901) versus NPH insulin in combination with oral treatment in type 2 diabetic patients. Diabetic Medicine 20: 545–51. Rosenstock J, Hassman DR, Madder RD Brazinsky, Farrell J, Khutoryansky N, Hale PM: (2004). Repaglinide versus nateglinide monotherapy. Diabetes Care 27: 1265–70. Ross SA, Zinman B, Campos RV, Strack T, Canadian Lispro Study Group (2001). A comparative study of insulin lispro and human regular insulin in patients with type 2 diabetes mellitus and secondary failure of oral hypoglycemic agents. Clinical Investigative Medicine 24: 292–8. Sandoval DA, Ertl AC, Richardson MA, Tate DB, Davis SN (2003). Estrogen blunts neuroendocrine and metabolic responses to hypoglycemia. Diabetes 52: 1749–55. Schernthaner G (2003). Gliclazide modified release: A critical review of pharmacodynamic, metabolic and vasoprotective effects. Metabolism 52 (Supplement 1): 29–34. Schernthaner G, Grimaldi A, Di Mario U, Drzewoski J, Kempler P, Kvapil M et al. (2004a). GUIDE study: double blind comparison of once-daily gliclazide MR and glimepiride in type 2 diabetic patients. European Journal of Clinical Investigation 34: 535–42. Schernthaner G, Kopp HP, Ristic S, Muzyka B, Peter L, Mitteregger G (2004b). Metabolic control in patients with type 2 diabetes using Humalog Mix50 injected three times daily: crossover comparison with human insulin 30/70. Hormone and Metabolic Research 36: 188–93. Schwartz NS, Clutter WE, Shah SD, Cryer PE (1987). Glycemic thresholds for activation of glucose counterregulatory systems are higher than the thresholds for symptoms. Journal of Clinical Investigation 79: 777–81.
REFERENCES
263
Segel SA, Paramore DA, Cryer PE (2002). Hypoglycemia-Associated Autonomic Failure in advanced type 2 diabetes. Diabetes 51: 724–33. Seltzer HS (1989). Drug-induced hypoglycemia. Endocrinology and Metabolism Clinics of North America 18: 163–82. Shamoon H, Friedman A, Canton C, Zacharowicz L, Hu M, Rossetti L (1994). Increased epinephrine and skeletal muscle responses to hypoglycemia in non-insulin-dependent diabetes mellitus. Journal of Clinical Investigation 93: 2562–71. Shorr RI, Ray WA, Daugherty JR, Griffin MR (1997). Incidence and risk factors for serious hypoglycemia in older persons using insulin or sulfonylureas. Archives of Internal Medicine 157: 1681–6. Spyer G, Hattersley A, Macdonald IA, Amiel S, MacLeod KM (2000). Hypoglycemic counterregulation at normal blood glucose concentrations in patients with well controlled type 2 diabetes. Lancet 356: 1970–4. Stahl M, Berger W (1999). Higher incidence of severe hypoglycaemia leading to hospital admission in type 2 diabetic patients treated with long-acting versus short-acting sulphonylureas. Diabetic Medicine 16: 586–90. Stephenson J, Fuller JH, on behalf of the EURODIAB IDDM Complications Study Group (1994). Microvascular and acute complications in IDDM patients: the EURODIAB IDDM complications study. Diabetologia 37: 278–85. Strange P, Schwartz SL, Graf RJ, Polvino W, Weston I, Marbury TC et al. (1999). Pharmacokinetics, pharmacodynamics, and dose-response relationship of repaglinide in type 2 diabetes. Diabetes Technology and Therapeutics 1: 247–56. ter Braak EWMT, Appelman AMMF, van de Laak M, Stolk RP, van Haeften TW, Erkelens DW (2000). Clinical characteristics of type 1 diabetic patients with and without severe hypoglycemia. Diabetes Care 23: 1467–71. ter Braak EWMT, Appelman AMMF, van Der Tweel I, Erkelens DW, van Haeften TW (2002). The sulfonylurea glibenclamide induces impairment of glucagon and growth hormone responses during mild insulin-induced hypoglycemia. Diabetes Care 25: 107–12. Tessier D, Dawson K, Tetrault JP, Bravo G, Meneilly GS (1994). Glibenclamide versus gliclazide in type 2 diabetes of the elderly. Diabetic Medicine 11: 974–80. The Diabetes Control and Complications Trial Research Group (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New England Journal of Medicine 329: 977–86. Thomson FJ, Masson EA, Leeming JT, Boulton AJM (1991). Lack of knowledge of symptoms of hypoglycaemia by elderly diabetic patients. Age and Ageing 20: 404–6. Trovati M, Carta Q, Cavalot F, Vitali S, Banaudi C, Lucchina PG et al. (1984). Influence of physical training on blood glucose control, glucose tolerance, insulin secretion and insulin action in noninsulin-dependent diabetic patients. Diabetes Care 7: 416–20. UK Hypoglycaemia Study Group (2007). Risk of hypoglycaemia in types 1 and 2 diabetes: effects of treatment modalities and their duration. Diabetologia 50: 1140–7. UK Prospective Diabetes Study (UKPDS) Group (1998a). Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 352: 854–65. UK Prospective Diabetes Study (UKPDS) Group (1998b). Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 352: 837–52. United Kingdom Prospective Diabetes Study Group (1998c). A 6 year randomised controlled trial comparing sulfonylurea, insulin and metformin therapy in patients with newly diagnosed type 2 diabetes that could not be controlled with diet therapy (UKPDS 24). Annals of Internal Medicine 128: 165–75. Vilsbøll T, Krarup T, Madsbad S, Holst JJ (2001). No reactive hypoglycaemia in type 2 diabetic patients after subcutaneous administration of GLP-1 and intravenous glucose. Diabetic Medicine 18: 144–9.
264
HYPOGLYCAEMIA IN TYPE 2 DIABETES AND IN ELDERLY PEOPLE
Weissberg-Benchell J, Antisdel-Lomaglio J, Seshadri R (2003). Insulin pump therapy. A meta-analysis. Diabetes Care 26: 1079–87. Woerle HJ, Meyer C, Popa EM, Cryer PE, Gerich JE (2003). Renal compensation for impaired hepatic glucose release during hypoglycemia in type 2 diabetes. Further evidence for hepatorenal reciprocity. Diabetes 52: 1386–92. Yki-Järvinen H, Dressler A, Ziemen M, the HOE 901/3002 study group (2000). Less nocturnal hypoglycemia and better post-dinner glucose control with bedtime insulin glargine compared with bedtime NPH insulin during insulin combination therapy in type 2 diabetes. Diabetes Care 23: 1130–6. Zammitt NN, Frier BM (2005). Hypoglycemia in type 2 diabetes: pathophysiology, frequency and effects of different treatment modalities. Diabetes Care 28: 2948–61. Zander M, Madsbad S, Madsen JL, Holst JJ (2002). Effect of 6-week course of glucagon-like peptide 1 on glycaemic control, insulin sensitivity and beta-cell function in type 2 diabetes: a parallel group study. Lancet 359: 824–30.
12 Mortality, Cardiovascular Morbidity and Possible Effects of Hypoglycaemia on Diabetic Complications Miles Fisher and Simon R. Heller
INTRODUCTION For patients with type 1 diabetes, insulin-induced hypoglycaemia is one of the most feared consequences of the disorder (Pramming et al., 1991). It is the major factor that restrains many patients from pursuing intensive insulin therapy and trying to achieve the levels of strict glycaemic control that are necessary to prevent the development of diabetic complications. Some fear the immediate lack of self control which can accompany the impairment of cognitive function during acute hypoglycaemia. Others are embarrassed by the dependence on other people for assistance during an episode of severe hypoglycaemia. Many patients share the worries expressed by some diabetes healthcare professionals about the possible long-term effects of recurrent hypoglycaemia on the brain (see Chapter 13). An additional factor which may dissuade many patients from improving their glycaemic control is the fear of dying during an episode of hypoglycaemia, especially when low blood glucose occurs during sleep. These anxieties may be shared by the patient’s relatives who may have witnessed previous episodes of nocturnal hypoglycaemia or convulsions, about which the patient has no recollection. Fear of hypoglycaemia was heightened in the 1980s by the publicity that surrounded the possible adverse effects of human insulin, and in particular the knowledge that some young people with type 1 diabetes had died suddenly and unexpectedly, the so-called ‘dead in bed syndrome’ (Campbell, 1991). It would be wrong for healthcare professionals to dismiss such fears as irrational. Many professionals will have first- or second-hand experience of the sudden death of a patient with type 1 diabetes in circumstances that have implicated acute hypoglycaemia. This chapter examines the epidemiology and causes of death from hypoglycaemia in patients with diabetes, including those risk factors that appear to be associated with sudden death. The ‘dead in bed syndrome’ is explored in detail, and comparisons drawn with other syndromes of sudden death in people who do not have diabetes. Putative mechanisms and risk factors for sudden death are described. Hypoglycaemia may also cause significant cardiovascular morbidity in people with diabetes, and the effects on heart disease and cardiovascular
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
266
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
disease are examined. Finally, the hypothesis that hypoglycaemia may worsen the chronic microvascular complications of diabetes is examined.
DEATHS ATTRIBUTABLE TO HYPOGLYCAEMIA Problems with Death Certification In people with diabetes the cause of death is often recorded inaccurately on the death certificate, even to the extent of omitting ‘diabetes’ altogether (Mulhauser et al., 2002; Waernbaum et al., 2006). Subsequent problems with analysis of death certificates can be compounded in the United Kingdom by the fact that ‘hypoglycaemia’ is not coded under one single heading. Furthermore, the exact coding of the cause of death is often left to the discretion of individual coding clerks who usually have no knowledge of the clinical details (Tattersall and Gale, 1993). These factors make it impossible to obtain any precise estimates of the frequency of sudden death, in contrast to records of the frequency of episodes of severe hypoglycaemia. Many episodes of hypoglycaemia, including nocturnal hypoglycaemia, are not recognised by patients or carers. If we add the difficulty of confirming hypoglycaemia at post-mortem, it is not surprising that considerable uncertainty and variation surround the estimated number of deaths that are attributed to hypoglycaemia in people with diabetes. However, since hypoglycaemia is so common, we can conclude that the risk of death during an individual episode is extremely low.
Problems of Establishing the Cause of Death at Post-mortem In attempting to establish a post-mortem diagnosis of hypoglycaemia, the pathologist needs to perform biochemical tests, examine the brain for evidence of hypoglycaemic brain damage, and exclude any other possible cause of death (Tattersall and Gale, 1993). Carbohydrate metabolism continues after death, and post-mortem changes in blood glucose can cause difficulties in confirming a hypoglycaemic death forensically. The continuing breakdown of glycogen (glycogenolysis) increases the blood glucose concentration in the inferior vena cava, so that the presence of a normal or high blood glucose concentration on the right side of the heart does not exclude ante-mortem hypoglycaemia (a false negative result for a diagnosis of hypoglycaemia). In the peripheral circulation, glucose continues to be utilised by red blood cells, so that the presence of a low glucose concentration does not necessarily indicate ante-mortem hypoglycaemia. Indeed low blood glucose is often found after death in those without diabetes (a false positive result for a diagnosis of hypoglycaemia). The measurement of the glucose concentration in the vitreous humour presents similar problems because of continued post-mortem glucose utilisation, and so this cannot be used to confirm ante-mortem hypoglycaemia (false positive). A normal or raised glucose concentration in the vitreous humour after death, however, excludes hypoglycaemia at the time of death (true negative). Thus, the sensitivities and specificities of blood and vitreous humour measurements of glucose in diagnosing ante-mortem hypoglycaemia are unknown. In addition to the biochemical problems in diagnosing hypoglycaemia after death, errors may be introduced by attribution bias of the pathologist performing the post-mortem. Death may be attributed to minor degrees of coronary heart disease, since it is so common in the
RISKS OF DEATH FROM HYPOGLYCAEMIA
267
diabetic population (false negative result for a diagnosis of hypoglycaemia). Alternatively, the pathologist, even when unsure may attribute death to hypoglycaemia rather than indicating no cause on a certificate (false positive).
CRUDE ESTIMATES OF MORTALITY FROM HYPOGLYCAEMIA Because of the problems detailed above, any estimate of mortality from hypoglycaemia will be crude. Published reports range from no deaths attributable to hypoglycaemia at one extreme to between 20% and 25% in reports from some Scandinavian centres. Most studies suggest that the proportion of deaths caused by hypoglycaemia is between 2% and 6%, a lower frequency than those associated with ketoacidosis. For a more detailed analysis, the reader is referred to the review by Tattersall and Gale (1993). If deaths caused by renal failure or coronary heart disease in people with diabetes continue to decline as diabetes care improves, then the relative proportion of deaths caused by hypoglycaemia may increase. This is particularly likely if intensive insulin therapy continues to be adopted more widely in an attempt to prevent or reduce microvascular disease (The DCCT Research Group 1991; The Diabetes Control and Complications Trial Research Group 1993).
RISKS OF DEATH FROM HYPOGLYCAEMIA The risk factors that are commonly cited as increasing the risk of death from hypoglycaemia are often anecdotal, and may owe more to the prejudices of individual clinicians than to scientific evidence. Those suggested are detailed in Box 12.1 and include alcohol abuse and/or inebriation (Arky et al., 1968; Kalimo and Olsson, 1980; Critchley et al., 1984; MacCuish, 1993), psychiatric illness or personality disorder (Shenfield et al., 1980; Tunbridge, 1981), self-neglect (Tunbridge, 1981), resistance to education (Shenfield et al., 1980), hypopituitarism following pituitary ablation therapy for proliferative retinopathy (Nabarro et al., 1979; Shenfield et al., 1980), and patients who have diabetes secondary to pancreatic disease (MacCuish, 1993).
Box 12.1
Possible risk factors for death from hypoglycaemia
• Alcoholism and/or inebriation • Psychiatric illness or personality disorder • Self-neglect; inanition • Fecklessness/resistance to education • Diabetes secondary to pancreatic disease • Hypopituitarism following pituitary ablation
268
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
SUDDEN DEATH The sudden and unexpected death of a young person is an infrequent event, and a series of deaths either from accidents or from natural causes creates considerable media interest. At a British inquest in the early 1990s, it was suggested that the frequency of sudden death in people with type 1 diabetes was increasing, that these deaths were caused by hypoglycaemia and that there might be a link with the clinical use of human insulin preparations. These comments received widespread coverage in the British media, and fuelled the controversy about hypoglycaemia associated with human insulin. Sudden death does occur, albeit very infrequently, in young people who do not have diabetes (Box 12.2), and examination of the recognised causes may give some insight into possible mechanisms of sudden death in those with type 1 diabetes (Box 12.3). Indeed, even the phenomenon of unexplained death in diabetes is not a new one. In the 1960s, Malins (1968) described 14 patients who had been attending his diabetes outpatient clinic who died in circumstances implicating hypoglycaemia and in whom no alternative cause of death was identified at autopsy. Eight were over 60 years of age, and the clinical records revealed a history of poor nutrition, treatment with a large dose of long-acting insulin, nocturnal hypoglycaemia, and the absence of an alert family member. This description is strikingly similar to the circumstances described by Tattersall and Gill (1991) more than 20 years later, although, since the patients described by Malins were much older, their deaths are more likely to have been related to established cardiovascular disease. Sudden, unexpected death has also been described in type 1 diabetic patients with advanced diabetic autonomic neuropathy (Ewing et al., 1991).
Unexplained Deaths of Type 1 Diabetic Patients Following the publicity generated by the assertion that there had been an increase in sudden deaths from acute hypoglycaemia, the British Diabetic Association (now Diabetes UK)
Box 12.2
Syndromes of sudden death in non-diabetic young people
• Hypertrophic obstructive cardiomyopathy (HOCM) • Coronary heart disease (severe coronary artery occlusion or myocardial infarction) • Other cardiac anatomical abnormalities (congenital anomalies of the coronary arteries, right ventricular dysplasia) • Syndromes of QT prolongation • Epilepsy • Phaeochromocytoma • Sudden death in water • Toxic substance abuse
SUDDEN DEATH
Box 12.3
269
Possible mechanisms contributing to sudden death
• Ventricular arrhythmias/fibrillation (HOCM, coronary artery occlusion) • Increased epinephrine (sport, phaeochromocytoma, sudden death in water) • Decreased potassium (sport, phaeochromocytoma) • Autonomic stimulation/bradycardia (cold water immersion) • Severe hyperkalaemia and red cell lysis (fresh water drowning) • Respiratory arrest (autonomic neuropathy) • Asystole (epilepsy)
requested notification of sudden, unexpected deaths of young people with type 1 diabetes. Cases were referred by the forensic chemist who had publicised the issue, relatives or friends who had heard of the television coverage, and physicians with an interest in diabetes who reported recent sudden deaths of patients under their care. Detailed analysis was published by Tattersall and Gill (1991) on behalf of the British Diabetic Association. A total of 53 cases were referred, but the analysis was confined to the 50 cases who were under 50 years of age. Five cases were excluded because a definite cause of death was identified at post-mortem, and in 11 cases, death was the result of suicide or self-poisoning. Six patients were thought to have died from ketoacidosis, two from hypoglycaemic brain damage, and in four cases the death was totally unexplained. The largest group comprised 22 patients who were classified as ‘dead in bed’, and in an accompanying editorial the term ‘dead in bed syndrome’ was used (Campbell, 1991) (Figure 12.1). Analysis of the 22 ‘dead in bed’ patients showed that they were aged between 12 and 43 years, with duration of diabetes from 3 to 27 years. All had been treated with human insulin, and three were taking four injections a day, 18 twice daily insulin and one was injecting insulin once a day. Information on diabetic complications was not available for all cases, but 13 had no complications and only four had severe complications, suggesting that undiagnosed autonomic neuropathy was not a factor. All died outside hospital, 19 were sleeping alone at the time of death, and 15 died during the night. Twenty patients were found lying in an undisturbed bed. Because 14 patients had a history of severe nocturnal hypoglycaemia, and most were apparently well on retiring to bed but were found dead in the morning, the scenario was consistent with an episode of severe or protracted nocturnal hypoglycaemia having precipitated sudden death. Although all had been taking human insulin at the time of death, most had been transferred from animal insulin between six months and two years earlier, and the authors concluded that no temporal relationship between the change in insulin species and the fatal event could be demonstrated. They proposed that ‘circumstantial evidence implicates nocturnal hypoglycaemia in many cases’. Neuropathological evidence of hypoglycaemia was rare however, suggesting that protracted neuroglycopenia had not occurred in most patients, and implicating sudden cardiac or respiratory arrest as a direct consequence of hypoglycaemia.
270
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
2 5
“Dead in bed” 22
Unexplained Diabetic ketoacidosis
11
Suicide Definite cause of death 6
4
Hypoglycaemic brain damage
Figure 12.1 Sudden deaths of 50 young type 1 diabetic patients. Data derived from Tattersall and Gill (1991)
A study in Sweden of over 4000 patients with type 1 diabetes diagnosed before the age of 14 and followed up for 13 years showed a three-fold increase in the standardised mortality rate in comparison to a non-diabetic population (Sartor and Dahlquist, 1995). In nine of the 33 deaths the subjects were found ’dead in bed’, eight of whom had gone to bed apparently in good health but subsequently were found dead. One subject had a cardiac arrest after her morning injection of insulin, and at autopsy one had lacerations inside the mouth, suggesting preceding convulsions. The authors suggested that hypoglycaemia was the most likely cause of death in all patients. Very little additional information was provided, and because glycated haemoglobin values could not be standardised it was not possible to ascertain whether an association existed between strict glycaemic control and sudden death. A further publication from the same Swedish Childhood Diabetes Register has extended the dataset for another ten years, and identified another eight subjects who were found deceased in bed at home by close relatives without any cause of death found at forensic autopsy (Dahlquist and Kallen, 2005). Another study from Norway of patients under the age of 40, identified 240 deaths from all causes and 16 cases that fulfilled the criteria of ‘dead in bed syndrome’ (Thordarson and Sovik, 1995). This represented 6.7% of all deaths in this age group. All were found in an undisturbed bed, and nine had been on regimens requiring multiple injections of insulin (eight were taking five injections a day, with one taking a total of seven injections a day). Frequent episodes of hypoglycaemia were documented in 12 cases, in ten of whom nocturnal hypoglycaemia had occurred. Although autopsy had been performed in 13 patients, a cause of death was not evident in any case. Again, the authors concluded that hypoglycaemia was the most likely precipitant of death. A similar study from Denmark showed an association between chronic alcohol abuse or acute alcohol intoxication and subjects who were found dead in bed (Borch-Johnsen and Helweg-Larsen, 1993). The number of subjects who were found dead in bed remained remarkably constant over a seven year period, and no association was observed with an increasing usage of human insulin in Denmark during that time (Figures 12.2 and 12.3).
SUDDEN DEATH
271
Figure 12.2 Number of sudden deaths and human insulin as % of total sales of insulin in Denmark from 1982 to 1988. Reproduced from Borch-Jensen and Helweg-Larsen (1993) with permission from John Wiley & Sons, Ltd
Figure 12.3 Number of deaths due to definite (closed bars) and possible (open bars) hypoglycaemia from 1982 to 1988. Reproduced from Borch-Jensen and Helweg-Larsen (1993) with permission from John Wiley & Sons, Ltd
A recent study from Norway reported on long-term mortality in all people with type 1 diabetes who had been diagnosed between 1973 and 1982, and who were less than 15 years of age at diagnosis (Skrivarhaug et al., 2006). Mortality was recorded from diabetes onset until the end of 2002. They reported 103 deaths with 17 sudden or unexpected deaths, and four of these met the criteria for ‘dead in bed’ (patients found dead in an undisturbed bed; observed to be in good health the day before; autopsy not informative). Common causes of death were acute metabolic complications of diabetes, violence and cardiovascular disease, including a 16 year old male with myocardial infarction confirmed by autopsy.
272
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
Importance of Young People Without Typical Cardiac Disease When assessing the risk of sudden death in people with diabetes, the risks have to be compared with people without diabetes. Sudden unexpected death can occur in any young person irrespective of whether they have diabetes or not. The critical issue is whether sudden death occurs more frequently in individuals with type 1 diabetes. Studies that have measured the frequency of sudden death in young people have reported rates around 1.3 to 8.5 per 100 000 patient-years. Since they are based on large numbers of subjects these estimates are probably more accurate than data reporting the risks in patients with type 1 diabetes. However, the figures suggest that the risk of sudden death is considerably greater in those with diabetes. Although it is difficult to be precise, the risk in patients with diabetes seems to be around three times higher.
RISK FACTORS FOR SUDDEN DEATH In the general population, the most frequent cause of sudden death is a cardiac arrhythmia, mostly related to coronary heart disease; it is likely that the same problem occurs in people with diabetes. Nevertheless, if sudden death is occurring more frequently in patients, then additional factors are probably responsible. Some of this increase may be a consequence of the more advanced or premature ischaemic heart disease which is associated with diabetes. In some people the development of hypoglycaemic convulsions may impose an additional insult, although the fact that most subjects were found with their bedclothes undisturbed is against the pre-terminal development of tonic-clonic convulsions. This scenario does not exclude other forms of seizure activity and a striking similarity exists between the syndromes of sudden death in epilepsy and diabetes (Brown et al., 1990; Nashef and Brown, 1996). A study using an implantable ECG recorder in 20 patients with epilepsy identified three patients with potentially fatal asystole, and permanent pacemakers were inserted in four patients (Rugg-Gunn et al., 2004). When considering other factors, the strongest candidates are probably coexisting autonomic neuropathy and hypoglycaemia.
Autonomic Neuropathy Autonomic neuropathy increases the risk of sudden death in patients with diabetes in some (Ewing et al., 1980) but not all (Sampson et al., 1990) studies. The exact cause of death remains uncertain although most groups that reported increased risk of death have suggested that a cardiac arrhythmia is responsible. Some groups have reported lengthened QT intervals in patients with autonomic neuropathy (Ewing and Neilson, 1990) highlighting the association between prolonged QT intervals and the risk of sudden death in other conditions such as the congenital long QT syndrome (Ewing et al., 1991). However, it seems unlikely that autonomic neuropathy alone could account for the greater risk of sudden death in young patients with diabetes. In those who died suddenly, autonomic function had seldom been formally tested. Some subjects had advanced diabetic complications and would probably have had some degree of autonomic neuropathy, but a significant proportion of those
RISK FACTORS FOR SUDDEN DEATH
273
who died had a relatively short duration of diabetes with no evidence of microvascular disease. However, other indicators of autonomic dysfunction, such as reduced heart rate variability and diminished baroceptor sensitivity, are often present in patients with type 1 diabetes even when formal cardiovascular tests of autonomic function are normal (Weston et al., 1996).
Hypoglycaemia All of the investigators who have reported sudden death in patients with type 1 diabetes have implicated hypoglycaemia as a contributing factor. Death had occurred during the night when hypoglycaemia is a common problem, and some had strict glycaemic control and had experienced nocturnal hypoglycaemia in the past (Box 12.4). The crucial question is whether any mechanism exists through which hypoglycaemia could cause sudden death. Hypoglycaemia can cause irreversible brain damage but those who suffer this complication require prolonged exposure to a low blood glucose and are unlikely to die suddenly. Previous authors have emphasised the likelihood of an arrhythmic death and have pointed out that the demonstration of a plausible mechanism by which hypoglycaemia caused cardiac arrhythmias would strongly implicate hypoglycaemia (Tattersall and Gale, 1993). How then does hypoglycaemia affect electrical activity in the heart? The evidence concerning the effects of hypoglycaemia on the electrocardiogram (ECG) has been obtained from different sources. First, there is anecdotal clinical evidence where arrhythmias have been detected during episodes of hypoglycaemia and resolved when blood glucose recovered. Second, there are experimental data where the electrocardiogram has been measured both in diabetic and non-diabetic subjects during controlled hypoglycaemia induced in the laboratory. Hypoglycaemia causes an increase in autonomic neural activity affecting both parasympathetic and sympathetic nerves, an increase in plasma epinephrine, and a fall in potassium. Simple electrocardiographic techniques have demonstrated flattening or inversion of the T wave and some studies have also reported prolongation of the QT interval (Fisher and Frier, 1993). A less consistent effect has been ST segment depression. In the presence of ischaemic heart disease it is not difficult to see how some of these changes might lead to malignant cardiac tachydysrhythmias and sudden death (see below). It is less easy to explain how even
Box 12.4
Possible risk factors for ‘dead in bed syndrome’
• Previous nocturnal hypoglycaemia • Living/sleeping alone • Intensive therapy • Multiple injections of insulin • Alcohol ingestion
274
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
these profound, but brief, physiological changes could precipitate a fatal cardiac event in those whose heart is otherwise healthy. Lengthening of the QT interval is associated with sudden death in other conditions, both congenital and acquired. The congenital long QT syndrome is an inherited disorder in which mutations within the genes coding for the membrane ion channels which contribute to the cardiac action potential cause profound lengthening of the QT interval (Jervell and LangeNeilsen, 1957; Jackman et al., 1988; Curran et al., 1995). Among the mutations that have been reported, one (within the SCN5A gene causing LQT3) leads to impaired sodium channel inactivation whereas others (LQT1-2, LQT5-6) produce abnormalities within potassium channels leading to decreased outward potassium current (Roden et al., 2002). Many of those people affected can be identified by marked QT lengthening on their surface ECG and are at high risk of developing fatal polymorphic ventricular tachycardia (Torsade de Pointe VT). However, members of some families are at risk for VT despite an ECG that appears normal. Triggers of VT (drugs that block K + channels, exercise, sudden noise, sleep) may vary according to the type of mutation. However, although the lifetime risk of death in the congenital long QT syndrome is around 70% without treatment, patients can clearly survive with a long QT interval for many years. At first sight it therefore seems unlikely that a short period of hypoglycaemia lasting for an hour or so could cause a malignant arrhythmia. Yet, under other conditions, shorter periods of altered depolarisation can precipitate ventricular tachycardia. In addition to congenital QT lengthening, certain therapeutic agents (including antiarrhythmic agents, antibiotics, antihistamines) can cause an acquired long QT syndrome and sudden death by blocking movement of K + ions through one of the cardiac ion channels (the rapid component of the delayed rectifier current, IKr coded by the HERG gene) responsible for cardiac repolarization (Roden et al., 2002). It appears that interactions between genetic factors (mutations and polymorphisms in the genes coding for proteins contributing to the cardiac action potential and its physiological regulation) and environmental effects (including age, sex, state of sympathoadrenal activation and K + concentration) can determine whether a cardiac arrhythmia is triggered. One can therefore hypothesise that during clinical episodes of hypoglycaemia, alterations in cardiac repolarisation may be sufficient to precipitate a fatal period of ventricular tachycardia. It has been demonstrated that the QT interval can lengthen during experimental hypoglycaemia both in diabetic and non-diabetic subjects (Marques et al., 1997) (Figure 12.4). Some subjects had quite pronounced increments in QT interval and a strong relationship was observed between the increase in QT interval and the rise in plasma epinephrine. Betablockade and potassium infusion both prevented QT lengthening during hypoglycaemia (Robinson et al., 2003). Epinephrine infusion in normal subjects also caused QT lengthening that was partially prevented by simultaneous potassium infusion (Lee et al., 2003). Similar changes in QTc have subsequently been described during clinical episodes of nocturnal hypoglycaemia in adults and children with type 1 diabetes (Robinson et al., 2004; Murphy et al., 2004). The major problem with the hypothesis is that because sudden death occurs so rarely, it is very difficult to test directly. Isolated case reports of transient cardiac dysrhythmias have not provided much additional useful information. There are reports of atrial fibrillation and supraventricular tachycardia occurring during clinical episodes of hypoglycaemia but there is no direct evidence that profound hypoglycaemia can cause a life threatening disturbance in cardiac rhythm.
RISK FACTORS FOR SUDDEN DEATH
Euglycaemia
275
Hypoglycaemia
Figure 12.4 Typical QT measurement with a screen cursor placement from one subject (a) during euglycaemia showing a clearly defined T wave, and (b) during hypoglycaemia showing prolonged repolarisation and prominent U wave. Horizontal: 700 ms epoch, vertical 1.33 mV full scale. Reproduced from Marques et al. (1997) with permission from John Wiley & Sons, Ltd
Another predisposing factor for an arrhythmia during hypoglycaemia might be an imbalance of autonomic activation during hypoglycaemia in people with diabetes. As mentioned above, hypoglycaemia leads to activation of the sympathetic and parasympathetic nervous system. Recent studies examining heart rate variability have described strong sympathetic cardiac activation with a concomitant increase in parasympathetic tone in normal subjects during insulin-induced hypoglycaemia (Schachinger et al., 2004), although no change in heart rate variability was demonstrated in another similar study (Laitinen et al., 2003). A further study in patients with type 1 diabetes and control subjects demonstrated that more prolonged hypoglycaemia resulted in a reduction of cardiac vagal outflow (Koivikko et al., 2005). Lee et al. (2004) appeared to refute the hypothesis that autonomic neuropathy contributes to hypoglycaemia-induced QT lengthening by demonstrating that those with autonomic neuropathy had the smallest increments in QT interval induced by experimental hypoglycaemia when compared to other diabetic groups. However, since these individuals also had the longest duration of diabetes, they unsurprisingly also experienced the smallest rise in sympathoadrenal activation. In the absence of a study comparing comparable levels of sympathoadrenal stimulation on cardiac electrophysiological responses, the degree to which autonomic neuropathy contributes to this phenomenon remains uncertain. Fatal events may be infrequent because the combination of factors that together lead to a fatal cardiac arrhythmia occur only rarely. Indeed severe hypoglycaemia, which causes a profound sympathoadrenal discharge without alerting the patient or partner, is itself relatively unusual. Thus, the ‘substrate’ for a lethal cardiac arrhythmia might be the combination of a severe hypoglycaemic attack at night (which fails to wake people unless symptoms are intense) and a cardiac conduction system affected by sub-clinical autonomic neuropathy in
276
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
an individual who has inherited polymorphisms in the LQT genes causing exaggerated QT lengthening during sympathoadrenal activation. In summary, the majority of sudden deaths in young patients with diabetes remain unexplained and we have hypothesised that these were deaths due to ventricular arrhythmia. These might have resulted from an increase in plasma epinephrine and a fall in potassium which accompany hypoglycaemia, so producing prolongation of the QT interval. It is possible that this occurs on a background of autonomic instability caused by early autonomic neuropathy.
What Do We Say to Patients? It is clear that the frequency of deaths resulting from hypoglycaemia is underestimated, and some deaths related to hypoglycaemia may be attributed to other causes at the time of certification. Nevertheless, when we consider that nocturnal hypoglycaemia is very common, affecting up to 60% of patients every night, we can reassure patients that sudden death as a consequence of acute hypoglycaemia is very rare. However, the available evidence prevents us stating that there is no risk of death from hypoglycaemia during sleep, or at other vulnerable times when treatment is not rendered promptly.
EFFECT OF HYPOGLYCAEMIA ON CARDIOVASCULAR DISEASE Acute hypoglycaemia provokes an intense haemodynamic response secondary to activation of the autonomic nervous system with the secretion of epinephrine (adrenaline) (DeRosa and Cryer, 2004). The heart rate increases over a period of 15 to 20 minutes, but rarely rises above 100 beats/minute. A modest but significant increase in systolic blood pressure is accompanied by a slight but significant fall in diastolic blood pressure (Fisher et al., 1987; Russell et al., 2001). The pulse pressure widens, with a substantial increase in cardiac output and a fall in total peripheral vascular resistance (Figure 12.5). These haemodynamic changes are relatively short-lived, and exert no significant after-effects on the 24-hour heart rate or blood pressure (Avogaro et al., 1994). In a person with a normal heart these haemodynamic changes are probably of no great significance, but in a patient who has underlying coronary heart disease the profound increase in cardiac workload may provoke a cardiac arrhythmia, myocardial ischaemia and even myocardial infarction (Box 12.5).
Arrhythmias and Coronary Heart Disease Occasional cardiac arrhythmias have been demonstrated in normal subjects during experimental hypoglycaemia studies. It would now be considered unethical to perform hypoglycaemia studies in patients with known heart disease, but many studies were performed in an earlier era both in diabetic and non-diabetic patients with coronary heart disease to examine the effects of acute hypoglycaemia (Fisher and Frier, 1993). Sinus bradycardia has been reported in a very small number of cases (Pollock et al., 1996; Navarro-Gutierrez et al.,
EFFECT OF HYPOGLYCAEMIA ON CARDIOVASCULAR DISEASE
277
Figure 12.5 Mean response of (a) heart rate, (b) systolic blood pressure and mean arterial blood pressure, and (c) left ventricular ejection fraction following intravenous injection of insulin at time 0. (R = autonomic reaction). Reproduced from Frier et al. (1987), with kind permission from Springer Science and Business Media
2003). Atrial fibrillation has been described in some patients and in addition there are several case reports of atrial fibrillation following hypoglycaemia in insulin-treated patients who had no overt evidence of heart disease (Collier et al., 1987; Baxter et al., 1990; Odeh et al., 1990; Navarro-Gutierrez et al., 2003). There is a single report of a transient ventricular tachycardia occurring during experimental hypoglycaemia in a non-diabetic patient with coronary heart disease, and ventricular tachycardia was recently documented in an elderly non-diabetic man who developed hypoglycaemia during emergency surgery (Chelliah, 2000). There have been case
278
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
Box 12.5
Cardiac effects of acute hypoglycaemia
• Increased heart rate • Widening of pulse pressure • Arrhythmias • Silent myocardial ischaemia • Angina • Myocardial infarction
reports, with ECG evidence, of ventricular ectopics, sustained ventricular tachycardia, ventricular fibrillation and asystole during hypoglycaemia in diabetic patients (Shimada et al., 1984; Burke and Kearney, 1999). Obviously this does not exclude the possibility of these arrhythmias occurring more frequently in clinical practice, as these arrhythmias will be fatal if uncorrected. In most instances it is unlikely that any precipitating cause of the arrhythmia would be sought; hypoglycaemia may not have been recognised and we have already alluded to the difficulties in establishing a putative diagnosis of hypoglycaemia at post-mortem.
Angina and Myocardial Ischaemia The provocation of angina and myocardial ischaemia by exercise is well documented in clinical practice. By contrast, acute hypoglycaemia, which provokes a more intense haemodynamic response and in particular a greater increase in plasma epinephrine, has rarely been documented as provoking anginal chest pain, either in the experimental situation or in anecdotal case reports. A literature search of over 6000 insulin tolerance tests recorded only two episodes of angina. This may reflect the fact that coronary heart disease would be considered a contraindication to insulin tolerance testing, and in clinical practice clinicians may accept higher ambient blood glucose concentrations in diabetic patients with known coronary heart disease to avoid hypoglycaemia. It is also possible that the haemodynamic changes of hypoglycaemia are so profound that they are frequently fatal in patients with coronary heart disease, and hypoglycaemia is probably overlooked as a provoking cause when determining cause of death. It is now well established that many episodes of ST segment depression on the ECG are not associated with angina, and constitute ‘silent ischaemia’. One case has been described during 24-hour ECG monitoring of hypoglycaemia that provoked silent ischaemia in a diabetic patient with suspected coronary heart disease (Pladziewicz and Nesto, 1989). More recently, 72-hour continuous glucose monitoring with simultaneous cardiac Holter monitoring was performed in 21 patients with coronary heart disease and insulin-treated type 2 diabetes (Desouza et al., 2003). A total of 26 episodes of symptomatic hypoglycaemia
WORSENING OF MICROVASCULAR COMPLICATIONS
279
were documented, with ten episodes of chest pain, four of which were associated with ECG abnormalities. Twenty eight episodes of asymptomatic hypoglycaemia were documented, with no episodes of chest pain, and two of these were associated with ECG abnormalities. Hypoglycaemia was therefore frequently associated with ECG abnormalities, some of which also were associated with chest pain, in this group of well-controlled patients.
Myocardial Infarction Myocardial infarction has rarely been documented as a consequence of hypoglycaemia (Fisher and Frier, 1993). In a series of non-diabetic patients with schizophrenia who were treated with hypoglycaemic shock therapy in the 1930s, 12 of 90 deaths were ascribed to cardiac causes, with the majority of deaths being caused by cerebral damage. It should be emphasised that this long-abandoned form of treatment of psychiatric disease necessitated prolonged and profound hypoglycaemia. Only a few cases have been published of myocardial infarction and hypoglycaemia in diabetic patients (Purucker et al., 2000; Chang et al., 2007). This possible association is very difficult to establish because of the problems described above. In addition, the release of stress hormones such as glucagon, cortisol and epinephrine will raise blood glucose and make the contribution of preceding hypoglycaemia almost impossible to confirm.
WORSENING OF MICROVASCULAR COMPLICATIONS Precipitation of acute vascular events (such as myocardial infarction or stroke) as a result of hypoglycaemia affecting macrovascular disease is relatively infrequent, considering how commonly episodes of hypoglycaemia occur in everyday life. It has been suggested that acute hypoglycaemia, by releasing vasoactive hormones and provoking changes in regional and capillary blood flow, might worsen established microvascular complications of diabetes (Frier and Hilsted, 1985). Although microvascular complications are recognised to be the consequence of chronic hyperglycaemia and can be prevented or delayed by strict glycaemic control, the effect of recurrent hypoglycaemia on an already compromised microvasculature may be deleterious and cause further damage (Box 12.6). Exposure to recurrent hypoglycaemia may precipitate capillary closure inducing localised tissue ischaemia and producing deterioration in retinopathy. A sudden fall in intraocular pressure occurs during hypoglycaemia, and could precipitate vitreous haemorrhage in patients with proliferative retinopathy, as friable new vessels are vulnerable to sudden changes in perfusion pressure or mechanical stresses. This could explain the occasional anecdotal reports of vitreous haemorrhage described by individual patients, often following nocturnal hypoglycaemia. Acute hypoglycaemia reduces renal plasma flow and glomerular filtration in normal subjects and diabetic patients without significant complications. In patients who have established nephropathy with glomerular sclerosis and arteriolar narrowing, the reduction in renal plasma flow may precipitate further closure of arterioles and progression in renal impairment.
280
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
Box 12.6
Postulated effects of acute hypoglycaemia in the microcirculation
• Changes in capillary blood flow • Increased coagulation factors • Platelet activation • Neutrophil activation • Increased free-radical activity
CONCLUSIONS • Biochemical and mild symptomatic hypoglycaemia occur commonly in the treatment of those with type 1 diabetes. Even severe episodes are not infrequent, but sudden and unexpected deaths from hypoglycaemia are rare. • There does appear to be an increased risk of sudden death in people with diabetes compared to those who do not have diabetes, which may become more numerous as greater attempts are made to control blood glucose more tightly. These deaths, referred to as the ‘dead in bed syndrome’, are probably related to hypoglycaemia through hypoglycaemia-induced tachydysrhythmias. • Experimental hypoglycaemia can provoke abnormalities of the ECG, which are recognised to be associated with sudden death in other conditions, and this observation offers a possible mechanism to explain the phenomenon. • Hypoglycaemia also produces changes in plasma viscosity and capillary perfusion that may increase the risk of myocardial ischaemia, although the clinical evidence that this is responsible for myocardial infarction in people with diabetes is limited. • The intense sympathoadrenal response provoked by severe hypoglycaemia may also provoke changes that could worsen established microvascular complications, although to date this is primarily hypothetical with little supportive evidence.
REFERENCES Arky RA, Veverbrants E, Abramson EA (1968). Irreversible hypoglycemia. A complication of alcohol and insulin. Journal of the American Medical Association 206: 575–8. Avogaro A, Crepaldi C, Miola M, Penzo M, Maran A, Zanin L et al. (1994). Sequelae of acute hypoglycaemia on 24 hour blood pressure and metabolic parameters in normal and type 1 (insulindependent) diabetic individuals. Diabetic Medicine 11: 573–7. Baxter MA, Garewal C, Jordan R, Wright AD, Nattrass M (1990). Hypoglycaemia and atrial fibrillation. Postgraduate Medical Journal 66: 981. Borch-Johnsen K, Helweg-Larsen K (1993). Sudden death and human insulin: is there a link? Diabetic Medicine 10: 255–9.
REFERENCES
281
Brown SW, Mawer GE, Lawler W, Taylor DC, Shorvon S, Betts TA et al. (1990). Sudden death and epilepsy. Lancet 355: 606–7. Burke BJ, Kearney TK (1999). Hypoglycaemia and cardiac arrest. Practical Diabetes International 16: 189–90. Campbell I (1991). Dead in bed syndrome: a new manifestation of nocturnal hypoglycaemia? Diabetic Medicine 8: 3–4. Chang J-H, Tseng C-F, Wang J-Y (2007). Hypoglycemia-induced myocardial infarction: an unusual adverse effect of sulfonylureas. International Journal of Cardiology 115: 414–16. Chelliah YR (2000). Ventricular arrhythmias associated with hypoglycaemia. Anaesthesia and Intensive Care 28: 698–700. Collier A, Matthews DM, Young RJ, Clarke BF (1987). Transient atrial fibrillation precipitated by hypoglycaemia: two case reports. Postgraduate Medical Journal 63: 895–7. Critchley JAJH, Proudfoot AT, Boyd SG, Campbell IW, Brown NS, Gordon A (1984). Deaths and paradoxes after intentional insulin overdosage. British Medical Journal 289: 225. Curran ME, Splawski I, Timothy KW, Vincent GM, Green ED, Keating MT (1995). A molecular basis for cardiac arrhythmias: HERG mutations cause long QT syndrome. Cell 80: 795–803. Dahlquist G, Kallen B (2005). Mortality in childhood-onset type 1 diabetes: a population-based study. Diabetes Care 28: 2384–7. DeRosa MA, Cryer PE (2004). Hypoglycemia and the sympathoadrenal system: neurogenic symptoms are largely the result of sympathetic neural, rather than adrenomedullary, activation. American Journal of Endocrinology 287: E32–41. Desouza C, Salazar H, Cheong B, Murgo J, Fonseca V (2003). Association of hypoglycemia and cardiac ischemia: a study based on continuous monitoring. Diabetes Care 26: 1485–9. Ewing DJ, Campbell IW, Clarke BF (1980). The natural history of diabetic autonomic neuropathy. Quarterly Journal of Medicine 49: 95–108. Ewing DJ, Neilson JM (1990). QT interval length and diabetic autonomic neuropathy. Diabetic Medicine 7: 23–6. Ewing DJ, Boland O, Neilson JM, Cho CG, Clarke BF (1991). Autonomic neuropathy, QT interval lengthening, and unexpected deaths in male diabetic patients. Diabetologia 34: 182–5. Fisher M, Gillen G, Dargie HJ, Inglis GC, Frier BM (1987). The effects of insulin-induced hypoglycaemia on cardiovascular function in normal man: studies using radionuclide ventriculography. Diabetologia 30: 841–5. Fisher M, Frier BM (1993). Effect on vascular disease. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M, eds. Edward Arnold, London: 355–61. Frier BM, Hilsted J (1985). Does hypoglycaemia aggravate the complications of diabetes? Lancet ii: 1175–7. Jackman WM, Friday KJ, Anderson JL, Aliot EM, Clark M, Lazzara R (1988). The long QT syndromes: a critical review, new clinical observations and a unifying hypothesis. Progress in Cardiovascular Diseases 31: 115–72. Jervell A, Lange-Neilsen F (1957). Congenital deaf-mutism, functional heart disease with prolongation of the QT interval and sudden death. American Heart Journal 54: 59–68. Kalimo H, Olsson Y (1980). Effects of severe hypoglycemia on the human brain. Neuropathological case reports. Acta Neurologica Scandinavica 62: 345–56. Koivikko ML, Salmela PI, Juhani Airaksinen KE, Tapanainen JS, Ruokonen A, Makikallio et al. (2005). Effects of sustained insulin-induced hypoglycemia on cardiovascular autonomic regulation in type 1 diabetes. Diabetes 54: 744–50. Laitinen T, Huopio H, Vauhkonen I, Camaro C, Hartikainen J, Laakso M et al. (2003). Effects of euglycaemia and hypoglycaemic hyperinsulinaemia on sympathetic and parasympathetic regulation of haemodynamics in healthy subjects. Clinical Science 105: 315–22.
282
MORTALITY, CARDIOVASCULAR MORBIDITY AND DIABETIC COMPLICATIONS
Lee S, Harris ND, Robinson RT, Yeoh L, Macdonald IA, Heller SR (2003). Effects of adrenaline and potassium on QTc interval and QT dispersion in man. European Journal of Clinical Investigation 33: 93–8. Lee SP, Yeoh L, Harris ND, Davies CM, Robinson RT, Leathard A et al. (2004). Influence of autonomic neuropathy on QTc interval lengthening during hypoglycemia in type 1 diabetes. Diabetes 53: 1535–42. MacCuish AC (1993). Treatment of hypoglycaemia. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M, eds. Edward Arnold, London: 212–21. Malins J (1968). Hypoglycaemia. In: Clinical Diabetes Mellitus. Eyre and Spottiswoode, London: 425–46. Marques JLB, George E, Peacey SR, Harris ND, Macdonald IA, Cochrane T, Heller SR (1997). Altered ventricular repolarization during hypoglycaemia in patients with diabetes. Diabetic Medicine 14: 648–54. Muhlhauser I, Sawicki PT, Blank M, Overmann H, Richter B, Berger M. (2002). Reliability of causes of death in persons with type I diabetes. Diabetologia 45: 1490–7. Murphy NP, Ford-Adams ME, Ong KK, Harris ND, Keane SM, Davies C et al. (2004). Prolonged cardiac repolarisation during spontaneous nocturnal hypoglycaemia in children and adolescents with type 1 diabetes. Diabetologia 47: 1940–7. Nabarro JDN, Mustaffa BE, Morris DV, Walport MJ, Kurtz AB (1979). Insulin deficient diabetes. Contrasts with other endocrine deficiencies. Diabetologia 16: 5–12. Nashef L, Brown S (1996). Epilepsy and sudden death. Lancet 348: 1324–5. Navarro-Gutierrez S, Gonzalez-Martinez F, Fernandez-Perez MT, Garcia-Moreno MT, Ballester-Vidal MR, Pulido-Morillo FJ (2003). Bradycardia related to hypoglycaemia. European Journal of Emergency Medicine 10: 331–3. Odeh M, Oliven A, Bassan H (1990). Transient atrial fibrillation precipitated by hypoglycemia. Annals of Emergency Medicine 19: 565–7. Pladziewicz DS, Nesto RW (1989). Hypoglycemia-induced silent myocardial ischemia. American Journal of Cardiology 63: 1531–2 Pollock G, Brady WJ Jr, Hargarten S, DeSilvey D, Carner CT (1996). Hypoglycemia manifested by sinus bradycardia: a report of three cases. Academic Emergency Medicine 3: 700–7. Pramming S, Thorsteinsson B, Bendtson I, Binder C (1991). Symptomatic hypoglycaemia in 411 type 1 diabetic patients. Diabetic Medicine 8: 217–22. Purucker E, Nguyen H-N, Lammert F, Koch A, Matern S (2000). Central pontine myelinosis and myocardial infarction following severe hypoglycemia. Intensive Care Medicine 26: 1406–7. Robinson RTCE, Harris ND, Ireland RH, Lee S, Newman C, Heller SR (2003). Mechanisms of abnormal repolarization during insulin-induced hypoglycemia. Diabetes 52: 1469–74. Robinson RT, Harris ND, Ireland RH, Macdonald IA, Heller SR (2004). Changes in cardiac repolarization during clinical episodes of nocturnal hypoglycaemia in adults with type 1 diabetes. Diabetologia 47: 312–5. Roden DM, Balser JR, George AL, Jr, Anderson ME (2002). Cardiac ion channels. Annual Review of Physiology 64: 431–75. Rugg-Gunn FJ, Simister RJ, Squirrell M, Holdright DR, Duncan JS (2004). Cardiac arrhythmias in focal epilepsy: a prospective long-term study. Lancet 364: 2212–9. Russell RR, Chyun D, Song S, Sherwin RS, Tamborlane WV, Lee FA et al. (2001). Cardiac responses to insulin-induced hypoglycemia in non-diabetic and intensively treated type 1 diabetic patients. American Journal of Endocrinology 281: E1029–36. Sampson MJ, Wilson S, Karagiannis P, Edmonds M, Watkins PJ (1990). Progression of diabetic autonomic neuropathy over a decade in insulin-dependent diabetics. Quarterly Journal of Medicine 75: 635–46. Sartor G, Dahlquist G (1995). Short-term mortality in childhood onset insulin-dependent diabetes mellitus: a high frequency of unexpected deaths in bed. Diabetic Medicine 12: 607–11.
REFERENCES
283
Schachinger H, Port J, Brody S, Linder L, Wilhelm FH, Huber PR et al. (2004). Increased highfrequency heart rate variability during insulin-induced hypoglycaemia in healthy humans. Clinical Science 106: 583–8. Skrivarhaug T, Bangstad H-J, Stene LC, Sandvik L, Hanssen KF, Joner G (2006). Long-term mortality in a nationwide cohort of childhood-onset type 1 diabetic patients in Norway. Diabetologia 49: 298–305. Shenfield GM, Bhalla IP, Elton RA, Duncan LJP (1980). Fatal coma in diabetes. Diabete et Metabolisme (Paris) 6: 151–5. Shimada R, Nakashima T, Nunoi K, Kohno Y, Takeshita A, Omae T (1984). Arrhythmia during insulin-induced hypoglycemia in a diabetic patient. Archives of Internal Medicine 144: 1068–9. Tattersall RB, Gale EAM (1993). Mortality. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. Frier BM and Fisher M, eds. Edward Arnold, London: 191–8. Tattersall RB, Gill GV (1991). Unexplained deaths of type 1 diabetic patients. Diabetic Medicine 8: 49–58. The DCCT Research Group (1991). Epidemiology of severe hypoglycemia in the Diabetes Control and Complications Trial. American Journal of Medicine 90: 450–9 The Diabetes Control and Complications Trial Research Group (1993). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New England Journal of Medicine 329: 977–86. Thordarson H, Sovik O (1995). Dead in bed syndrome in young diabetic patients in Norway. Diabetic Medicine 12: 782–7. Tunbridge WMG (1981). Factors contributing to deaths of diabetics under fifty years of age. Lancet ii: 569–72. Waernbaum I, Blohme G, Ostman J, Sundkvist G, Eriksson JW, Arnqvist HJ et al. (2006). Excess mortality in incident cases of diabetes mellitus aged 15 to 34 years at diagnosis: a population-based study (DISS) in Sweden. Diabetologia 49: 653–9. Weston PJ, Panerai RB, McCullough A, McNally PG, James MA, Potter JF et al. (1996). Assessment of baroceptor-cardiac reflex sensitivity using time domain analysis in patients with IDDM and the relation to left ventricular mass index. Diabetologia 39: 1385–91.
13 Long-term Effects of Hypoglycaemia on Cognitive Function and the Brain in Diabetes Petros Perros and Ian J. Deary
INTRODUCTION Almost one-third of all diabetic patients treated with insulin experience one or more episodes of severe hypoglycaemia every year (MacLeod et al., 1993; ter Braak et al., 2000; PedersenBjergaard et al. 2004), this being defined as an episode that requires the help of another person to effect recovery. Strict glycaemic control and intensified insulin treatment are associated with a three-fold increase in the probability of developing severe hypoglycaemia (see Chapter 8). In this chapter the effects of diabetes on the brain are reviewed, with an emphasis on the chronic complications of hypoglycaemia. ‘Hypos’ are usually perceived as a temporary and reversible complication of insulin therapy by people with diabetes and their relatives. However, severe and prolonged hypoglycaemia lasting for several hours can cause serious and permanent brain damage and, rarely, can be fatal (Malouf and Brust, 1985; Yoneda and Yamamoto, 2005). Fortunately, such devastating complications are rare. The vast majority of people who experience an episode of severe hypoglycaemia appear to make a full recovery. However, it is possible that repeated exposure to severe hypoglycaemia may have subtle progressive long-term effects on brain function and mental functions of people with type 1 diabetes. Currently, there is evidence for and against the possibility that recurrent episodes of severe, and apparently reversible, hypoglycaemia in adult patients with type 1 diabetes may have a small detrimental effect on mental capacities. Thus, the brain may not be immune from diabetic complications. The concept of ‘diabetic encephalopathy’, that is, a disorder of the brain associated with some aspects of diabetes, is gaining acceptance. Its causes are complex, and may be related to several factors, only one of which is hypoglycaemia (Dejgaard et al., 1991; McCall, 1992; Biessels et al., 1994; Makimattila et al., 2004).
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
286
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
COGNITIVE FUNCTION AND HYPOGLYCAEMIA Hypoglycaemia and Mental Functions in Children and Adolescents There is controversy about whether repeated episodes of severe hypoglycaemia have lasting effects on the thinking skills of children (Ryan et al., 2005). In addition to the severity of hypoglycaemia, the age of the individual is important in determining the potential impact of hypoglycaemia on the brain (Ack et al., 1961). The human brain develops rapidly until the age of five years, and during this critical period any insult can have long-lasting effects. In diabetic children important risk factors for the development of later cognitive impairment are as follows (Ryan, 1988; 1997): • early onset of diabetes; • long duration of diabetes; • poor metabolic control; • severe hypoglycaemia. Children with type 1 diabetes who suffer repeated and severe hypoglycaemia while younger than five years old have lower mental abilities later on in life, and may show more difficult behaviour (Ryan et al., 1984; Rovet et al., 1987; Golden et al., 1989; Hershey et al., 2005; Ryan et al., 2005). The combination of an early onset of diabetes (before five years of age) and recurrent severe hypoglycaemia appears to be associated with reduced attention, psychomotor efficiency, and spatial memory in adolescence (Rovet and Alvarez, 1997; Bjorgaas et al., 1997; Hershey et al., 2005). Adolescents who had developed type 1 diabetes after the age of five years have been shown to have lower verbal IQ than their peers, but this may be related in part to learning-related problems at school and loss of formal education rather than with hypoglycaemia (Fallstrom, 1974; Ryan et al., 2005). Most of the studies mentioned above are cross-sectional, that is, they have tested groups of children, with and without diabetes, and have tried to review the children’s clinical records to estimate the amount of previous hypoglycaemia experienced by each child with diabetes. A more robust type of study is one in which groups of children are followed prospectively. One such study is ongoing in Melbourne, Australia (Northam et al., 1995). Over 100 children with newly diagnosed type 1 diabetes have been compared with a matched control group of non-diabetic children. No differences in mental abilities or in educational attainments were discernible between the two groups. Therefore, when children develop type 1 diabetes they do not begin with any mental decrements when they are compared with their non-diabetic peers. An initial report from this invaluable study indicated that within two years of the development of diabetes the mental abilities of the diabetic children may begin to lag behind their nondiabetic peers (Northam et al., 1998). Six years after diagnosis, diabetic children performed worse than non-diabetic controls across a range of cognitive performance tests, while severe hypoglycaemia was associated with poorer verbal and IQ scores (Northam et al., 2001). However, the roles of hyper- and hypoglycaemia and other possible effects of having diabetes in promoting these changes, such as increased school absence, remain to be elucidated. Another study of 41 children with early onset (< 6 years) type 1 diabetes found no association between severe hypoglycaemia and cognition (Strudwick et al., 2005). Their episodes of severe hypoglycaemia were recorded prospectively and their scores on IQ,
COGNITIVE FUNCTION AND HYPOGLYCAEMIA
287
memory, behaviour problems and depression did not differ from those of 43 non-diabetic peers. Intensive insulin therapy is associated with more severe hypoglycaemia and might be associated with cognitive decrements. An 18-month-longitudinal study of 142 children aged 6 to 15, who were randomised to either intensive therapy or conventional care, found no association between the occurrence or frequency of severe hypoglycaemia and a wide range of cognitive functions (Wysocki et al., 2003). Examination of electroencephalograms (EEGs), the electrical signals that can be detected from living brains (Haumont et al., 1979), and of visual evoked potentials (signals generated in the brain in response to a stimulus) (Seidl et al., 1996), has found that abnormalities are commoner in children with early-onset diabetes who have had recurrent severe hypoglycaemia. The brain’s electrical responses to stimuli are significantly slowed in almost three-quarters of adolescents with type 1 diabetes (Uberall et al., 1996). However, this same study found no differences in the mental ability of children with diabetes when they were compared to non-diabetic controls, and the neurophysiological changes in the diabetic children were not related to age at onset of diabetes, duration of diabetes, quality of metabolic control or the presence of peripheral neuropathy. In children, repeated exposure to severe hypoglycaemia has its most deleterious effects on the front and central regions of the brain’s cerebral hemispheres (Bjorgaas et al., 1996). During controlled, modest hypoglycaemia induced in the laboratory, the EEGs of children with diabetes were more disturbed than those of non-diabetic children (Bjorgaas et al., 1998). There are few brain-imaging studies examining the effects of severe hypoglycaemia on children. One small study using single photon emission tomography imaging found some evidence of mild dominant hemisphere dysfunction in diabetic children with a history of severe hypoglycaemia compared to an age-matched diabetic group with no history of severe hypoglycaemia (Tupola et al., 2004). In summary, there is convincing evidence to suggest that children with type 1 diabetes who have repeated exposure to severe hypoglycaemia, especially when this occurs below the age of five years, will subsequently have lower mental ability levels with evidence of detrimental effects on the physiological activity of their brains.
Evidence for Neuropsychological Deterioration Following Repeated Hypoglycaemia in Adults Adults with type 1 diabetes perform less well on mental ability tests than non-diabetic subjects (Ryan, 1988; Ryan et al., 2005), but the differences are subtle and the underlying causes unclear. This is a complex and difficult area of clinical research with a number of possible causative factors that are hard to tease apart; these include the metabolic disturbances of diabetes and its treatment, and the social and educational impact of chronic illness on intelligence. A few carefully-controlled studies have focused on adult subjects with insulintreated diabetes who have a history of severe recurrent hypoglycaemia. These patients seem to recover mentally and physically after each episode of hypoglycaemia, but when they are tested in the laboratory with standardised mental tests they display subtle chronic impairment of some mental functions. Abnormal neurological symptoms and signs are usually absent. The evidence from the small number of retrospective studies that are available indicates an association between a history of recurrent severe hypoglycaemia and a modest reduction in IQ (Deary, 1993). The main findings from some of the more influential studies can be summarised as follows:
288
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
Figure 13.1 Pre-morbid (tinted bars) and present (white bars) IQ levels for Group A (subjects with type 1 with no history of severe hypoglycaemia) and Group B (subjects with type 1 with at least five episodes of severe hypoglycaemia). Pre-morbid versus present IQ comparison for Group A is non-significant, comparison for Group B is significant at p < 0001. Reproduced from Langan et al. (1991). With kind permission from Springer Science and Business Media
1. Wredling et al. (1990) performed a carefully controlled study in two small groups of patients with type 1 diabetes, with and without histories of recurrent severe hypoglycaemia. They demonstrated impaired performance on a number of mental function tests in the group with a history of severe hypoglycaemia. The study design could not exclude the possibility that the patients with a history of hypoglycaemia had a lower pre-morbid IQ. 2. Langan et al. (1991) conducted a study in 100 patients with type 1 diabetes, using more detailed tests of cognitive functions. Within this sample of people with diabetes, the group of patients with more than five episodes of severe hypoglycaemia displayed a small, but significant, decline in IQ (averaging about six IQ points) compared to the diabetic patients who had experienced no episodes of severe hypoglycaemia (Figure 13.1). Pre-morbid IQ was similar in patients with and without severe hypoglycaemia, thus strengthening the hypothesis that repeated, severe hypoglycaemia was responsible for the lower IQ (Langan et al., 1991; Deary et al., 1993). Taking the 100 diabetic patients as a whole, they had lower IQs than healthy, non-diabetic subjects with similar ages and social and educational backgrounds (Deary et al., 1993). Impaired performance IQ was closely associated with repeated, severe hypoglycaemia. Making decisions and initiating responses appeared to be affected specifically by recurrent severe hypoglycaemia (Deary et al., 1992). Verbal IQ was lower in people with type 1 diabetes compared to healthy control subjects, regardless of their history of hypoglycaemia. This may result from the social impact of the disorder (Deary et al., 1993). 3. The results of Langan et al. (1991) have been confirmed by another team of researchers (Lincoln et al., 1996) using an identical study design. 4. A small group of patients with type 1 diabetes has been described (Gold et al., 1994), in which the individuals have suffered many episodes of severe hypoglycaemia over several years of treatment with insulin, and have subsequently developed severe mental and memory problems and devastating social and psychological deficits, causing premature retirement from employment and disrupting social and family life.
FUNCTIONAL EFFECTS OF HYPOGLYCAEMIA
289
Retrospective studies suggest that recurrent severe hypoglycaemia has a detrimental effect on cognitive functions. By contrast, the limited evidence from prospective studies of intensified insulin therapy, namely the Diabetes Control and Complications Trial (DCCT) (The Diabetes Control and Complications Trial Research Group, 1996) and the Stockholm Diabetes Intervention Study (Reichard and Pihl, 1994), appears to indicate that cognitive function does not deteriorate in patients who suffer recurrent hypoglycaemia, at least in the timescale (less than ten years) of these studies. It cannot be concluded for certain that recurrent severe hypoglycaemia causes significant long-term effects on cognitive function (Deary, 1997). Indeed, there is better evidence to suggest that chronic hyperglycaemia is a more likely cause of cognitive decrements in people with diabetes (Ferguson et al., 2003; Ryan et al., 2005; Ferguson et al., 2005), and a recent meta-analysis of the effects of type 1 diabetes on cognitive performance found that, overall, repeated severe hypoglycaemia was not associated with cognitive decrements (Brands et al., 2005). However, showing no effect of repeated severe hypoglycaemia on cognitive functions is not equivalent to there being no effect on the brain. In one study of diabetic patients with and without histories of severe hypoglycaemia, no cognitive test score difference was evident, but the severe hypoglycaemia group had EEG changes indicative of decreased vigilance (Howorka et al., 2000). The benefits of strict glycaemic control in reducing the microvascular complications of diabetes are undoubted, but there is a price to pay: a substantial increase in the risk of severe hypoglycaemia. Within the timescale of the original study, the DCCT cohort did not suffer a detrimental effect in cognitive function. The results of the long-term follow-up (average 18 years) of 75% of the original cohort (in the Epidemiology of Diabetes Interventions and Complications Study – EDIC), have shown that exposure to episodes of hypoglycaemic coma or seizure had no significant effect on cognitive function (The DCCT/EDIC Research Group, 2007). This is very reassuring, but the DCCT participants were young, highly motivated, of above average intelligence, free of advanced complications with no history of severe hypoglycaemia before entering the study, and they received a very high level of support from health professionals. In most diabetes outpatient clinics where resources are limited, such model patients are not the norm. It seems entirely justifiable to aim for strict glycaemic control for patients who fit the entry criteria used in the DCCT. It is probably also appropriate to extrapolate the lessons of the DCCT to older patients with more advanced diabetic complications and reasonable life expectancy, who have not previously experienced recurrent severe hypoglycaemia. There still remains a sizeable group of patients with type 1 diabetes whose glycaemic control is sub-optimal by the standards of the DCCT, and yet they have suffered recurrent severe hypoglycaemia in the past. The targets of glycaemic control should be set less rigidly for these patients, who are entitled to be informed of the potential risks of further hypoglycaemia on cognitive function.
FUNCTIONAL EFFECTS OF HYPOGLYCAEMIA Hypoglycaemia-induced Neurological Syndromes Hypoglycaemia can cause a wide range of neurological symptoms and clinical signs, which can be subtle or severe, reversible or permanent. The effects of hypoglycaemia on the brain depend on several factors:
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
290
• the blood glucose nadir reached during hypoglycaemia; • the duration of hypoglycaemia; • the frequency of hypoglycaemia; • the presence of previous brain insults (e.g. head injury, chronic alcohol abuse).
Reversible Effects of Hypoglycaemia on the Brain An acute fall in blood glucose causes mental slowness which, if untreated, can proceed to loss of consciousness. Recovery is usually rapid (within 30 to 45 minutes) after the blood glucose concentration returns to normal. Patients, however, often complain of headache, malaise and memory problems for several hours, and although most aspects of intellectual performance recover within a day of the event, altered mood may take much longer to recover (Strachan et al., 2000). In some patients, hypoglycaemia triggers stereotypical responses (Box 13.1). Diagnostic confusion may arise because of an atypical presentation, a post-ictal state, and if it is measured, blood glucose concentration may be either in the normal range or even elevated by the time of arrival at hospital, because of the compensatory counterregulatory response.
Convulsions and Associated Morbidity Focal or generalised convulsions can be precipitated by hypoglycaemia and have been estimated to occur with a frequency of two convulsions per 100 diabetic patients per year in
Box 13.1
Transient neuropsychological manifestations of severe hypoglycaemia
Neurological • Focal or generalised convulsions • Hemiparesis • Focal neurological syndromes Psychosocial • Mental slowness • Inappropriate behaviour • Automatic behaviour • Aggressive behaviour
FUNCTIONAL EFFECTS OF HYPOGLYCAEMIA
291
up to 10% of all patients treated with insulin (MacLeod et al., 1993). There is an obvious risk of injury during convulsions (Hepburn et al., 1989), including: • fracture-dislocation of joints; • vertebral compression fractures (Figure 13.2); • soft tissue injury; • head injury. Idiopathic epilepsy (which occurs in insulin-treated diabetic patients with the same frequency as in the non-diabetic population) may be misdiagnosed and patients may be treated unnecessarily with anticonvulsant drugs, which are thought to be ineffective in
Figure 13.2 Lateral X-ray of thoracic spine demonstrating a vertebral compression fracture sustained during a hypoglycaemia-induced convulsion (courtesy of Professor B.M. Frier)
Figure 13.3 EEG changes during hypoglycaemia in a child with type 1 diabetes who experienced hypoglycaemia-induced convulsions. There is progressive slow activity leading to epileptiform spikes at blood glucose of 2.1 mmol/l. Reproduced from Bjorgaas et al. (1998) with permission from John Wiley & Sons, Ltd
FUNCTIONAL EFFECTS OF HYPOGLYCAEMIA
293
preventing hypoglycaemia-induced convulsions. The distinction between idiopathic epilepsy and hypoglycaemia-induced convulsions can be difficult. The EEG is often unhelpful as changes occur with acute hypoglycaemia (Figure 13.3), and abnormalities can persist for several days following an episode of hypoglycaemia. EEG examination should therefore be deferred for at least a week and blood glucose should be estimated at the time of examination. Permanent EEG abnormalities have been identified in 30–80% of diabetic patients (Haumont et al., 1979; Pramming et al., 1988). Cerebral oedema is a dreaded complication of severe insulin-induced hypoglycaemia and should be suspected if further deterioration or false localising signs ensue (MacCuish, 1993). Hypoglycaemia-associated cerebral oedema is often very resistant to treatment and is usually fatal. Urgent imaging of the brain is imperative to exclude other potentially remediable causes of neurological abnormalities or coma.
Permanent Neurological Effects of Hypoglycaemia on the Brain In rare cases, severe and protracted hypoglycaemia can cause permanent brain damage, but this has often been associated with excessive consumption of alcohol (Arky et al. 1968) (see Chapter 5) and is occasionally the sequel of attempted suicide or unintentional insulin overdose. Some patients survive but remain in a persistent vegetative state (Agardh et al., 1983). Some recover partially with focal neurological deficits such as hemiparesis, ataxia or severe memory loss (Malouf and Brust, 1985; Lins and Adamson, 1993) (Box 13.2). Patients
Box 13.2 Long-term neuropsychological manifestations of severe insulin-induced hypoglycaemia Neurological • Persistent vegetative state • Hemiparesis • Focal abnormalities (motor, sensory) • Brainstem syndrome • Ataxia; choreoathetosis • Epilepsy Psychological • Cognitive impairment • Behavioural abnormalities • Automatism; psychosis • Psychosocial problems
294
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
with neurological complications of severe hypoglycaemia are usually admitted to hospital. Patients who require hospital admission for treatment of severe hypoglycaemia have been observed to have a high incidence of psychiatric disturbance and increased mortality within a few months of discharge (Hart and Frier, 1998).
STRUCTURAL AND FUNCTIONAL CHANGES IN THE CENTRAL NERVOUS SYSTEM Structural Changes of the Brain in Diabetes Hypertension and hyperlipidaemia are common in diabetes and cerebrovascular disease is a recognised macrovascular complication. Atheromatous cerebral artery occlusion involving major vessels, embolism from cervical arteries, and lacunar strokes are more extensive and occur at an earlier age in diabetic patients compared with the non-diabetic population (McCall, 1992; Mankovsky and Ziegler, 2004). It is uncertain whether microvascular disease affects the brain. Following the death of a group of young patients with long-standing type 1 diabetes, meningeal fibrosis, pseudocalcinosis and diffuse degeneration of grey and white matter were observed in their brains (Reske-Nielsen et al., 1965). However, these patients had uraemia and hypertension secondary to renal failure with diabetic nephropathy, and the neuropathological changes could not be attributed to diabetes per se. Despite the vulnerability of retinal vessels to microvascular disease, the cerebral microcirculation appears to be protected from diabetic microangiopathy. However, subtle changes in cerebral capillaries have been described (increased endothelial basal membrane thickness and, infrequently, microaneurysms) using sensitive techniques in specimens from the brains of diabetic subjects (Johnson et al., 1982). The premise that the brain is not susceptible to microvascular diabetic complications is as yet unproven. This is an important consideration because of the hypothesis that the haemodynamic and haemorrheological changes induced by hypoglycaemia may precipitate ischaemia in tissues with established disease of the macro- and microvasculature (Fisher and Frier, 1993) (see Chapter 12). The relatively new analytical technique of voxel-based morphometry allows the density of white and grey matter to be studied in different brain regions. Young adults with diabetes have lower grey matter density in various brain regions, and a history of severe hypoglycaemia was associated with less grey matter density in areas of the brain that support memory and language processing (Musen et al., 2006).
Effect of Hypoglycaemia on Cerebral Blood Flow and Structure Hypoglycaemia promotes a redistribution of regional cerebral blood flow (Tallroth et al., 1992; MacLeod et al., 1994; Kennan et al., 2005) which may encourage localised neuronal ischaemia, particularly if the cerebral macro- or microcirculation is already compromised in subjects with type 1 diabetes. Using techniques such as Single Photon Emission Tomography, the blood flow to the frontal lobes has been shown to be increased during acute hypoglycaemia in non-diabetic subjects (Tallroth et al., 1992). In patients with a history of previous severe hypoglycaemia (MacLeod et al., 1994), and in patients with impaired hypoglycaemia awareness (MacLeod et al., 1996), this altered pattern in
STRUCTURAL AND FUNCTIONAL CHANGES IN CENTRAL NERVOUS SYSTEM
295
Figure 13.4 Schematic representation of regions of interest in the brain in a neuroanatomical template used for transaxial (horizontal) slices at the level of the basal ganglia, examining cerebral blood flow. By using single photon emission tomography (SPET) during acute hypoglycaemia, increased uptake of isotope in the frontal area indicated increased blood flow while in the calcarine area it was reduced compared to euglycaemia, thus demonstrating redistribution of regional blood flow (MacLeod et al., 1994)
regional cerebral blood flow appears to be a permanent sequel (Figure 13.4). This permanent increase in regional cerebral blood flow to the frontal lobes may be an adaptive response to protect an area of the brain that is most vulnerable to the effects of hypoglycaemia. This susceptibility of the frontal areas has been shown by other techniques, including EEG (Pramming et al., 1988), and tests of cognitive function (see Chapter 2). Neuropathological observations have indicated that the brain is susceptible to neuroglycopenia in a rostro-caudal direction with the cerebral cortex and hippocampus being most sensitive and the brainstem and spinal cord being most resistant (Auer et al., 1984) (Figure 13.5). Other imaging techniques of the brain have yielded complementary information about abnormal brain structure in diabetes (Figure 13.6). Studies using CT and MRI scanning have shown a high prevalence of cerebral atrophy in people with diabetes (36–53% compared to 12% in age-matched non-diabetic controls), which occurs earlier in life than in non-diabetic control subjects and tends to be more extensive (Figure 13.7) (Araki et al., 1994). Ventricular enlargement also occurs more frequently in patients with diabetes than in healthy controls (Lunetta et al., 1994). Studies of the brains of people with diabetes using magnetic resonance imaging (MRI) demonstrated a high prevalence (69% in type 1 diabetes versus 12% in healthy non-diabetic subjects) of small periventricular high-intensity lesions known as ‘leukoaraiosis’ (Dejgaard et al., 1991). Leukoaraiosis is an age-related radiological finding that is also associated with hypertension, vascular disease, dementia and demyelination (Pantoni and Garcia, 1996). In a recent study using MRI, small subcortical white matter lesions were present in about a third of diabetic patients (Ferguson et al., 2003).
296
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
Figure 13.5 Diagram indicating the sensitivity of regions of the brain to acute neuroglycopenia. The cortex and hippocampus are most vulnerable and the brainstem and spinal cord are most resistant
Pathologically, leukoaraiosis has non-specific features consisting of areas of gliosis, loss of myelin sheaths and increased water content (Awad et al., 1986). The significance of leukoaraiosis in diabetes is unknown, but may represent localised ischaemia (Brands et al., 2004). In one study it was associated with advanced microvascular diabetic complications (Dejgaard et al., 1991) (Box 13.3). Recently, a high incidence of cerebral atrophy (33%), cerebellar atrophy (11%) and leukoaraiosis (56%) was observed in diabetic patients with the 3243 mitochondrial tRNA mutation (Suzuki et al., 1996). Some abnormal patterns of the appearance of MRI scans of the brain are shown schematically in Figure 13.8.
Structural Changes Associated with Hypoglycaemia (Box 13.4) Human subjects who have succumbed to severe hypoglycaemia have been studied at postmortem, and are shown to have areas of cortical necrosis, particularly in the frontal lobes and hippocampus, with relative sparing of the hindbrain (Auer et al., 1984). Cortical and hippocampal atrophy and ventricular enlargement have been described in long-term survivors of severe hypoglycaemia (McCall, 1992). The neurohistological features, however, are non-specific and are similar to those of anoxic brain damage. Human studies are further confounded by the fact that many subjects have suffered secondary brain damage as a result of cardiorespiratory collapse (Patrick and Campbell, 1990). In hypoglycaemic brain damage there is selective neuronal acidophilia with shrinkage of the cells which have a bright red
STRUCTURAL AND FUNCTIONAL CHANGES IN CENTRAL NERVOUS SYSTEM
297
Figure 13.6 Common asymptomatic neurological abnormalities observed with MRI in patients with type 1 diabetes: (a) cortical atrophy; (b) ventricular dilatation; (c) leukoaraiosis
cytoplasm (Figure 13.9). These cannot be differentiated from ischaemic neurones, but the pattern of neuronal injury characterises hypoglycaemic damage with cells in specific layers of the cortex being destroyed. A few case reports have described abnormalities of brain structure detected by CT scanning or MRI, associated with focal neurological deficit following one or more episodes of severe hypoglycaemia. Marked global cerebral atrophy has been described in a young patient with type 1 diabetes within a few months of a severe episode of hypoglycaemia that
298
Box 13.3
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
Structural abnormalities of the brain associated with diabetes
Gross pathology Severe cerebral atheroma Cerebral infarction Lacunar strokes
Histological abnormalities∗ Meningeal fibrosis Pseudocalcinosis Diffuse degeneration of grey and white matter Increased endothelial basement membrane thickness Microaneurysms
Abnormal imaging (Figure 13.6) Cortical atrophy Ventricular dilatation Leukoaraiosis ∗
Some changes were observed in patients who died with coexisting uraemia and hypertension – changes may not be specific to diabetes.
Figure 13.7 The prevalence of brain atrophy with increasing age as demonstrated by MRI. This is more common in diabetic subjects (type 1 and type 2) at an earlier age. Reproduced from Araki et al. (1994), with permission from Springer Science and Business Media
STRUCTURAL AND FUNCTIONAL CHANGES IN CENTRAL NERVOUS SYSTEM
299
Box 13.4 Structural abnormalities of the brain associated with profound hypoglycaemia Lethal hypoglycaemia Cortical necrosis Hippocampal necrosis Survivors of severe hypoglycaemia with gross neurological deficit Cortical atrophy Hippocampal atrophy Ventricular dilatation Patients with severe recurrent hypoglycaemia and no neurological signs Cortical atrophy
Figure 13.8 Diagrammatical representations of the patterns of abnormal appearance observed in MRI scans of brains in subjects with type 1 diabetes
300
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
Figure 13.9 Histopathological appearance of neurones in layer 2 of the parietal cortex destroyed by exposure to severe hypoglycaemia in a fatal case of a patient with type 1 diabetes, showing pronounced shrinkage of neurones which appeared acidophilic and were stained bright red (not demonstrable in black and white print). Photograph by courtesy of Dr G.A. Lammie
was associated with severe neurological deficit and cortical blindness (Gold and Marshall, 1996). Following severe hypoglycaemia, lesions have been located in the hippocampus in diabetic patients with severe amnesia (Chalmers et al., 1991; Boeve et al., 1995). A lesion with similar appearance on MRI (Figure 13.10) was found in the pons of a patient with persistent ataxia and hemiparesis after an episode of severe hypoglycaemia (Perros et al., 1994). Using fluid-attenuated inversion recovery (FLAIR) sequences on MRI or diffusion weighted images (DWI), more and earlier structural abnormalities can be seen in patients who have suffered severe hypoglycaemia (Finelli, 2001). Some of these changes disappeared after 14 days, coinciding with an improvement in the patient’s condition (Maekawa et al., 2005). The neuropathology of mild cognitive impairment (in the absence of abnormal neurological signs) associated with recurrent severe hypoglycaemia is unknown, but may be either a milder form of structural neuronal damage, similar to that described in lethal cases, or a functional (metabolic) defect. In support of the former hypothesis is a study using brain MRI, in which a group of 11 diabetic patients with a history of severe recurrent hypoglycaemia had a high prevalence of cortical atrophy (45%) compared to none in a matched diabetic control group (Perros et al., 1997). A subsequent larger study (Ferguson et al., 2003) found a strong association between leukoaraiosis and retinopathy, but not with hypoglycaemia, suggesting that leukoaraiosis represents a microvascular complication of hyperglycaemia. A meta-analysis of several studies on the relationship between type 1 diabetes and cognitive impairment confirmed that such an association exists and is associated with microvascular complications (Brands et al., 2005). Therefore, it appears that subtle structural changes (leukoaraiosis) frequently seen on brain imaging of diabetic subjects are more likely to be related to microvascular complications of poor control rather than hypoglycaemia. This
STRUCTURAL AND FUNCTIONAL CHANGES IN CENTRAL NERVOUS SYSTEM
301
Figure 13.10 MRI scan showing an irregular area of high signal intensity in the left pons in a patient with type 1 diabetes who suffered permanent ataxia and hemiparesis following a single episode of severe hypoglycaemia. Reproduced from Perros et al. (1994) with permission from The American Diabetes Association
premise is supported by the results of a recent functional MRI study of people with type 1 diabetes with proliferative diabetic retinopathy (Wessels et al., 2006).
Mechanisms of Hypoglycaemia-induced Brain Injury The principal mechanism by which hypoglycaemia leads to its acute neuropsychological manifestations is thought to be the direct effect of lack of glucose on neurones, causing energy failure. Cerebral glycogen stores (albeit limited) may be important in curtailing the effects of hypoglycaemia, though the importance of this glucose source in human subjects is unknown (Gruetter et al., 2003; McCall, 2004). Additional alterations in the cerebral circulation induced by hypoglycaemia may cause transient and localised ischaemia, provoking focal neurological abnormalities such as hemiparesis. Less is known about the pathogenesis of permanent neurological damage following severe prolonged hypoglycaemia. In animal
302
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
models, activation of postsynaptic neurocytotoxin receptors by neurotransmitters (glutamate and N-acetyl aspartate) released from presynaptic neurones as a result of hypoglycaemia, appear to be an important cause of neuronal death (Cotman and Iversen, 1987; Choi, 1990; Auer, 2004). Increased influx of calcium, which may be linked to stimulation of neurocytotoxin receptors, is also toxic and can cause cell death (Siesjo and Bengsston, 1989). These mechanisms may explain the selective nature of hypoglycaemia-induced neuronal damage which spares glial and vascular tissue in the brain.
Evidence for Diabetic Encephalopathy Considerable evidence indicates an association between neuropsychological dysfunction and diabetes. The nature of this association is unclear but four main contributing factors have been identified: • poor glycaemic control; • cerebrovascular disease; • hypoglycaemia; • the psychosocial impact of diabetes per se. Hypoglycaemia is of particular importance because it is potentially avoidable, and the subtle cumulative effects on cognitive function may not be noticed until its severity compromises the social and psychological functioning of the affected individual. The misplaced enthusiasm with which some health professionals (and patients) pursue and implement strict glycaemic control when this may not be prudent or appropriate (such as in people with impaired awareness of hypoglycaemia), may place some people at risk of developing diabetic encephalopathy. In a clinical context, severe hypoglycaemia is encountered in three broad categories of patients: • patients with type 1 diabetes who have strict glycaemic control with no or minimal microvascular complications; • patients with long duration of type 1 diabetes, moderate or poor glycaemic control (often due to inadequate diabetes self-management, erratic lifestyle, inappropriate insulin dose or regimen, coexistent social and psychological problems), associated with advanced microvascular complications; • patients who have suffered a single devastating episode of hypoglycaemia as a result of deliberate or accidental overdose of insulin or sulphonylurea. Whereas the evidence so far suggests that younger patients in the first category (resembling the highly selected population of patients with type 1 diabetes studied in the DCCT) may not be susceptible to cumulative cognitive deterioration (Reichard and Pihl, 1994; The Diabetes Control and Complications Trial Research Group, 1996; 1997), in clinical practice a sizeable proportion of patients belongs to the second category. They have elevated glycated haemoglobin concentrations and established microvascular complications. It has been suggested that hypoglycaemia can aggravate established micro- and macrovascular
REFERENCES
303
disease (Fisher and Frier, 1993) and potentiate the risk of hypoglycaemia-induced damage to the brain. The evidence from retrospective studies suggests that chronic deterioration in cognitive function may be a real risk should the conclusions of the DCCT be applied indiscriminately to these patients (Deary and Frier, 1996). The effects of diabetes on the brain have been reviewed by Ryan (2006), who suggests that there is little evidence to support a classical ‘diabetic encephalopathy’. Although cognitive dysfunction does exist, in most people with type 1 diabetes the changes are subtle and represented principally as mental slowing, similar to that observed with ageing. This may be a manifestation of chronic hyperglycaemia, and not recurrent exposure to severe hypoglycaemia.
CONCLUSIONS • It is vital that every effort is made to avoid exposure to severe hypoglycaemia in very young children with type 1 diabetes. • The targets for glycaemic control should be set flexibly and individually for patients with a history of recurrent severe hypoglycaemia. • The brain, like the retina, kidney and peripheral nervous system, can be regarded as a target organ in diabetes. • Hypoglycaemia should be considered as a possible diagnosis in all diabetic patients presenting with any neurological syndrome. • Hypoglycaemia should be considered in insulin-treated diabetic patients who present with a convulsion. Cerebral oedema should be sought if a patient does not quickly recover consciousness after treatment. • A wide range of relatively minor and non-specific abnormalities on brain imaging that resemble the changes of normal ageing, are common in diabetic patients. In otherwise asymptomatic patients these abnormalities do not necessarily warrant further investigation. • The pathogenesis of diabetic encephalopathy is as yet unknown, but hypoglycaemia probably plays a significant contributory role. • Research in this area and the application of new imaging techniques of the brain are likely to shed further light on this important complication of diabetes.
REFERENCES Ack M, Miller I, Weil WB (1961). Intelligence of children with diabetes mellitus. Pediatrics 28: 764–70. Agardh C-D, Rosen I, Ryding E (1983). Persistent vegetative state with high cerebral blood flow following profound hypoglycemia. Annals of Neurology 14: 482–6. Araki Y, Nomura M, Tanaka H, Yamamoto T, Tsukaguchi I, Nakamura H (1994). MRI of the brain in diabetes mellitus. Neuroradiology 36: 101–3. Arky RA, Veverbrants E, Abramson EA (1968). Irreversible hypoglycemia. A complication of alcohol and insulin. Journal of the American Medical Association 206: 575–8.
304
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
Auer RN, Wieloch T, Olsson Y, Siesjo BK (1984). The distribution of hypoglycemic brain damage. Acta Neuropathologica (Berlin) 64: 177–91. Auer RN (2004). Hypoglycemic brain damage. Metabolic Brain Disease 19: 169–75. Awad IA, Johnson PC, Spetzler RF, Hodak JA (1986). Incidental subcortical lesions identified on magnetic resonance imaging in the elderly, II: postmortem pathological correlations. Stroke 17: 1090–7. Biessels GJ, Kappelle AC, Bravenboer B, Erkelens DW, Gispen WH (1994). Cerebral function in diabetes mellitus. Diabetologia 37: 643–50. Bjorgaas M, Sand T, Gimse R (1996). Quantitative EEG in type 1 diabetic children with and without episodes of severe hypoglycemia: a controlled, blind study. Acta Neurologica Scandinavica 93: 398–402. Bjorgaas M, Gimse R, Vik T, Sand T (1997). Cognitive function in type 1 diabetic children with and without episodes of severe hypoglycaemia. Acta Paediatrica 86: 148–53. Bjorgaas M, Sand T, Vik T, Jorde R (1998). Quantitative EEG during controlled hypoglycaemia in diabetic and non-diabetic children. Diabetic Medicine 15: 30–7. Boeve BF, Bell DG, Noseworthy JH (1995). Bilateral temporal lobe MRI changes in uncomplicated hypoglycemic coma. Canadian Journal of Neurological Science 22: 56–8. Brands AM, Kessels RP, de Haan EH, Kappelle LJ, Biessels GJ (2004). Cerebral dysfunction in type 1 diabetes: effects of insulin, vascular risk factors and blood-glucose levels. European Journal of Pharmacology 490: 159–68. Brands AM, Biessels GJ, de Haan EH, Kappelle LJ, Kessels RP (2005). The effects of type 1 diabetes on cognitive performance: a meta-analysis. Diabetes Care 28: 726–35. Chalmers JC, Risk MTA, Kean DM, Grant R, Ashworth B, Campbell IW (1991). Severe amnesia after hypoglycemia. Clinical, psychometric, and magnetic resonance imaging correlations. Diabetes Care 14: 922–5. Choi DW (1990). Methods for antagonizing glutamate neurotoxicity. Cerebrovascular Brain Metabolism Reviews 2: 105–47. Cotman CW, Iversen LL (1987). Excitatory amino acids in the brain-focus on NMDA receptors. Trends in Neuroscience 10: 263–5. Deary IJ, Langan SJ, Graham KS, Hepburn D, Frier BM (1992). Recurrent severe hypoglycemia, intelligence, and speed of information processing. Intelligence 16: 337–59. Deary IJ (1993). Neuropsychological manifestations. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. BM Frier and M Fisher, eds. Edward Arnold, London: 337–46. Deary IJ, Crawford JR, Hepburn DA, Langan SJ, Blackmore LM, Frier BM (1993). Severe hypoglycemia and intelligence in adult patients with insulin-treated diabetes. Diabetes 42: 341–4. Deary IJ, Frier BM (1996). Severe hypoglycaemia and cognitive impairment in diabetes. British Medical Journal 313: 767–9. Deary IJ (1997). Hypoglycemia-induced cognitive decrements in adults with type 1 diabetes: a case to answer? Diabetes Spectrum 10: 42–7. Dejgaard A, Gade A, Larsson H, Balle V, Parving A, Parving H-H (1991). Evidence for diabetic encephalopathy. Diabetic Medicine 8: 162–7. Fallstrom K (1974). On the personality structure in diabetic school children aged 7–15 years. Acta Paediatrica Scandinavica 251 (suppl 1): 1–70. Ferguson SC, Blane A, Perros P, McCrimmon RJ, Best JJ, Wardlaw J et al. (2003). Cognitive ability and brain structure in type 1 diabetes: relation to microangiopathy and preceding severe hypoglycemia. Diabetes 52: 149–56. Ferguson SC, Blane A, Wardlaw J, Frier BM, Perros P, McCrimmon RJ, Deary IJ (2005). Influence of an early-onset age of type 1 diabetes on cerebral structure and cognitive function. Diabetes Care 28: 1431–7. Finelli PF (2001). Diffusion-weighted MR in hypoglycemic coma. Neurology 57: 933–5.
REFERENCES
305
Fisher M, Frier BM (1993). Effect on vascular disease. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. BM Frier and M Fisher, eds. Edward Arnold, London, 355–61. Gold AE, Deary IJ, Jones RW, O’Hare JP, Reckless JPD, Frier BM (1994). Severe deterioration in cognitive function and personality in five patients with long-standing diabetes: a complication of diabetes or a consequence of treatment? Diabetic Medicine 11: 499–505. Gold AE, Marshall SM (1996). Cortical blindness and cerebral infarction associated with severe hypoglycemia. Diabetes Care 19: 1001–3. Golden MP, Ingersol GM, Brack CJ, Russell BA, Wright JC, Huberty TJ (1989). Longitudinal relationship of asymptomatic hypoglycemia to cognitive function in IDDM. Diabetes Care 12: 89–93. Gruetter R, Adriany G, Choi IY, Henry PG, Lei H, Oz G (2003). Localized in vivo 13C NMR spectroscopy of the brain. NMR in Biomedicine 16: 313–38. Hart SP, Frier BM (1998). Causes, management and morbidity of acute hypoglycaemia in adults requiring hospital admission. Quarterly Journal of Medicine 91: 505–10. Haumont D, Dorcy H, Pelc S (1979). EEG abnormalities in diabetic children. Influence of hypoglycemia in children and adolescents with IDDM. Clinical Pediatrics 18: 750–3. Hepburn DA, Steel JM, Frier BM (1989). Hypoglycemic convulsions cause serious musculoskeletal injuries in patients with IDDM. Diabetes Care 12: 32–4. Hershey T, Perantie DC, Warren SL, Zimmerman EC, Sadler M, White NH (2005). Frequency and timing of severe hypoglycemia affects spatial memory in children with type 1 diabetes. Diabetes Care 28: 2372–7. Howorka K, Pumprla J, Saletu B, Anderer P, Krieger M, Schabmann A. (2000). Decrease of vigilance assessed by EEG-mapping in type 1 diabetic patients with history of recurrent severe hypoglycaemia. Psychoneuroendocrinology 25: 85–105. Johnson PC, Brendel K, Meezan E (1982). Thickened cerebral cortical capillary basement membranes in diabetics. Archives of Pathological Laboratory Medicine 106: 214–7. Kennan RP, Takahashi K, Pan C, Shamoon H, Pan JW (2005). Human cerebral blood flow and metabolism in acute insulin-induced hypoglycemia. Journal of Cerebral Blood Flow and Metabolism 25: 527–34. Langan SJ, Deary IJ, Hepburn DA, Frier BM (1991). Cumulative cognitive impairment following recurrent severe hypoglycaemia in adult patients with insulin-treated diabetes mellitus. Diabetologia 34: 337–44. Lincoln NB, Faleiro RM, Kelly C, Kirk BA, Jeffcoate WJ (1996). Effect of long-term glycemic control on cognitive function. Diabetes Care 19: 656–8. Lins PE, Adamson U (1993). Neurological manifestations of hypoglycaemia. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. M Fisher and BM Frier, eds. Edward Arnold, London, 347–54. Lunetta M, Damanti AR, Fabbri G, Lombardo M, Di Mauro M, Mughini L (1994). Evidence by magnetic resonance imaging of cerebral alterations of atrophy type in young insulin-dependent diabetic patients. Journal of Endocrinological Investigation 17: 241–5. MacCuish AC (1993). Treatment of hypoglycaemia. In: Hypoglycaemia and Diabetes: Clinical and Physiological Aspects. BM Frier and M Fisher, eds. Edward Arnold, London, 212–21. MacLeod KM, Hepburn DA, Frier BM (1993). Frequency and morbidity of severe hypoglycaemia in insulin-treated diabetic patients. Diabetic Medicine 10: 238–45. MacLeod KM, Hepburn DA, Deary IJ, Goodwin GM, Ebmeier KP, Frier BM (1994). Regional cerebral blood flow in IDDM patients: effects of diabetes and of recurrent severe hypoglycaemia. Diabetologia 37: 257–63. MacLeod KM, Gold AE, Ebmeier KP, Hepburn DA, Deary IJ, Goodwin GM, Frier BM (1996). The effects of hypoglycemia on relative cerebral blood flow distribution in patients with type 1 (insulin-dependent) diabetes and impaired hypoglycemia awareness. Metabolism 45: 974–80. Maekawa S, Aibiki M, Kikuchi K, Kikuchi S, Umakoshi K (2005). Time related changes in reversible MRI findings after prolonged hypoglycemia. Clinical Neurology and Neurosurgery 108: 511–13.
306
LONG-TERM EFFECTS OF HYPOGLYCAEMIA
Makimattila S, Malmberg-Ceder K, Hakkinen AM, Vuori K, Salonen O, Summanen P et al. (2004). Brain metabolic alterations in patients with type 1 diabetes-hyperglycemia-induced injury. Journal of Cerebral Blood Flow and Metabolism 24: 1393–9. Malouf R, Brust JC (1985). Hypoglycemia: causes, neurological manifestations, and outcome. Annals of Neurology 17: 421–30. Mankovsky BN, Ziegler D (2004). Stroke in patients with diabetes mellitus. Diabetes Metabolism Research and Reviews 20: 268–87. McCall AL (1992). The impact of diabetes on the CNS. Diabetes 41: 557–70. McCall AL (2004). Cerebral glucose metabolism in diabetes mellitus. European Journal of Pharmacology 1490: 147–58. Musen G, Lyoo IK, Sparks CR, Weinger K, Hwang J, Ryan CM et al. (2006). Effects of type 1 diabetes on gray matter density as measured by voxel-based morphometry. Diabetes 55: 326–333. Northam E, Anderson P, Werther G, Adler R, Andrewes D (1995). Neuropsychological complications of insulin-dependent diabetes in children. Child Neuropsychology 1: 74–87. Northam EA, Anderson PJ, Werther GA, Warne GL, Adler RG, Andrewes D (1998). Neuropsychological complications of IDDM in children 2 years after disease onset. Diabetes Care 21: 379–84. Northam EA, Anderson PJ, Jacobs R, Hughes M, Warne GL, Werther GA (2001). Neuropsychological profiles of children with type 1 diabetes 6 years after disease onset. Diabetes Care 24: 1541–6. Pantoni L, Garcia JH (1996). The significance of cerebral white matter abnormalities 100 years after Binswanger’s report. Stroke 26: 1293–301. Patrick AW, Campbell IW (1990). Fatal hypoglycaemia in insulin-treated diabetes mellitus: clinical features and neuropathological changes. Diabetic Medicine 7: 349–54. Pedersen-Bjergaard U, Pramming S, Heller SR, Wallace TM, Rasmussen AK, Jorgensen HV et al. (2004). Severe hypoglycaemia in 1076 adult patients with type 1 diabetes: influence of risk markers and selection. Diabetes Metabolism Research and Reviews 20: 479–86. Perros P, Sellar RJ, Frier BM (1994). Chronic pontine dysfunction following insulin-induced hypoglycemia in an IDDM patient. Diabetes Care 17: 725–7. Perros P, Deary IJ, Sellar RJ, Best JJK, Frier BM (1997). Brain abnormalities demonstrated by magnetic resonance imaging in adult IDDM patients with and without a history of recurrent severe hypoglycemia. Diabetes Care 20: 1013–8. Pramming S, Thorsteinsson B, Stigsby B, Binder C (1988). Glycaemic threshold for changes in electroencephalograms during hypoglycaemia in patients with insulin dependent diabetes. British Medical Journal 296: 665–7. Reichard P, Pihl M (1994). Mortality and treatment side-effects during long-term intensified conventional insulin treatment in the Stockholm Diabetes Intervention Study. Diabetes 43: 313–7. Reske-Nielsen E, Lundbaek K, Rafaelson OJ (1965). Pathological changes in the central and peripheral nervous system of young long-term diabetics. Diabetologia 1: 233–41. Rovet JF, Ehrlich RM, Hoppe M (1987). Intellectual deficits associated with early onset of insulindependent diabetes mellitus in children. Diabetes Care 10: 510–5. Rovet J, Alvarez M (1997). Attentional functioning in children and adolescents with IDDM. Diabetes Care 20: 803–10. Ryan C, Vega A, Longstreet C, Drash A (1984). Neuropsychological changes in adolescents with insulin-dependent diabetes. Journal of Consulting Clinical Psychology 52: 335–42. Ryan CM (1988). Neuropsychological complications of type 1 diabetes. Diabetes Care 11: 86–93. Ryan CM (1997). Effects of diabetes mellitus on neuropsychological functioning: a lifespan perspective. Seminars in Clinical Neuropsychiatry 2: 4–14. Ryan CM, Gurtunca N, Becker D. (2005). Hypoglycemia: a complication of diabetes therapy in children. Pediatric Clinics of North America 52: 1705–33. Ryan CM (2006). Diabetes and brain damage: more (or less) than meets the eye? Diabetologia 49: 2229–33.
REFERENCES
307
Seidl R, Birnbacher R, Hauser E, Bernert G, Freilinger M, Schober E (1996). Brainstem auditory evoked potentials and visually evoked potentials in young patients with IDDM. Diabetes Care 19: 1220–4. Siesjo BK, Bengsston F (1989). Calcium fluxes, calcium antagonists, and calcium related pathology in brain ischemia, hypoglycemia and spreading depression: a unifying hypothesis. Journal of Cerebral Blood Flow and Metabolism 9: 127–40. Strachan MWJ, Deary IJ, Ewing FME, Frier BM (2000). Recovery of cognitive function and mood after severe hypoglycemia in adults with insulin-treated diabetes. Diabetes Care 23: 305–12. Strudwick SK, Carne C, Gardner J, Foster JK, Davis EA, Jones TW. (2005). Cognitive functioning in children with early onset type 1 diabetes and severe hypoglycemia. Journal of Pediatrics 147: 680–5. Suzuki Y, Hata, T, Miyaoka H, Atsumi Y, Kadowaki H, Taniyama M et al. (1996). Diabetes with the 3243 mitochondrial tRNA Leu(UUR) mutation. Diabetes Care 19: 739–84. Tallroth G, Ryding E, Agardh C-D (1992). Regional cerebral blood flow in normal man during insulin-induced hypoglycemia and in the recovery period following glucose infusion. Metabolism 41: 717–21. The Diabetes Control and Complications Trial Research Group (1996). Effects of intensive diabetes therapy on neuropsychological function in adults in the Diabetes Control and Complications Trial. Annals of Internal Medicine 124: 379–88. The Diabetes Control and Complications Trial Research Group (1997). Hypoglycemia in the Diabetes Control and Complications Trial. Diabetes 46: 271–86. The Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study Research Group (2007). Long-term effect of diabetes and its treatment on cognitive function. New England Journal of Medicine 356: 1842–52. ter Braak EWMT, Appelman AMMF, van de Laak MF, Stolk RP, van Haeften TW, Erkelens DW (2000). Clinical characteristics of type 1 diabetic patients with and without severe hypoglycemia. Diabetes Care 23: 1467–71. Tupola S, Salonen I, Hannonen R, Verho S, Saar P, Riikonen R. (2004). Comparison of regional cerebral perfusion, EEG and cognitive functions in type 1 diabetic children with and without severe hypoglycaemia. European Journal of Paediatrics 163: 335–6. Uberall MA, Renner C, Edl S, Parzinger E, Wenzel D (1996). VEP and ERP abnormalities in children and adolescents with prepubertal onset of insulin-dependent diabetes mellitus. Neuropediatrics 27: 88–93. Wessels AM, Rombouts SARB, Simsek S, Kuijer JPA, Kostense PJ, Barkhof F et al. (2006). Microvascular disease in type 1 diabetes alters brain activation. A functional Magnetic Resonance Imaging study. Diabetes 55: 334–40. Wredling R, Levander S, Adamson U, Lins P-E (1990). Permanent neuropsychological impairment after recurrent episodes of severe hypoglycaemia in man. Diabetologia 33: 152–7. Wysocki T, Taylor A, Harris MA, Jackson SC, Mauras N, White NH, Fox L. (2003). Absence of adverse effects of severe hypoglycaemia on cognitive function in school-aged children with diabetes over 18 months. Diabetes Care 26: 1100–5. Yoneda Y, Yamamoto S (2005). Cerebral cortical laminar necrosis on diffusion-weighted MRI in hypoglycaemic encephalopathy. Diabetic Medicine 22: 1098–100.
14 Living with Hypoglycaemia Brian M. Frier
INTRODUCTION Hypoglycaemia is recognised to be the single major limiting factor in achieving and maintaining good glycaemic control in people with insulin-treated diabetes. Because hypoglycaemia can occur at any time of day or night, is often unpredictable, affects intellectual and physical performance and disrupts the life of the affected individual and others, its effects can impinge on every aspect of everyday living. Irrespective of the causes and risk factors for hypoglycaemia, the effects on the affected individual are generally unpleasant, frightening and can have wide ramifications which include psychological sequelae (Box 14.1). Adverse experiences of severe hypoglycaemia can influence the subsequent behaviour of an individual as he or she attempts to avoid further events, and the effect on a patient’s self-care of diabetes may encourage poor glycaemic control.
Box 14.1
Psychological consequences of hypoglycaemia Short-term
Long-term
• Anxiety
• Stress
• Transient cognitive dysfunction
• Avoidance behaviour
• Aversion
• Obsessive self-monitoring
• Depersonalisation
• Relationship conflicts
• Loss of control
• Guilt, frustration
• Guilt, frustration
• Work/school problems
• Embarrassment
• Social isolation
• Dependence on others
• ? Permanent cognitive dysfunction
• Accidents
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
Edited by B.M. Frier and M. Fisher
LIVING WITH HYPOGLYCAEMIA
310
PSYCHO-SOCIAL EFFECTS Fear of Hypoglycaemia In addition to the subjective experience of symptoms and physical changes induced by acute hypoglycaemia, the effects on cognitive and non-cognitive functions may be very disabling, leading to loss of control of events and reliance on others for assistance. Some reasons why hypoglycaemia is feared are listed in Box 14.2. The emotional consequences of living with the ever-present risk of hypoglycaemia can affect the personal lives of both the affected individual and other members of the family. It is not surprising that most individuals with recurrent exposure to severe hypoglycaemia develop an aversion to it. Many rate their fear of severe hypoglycaemia as equivalent to their concern about developing serious long-term complications of diabetes (Pramming et al., 1991) (Figure 14.1). In a group of 60 people with type 1 diabetes, 11 (20%) described severe hypoglycaemia as being the most frightening event in their lives (Sanders et al., 1975) and many associated this with feelings of insecurity, tension and depression. Many people with insulin-treated diabetes who have experienced frequent severe hypoglycaemia suffer higher levels of psychological distress, including increased anxiety, depression and fear of future hypoglycaemia (Wredling et al., 1992; Gold et al., 1994a). Fear of hypoglycaemia is also a common source of anxiety for relatives, and may strain marital and family relationships. Spouses have a greater fear of hypoglycaemia, and report experiencing sleep disturbance through worrying about nocturnal hypoglycaemia when compared with the spouses of those who do not suffer severe hypoglycaemia (Gonder-Frederick et al., 1997; Jorgensen et al., 2003). The negative consequences of hypoglycaemia not only affect spouses, but also the parents of children with type 1 diabetes (Clarke et al., 1998), the children of diabetic parents and other family members. Two thirds of a group of 60 spouses of people with type 1 diabetes said that the risk of severe hypoglycaemia was a major source of concern to them, and when their partner is late, one in five considered hypoglycaemia
Box 14.2
Reasons why hypoglycaemia is feared
• Loss of control • Personal embarrassment • Unpleasant symptoms and effects on mood • Risk of losing consciousness • Risk of injury to self (and others) • Potential dependence on others for help • Risk of occurrence during sleep (without wakening) • Warning symptoms may be inadequate or absent • Potential impact on daily activities
PSYCHO-SOCIAL EFFECTS
311
Figure 14.1 Attributes towards different aspects of diabetes indicated by 411 patients with type 1 diabetes using a visual analogue scale. Reproduced from Pramming et al. (1991) with permission from John Wiley & Sons, Ltd
to be the principal cause (Stahl et al., 1998). About 10% felt that severe hypoglycaemia was ‘always’ a burden. Similar worries may afflict a child who has previously discovered a comatose, hypoglycaemic diabetic parent. Anecdotal evidence would suggest that the stress of dealing with episodic severe hypoglycaemia can sometimes lead to marital breakdown. Cox et al. (1987) have devised a simple questionnaire, the Hypoglycaemia Fear Survey, which can assess an individual’s fear of hypoglycaemia. In addition to measuring fear, it is possible to assess the way that people worry about hypoglycaemia and the behavioural responses that they take to avoid it. In some patients, hypoglycaemia may profoundly influence both the impact of diabetes on their daily life, and their approach to self-management. The consequences of evasion and subsequent fear of hypoglycaemia may promote a phobia and so encourage behavioural changes that maintain a high blood glucose to avoid future hypoglycaemia. Although the psychological consequences rarely produce frank psychiatric illness, the long-term effect of hypoglycaemia on subsequent behaviour may be much greater in many patients than is recognised by clinicians. This may partly explain the resistance shown by some individuals to therapeutic recommendations to improve their glycaemic control. Fear of hypoglycaemia was cited as the main reason why many young patients with type 1 diabetes were not enthusiastic about attaining strict glycaemic control despite the findings of the Diabetes Control and Complications Trial (DCCT) (Thompson et al., 1996).
LIVING WITH HYPOGLYCAEMIA
312
Box 14.3
Effects of hypoglycaemia on non-cognitive psychological function
Mood • Tense-tiredness • Unhappiness; anger • Pessimism about life expectations Fear • Related to frequency and severity of hypoglycaemia • Affects personality traits • Promotes anxiety (phobia) • ? Behavioural modification Behaviour • Irritability; emotional lability • Hostility; aggression (adults) • Naughtiness (children)
The non-cognitive effects of acute hypoglycaemia on mood and behaviour (Box 14.3) have been reviewed by Gold et al. (1997) and are described in Chapter 2. The disruption to domestic life caused by an episode of severe hypoglycaemia, the feeling of helplessness to prevent further episodes, and the increased tension and anxiety that is engendered by hypoglycaemia, both in the person with diabetes and in their relatives, is not conducive to a relaxed home environment. The psycho-social implications and consequences of hypoglycaemia on the family and on home life are little understood by many health professionals, who do not empathise with the domestic problems presented by hypoglycaemia.
EXERCISE Regular exercise has long-term health benefits for people with insulin-treated diabetes (Wassermann and Zinman, 1994). Exercise increases insulin sensitivity, helps to avoid weight gain and is beneficial for several metabolic parameters, including lipids and cardiovascular risk factors (Lehmann et al., 1997). However, exercise may not improve overall glycaemic control unless it is very frequent and intense (as pursued by many athletes), and exercise of moderate intensity increases the risk of hypoglycaemia during and after physical activity in people with type 1 diabetes. This is caused partly by the increase in insulin
EXERCISE
313
sensitivity and partly by contraction-mediated activation of glucose utilisation in skeletal muscle (Pierce, 1999). In the non-diabetic individual, exercise promotes the release of counterregulatory hormones, especially epinephrine (adrenaline), and inhibits insulin secretion. This stimulates the hepatic output of glucose, initially through hepatic glycogenolysis, but if exercise is sustained, gluconeogenesis is also promoted. Glucose is utilised by skeletal muscle, and so blood glucose concentration does not alter. In people with insulin-treated diabetes, the prevailing plasma concentration of insulin is independent of, and cannot be suppressed by, exercise, and this determines the metabolic consequences of exercise (Wassermann and Zinman, 1994). If plasma insulin is low, the peripheral uptake of glucose by muscle is reduced and hyperglycaemia results from exercise. If plasma insulin is high, hepatic output of glucose is inhibited, peripheral utilisation by muscle is stimulated, and the blood glucose falls, resulting in hypoglycaemia (see Chapter 3). The temporal relationship between the time of exercise, the time of administration of insulin, and its time-action profile, are therefore major determinants of the metabolic outcome with respect to the blood glucose response (Riddell and Perkins, 2006). Other factors of relevance include the time of the ingestion of food, the nature of the food consumed, the intensity, nature and duration of the exercise, and the site of insulin injection. Exercise of a limb into which insulin has recently been injected will increase the rate of absorption by muscle action (Box 14.4). The risk of hypoglycaemia is increased if intramuscular injection is made inadvertently (Frid et al., 1990).
Prevention of Hypoglycaemia Following Exercise The potential risk of associated hypoglycaemia requires the adoption of differing strategies to prevent an undesirable fall in blood glucose. The individual can either ingest additional short-acting carbohydrate, in liquid or solid form, or reduce the dose of insulin in anticipation of a period of physical activity. Both measures may be necessary and may be determined by the duration and the intensity of the exercise intended. Strenuous anaerobic exercise as a short burst of activity, such as a sprint or a game of squash, usually requires the
Box 14.4 Factors influencing blood glucose response to exercise in people with insulin-treated diabetes • Time of previous insulin administration • Type of insulin used; insulin regimen • Site of insulin injection • Time of previous meal or snack • Nature and quality of food consumed before exercise • Duration and nature (intensity) of exercise • Time of day of exercise
314
LIVING WITH HYPOGLYCAEMIA
prophylactic consumption of short-acting carbohydrate in advance of the exercise. Protracted physical activity (aerobic exercise) lasting for several hours, such as hill walking or marathon running, requires a substantial reduction in total insulin dose, in addition to an increase in consumption of carbohydrate. It is very difficult to advocate general measures that will be applicable to everyone and all types of exercise, because the response to exercise can be idiosyncratic and depends partly on the overall quality of glycaemic control. The risk of inducing hypoglycaemia with exercise is obviously heightened in individuals who have strict glycaemic control and whose blood glucose is lower at the start of exercise (particularly 5.0 mmol/l or less), compared to those with moderate hyperglycaemia; and in people with poor glycaemic control with high blood glucose, there is a risk of inducing ketosis with strenuous exercise. Some degree of trial and error may be necessary to assess the effect of specific activities on blood glucose in individuals. A fall in blood glucose may not occur during, or immediately following physical exertion, but may be delayed for several hours, sometimes occurring up to 15 hours later (MacDonald, 1987). If the exercise is taken in the late afternoon or early evening, hypoglycaemia may occur during the night or even the following day. The response obviously depends on factors such as the efficiency of mobilisation of glucose from glycogen stores, how effectively these are replenished after exercise, the magnitude of the coexisting hormonal response and sympatho-adrenal activation, and the prevailing plasma insulin concentration during and after exercise. The fall in blood glucose was less after high intensity exercise in healthy individuals with type 1 diabetes than with moderate intensity exercise, although the former expended more work (Guelfi et al., 2005). Plasma levels of catecholamines, growth hormone and lactate were higher during early recovery from high intensity exercise. Short periods of high intensity exercise (10–15 minutes) promote a rise in post-exercise blood glucose in people with insulin-treated diabetes, irrespective of the quality of glycaemic control (Marliss and Vranic, 2002), and this strategy might be employed to counter a fall in blood glucose after conventional exercise, although most people could not tolerate high intensity exercise for 15 minutes. The value of a short (10 second) maximal sprint immediately after a period of moderate-intensity exercise on a cycle ergometer has been studied in young healthy adults with type 1 diabetes, to examine the potential of this manoeuvre in countering a fall in blood glucose (Bussau et al., 2006). The short cycling sprint prevented a fall in blood glucose for 120 minutes, whereas the blood glucose declined during a period of rest after the moderateintensity exercise. This brief period of intense physical exercise therefore reduced the risk of early post-exercise hypoglycaemia. This may help to stabilise blood glucose after exercise, at least in the short term. For some people one of the safest times of day to exercise is in the fasting state (before breakfast) and before the administration of morning insulin, because plasma insulin should be relatively low at this time of day (Ruegemer et al., 1990). In the fasting state the fall in blood glucose during moderate exercise is small or absent in subjects with insulin-treated diabetes; the prophylactic ingestion of carbohydrate may cause an unwanted rise in blood glucose, and may not therefore be necessary (Soo et al., 1996). The use of continuous subcutaneous insulin infusion (CSII) does not avoid the risk of exercise-induced hypoglycaemia, and interrupting the basal insulin infusion may be insufficient to prevent a fall in blood glucose after postprandial exercise if hyperinsulinaemia is present (Edelmann et al., 1986). In wellcontrolled patients using CSII the insulin dose has to be reduced before, during and after exercise to minimise the risk of acute and late hypoglycaemia (Sonnenberg et al., 1990). Measures to prevent hypoglycaemia occurring in relation to exercise are shown in Box 14.5.
EXERCISE
Box 14.5
315
Measures to prevent hypoglycaemia induced by exercise
• Take extra carbohydrate (20–30 g short-acting) before, and possibly during, exercise (especially if prolonged). • Reduce insulin dose before exercise. • Monitor blood glucose frequently. • Avoid peak absorption and time of action of insulin for strenuous exercise (2–4 hours after soluble insulin). • Use anterior abdominal wall for injection of insulin (avoid active limbs). • Avoid exercise if blood glucose is high (> 150 mmol/l) especially if ketosis is present. • Learn the glycaemic response to different types of exercise. • Carry identification re insulin therapy.
Strenuous exercise may sometimes be unpremeditated, as in an emergency situation. In addition, it is very easy for individuals to become distracted during activities such as home decorating or gardening, during which they may work much harder or for longer than originally intended. It is therefore essential that all people with insulin-treated diabetes carry a supply of glucose tablets or an alternative source of quick-acting refined carbohydrate at all times to counter a sudden decline in blood glucose.
Sport Some sports are inherently dangerous and may be inadvisable for an individual who is at high risk of developing hypoglycaemia. Dangerous activities are those involving height, water, extremes of climate and exposure to inhospitable terrain. Activities such as hand-gliding and parachuting, sub aqua-diving or unaccompanied rock climbing are usually proscribed for people treated with insulin. Boxing is inadvisable because a decline of blood glucose into the hypoglycaemic range will impair performance and increase vulnerability to sustain a head injury, and it may be difficult for the hypoglycaemic boxer to distinguish early symptoms of hypoglycaemia from those generated by the physical activity. Apart from the risk to the individual, the safety of others must be considered, and an individual experiencing acute hypoglycaemia should not put them at risk. However, with adequate precautions and careful preparation, people with insulin-treated diabetes can tackle most sports safely, and many athletes or professional sportsmen and women who have diabetes have achieved distinction at the highest levels of sporting prowess. For the average individual, for whom sport is principally a form of recreation and a means of obtaining regular exercise, measures to avoid hypoglycaemia are relatively straightforward, as described earlier.
316
LIVING WITH HYPOGLYCAEMIA
Most team sports, such as football and hockey, and competitive games such as squash or tennis usually have a predictable duration, but other activities such as swimming, cycling or running may be much more variable. Endurance sports and protracted and demanding physical activities require more elaborate planning. Sustained exercise such as hill walking requires a premeditated reduction in total insulin dose of at least 20–30%. Teenagers with diabetes attending an outdoor activity holiday in Scotland reported that hill walking, canoeing and mountain biking, all activities that involved intense exercise of long duration, were those most commonly associated with frequent and severe hypoglycaemia (Thompson et al., 1999). Long distance running requires a considerable reduction in insulin dose. In 13 runners with insulin-treated diabetes who participated in the New York marathon the total insulin dose was reduced by a mean of 38% (Grimm and Muchnick, 1993). The frequent ingestion of beverages and snacks that are rich in carbohydrate is also necessary. A personal account of a marathon run (Kjeldby, 1997) emphasised the difficulty in determining how much intermediate-acting insulin to inject in the evening after the run, and the necessity to do frequent measurements of blood glucose over the next 24 to 48 hours to avoid delayed hypoglycaemia. Outward Bound mountain courses and holidays for young people with type 1 diabetes, which include rock climbing, canoeing, horse-riding, caving and mountain expeditions, have been described by Hillson (1984; 1987) who has detailed the sort of measures necessary for participants to avoid and to treat hypoglycaemia (Box 14.6). Anticipation of potential hazards for people with diabetes at risk of hypoglycaemia must be considered for all activities, with consideration given to the timing of meals and administration of insulin, travelling time
Box 14.6 Measures to prevent hypoglycaemia in outdoor activities and holidays (derived from Hillson, 1984; 1987) • Reduce total insulin dose (by 10–15%). • Ensure a good intake of high-fibre carbohydrate with plentiful quick-acting carbohydrate. • Increase carbohydrate at main meals and double the amount taken at snacks, or the number of snacks between main meals. • Consume glucose tablets or drinks immediately before climbing up or down anything high, or during water activities. • Carry glucose at all times and keep by the bed at night. • Monitor blood glucose four times daily and respond appropriately to the results. Aim for a blood glucose of ∼100 mmol/l. Blood glucose may be difficult to measure in cold or wet weather. • Take an hourly snack during prolonged exercise such as cycling or mountain walking. • Take a large pre-bedtime snack to avoid delayed hypoglycaemia.
DRIVING
317
and how much energy is likely to be expended. It may be difficult to distinguish between physical exhaustion and hypoglycaemia, both of which may coexist (Hillson, 1984), and hypothermia can be induced by hypoglycaemia as well as by cold and wet conditions. Coldinduced hypoglycaemia has been associated particularly with water sports such as canoeing and windsurfing, especially in adverse weather (Thompson et al., 1999). It may be necessary to reduce insulin dosage and increase carbohydrate intake at summer camps for children with diabetes (Frost et al., 1986; Braatvedt et al., 1997), because frequent hypoglycaemia is associated with the sudden increase in energetic activities after the children arrive at camp. In children, this policy may have to be applied to holidays in general (see Chapter 9). For a review of the management of sporting activities involving children with type 1 diabetes, a recent article from Canada can be recommended (Riddell and Iscoe, 2006). Dr Ian Gallen, a specialist in diabetes at Wycombe Hospital, Buckinghamshire, UK, has developed a detailed interest in how to manage insulin-treated diabetes in relation to sport (Gallen, 2004), and information and advice about different sports is provided on a valuable website that he has created at www.runsweet.com.
Recreation Strenuous and protracted exercise is not confined to sport and may occur during recreational activities, such as prolonged and vigorous dancing. These social events may also involve the consumption of alcohol, another potential cause of promoting and protracting hypoglycaemia. Some ‘recreational’ drugs such as amphetamines have been associated with promoting frenetic behaviour and increased metabolic rate, which may then induce hypoglycaemia in people treated with insulin (Jenks and Watkinson, 1998). Young people with type 1 diabetes who attend clubs or parties often avoid the potential risk and embarrassment of hypoglycaemia by not taking their insulin before the social event. Although this may seem to be a pragmatic approach, the problem with this strategy is that exercise may worsen the pre-existing hyperglycaemia, and could promote development of ketoacidosis. A modest reduction of insulin dose, combined with appropriate high carbohydrate snacks and the judicious consumption of alcohol, should avoid hypoglycaemia, although this requires forward planning and may not be conducive to the spontaneity of social activity desired by many young adults. Similarly, unpremeditated and energetic sexual intercourse can precipitate unexpected hypoglycaemia, depending on the related metabolic circumstances, and care should be taken to avoid this hazard if possible! It is easy to see why meticulous self-care of diabetes could inhibit the social activities of teenagers and young adults and interferes with late nights and parties. It is also clear why many people do not strive to achieve good glycaemic control in this situation. An episode of severe hypoglycaemia will ruin a social outing, but chronic hyperglycaemia is equally undesirable.
DRIVING For people with diabetes who are treated with insulin, the potential risks of hypoglycaemia are always present and have therefore influenced the ways in which modern society regulates and restricts their activities. This principally affects driving licences and some forms of employment. Although most of these restrictions are reasonable and important
318
LIVING WITH HYPOGLYCAEMIA
for public safety, much lay, and even medical, ignorance exists about hypoglycaemia and its effects, so that discriminatory practices still occur, particularly with regard to employment.
Effect of Hypoglycaemia on Driving Driving is a common and everyday activity that demands complex psychomotor skills, including good visuo-spatial functions, rapid information processing, vigilance and satisfactory judgment. Because hypoglycaemia rapidly interferes with cognitive functions, even modest degrees of neuroglycopenia may affect driving skills, without necessarily provoking symptomatic awareness of hypoglycaemia. Seminal studies using a sophisticated driving simulator have examined the driving abilities of drivers with type 1 diabetes at different blood glucose concentrations, maintained by a glucose clamp (Cox et al., 1993; Cox et al., 2000). Driving performance started to deteriorate when blood glucose declined below 3.8 mmol/l, and typical driving deficiencies included speeding and inappropriate braking, driving off the road, crossing the centre line, ignoring ’STOP’ signs and causing an increased number of ‘crashes’. Allowing for the artificial conditions of a driving simulator, it is evident that hypoglycaemia has an adverse effect on driving performance. A particularly disconcerting observation in these studies was that none of the drivers took action to treat hypoglycaemia until their blood glucose had declined to < 28 mmol/l, and then only 30% of the participants responded (Cox et al., 2000). Many of the drivers did not experience any warning symptoms of hypoglycaemia, and fewer than 25% said they did not feel competent to drive when their blood glucose was low (Cox et al., 1993; Cox et al., 2000). The driving simulator studies also demonstrated that driving has a substantial metabolic demand that can lower blood glucose (Cox et al., 2001, Cox et al., 2002), leading the authors to recommend that a prophylactic snack should be consumed before driving if blood glucose is 5.0 mmol/l or less. Various studies have shown that many drivers with insulin-treated diabetes believe that it is safe to drive when their blood glucose is low (Weinger et al., 1999; Clarke et al., 1999; Graveling et al., 2004); this misperception may be influenced by progressive neuroglycopenia. Hypoglycaemia can impair cognitive function and judgment without necessarily provoking warning symptoms or altering consciousness. Driving can therefore continue while apparently not being under conscious control, a condition which is given the strictly legal term of ‘automatism’ (there are no medical publications about ‘automatism’), and irrational and compulsive behaviour during hypoglycaemia has been described by insulin-treated diabetic drivers (Frier et al., 1980). The police, suspecting alcohol and inebriation to be the cause of a driver’s altered behaviour and symptoms, have on occasion arrested diabetic drivers who are hypoglycaemic.
Risk of Accidents and Restriction of Driving Licences Hypoglycaemia can adversely affect the ability to drive, and in individual cases hypoglycaemia has been implicated as a precipitating cause of road traffic accidents, causing the occasional fatality. In a study of insulin-treated drivers in Northern Ireland, the number of hypoglycaemic episodes that occurred while driving in the preceding year was shown to be associated with the total number of accidents during the previous five years (Stevens et al.,
DRIVING
319
1989), consistent with Scottish studies which showed a greater rate of accidents among diabetic drivers who experienced hypoglycaemia while driving (Frier et al., 1980; Eadington and Frier, 1989; MacLeod et al., 1993). It is difficult to quantitate how often hypoglycaemia occurs during driving and how often this precipitates a road traffic accident, particularly as hypoglycaemic incidents in which the diabetic driver is killed are seldom identified after the event. In the UK, around a third of insulin-treated diabetic drivers have admitted to experiencing hypoglycaemia while driving (Frier et al., 1980; Stevens et al., 1989; Eadington and Frier, 1989; Graveling et al., 2004). The rate of hypoglycaemia-induced accidents is extremely difficult to evaluate and is, of necessity, anecdotal. Most road accidents have several contributory factors, and it may be difficult to isolate hypoglycaemia as being the principal cause. Studies in the UK have suggested that the accident rate of diabetic drivers is very similar to non-diabetic drivers (Stevens et al., 1989; Eadington and Frier, 1989), and this premise is supported by studies from Germany (Chantelau et al., 1990) and the USA (Songer et al., 1988) (Table 14.1). In an assessment of medical factors causing road traffic accidents, a study in Iceland showed that disorders such as diabetes were not over-represented (Gislason et al., 1997). However, one American study has observed a ‘slight increase’ in the risk of motor vehicle accidents in diabetic drivers (Hansotia and Broste, 1991), but considered this increase to be insufficient to ‘warrant further restrictions on driving privileges’. These studies have been criticised for being retrospective, for excluding fatal accidents and being influenced by the removal of diabetic drivers who have ceased driving either by their own volition or through the efforts of the regulatory authorities. Police notifications in the UK to the Driver and Vehicle Licensing Agency (DVLA) of serious accidents associated with hypoglycaemia have risen steadily in recent years, which probably represent an increase in identification and reporting, rather than an increasing risk of hypoglycaemia-related road traffic accidents, but several fatalities associated with hypoglycaemic drivers are reported annually. Most licensing authorities in states in the European Community issue ordinary driving licences to people with insulin-treated diabetes that are restricted in duration, and are subject to medical review. Other than visual impairment, the principal factors that commonly lead to an ordinary driving licence being revoked are related to hypoglycaemia. Impaired awareness of hypoglycaemia with its increased risk of severe hypoglycaemia (see Chapter 7), and recurrent severe hypoglycaemia during waking hours clearly present hazards to safe driving by diminishing medical fitness to drive and are common reasons for driving licences being revoked.
Table 14.1 Hypoglycaemia-related road traffic accidents: rates per mileage driven
Hypo-related accidents
Total mileage (X 106 )
166
8
9
10.5
090
241
2
10
5.7
176
127
1
2
1.5
130
n Eadington and Frier (1989) Chantelau et al. (1990) Songer et al. (1988)
Hypo-related accidents per 106 miles
Time (years)
320
LIVING WITH HYPOGLYCAEMIA
Vocational Driving Licences A more stringent approach towards vocational licences, i.e., those for large goods vehicles (LGV) and passenger carrying vehicles (PCV), has been adopted by the European Community, and for several years insulin-treated drivers have been debarred from holding vocational driving licences in most European countries. However, there is a wide international variation in the policies of governments towards vocational licensing for diabetic drivers (DiaMond Project Group on Social Issues, 1993) and even between states in the USA (Gower et al., 1992). The Federal Highways Administration in North America has explored a waiver scheme for insulin-treated diabetic drivers of commercial trucks and those who drive between states, provided they meet strict medical criteria and are free from severe hypoglycaemia, but regulations differ between States. In Europe, as a consequence of the second EC Directive on driving in 1991, there was a reclassification of vocational licences; people with insulin-treated diabetes are now allowed ‘in exceptional circumstances’ to drive commercial vehicles, such as vans or lorries, weighing between 3.5 and 7.5 tonnes (3500 to 7500 kg) (with a C1 licence), and mini buses carrying up to 16 passengers (with a D1 licence) for their employment, subject to more stringent annual review of medical fitness to drive. This concession does not include D1 licences in the UK, although an aberration remains in British legislation that allows voluntary drivers with insulin-treated diabetes to drive minibuses, such as for charity work or for voluntary organizations. The main medical concern is the risk of hypoglycaemia affecting drivers of these larger vehicles. The need to safeguard public safety has to be balanced against the rights of the individual with diabetes, but this issue has aroused considerable controversy. Taxi driving is controlled by local authorities and for drivers with diabetes who work for emergency services (such as the police, fire and ambulance services), driving restrictions are determined by the employer, usually with advice from occupational health physicians.
Advice for Diabetic Drivers Although this chapter is primarily concerned with hypoglycaemia, there are various reasons why an individual driver who is taking insulin may be advised to cease driving, albeit temporarily (Box 14.7). Cox et al. (1994) have claimed that blood glucose awareness training in a small number of people with impaired awareness of hypoglycaemia led to fewer road traffic accidents in subsequent years, suggesting an indirect benefit of this approach to improving the recognition of blood glucose fluctuations (see Chapter 7). Prevention of hypoglycaemia while driving is essential (Box 14.8) and it is important for the driver to plan each journey (no matter how short) in advance. Blood glucose testing is advisable before driving, and at intervals of about two hours during long journeys, and rest periods for snacks and meals should be taken. Unfortunately, this practice is not common. A survey in Edinburgh showed that 50% of 202 insulin-treated diabetic drivers never test blood glucose in relation to driving, and only 14% do this regularly, most of these individuals having impaired awareness of hypoglycaemia (Graveling et al., 2004). Around half of those questioned admitted to a variety of unsafe practices with respect to driving.
DRIVING
Box 14.7
321
Diabetic drivers – Reasons to cease driving
Hypoglycaemia • People with newly diagnosed type 1 diabetes or any patient commencing treatment with insulin, should cease driving until glycaemic control and vision are stable. • Recurrent hypoglycaemia (especially if severe). • Impaired awareness of hypoglycaemia (if disabling). Other • Reduced (corrected) visual acuity for distance (worse than 6/12 on Snellen chart) in both eyes. Care required after use of mydriatic for eye examination. • Sensori-motor peripheral neuropathy with loss of proprioception. • Severe peripheral vascular disease; lower limb amputation (hand controls and automatic transmission may be feasible).
Box 14.8
Advice for diabetic drivers regarding hypoglycaemia
• If hypoglycaemia occurs while driving, stop the vehicle in a suitable location; leave the driver’s seat. • Always keep an emergency supply of readily accessible fast-acting carbohydrate (e.g. glucose tablets or sweets) in the vehicle. • Check blood glucose before driving (even on short journeys) and estimate at regular intervals on long journeys. • Take regular meals and snacks, and rest periods on long journeys; avoid alcohol. • If hypoglycaemia is experienced, do not drive until 45 minutes after blood glucose is restored to normal (delayed recovery of cognitive function). • Carry personal identification indicating ‘diabetes’ in case of injury in a road traffic accident.
A supply of both quick-acting and more substantial carbohydrate should be kept constantly in the vehicle in case of unexpected delays or emergencies (traffic jams, breakdowns) or unpremeditated exercise such as changing a wheel. If hypoglycaemia occurs while driving, the driver should stop the vehicle, switch off the engine and leave the driver’s seat, as in British law a charge of driving under the influence of a drug (insulin) can be made even
322
LIVING WITH HYPOGLYCAEMIA
if the car is stationary. It is also important that driving is not recommenced immediately after normoglycaemia is restored. In this situation, blood glucose does not accurately reflect the glucose concentration in the brain, with the rise in intra-cerebral glucose lagging behind that in the peripheral blood. The complete recovery of cognitive function following hypoglycaemia takes up to 45 minutes after blood glucose has returned to normal (Chapter 2), and so an interval of this duration should be allowed before driving is recommenced.
Medico-legal Aspects Physicians who specialize in diabetes are often required to provide medical reports in relation to road traffic accidents involving drivers with insulin-treated diabetes, in whom hypoglycaemia has been implicated as a possible cause. A detailed history of the circumstances should be taken from the diabetic driver to identify whether hypoglycaemia was likely at the time of the accident, as contemporaneous blood glucose measurements are seldom available. Occasionally, blood glucose has been measured at the scene of the accident by paramedical ambulance staff or on subsequent admission to hospital. However, any significant delay before the blood glucose is estimated may obscure the glycaemic status at the time of the accident, through the effect of counterregulatory hormones released by the stress of the accident and/or hypoglycaemia per se, or as a result of treatment. The presentation of a convincing story of hypoglycaemia preceding the accident has to be accompanied by a careful description of the potential effects of hypoglycaemia on cognitive function and behaviour, comprehensible to a lay person. Although this mitigating factor may not allow legal charges to be dismissed, in my experience the penalty may be substantially reduced if hypoglycaemia is accepted to be the principal problem that has affected the individual’s driving ability and precipitated the accident. However, this must not be considered to be a foregone conclusion, as the legal view of hypoglycaemia occurring in a person treated with insulin (or an oral hypoglycaemic drug) is that this represents ‘careless’ or ‘reckless’ behaviour on their part and is therefore the ‘fault’ of the individual, even though in clinical practice no specific cause can be determined for many episodes of hypoglycaemia. Although the difficulty of always being able to test blood glucose before driving is recognised in clinical practice, when this has a serious outcome, the judiciary may take a much stricter view. In a case in Edinburgh Sheriff Court in 2000, which involved a fatal car accident caused by a hypoglycaemic diabetic driver, who had not measured his blood glucose before driving, the Sheriff commented: ‘There is a risk associated with diabetes and driving, and as a consequence, there is a need to monitor your blood sugar level. There is some culpability on your part’. The driver was found guilty of dangerous driving. By contrast, spontaneous hypoglycaemia is an accepted defence, and one of my patients with insulin-treated diabetes had charges of dangerous driving dismissed when it was shown that at the time of the offence he had developed undiagnosed and untreated Addison’s disease – a relatively rare but recognised cause of increased and unpredictable severe hypoglycaemia in type 1 diabetes. Medico-legal aspects of hypoglycaemia and diabetes have been reviewed elsewhere (Frier and Maher, 1988).
TRAVEL Many of the measures recommended for longer car journeys (Box 14.8) are appropriate to long distance travel, irrespective of the mode of transport used. Forward planning is
EMPLOYMENT
323
essential to avoid hypoglycaemia, with emphasis on adjustment of insulin dose (or regimen) if necessary, carrying equipment for blood glucose monitoring and ensuring an adequate supply both of quick-acting carbohydrate and of non-perishable emergency rations in case suitable food is not available during travel. Standard airline meals are often low in unrefined carbohydrate and may be unpalatable. Advice for travel and holidays is available from various sources, but with respect to avoiding (and treating) hypoglycaemia, some practical points can be made. • For long-distance air travel, crossing several time zones, frequent administration of shortacting (soluble or analogue) insulin is much simpler to use than attempting to modify the times of administration and dosage of intermediate-acting insulins. Disposable insulin pens are also very useful for this purpose. Rapid-acting insulin analogues have the advantage that their administration can be delayed until the food on offer is available and its palatability assessed, or can be taken after the meal, providing greater flexibility for treatment. • Some blood glucose meters are inaccurate in the hypoglycaemic range and many do not give accurate readings at high altitude or at extremes of temperature. Visually-read strips for blood glucose estimation may therefore be necessary in some situations. It is advisable to carry a spare blood glucose monitor in case of equipment failure. • A supply of quick-acting carbohydrate is essential, but should be stored appropriately. Dextrose tablets may disintegrate or become very hard in hot and humid climates unless wrapped in silver foil or stored in a suitable container, and chocolate will melt if the temperature is high. At very cold temperatures, the wrapper may become welded to the chocolate. Cartons of orange juice cannot be re-used once opened, and so a plastic bottle or container with a screw top is preferable. Sealed packets of powdered glucose may be more suitable to carry in hot humid climates. • Travelling companions should carry a supply of quick-acting carbohydrate (and glucagon) for emergency use. The nature of the travel undertaken, how much energy is expended, the quality and nature of food and the risk of intermittent illness (such as travel sickness or gastroenteritis) are all potential factors that can influence blood glucose and potentially induce hypoglycaemia. Although many situations are predictable, the most important measure is frequent monitoring of blood glucose so that sensible adjustments in insulin therapy and ingestion of food can be made.
EMPLOYMENT The risk of developing acute hypoglycaemia and its consequences (mainly in people with insulin-treated diabetes) provide reasons why some forms of employment are not available to individuals who require insulin therapy. Employment prospects are often restricted where the threat of hypoglycaemia poses a risk to the diabetic worker or to his or her colleagues, and to the general public. With some occupations, such as a train or bus driver, or a commercial airline pilot, any risk of hypoglycaemia is unacceptable. In other areas the potential risks of hypoglycaemia may be less well defined, and restrictions to employment have been
324
LIVING WITH HYPOGLYCAEMIA
established by individual industries or firms, rather than by legislation. Although medical advice has usually been sought, this has not always been well informed, and may not have involved physicians with expertise in diabetes or of occupational health. Some restrictions have been challenged successfully, with one example in the UK being the reinstatement of several active firefighters, based on individual medical assessment. People with insulin-treated diabetes are not usually permitted to work alone in isolated or dangerous areas or at unprotected heights. They are also debarred from serving in the armed forces. This is based on the grounds that all service personnel (including noncombatants) could be involved in a conflict at short notice, and maintaining provision of insulin and appropriate dietary requirements could present difficulties in a wartime situation. Employment is not usually permitted in emergency teams, civil aviation, work in the offshore oil industry, and in many forms of commercial driving (Waclawski, 1989). A list of jobs in which the employment of people with insulin-treated diabetes (both types 1 and 2) is restricted is shown in Table 14.2. The civil aviation authority in the UK does not permit diabetic individuals who are treated with insulin or sulphonylureas to fly commercial aircraft, or to work as air traffic controllers, although in the USA an air traffic controller appealed successfully against dismissal on grounds of discrimination. In the European Community, discussions are proceeding to produce common airworthiness regulations for pilots with medical disorders, including diabetes.
Table 14.2 Employment restrictions placed on diabetic workers treated with insulin in UK (adapted from Waclawski, 1989) Vocational driving
Large goods vehicles (LGV licences) Passenger carrying vehicles (PCV licences) Locomotives and underground trains Professional drivers (chauffeurs) Taxi drivers (variable: depends on local authority)
Civil aviation
Commercial pilots; flight engineers Aircrew Air traffic controllers
National and emergency services
Armed forces (Army, Navy, Air Force) Police force Fire brigade or Rescue services Prison and Security services
Dangerous areas for work
Offshore oil-rig work Moving machinery Incinerator loading Hot-metal areas Work on railway tracks Coal mining
Work at heights
Overhead linesmen Crane driving Scaffolding/high ladders or platforms
EMPLOYMENT
325
Hypoglycaemia at Work
% of episodes of severe hypoglycaemia
Although many anecdotal accounts exist of severe hypoglycaemia affecting employees with insulin-treated diabetes while at work, this does not appear to be a widespread problem. Although isolated episodes of severe hypoglycaemia occurring in the work place are inevitable, it appears that most hypoglycaemia is mild, quickly self-treated and does not cause disruption. The times of day at which hypoglycaemia is most common were observed in a prospective study of 60 patients with type 1 diabetes, half of whom had impaired awareness of hypoglycaemia (Gold et al., 1994b). Most episodes of severe hypoglycaemia occurred during the evening or night, or in the early morning before the subjects went to work (Figure 14.2). The higher frequency of severe hypoglycaemia in the evening or during the night was not attributable to the insulin regimens being used, and may be related to relaxed vigilance or different behaviour in the evening, when at home. A study of 243 people with insulin-treated diabetes in full-time employment in Edinburgh, conducted prospectively for one year, indicated that the frequency of severe hypoglycaemia at work was low (15% of all episodes, affecting 11% of workers) (Leckie et al., 2005). Although severe hypoglycaemia occurred at home and at other times, the hypoglycaemia experienced in the workplace seldom caused disruption or serious morbidity. People treated with insulin may be more vigilant while they are at work and actively try to avoid the development of low blood glucose, working activities may be more regular than at home or at weekends, or they selftreat low blood glucose promptly, so avoiding significant neuroglycopenia. In the Edinburgh study, very few of the people in employment had impaired awareness of hypoglycaemia (a major risk factor for severe hypoglycaemia), and overall glycaemic control was not strict, suggesting that employees with insulin-treated diabetes may deliberately avoid having a low HbA1c to limit their hypoglycaemia risk (Leckie et al., 2005). Shift work is generally not a contraindication to the employment of people with type 1 diabetes. However, occasionally a frequent change of shift rota can cause difficulties
100
80 p = 0.03
60 40 20
0
NS
p = 0.05
NS
8 am–1 pm
1 pm–6 pm
6 pm–12 MN
00–8 am
Figure 14.2 Percentages of total number of episodes of severe hypoglycaemia occurring at different times of day in patients with type 1 diabetes with normal (solid bars) and impaired awareness of hypoglycaemia (hatched bars). Reproduced from Gold et al. (1994) with permission from The American Diabetes Association
326
LIVING WITH HYPOGLYCAEMIA
with glycaemic control, although this usually causes deterioration in control rather than hypoglycaemia (Poole et al., 1992). Measures to avoid and/or treat hypoglycaemia at work are no different from any other time or circumstance, although in some jobs, blood glucose monitoring may not be feasible while at work and break times for snacks and meals may be variable. It is essential that workmates or colleagues are familiar with the emergency treatment of diabetes and that a supply of quick-acting carbohydrate is available in the work place. It is advisable that the individual’s employer is aware that he or she has insulintreated diabetes, although some people conceal this fact, fearing dismissal or discrimination. Legislation to avoid this is in place in many Western countries, but widespread ignorance remains about the nature of diabetes, its treatment and the possible side-effects, and this has to be confronted by those with specialist expertise. Similar arrangements for emergency treatment should be in place in schools (see Chapter 9), colleges and other venues of tertiary education, including university accommodation. In separate cases in Edinburgh (Strachan et al., 2000), two young male students with type 1 diabetes died tragically after developing severe hypoglycaemia while living in university halls of residence. One of the students was left lying unconscious for two days on the floor of his room, despite being observed by a member of the domestic staff who thought he was asleep or drunk. Although the warden had known that the student had diabetes, other members of staff were not aware. At the subsequent fatal accident enquiry, the Sheriff discussed the dichotomy between maintaining the privacy and confidentiality of the individual, and the duty of care owed to young people living in university accommodation. As a result of his report and recommendations, the facilities and practical handling of students with type 1 diabetes in residencies were modified in Scottish universities. However, these cases highlight the importance of disclosure of diabetes, and its potential metabolic problems (especially hypoglycaemia), to the appropriate authorities and those who may potentially be required to render emergency assistance.
Specialist Medical Reports Physicians who specialise in diabetes are often required to provide medical reports for employers, either related to the suitability of specific types of work for a person with type 1 diabetes or to their capability of performing the job. It is often necessary to advocate on the behalf of patients when a problem at work is specifically related to some aspect of their diabetic management, such as hypoglycaemia. The introduction of insulin therapy sometimes has to be delayed because of the impact that this will have on the individual’s employment. By strict dietary measures, an oil tanker driver (with a LGV licence) attending my clinic, who had longstanding type 2 diabetes, managed to maintain adequate glycaemic control on the maximum dose of combined oral antidiabetic therapy for nearly two years, delaying conversion to insulin. He was anxious to protect his pension rights until his retirement, at which point insulin was commenced. However, procrastination with starting insulin therapy is usually inadvisable in type 2 diabetes, and is not possible in those presenting with newlydiagnosed type 1 diabetes. Sympathetic medical counselling may be necessary to recommend an alternative occupation. This is often necessary in people who hold vocational driving licences, such as bus drivers or lorry drivers.
School and Academic Examinations It is recognised that school pupils and students with type 1 diabetes may be disadvantaged during school or college examinations by unpredictable fluctuations in blood glucose
PRISON AND POLICE CUSTODY
327
concentrations, with pronounced cognitive impairment and mood changes occurring when blood glucose falls below 3.5 mmol/l. Fluctuations in blood glucose may be exacerbated by the stress associated with an exam, and sometimes children are unable to consume adequate carbohydrate in advance of an academic test because of pre-exam anxiety and temporary loss of appetite. The consumption of glucose by the brain is very substantial during a period of intense intellectual activity and concentration, and this may cause the blood glucose to fall considerably and provoke hypoglycaemia. This effect may be greater in the child than in an adult, because of the larger size of the brain relative to the rest of the body, and its high energy requirements. The idiosyncratic nature of this effect in individuals, and the variability in different circumstances and at different times of day, makes the accurate determination of the insulin requirement very difficult in advance of an exam, but generally a lower dose of insulin should be given. However, an excessive reduction in insulin dose to avoid hypoglycaemia may lead to undesirable hyperglycaemia, causing thirst and polyuria. Although a pupil or student with type 1 diabetes can (and should) measure blood glucose immediately before an exam, it is generally impractical to do repeated monitoring during the exam, where the use of a blood glucose meter or the consumption of food or glucose drinks may be prohibited. It may be necessary to request special dispensation in advance. During an exam, because of multiple distractions and the need to focus on the exam itself, the affected individual may not identify the usual warning symptoms of hypoglycaemia. In addition, because many of the typical symptoms of hypoglycaemia (such as tremor, sweating, pounding heart, feelings of anxiety) are caused by stimulation of the autonomic nervous system (see Chapter 2), even when they are generated in this situation, they may be incorrectly attributed by the student (or by an observer) to exam-induced anxiety; the falling blood glucose is therefore not identified and corrective treatment is not given. Neuroglycopenic symptoms such as difficulty concentrating, feeling lightheaded and drowsiness, may also be ignored in an exam situation, and the cognitive dysfunction may contribute to the difficulty in initiating self-treatment. Hypoglycaemia affects most cognitive functions, with the severity depending on the depth and duration of the low blood glucose, and the complexity of the tasks being undertaken (Chapter 2). All of the major cognitive domains are affected, including memory, attention, concentration, reasoning, problem-solving ability, abstract thought, rapid decision-making and judgment, and skills like hand-eye co-ordination are impaired. In addition, hypoglycaemia provokes major mood changes, including tense-tiredness, pessimism, irritability and sometimes anger, which are not conducive to enhancing exam performance. Hyperglycaemia can also affect cognitive function, but in more subtle ways and with less dramatic impact than hypoglycaemia, and is also associated with negative mood changes.
PRISON AND POLICE CUSTODY Police Custody and Hypoglycaemia Many features of acute hypoglycaemia simulate those of alcoholic inebriation, so that people with insulin-treated diabetes have occasionally been arrested by the police under the mistaken impression that they are drunk or under the influence of drugs. The danger of this situation is often compounded by their detention at a police station, with confinement in a cell, instead of receiving treatment or urgent transfer to hospital. Unfortunately, the consumption of
328
LIVING WITH HYPOGLYCAEMIA
alcohol can promote hypoglycaemia in people treated with insulin, and may be a contributory factor to inducing the low blood glucose, so causing further difficulty with identification of the underlying metabolic problem. This emphasises the importance of an individual with insulin-treated diabetes carrying some form of identification to indicate that they are taking insulin and may be at risk of developing hypoglycaemia-induced coma. In addition to the risk of being arrested during an episode of hypoglycaemia because of aggressive or abnormal behaviour, low blood glucose may develop while in custody. The police may have limited comprehension of the needs of a person with diabetes and the risks of hypoglycaemia. The young male patient with undiagnosed Addison’s disease, described earlier, who was arrested on a driving charge, was profoundly neuroglycopenic when taken into custody. He was detained without treatment for two hours. When his father arrived at the police station, he recognised immediately that his son was severely hypoglycaemic and needed emergency treatment with dextrose. This type of situation is clearly alarming, and potentially could have a fatal outcome. An initiative in Edinburgh has liaised successfully with the local police force to improve the way in which people with insulin-treated diabetes are handled while in custody (Barclay et al., 2007).
Management of Diabetes in Prison The general problems of managing diabetes in prison have been examined in two British studies (Gill and MacFarlane, 1989; MacFarlane et al., 1992) and recommendations have been made to improve the care of diabetes in prison, by the American Diabetes Association (Eichold, 1989) and by Diabetes UK (Gill et al., 1992). Imprisonment causes particular problems for the management of diabetes, which are conducive to the development of hypoglycaemia. These include the following: • an inadequate or inappropriate prison diet; • long ‘lock-up’ periods necessitated by prison routine; • solitary confinement for individual prisoners; • restrictions in the time and place of insulin administration; • the use of some insulin regimens (e.g. basal-bolus and/or injection of bedtime isophane insulin) are precluded by prison routine; • a long time interval between the evening meal and breakfast the following morning (sometimes over 12 hours); • no blood glucose monitoring facilities being allowed in cells; • lack of medical knowledge among most prison officers with few personnel having any medical training. Many of these problems predispose to a risk of nocturnal hypoglycaemia, and actively discourage any attempt at achieving strict glycaemic control. Various measures can be suggested to try and prevent hypoglycaemia in prisoners with insulin-treated diabetes:
REFERENCES
329
• availability of dextrose tablets (or an alternative source of carbohydrate, e.g. biscuits) in cells; • provision of a late evening snack; • avoidance of solitary confinement if possible; • sharing cells with prisoner(s) who can recognise hypoglycaemia; • arranging access to specialist advice on management of diabetes.
CONCLUSIONS • Fear of hypoglycaemia is common and may influence self-management of blood glucose by individual patients. Worries about hypoglycaemia extend to relatives, spouses and partners of the person with diabetes, and recurrent hypoglycaemia can disrupt family life. • The risk of hypoglycaemia occurring during exercise depends on the prevailing plasma concentrations of insulin and glucose and the duration and intensity of the physical activity. Strategies to avoid a fall in blood glucose include the ingestion of additional carbohydrate and a reduction in insulin dose. The contributory effect of alcohol may be important. Sport, recreational activities and travel all require forward planning and preventative measures. Some dangerous recreational activities should be avoided. • Although the risk of hypoglycaemia-related driving accidents is difficult to quantitate, hypoglycaemia is a potential hazard when driving, and impaired awareness of hypoglycaemia may cause revocation of the driving licence. The diabetic driver must carry a supply of glucose in the vehicle and take suitable precautions to ensure safe driving on all journeys, including blood glucose monitoring. • Hypoglycaemia at work is uncommon but the potential risk debars people with insulintreated diabetes from certain occupations. Discrimination by employers against people with type 1 diabetes still occurs. • The problems of managing diabetes in prison include inadequate facilities for preventing hypoglycaemia, especially overnight. Retention of diabetic individuals in police custody may cause difficulties by failure of the custodians to recognise hypoglycaemia and its similarity to the features of inebriation.
REFERENCES Barclay JIP, Plastow B, Frier BM (2007). You’re nicked! A project in improving diabetes care in police custody. Diabetic Medicine (abstract), 24 (Suppl.1): 89. Braatvedt GD, Mildenhall L, Patten C, Harris G (1997). Insulin requirements and metabolic control in children with diabetes mellitus attending a summer camp. Diabetic Medicine 14: 258–61. Bussau VA, Ferreira LD, Jones TW, Fournier PA (2006). The 10-s maximal sprint. A novel approach to counter an exercise-mediated fall in glycemia in individuals with type 1 diabetes. Diabetes Care 29: 601–6.
330
LIVING WITH HYPOGLYCAEMIA
Chantelau E, Dannbeck S, Kimmerle R, Ross D (1990). Zur verkehrstüchtigkeit insulin-behandelter diabetiker. Münchener Medizinische Wochenschrift 132: 468–71. Clarke WL, Gonder-Frederick LA, Miller S, Richardson T, Snyder A (1998). Maternal fear of hypoglycemia in their children with insulin-dependent diabetes mellitus. Journal of Pediatric Endocrinology and Metabolism 4: 189–94. Clarke WL, Cox DJ, Gonder-Frederick LA, Kovatchev B (1999). Hypoglycemia and the decision to drive a motor vehicle by persons with diabetes. Journal of the American Medical Association 282: 750–4. Cox DJ, Irvine A, Gonder-Frederick L, Nowacek G, Butterfield J (1987). Fear of hypoglycemia: quantification, validation and utilization. Diabetes Care 10: 617–21. Cox DJ, Gonder-Frederick L, Clarke W (1993). Driving decrements in type 1 diabetes during moderate hypoglycemia. Diabetes 42: 239–43. Cox DJ, Gonder-Frederick L, Julian DM, Clarke W (1994). Long-term follow-up evaluation of blood glucose awareness training. Diabetes Care 17: 1–5. Cox DJ, Gonder-Frederick LA, Kovatchev BP, Julian DM, Clarke WL (2000). Progressive hypoglycemia’s impact on driving simulation performance. Occurrence, awareness and correction. Diabetes Care 23: 163–70. Cox DJ, Gonder-Frederick LA, Kovatchev BP, Clarke WL (2001). Self-treatment of hypoglycemia while driving. Diabetes Research and Clinical Practice 54: 17–26. Cox DJ, Gonder-Frederick LA, Kovatchev BP, Clarke WL (2002). The metabolic demands of driving for drivers with type 1 diabetes mellitus. Diabetes Metabolism Research and Reviews 18: 381–5. DiaMond Project Group on Social Issues (1993). Global regulations on diabetics treated with insulin and their operation of commercial motor vehicles. British Medical Journal 307: 250–3. Eadington DW, Frier BM (1989). Type 1 diabetes and driving experience: an eight-year cohort study. Diabetic Medicine 6: 137–41. Edelmann E, Staudner V, Bachmann W, Walter H, Haas W, Mehnert H (1986). Exercise-induced hypoglycaemia and subcutaneous insulin infusion. Diabetic Medicine 3: 526–31. Eichold S (1989). Diabetes in correctional institutions. Diabetes Care 12: 509–10. Frid A, Östman J, Linde B (1990). Hypoglycemia risk during exercise after intramuscular injection of insulin in thigh in IDDM. Diabetes Care 13: 473–7. Frier BM, Matthews DM, Steel JM, Duncan LJP (1980). Driving and insulin-dependent diabetes. Lancet 1: 1232–4. Frier BM, Maher G (1988). Diabetes and hypoglycaemia: medico-legal aspects of criminal responsibility. Diabetic Medicine 5: 521–6. Frost GF, Hodges S, Swift PGF (1986). Dietary carbohydrate deficits and hypoglycaemia in the young diabetic on holiday. Diabetic Medicine 3: 250–2. Gallen IW (2004). Helping the athlete who has type 1 diabetes. British Journal of Diabetes and Vascular Disease 4: 87–92. Gill GV, MacFarlane IA (1989). Problems of diabetics in prison. British Medical Journal 298: 221–3. Gill GV, MacFarlane IA, Tucker NH (1992). Diabetes care in British prisons: existing problems and potential solutions. Diabetic Medicine 9: 109–13. Gislason T, Tomasson K, Reynisdottir H, Björnsson JK, Kristbjarnarson H (1997). Medical risk factors amongst drivers in single-car accidents. Journal of International Medicine 241: 213–9. Gold AE, Deary IJ, Jones RW, O’Hare JP, Reckless JPD, Frier BM (1994a). Severe deterioration in cognitive function and personality in five patients with long-standing diabetes: a complication of diabetes or a consequence of treatment? Diabetic Medicine 11: 499–505. Gold AE, MacLeod KM, Frier BM (1994b). Frequency of severe hypoglycemia in patients with type 1 diabetes with impaired awareness of hypoglycemia. Diabetes Care 17: 697–703. Gold AE, Deary IJ, Frier BM (1997). Hypoglycaemia and non-cognitive aspects of psychological function in insulin-dependent (Type 1) diabetes mellitus (IDDM). Diabetic Medicine 14: 111–18.
REFERENCES
331
Gonder-Frederick L, Cox D, Kovatchev B, Julian D, Clarke W (1997). The psychosocial impact of severe hypoglycemic episodes on spouses of patients with IDDM. Diabetes Care 20: 1543–6. Gower IF, Songer TJ, Hylton H, Thomas NL, Ekoe J-M, Lave LB, LaPorte RE (1992). Epidemiology of insulin-using commercial motor vehicle drivers. Diabetes Care 15: 1464–7. Graveling AJ, Warren RE, Frier BM (2004). Hypoglycaemia and driving in people with insulin-treated diabetes: adherence to recommendations for avoidance. Diabetic Medicine 21: 1014–9. Grimm J-J, Muchnick S (1993). Type 1 diabetes and marathon running. Diabetes Care 16: 1624 (letter). Guelfi KJ, Jones TW, Fournier PA (2005). The decline in blood glucose level is less with intermittent high-intensity compared with moderate exercise in individuals with type 1 diabetes. Diabetes Care 28: 1289–94. Hansotia P, Broste SK (1991). The effect of epilepsy or diabetes mellitus on the risk of automobile accidents. New England Journal of Medicine 324: 22–6. Hillson RM (1984). Diabetes Outward Bound mountain course, Eskdale, Cumbria. Diabetic Medicine 1: 59–63. Hillson RM (1987). British Diabetic Association activities for young people – safety while adventuring. Practical Diabetes 4: 233–4. Jenks J, Watkinson M (1998). Minimising the risks of amphetamine use for young adults with diabetes. Journal of Diabetes Nursing 2: 179–82. Jorgensen HV, Pedersen-Bjergaard U, Rasmussen AK, Borch-Johnsen K (2003). The impact of severe hypoglycemia and impaired awareness of hypoglycemia on relatives of patients with type 1 diabetes. Diabetes Care 26: 1106–9. Kjeldby B (1997). Running the New York marathon with diabetes. British Medical Journal 314: 1053–4. Leckie AM, Graham MK, Grant JB, Ritchie PJ, Frier BM (2005). Frequency, severity, and morbidity of hypoglycemia occurring in the workplace in people with insulin-treated diabetes. Diabetes Care 28: 1333–8. Lehmann R, Kaplan V, Bingisser R, Bloch KE, Spinas GA (1997). Impact of physical activity on cardiovascular risk factors in IDDM. Diabetes Care 20: 1603–11. MacDonald MJ (1987). Postexercise late-onset hypoglycemia in insulin-dependent diabetic patients. Diabetes Care 10: 584–8. MacFarlane IA, Gill GV, Masson E, Tucker NH (1992). Diabetics in prison: can good diabetic care be achieved? British Medical Journal 304: 152–5. MacLeod KM, Hepburn DA, Frier BM (1993). Frequency and morbidity of severe hypoglycaemia in insulin-treated diabetic patients. Diabetic Medicine 10: 238–45. Marliss EB, Vranic M (2002). Intense exercise has unique effects on both insulin release and its roles in glucoregulation: implications for diabetes. Diabetes 51: Suppl. S271–83. Pierce NS (1999). Diabetes and exercise. British Journal of Sports Medicine 33: 161–73. Poole C, Wright A, Nattrass M (1992). Control of diabetes mellitus in shift workers. British Journal of Industrial Medicine 49: 513–5. Pramming S, Thorsteinsson B, Bendtson I, Binder C (1991). Symptomatic hypoglycaemia in 411 type 1 diabetic patients. Diabetic Medicine 8: 217–2. Riddell MC, Iscoe KE (2006). Physical activity, sport and pediatric diabetes. Pediatric Diabetes 7: 60–70. Riddell MC, Perkins BA (2006). Type 1 diabetes and vigorous exercise: applications of exercise physiology to patient management. Canadian Journal of Diabetes 30: 63–71. Ruegemer JJ, Squires RW, March HM, Haymond MW, Cryer PE, Rizza RA, Miles JM (1990). Differences between pre-breakfast and late afternoon glycemic responses to exercise in IDDM patients. Diabetes Care 13: 104–10. Sanders K, Mills J, Martin FIR, Horne DJD (1975). Emotional attitudes in adult insulin-dependent diabetics. Journal of Psychosomatic Research 19: 241–6.
332
LIVING WITH HYPOGLYCAEMIA
Songer TJ, LaPorte RE, Dorman JS, Orchard TJ, Cruickshank KL, Becker DJ, Drash AL (1988). Motor vehicle accidents and IDDM. Diabetes Care 11: 701–7. Sonnenberg GE, Kemmer FW, Berger M (1990). Exercise in type 1 (insulin-dependent) diabetic patients treated with continuous subcutaneous insulin infusion. Prevention of exercise induced hypoglycaemia. Diabetologia 33: 696–703. Soo K, Furler SM, Samaras K, Jenkins AB, Campbell LV, Chisholm DJ (1996). Glycemic responses to exercise in IDDM after simple and complex carbohydrate supplementation. Diabetes Care 19: 575–9. Stahl M, Berger W, Schaechinger H, Cox DJ (1998). Spouse’s worries concerning diabetic partner’s possible hypoglycaemia. Diabetic Medicine 15: 619–20 (letter). Stevens AB, Roberts M, McKane R, Atkinson AB, Bell PM, Hayes JR (1989). Motor vehicle driving among diabetics taking insulin and non-diabetics. British Medical Journal 299: 591–5. Strachan MWJ, MacCuish AC, Frier BM (2000). The care of students with insulin-treated diabetes mellitus living in university accommodation: scope for improvement? Diabetic Medicine 17: 70–3. Thompson CJ, Cummings JFR, Chalmers J, Gould C, Newton RW (1996). How have patients reacted to the implications of the DCCT? Diabetes Care 19: 876–8. Thompson C, Connacher A, Rewt D, Newton R (1999). Exercise in children and adolescents. In: Exercise and Sport in Diabetes. Burr B and Nagi D, eds. John Wiley & Sons Ltd, Chichester: 83–96. Waclawski ER (1989). Employment and diabetes: a survey of the prevalence of diabetic workers known by occupation physicians, and the restrictions placed on diabetic workers in employment. Diabetic Medicine 6: 16–9. Wasserman DH, Zinman B (1994). Exercise in individuals with IDDM. Diabetes Care 17: 924–7. Weinger K, Kinsley BT, Levy CJ, Bajaj M, Simonson DC, Cox DJ et al. (1999). The perception of safe driving ability during hypoglycemia in patients with type 1 diabetes mellitus. American Journal of Medicine 107: 246–53. Wredling RAM, Theorell PGT, Roll HM, Lins PES, Adamson UKC (1992). Psychosocial state of patients with IDDM prone to recurrent episodes of severe hypoglycemia. Diabetes Care 15: 518–21.
Index Note: Figures and Tables are indicated by italic page numbers, and Boxes by emboldened numbers abdominal circumference, measurement in diabetic pregnancy, 223, 226 abdominal pain (symptom), in children, 32, 196 Aberdeen Maternity Hospital, combined diabetic antenatal charts, 224–225 acarbose, nocturnal hypoglycaemia affected by, 94 ACE, serum levels and risk of severe hypoglycaemia, 70–71, 174, 201 ACE inhibitors, and severe hypoglycaemia, 71 acetazolamide, 133 acquired hypoglycaemia syndromes, and risk of severe hypoglycaemia, 68–70 activity holidays, 316 Addison’s disease, 74, 102, 130, 206 and dangerous driving charge, 322, 328 adipose tissue, metabolic pathways, 2, 3, 4, 5 ADOPT (A Diabetes Outcome Progression Trial), 247 adrenal system, hormonal changes during hypoglycaemia, 13, 15, 21, 129, 142 adrenaline, see epinephrine adrenocorticotrophic hormone (ACTH), changes during hypoglycaemia, 13, 14 aerobic exercise, 314 ageing blood glucose thresholds affected by, 239, 241 counterregulatory responses affected by, 241–242 symptoms of hypoglycaemia affected by, 239–240 aggressiveness (symptom), 32, 32, 33, 196, 312 mistaken for alcohol inebriation, 328 air traffic controllers, 324 air travel, 323 airline pilots, 323, 324 alanine, as glucose precursor, 2, 124, 125 alcohol consumption
Hypoglycaemia in Clinical Diabetes, 2nd Edition. © 2007 John Wiley & Sons, Ltd
advice on, 108 biochemical effects, 104–105 and hypoglycaemia, 7, 62, 75, 103–109, 131, 328 in type 2 diabetes, 246 alcohol intoxication brain damage due to, 293 death due to, 272 hypoglycaemic symptoms mistaken for, 62, 103, 318, 327–328 aldosterone, changes during hypoglycaemia, 15 alpha-glucosidase inhibitors, 247 American Diabetes Association, on blood glucose lower limit, 173 amino acids, as glucose precursors, 2, 124, 125 anaerobic exercise, 313–314 analogue insulins, see insulin analogues angina, 278 angiotensin-converting enzyme (ACE) activity, risk of severe hypoglycaemia affected by, 70–71, 174, 201 angiotensin, changes during hypoglycaemia, 15 anger feelings during hypoglycaemia, 43, 312, 328 antecedent hypoglycaemia autonomic responses affected by, 159, 173 cognitive dysfunction affected by, 41, 159 counterregulatory response affected by, 129, 156–161 and impaired awareness of hypoglycaemia, 160 subsequent hypoglycaemia affected by, 35, 41, 156–161 antecedent nocturnal hypoglycaemia caffeine-associated reduction, 110 cause of impaired awareness of hypoglycaemia, 88–89, 160, 166 effects, 157, 160
Edited by B.M. Frier and M. Fisher
334 anterior pituitary gland, hormonal changes during hypoglycaemia, 13–14, 13 anxiety about hypoglycaemia, 44, 49, 172, 310–311, 312 and counterregulatory deficiencies, 135 as warning symptom, 26, 30, 31 240 argumentativeness (symptom), in children, 32 armed forces, employment in, 324 asymptomatic hypoglycaemia definition, 50, 51 frequency, 55–57, 145 and impaired awareness of hypoglycaemia, 154–155 atrial fibrillation, 277 auditory processing, effect of hypoglycaemia, 42 automatism, 318 autonomic dysfunction, and impaired awareness of hypoglycaemia, 110, 152–153 autonomic nervous system, 10, 11 activation during hypoglycaemia, 10–12, 21 autonomic neuropathy blood glucose threshold(s), 151 death due to, 272–273, 275, 276 and impaired awareness of hypoglycaemia, 152–153 autonomic symptoms, 30, 31, 142, 143, 196 blood glucose threshold(s), 29, 50, 142, 151, 173 in children, 32 196, 327 compared with neuroglycopenic symptoms, 144 and detection of hypoglycaemia, 35, 142–143 in exam situations, 327 in older people, 240 awareness of hypoglycaemia, 142–145, see also impaired awareness of hypoglycaemia classification of, 146–147 enhancement by caffeine, 110 external cues, 145 factors influencing, 144 internal cues, 36, 142–144, 145 behavioural disturbance (symptom), in children, 32, 33, 43 beta-adrenergic sensitivity, effect of impaired awareness of hypoglycaemia, 153 beta-agonists, nocturnal hypoglycaemia affected by, 94
INDEX beta-endorphins, changes during hypoglycaemia, 13, 14 beverages, hypoglycaemia treatment using, 114, 208 biochemical definitions (of hypoglycaemia), 49–50, 192 Blood Glucose Awareness Training (BGAT) programme, 34, 164 blood glucose monitoring detection of asymptomatic hypoglycaemia by, 51, 113–114, 121–122, 145, 182 for impaired awareness patients, 164 practical problems, 182 target ranges, 184, 223 training in use, 182–183 unwillingness to do, 184 workplace, 326 blood glucose thresholds autonomic responses, 29, 50, 142, 151, 173, 196 cognitive dysfunction, 37, 38, 39, 41, 50, 142, 159, 173 counterregulatory responses, 9, 50, 126, 134, 142, 152 effect of antecedent hypoglycaemia, 159 Diabetes UK recommendation, 50, 114, 173, 192 driving performance deterioration, 318 effect of age, 239, 241 neurophysiological dysfunction, 50, 142 nocturnal hypoglycaemia, 84 85 severe neuroglycopenia, 142 warning symptoms, 29, 50, 142, 149–152, 196 effect of antecedent hypoglycaemia, 159, 173 effect of strict glycaemic control, 41, 50, 150, 152, 155 blood glucose variability, and risk of hypoglycaemia, in type 2 diabetes, 246 blood pressure, changes due to hypoglycaemia, 16, 276, 277 blood-to-brain glucose transport, 6, 133, 155 brain counterregulatory responses, 124, 125–126 effects of glucose deprivation on metabolism, 5–7 regional variation, 6, 179 reversible effects of hypoglycaemia, 290 structural changes, 294 hypoglycaemia-associated, 296–301
INDEX brain damage causes, 6, 116, 203, 205, 273, 285, 293 deaths due to, 268, 270 brain fuel, 5, 6, 191, 203 transport mechanism, 6, 132 and counterregulatory response failure, 132–133 brain glycogen supercompensation hypothesis, 133 brain injury, hypoglycaemia-induced, 301–302 brain metabolism alteration due to hypoglycaemia, 133, 176 children, 191, 192, 202 ‘brittle’ diabetes, 130, 136 Bucharest-Dusseldorf Study, 65, 66 bus drivers, 323, 326, 329 C-peptide negativity, and risk of severe hypoglycaemia, 72, 174 caffeine enhancement of hypoglycaemia awareness by, 110, 163–164 pharmacological actions, 109, 133, 134, 162 sleep pattern affected by, 111 symptoms intensity affected by, 35, 110–111 carbohydrate/insulin mismatch, 62, 101 cardiac arrhythmias and coronary heart disease, 276–278 death due to, 272 and hypoglycaemia, 273–276 cardiovascular disease, effect of hypoglycaemia, 276–279 catecholamines changes during hypoglycaemia, 8, 9, 13, 15 effect of alcohol, 106 107, 108 causes of hypoglycaemia, 7, 61–63, 102 alcohol consumption, 62, 103–109 patient error, 62, 103 central nervous system (CNS) action of insulin on, 125–126 effects of glucose deprivation on metabolism, 5–7 structural changes in diabetes, 294 cerebral adaptation, 6, 41, 151, 155, 177–179 cerebral atrophy, 295, 297, 298, 299, 300 prevalence with increasing age, 298 cerebral blood flow changes during hypoglycaemia, 16, 17, 160, 294–296
335 cerebral glucose sensors, 10, 13, 177 cerebral glycogen metabolism, 133, 301 cerebral oedema, 116, 293 cerebral trauma, children, 204–205 childhood hypoglycaemia cognitive impairment caused by, 204–206, 234, 286–287 long-term effects, 204–205 short-term effects, 204 consequences, 204–207 Nocturnal, 83–85, 197, 203–206 preventative management of, 206–208 risk factors, 197–202 age, 200 clinic structure/size, 201 diet, 199 genetics, 201 intensive insulin therapy, 198–199 physical activity, 200–201, 315 sleep, 198 strict glycaemic control, 197–198 children counterregulatory responses, 133–134, 202–203 definition of hypoglycaemia, 191–193 learning ability affected by hypoglycaemia, 203 sporting activities, 200–201, 317 and strict glycaemic control, 185, 205 symptoms of hypoglycaemia, 32–33, 196–197 chlorpropamide, 247, 249, 251 choice/decision reaction time, effect of hypoglycaemia, 38, 39, 179, 203 clinical definitions (of hypoglycaemia), 50–51, 112–114, 172–173 cognitive dysfunction blood glucose thresholds, 37, 38, 39, 41, 142, 159, 173 factors affecting, 40–41, 103, 161 hypoglycaemia-induced, 37–43 in children and infants, 202–204, 232, 286–287, 326, 327 in examinations [academic], 326, 327 and nocturnal hypoglycaemia, 89, 204 real-life implications, 41–43 cold-induced hypoglycaemia, 316 college examinations, 326–327 colleges, emergency treatment arrangements, 326
336 complications due to diabetes, see also microvascular complications in pregnancy, 230–234 risks to fetus/infant, 231–233 risks to mother, 230–232 concentration difficulty (symptom), 26, 27, 29, 30, 31, 36, 143 children, 196, 327 older people, 239 in pregnancy, 220 confusion (symptom), 26, 30, 31, 36, 196 in older people, 239 congenital malformations incidence in diabetic pregnancy, 221 reduction of risks, 229, 232 continuous glucose monitoring systems (CGMS), 57, 112–114 exercise in childhood study, 200 hypoglycaemia defined, 113 limitations, 113, 114 nocturnal hypoglycaemia studied, 85, 114, 219 treatment of counterregulatory failure, 135 use in detection of (asymptomatic) hypoglycaemia, 113–114, 145, 183, 245 continuous subcutaneous insulin infusion (CSII) compared with multiple injection therapy, 181, 182 adolescents, 208 in type 2 diabetes, 254–255 and exercise, 200, 314 exercise in childhood study, 200 and nocturnal hypoglycaemia, 95, 228 and severe hypoglycaemia, 65, 174, 180, 181 in children and adolescents, 208 treatment of counterregulatory failure, 135 convulsions, 290–293 coordination, lack of (symptom), 26, 29, 30, 31, 196 in older people, 239 cornstarch, nocturnal hypoglycaemia affected by, 93, 208 coronary heart disease, and cardiac arrhythmias, 276–278 cortical atrophy, 297 299, 300 cortisol responses during hypoglycaemia, 9, 13, 15, 130 in children, 202 defects, 130–131 effect of alcohol, 104, 107
INDEX counterregulatory responses, 7–10, 21, 123–126 age-related changes, 241–242 blood glucose threshold(s) for, 9, 50, 126, 134, 142, 152 effect of age, 240, 241 defective, 126–133 causes, 127–128, 129, 131 effect of intensive insulin therapy, 174–175 mechanisms, 131–134 in pregnant women, 221, 222 risk factors, 127, 173 and severe hypoglycaemia, 69 treatment of, 135 effect of alcohol, 62, 103, 105–106 effect of exercise, 63, 313 effect of posture, 87 and nocturnal hypoglycaemia, 86–88, 202 children, 202 normal, 7–10, 123–126 stimulus for, 9, 124–125, 177 principal metabolic effects, 124–125 suppression during sleep, 73, 87–88, 202 in type 2 diabetes, 242–245 cross-sectional studies impaired awareness of hyperglycaemia, 147, 148 severe hypoglycaemia, children, 193 194, 286 dangerous activities/sports, 315, 316 dangerous driving offence, 322 dangerous jobs dealing with, 185 restrictions, 324 ‘dawn phenomenon’, 91, 198 ‘dead in bed’ syndrome, 83, 89, 265, 269–271, 273 possible risk factors, 273 deaths due to hypoglycaemia, see also sudden death certification, 266 estimates of incidence, 267 post-mortem diagnosis, 266–267 risk factors, 267–268 decision reaction time, effect of hypoglycaemia, 38, 39, 203 definitions (of hypoglycaemia), 49–51, 112–114, 172–173 children, 191–193 detection of (asymptomatic) hypoglycaemia, 51, 113–114, 121–122, 145, 182
INDEX dextrose (glucose) intravenous solution, 115–116 Diabetes Control and Complications Trial (DCCT) children, severe hypoglycaemia, 194, 197 cognitive function affected by hypoglycaemia, 289 complications of type 1 diabetes, 49, 63 definition of severe hypoglycaemia, 51 HbA1c as risk factor for severe hypoglycaemia, 67–68, 121 hypoglycaemia in pregnancy, 217 intensive compared with conventional insulin therapy, 65, 66, 172 in pregnant women, 226–227, 228 nocturnal hypoglycaemia, 73 risk factors examined, 64 subjects in trial, 52, 65, 302 vascular complications reduced, 183 Diabetes in Early Pregnancy (DIEP) study, 219 Diabetes UK on blood glucose lower limit, 50, 114, 173, 191 on sudden unexpected deaths, 268 diabetic encephalopathy, 285, 302–303 diabetic ketoacidosis (DKA) deaths due to, 267, 269, 270 as side effect of intensive insulin therapy, 179 worry about, 205 diabetic pregnancy, see also pregnancy clinical management in, 223–229 combined diabetic antenatal charts, 224−225 counterregulatory responses, 221, 222 frequency of hypoglycaemia, 217–220 insulin regimens in, 226–229 management of delivery, 230 reasons for greater risk of hypoglycaemia, 220–222 timing of delivery, 230 diet, as risk factor in children, 199 dietary advice/intervention in diabetic pregnancy, 228 in nocturnal hypoglycaemia, 92–94, 208 dipeptidyl peptidase IV (DPP-IV) inhibitors, 255 disobedience (symptom), in children, 33, 312 dizziness (symptom), 26, 27, 29 children, 32 196 older people, 240 Dose Adjustment For Normal Eating (DAFNE) trial, 172
337 doses of insulin, effects on hypoglycaemic risk, 73 Driver and Vehicle Licensing Agency (DVLA), notifications, 319 driving advice for diabetic drivers, 320–322 advice on alcohol consumption, 108 effect of hypoglycaemia, 41–42, 231, 318 and risk of accidents, 318–319 medico-legal aspects, 322 by pregnant women, 231 reasons to cease, 321 driving licences restriction of, 318 vocational, 320 drowsiness (symptom), 26, 30, 31, 196 children, 32 196, 196, 327 difficulty in interpreting, 144 older people, 240 drug, insulin [treated in law] as, 321–322 duration of insulin therapy and defective responses to hypoglycaemia, 131, 132 and mild hypoglycaemia 55 impaired awareness of hypoglycaemia affected by, 69, 147, 148 as risk factor for severe hypoglycaemia, 64, 68, 69–70, 102 type 2 diabetes, 252, 253, 254 early-warning symptoms, 27, 29, 30, 144 actions taken by patient, 114 Edinburgh Hypoglycaemia Scale, 37 symptom groupings and classification, 30, 31 education/training programmes, 34, 66, 67, 121, 122, 135, 183 for children, 207–208 elderly people, see older people electrocardiogram (ECG), QT interval measurements, 274, 275 electroencephalograms (EEG) children with diabetes, 287, 292 limitations in diagnosing convulsions, 295 emergency services employment, 320, 324 and severe hypoglycaemia, 58, 60, 115–116 type 1 compared with type 2 diabetes, 255 emotions during hypoglycaemia, 43–44, 312–313
338 employment aspects, 323–326 restricted jobs [listed], 324 specialist medical reports, 326 workplace-occurring hypoglycaemia, 325–326 encephalopathy, diabetic, 285, 302–303 endocrinopathies, and risk of severe hypoglycaemia, 74, 131, 330 endogenous insulin secretion absence affecting risk of severe hypoglycaemia, 72 effect of strict glycaemic control, 174 Epidemiology of Diabetes Interventions and Complications (EDIC) Study, 289 epinephrine (adrenaline) actions, 124, 142, 245, 276 defective response, 8, 69, 129, 130, 131, 132 136 during sleep, 87, 88, 203 effect of alcohol, 106, 107 response to hypoglycaemia, 124, 275, 276 blood glucose threshold, 125 in children, 134, 202 effect of glycaemic control, 178 in older people, 241, 242 type 2 diabetes, 244, 245 euglycaemia, 1 mechanisms maintaining, 1–5 EURODIAB IDDM Complications Study, 64, 68, 73 exam-induced anxiety, and symptoms of hypoglycaemia, 326, 327 examinations [academic], 326–327 Exenatide, 255 exercise benefits, 63, 312 counterregulatory response affected by, 63, 313 metabolic consequences, 312 safest time for, 314 exercise-induced hypoglycaemia, 63, 314 children, 200–201 delayed, 316 factors affecting, 313 management/prevention of, 206–207, 313–315 in type 2 diabetes, 246 fasting, metabolic pathways, 3–4 fatty acids
INDEX metabolic alterations, 3, 5, 9, 125 effect of alcohol, 104 fear of hypoglycaemia, 43–44, 74, 121, 172, 310–311 children, 206, 311 in pregnancy, 221 reasons for, 310 flushing (clinical sign), 17 fetus, effects of maternal hypoglycaemia, 231–233 folate/folic acid supplement, 229 frequency of hypoglycaemia, 51–60 asymptomatic hypoglycaemia, 55–57 in diabetic pregnancy, 219–222 factors affecting various studies, 51–52 mild hypoglycaemia, 52–55, 246 severe hypoglycaemia, 57–60, 122, 123, 198, 246 children, 194−195 pregnant women, 219, 220 in type 1 diabetes, 51–60 in type 2 diabetes, 246–256 and incretin mimetics, 255 and insulin, 253–256 and oral antidiabetic agents, 247–250 frequency of sudden death, 272 functional changes during hypoglycaemia, 18–20 gastric emptying, factors affecting, 20, 74, 102 gastrointestinal system, changes in blood flow during hypoglycaemia, 16–17, 17 gastroparesis diabeticorum, 74 genetic predisposition to hypoglycaemia, 70–72, 174, 201 genetically engineered insulin, 72, 161–162 gestational diabetes, 220 insulin management in, 228, 229 screening for, 233 glibenclamide, 247, 249, 251 gliclazide, 249 modified-release form, 249, 249 glimepiride, 249, 249 glipizide, 247, 249 glucagon actions, 2, 3, 124 response to hypoglycaemia, 8, 13, 15, 125, 127, 136 in children, 201 effect of alcohol, 104 impairment of, 127–129, 131, 132
INDEX in older people, 241, 242 type 2 diabetes, 243, 244, 244 in treatment of hypoglycaemia, 115, 207, 229 glucagon-like peptide 1 (GLP-1) analogues, 255 gluconeogenesis, 2, 3, 123 alcohol-induced suppression of, 62, 104 glucose gel (for treatment of hypoglycaemia), 115, 230 glucose homeostasis, 1–5 changes during pregnancy, 218 effect of exercise in childhood, 200–201 in fasting, 3–4 in fed (post-prandial) state, 4, 5 24-hour glucose and insulin profiles, 61 glucose meters, limitations, 111 glucose-sensing neurones, 10, 13 glucose transporters (GLUTs), increase in activity, 6, 132, 155 glycaemic control, see also strict glycaemic control in diabetic pregnancy, 223 as risk factor for severe hypoglycaemia, 67–68, 102 in children, 197–198 glycaemic thresholds, see blood glucose thresholds glycated haemoglobin (HbA1c concentration, see also strict glycaemic control in asymptomatic hypoglycaemia studies, 56, 57 in DCCT, 65, 147, 184 intensive insulin therapy patients, 65, 66, 67, 68, 147 in mild hypoglycaemia studies 53, 54, 55 in nocturnal hypoglycaemia studies, 84 in severe hypoglycaemia studies, 59, 65, 66, 67, 68, 102 glycerol, as glucose precursor, 2, 3, 124, 125 glycogen, 2, 3 glycogenolysis, 2, 3, 123 continuation after death, 266 glycolysis, 2 growth hormone (GH) responses during hypoglycaemia, 9, 13, 14, 130 in children, 202–203 defects, 130–131 effect of alcohol, 104–105, 106, 107, 131 type 2 diabetes, 244
339 haemodynamic changes during hypoglycaemia, 15–16, 276 headache (symptom), 26, 30 children, 32 196 in pregnancy, 221 heart rate change due to hypoglycaemia, 16, 19, 276, 277 as symptom, 26, 29, 31, 142, 143 hemiparesis, hypoglycaemic, 205, 290, 293 hepatic autoregulation, 9, 10, 123 hepatic glucose production, 3, 4, 313 effect of impaired glucagon response, 128–129 holidays, planning for, 316, 323 home blood glucose monitoring, 55–56, 65 anxiety caused by, 172 driving, 322–323 in diabetic pregnancy, 223, 225 hormonal changes during hypoglycaemia, 10–15 counterregulation, 9, 21, 123–126, 175 during pregnancy, 218 hospital admissions/referrals, 58, 60, 62, 115, 116 type 2 diabetes, 249 human insulin, 72, 161–163 compared with animal insulins, 162 controversy over introduction, 72, 162, 268 and counterregulatory responses, 135 hunger (symptom), 26, 27, 29, 30, 31, 36, 196 children, 32, 196, 196 Hvidore Study Group, 199 hyperinsulinaemia effect of intensive insulin therapy, 181 nocturnal, 198 hyperinsulinaemic clamp technique alcohol influence studies, 105, 107 cognition studies, 37–38, 41 counterregulatory responses, 241, 244 in pregnant women, 221 Hypoglycaemia-Associated Autonomic Failure (HAAF), 69, 129, 130, 153–154, 244 risk-reduction strategies, 135 Hypoglycaemia Fear Survey (HFS), 43–44, 74, 311 hypoglycaemic hemiplegia, 205, 290, 293 hypoglycaemia unawareness, see impaired awareness of hypoglycaemia Hypoglycemia Anticipation, Awareness and Treatment Training (HAATT) programme, 34
340 hypopituitarism, 74, 102, 130 hypothalamus, hormonal changes during hypoglycaemia, 13–14, 13 hypothermia, and hypoglycaemia, 317 iatrogenic causes of hypoglycaemia, 7 idiopathic epilepsy, misdiagnosis of, 291–293 illicit drinks, intoxication due to, 104 impaired awareness of hypoglycaemia and asymptomatic hypoglycaemia, 154–155 cognitive function affected by, 161 definition(s), 141, 145–146 effect of alcohol, 62, 104, 108 effect of duration of insulin therapy, 69, 147, 148 effect of strict glycaemic control, 147, 173, 176 mental performance affected by, 41, 146 methods of assessment, 145–147 nocturnal hypoglycaemia, 73, 88–89 pathogenesis, 149–161 in pregnant women, 221 prevalence, 147–149 as risk factor for severe hypoglycaemia, 69, 152, 161 treatment of, 163–165 in type 2 diabetes, 245–246 and undetected hypoglycaemia, 145 and worry, 44 incidence, see frequency of hypoglycaemia incretin mimetics, and hypoglycaemia, 255 infants, see also children developmental effects of maternal hypoglycaemia, 233–234 prevalence of severe hypoglycaemia, 194–195 inhaled insulin, 72, 254 injection site problems, 101, 135 insulin actions, 2, 5, 7 24-hour profile in glucose homeostasis, 61 insulin analogues, 72, 94–95, 135 combined with metformin in type 2 diabetes, 254 and nocturnal hypoglycaemia, 94–95, 183 in children, 206 risk of hypoglycaemia affected by, 72, 174 in children, 199–200, 208–209 in pregnancy, 227 type 2 diabetes, 254 insulin delivery systems, see also continuous subcutaneous insulin infusion (CSII)
INDEX for children, 208 limitations, 86 for pregnant women, 226–229 risk factors affected by, 65, 101–102 insulin detemir, 95, 254 insulin glargine, 95, 208, 228, 254 insulin glulisine, 254 insulin-like growth factor 1 (IGF-1), glucagon response to hypoglycaemia affected by, 127 insulin lispro, 72, 95 children and adolescents, 208 in diabetic pregnancy, 227, 228 type 2 diabetes, 254 insulin requirements for children, 197 in pregnancy, 219 insulin resistance in adolescents, 197 in pregnancy, 218 insulin sensitivity, factors affecting, 102, 106, 181 intellectual activity, glucose consumption caused by, 327 intelligence quotient (IQ) effect of hypoglycaemia, 288 in children, 205, 234, 286 intensive insulin therapy cognitive dysfunction affected by, 41 contraindication for impaired awareness patients, 184 HbA1c levels, 65, 66, 67, 68, 147 and impaired awareness of hypoglycaemia, 147, 173 nocturnal hypoglycaemia affected by, 84, 85 pregnant women, 227 as risk factor for severe hypoglycaemia, 63–67, 102, 173–177, 285, 287, see also strict glycaemic control in children and adolescents, 199–200, 208 risk of hyperinsulinaemia, 181 risk of ketosis, 179, 180 International Society for Paediatric and Adolescent Diabetologists (ISPAD) classification of hypoglycaemia, 192, 193 guidelines, 206, 209 interstitial glucose concentration, relationship with blood glucose, 113, 114 interstitial glucose monitoring, 57, 112–114, see also continuous glucose monitoring systems
INDEX limitations, 112, 113 use in detection of hypoglycaemia, 113–114, 121–122 interstitial hypoglycaemia, 112, 113 intravenous injection (in treatment of hypoglycaemia), 115–116 irritability (symptom), 43, 312 in children, 32, 32, 33, 196, 327 isophane (NPH) insulin, 53, 54, 56, 57, 59, 66 compared with insulin analogues, 95, 208, 254 and nocturnal hypoglycaemia, 95, 163 children and adolescents, 208 in pregnancy, 227, 228 ketones metabolism, in brain, 6 production, during fasting, 3 ketosis effect of strenuous exercise, 314 as side effect of intensive insulin therapy, 179, 180 kidney, changes in blood flow during hypoglycaemia, 17, 18 (kidney), contribution to gluconeogenesis, 2 labour, management of diabetes during, 230, 231 lactate, as glucose precursor, 2, 124, 125 language comprehension, effect of hypoglycaemia, 42 large goods vehicle (LGV) licences, 320, 324, 326 learning ability of children, effect of hypoglycaemia, 203 leukoaraiosis, 295–296, 297, 299, 300–301 lifestyle management, in diabetic pregnancy, 229 lifestyle moderators, see also alcohol consumption; caffeine; exercise effect on hypoglycaemia type 1 diabetes, 103–111 type 2 diabetes, 246 light-headedness (symptom), 26, 27, 29, see also dizziness (symptom) children, 327 in older people, 240 limited life expectancy patients, unsuitability for strict glycaemic control, 184, 185 lipolysis, 9, 123, 124 alcohol-induced suppression of, 104
341 liver changes in blood flow during hypoglycaemia, 17 metabolic pathways, 2, 3, 3, 4, 5 long distance running, 316 long QT syndrome, 274 lorry drivers, 320, 327 management of hypoglycaemia, 114–116 children, 207–209 pregnant women, 230 marathon running, 316 maternal hypoglycaemia risks to fetus/infant, 232–234 risks to mother, 231 memory impairment, hypoglycaemia-associated, 42–43, 111 mental arithmetic, effect of hypoglycaemia, 38 mental performance impairment, hypoglycaemia-impaired, 37–40, see also cognitive dysfunction in children, 204–205, 286–287 factors affecting, 40–41 real-life implications, 41–43 metformin, frequency of hypoglycaemia, 247, 251 microvascular complications in diabetic pregnancy, 232 effect of recurrent hypoglycaemia, 279 effect of strict glycaemic control, 63, 171, 184, 197, 289 frequency of severe hypoglycaemia affected by, 73–74 reduction in prevalence, 122 mild hypoglycaemia attitudes to, 325 definition(s), 50, 51, 112–114, 172, 192, 193, 246 frequency, 52–55, 174 management of, 114–115, 207 minibus drivers, 320 model for occurrence and avoidance of hypoglycaemia, 34 moderate hypoglycaemia, 51, 172, 193 moderators of hypoglycaemia type 1 diabetes, 103–111 type 2 diabetes, 245–246 monitoring, 111–114 continuous systems, 57, 85, 112–114
342 mood changes due to hypoglycaemia, 43, 314 in examinations [academic], 326, 327 recovery from, 293 morbidity of hypoglycaemia, 255–256 type 2 diabetes, 256 mortality, 270 risk factors, 267–272 motor skills, effect of hypoglycaemia, 38 multiple injection therapy children and adolescents, 199 compared with CSII, 180, 181 adolescents, 208 and nocturnal hypoglycaemia, 84, 85 and severe hypoglycaemia, 65, 181, 182 treatment of counterregulatory failure, 135 muscle changes in blood flow during hypoglycaemia, 17−18, 17 metabolic pathways, 2, 3, 4, 5, 200 sympathetic activity, 11–12 myocardial infarction, 279 myocardial ischaemia, 278–280 nateglinide, 250 nausea (symptom), 26, 30 in children, 32 196 neonates, risks of maternal hypoglycaemia, 232–234 nephropathy, 73, 279 in diabetic pregnancy, 227, 232 reduction in risk, 184 nervous system, see autonomic nervous system; central nervous system (CNS) nervousness (symptom), 26, 27, 29, 31 neuroendocrine activation during hypoglycaemia, 13–15 neuroglycopenic symptoms, 30, 31, 143–144, 196 blood glucose threshold(s), 29, 50, 142, 151 in children, 32 196, 327 compared with autonomic symptoms, 144 confusion with alcohol intoxication, 62 and detection of hypoglycaemia, 35–36, 142–143 in exam situations, 326 in older people, 240 neurological symptoms, in older people, 240, 240 neurological syndromes, hypoglycaemia-induced, 289–290
INDEX neurophysiological dysfunction, blood glucose thresholds, 50, 142 neuropsychological deterioration due to severe hypoglycaemia evidence for, 289–290 long-term manifestations, 295–296 transient manifestations, 290 nocturnal hypoglycaemia advice to patients, 276 caffeine-associated reduction, 110 causes, 74, 86–88 in children, 197 and cognitive function, 204 effect of exercise, 201, 209 clinical solutions, 92–95 continuous subcutaneous insulin infusion, 95, 183 dietary approach, 92–94, 208 pharmaceutical interventions, 94 timing and type of insulin, 94–95, 163, 183, 206, 227, 228 consequences, 88–89 effect of exercise, 200, 315 epidemiology, 63, 76, 83–85 frequency, 84 and impaired awareness, 74, 88–89, 200 neurological consequences, 89 prediction of, 90 in children, 206 in pregnancy, 219 sudden death associated with, 83, 89, 267, 271–273, 274, 278 typical overnight glucose profiles, 85 122 nocturnal insulin requirements, 95 for children, 199 novel insulins, see insulin analogues obese people, counterregulatory responses, 134 observational data, effect of intensive therapy on severe hypoglycaemia, 66, 72 ‘odd behaviour’ (symptom), 30, 31, 32, 43, 196 in children, 32 mistaken for alcohol inebriation, 327 offshore oil industry, employment in, 324 older people counterregulatory responses, 134, 241–242 symptoms of hypoglycaemia, 33, 239–240, 240 blood glucose thresholds, 240, 241 unsuitability for strict glycaemic control, 184, 185
INDEX oral antidiabetic agents, see also alpha-glucosidase inhibitors; metformin, frequency of hypoglycaemia; sulphonylureas; thiazolidinediones hypoglycaemic effects, 247–252 compared with insulin therapy, 251–253 oxytocin, changes during hypoglycaemia, 13, 14–15 pallor (clinical sign), 17 palpitations (symptom), 30, 31, 36, 196 difficulty in interpreting, 37 pancreas, hormonal changes during hypoglycaemia, 13, 15, 21 pancreatic polypeptide, changes during hypoglycaemia, 13, 15 parasympathetic nervous system, 10, 11 activation during hypoglycaemia, 12, 143 passenger carrying vehicle (PCV) licences, 320, 324 patient ‘error’, as ‘cause’ of hypoglycaemia, 62, 103 peripheral autonomic neuropathy and impaired awareness of hypoglycaemia, 152–153 and severe hypoglycaemia, 73–74, 110 pessimism during hypoglycaemia, 43, 312 phobia about hypoglycaemia, 43, 312 physical activity (children) management of, 208–209 as risk factor, 200–201 physical exhaustion, and hypoglycaemia, 317 physically demanding jobs, and strict glycaemic control, 185 physiological responses to hypoglycaemia, 15–20, 142, 143 pituitary failure, 130 pituitary function, effect of hypoglycaemia, 13 pituitary gland, hormonal changes during hypoglycaemia, 13–15, 21, 124 placenta, in mother with diabetes, 218 police custody, hypoglycaemia while in, 327–329 positron emission tomography (PET), brain metabolism studies, 6, 132, 161, 177 post-mortem diagnosis, 266–269 posterior pituitary gland, hormonal changes during hypoglycaemia, 13, 14–15 postprandial metabolic pathways, 3, 4 posture, counterregulatory responses affected by, 87
343 potassium levels, fall due to hypoglycaemia, 273, 276 ‘pounding’ heart (symptom), 26, 29, 31, 142, 143 children, 32 older people, 240 in pregnancy, 221 prandial glucose regulators, 249 pregnancy, see also diabetic pregnancy average age of mothers with diabetes, 217 frequency of hypoglycaemia, 219–222 lifestyle management in, 229 maternal complications due to diabetes, 230–231 metabolic changes during, 217–219 organisation of clinical care, 223–226 pre-conception care/counselling, 223 prevalence of hypoglycaemia mild hypoglycaemia, 54, 246 nocturnal hypoglycaemia, 83, 193, 196 severe hypoglycaemia, 121, 198 in pregnancy, 220, 227 in type 2 diabetes, 253, 256 previous history of hypoglycaemia, see also antecedent hypoglycaemia as risk factor, 64, 68, 69, 102, 157–160 in pregnancy, 221 and worry, 44 prison, management of diabetes in, 330–331 prison officer, as restricted occupation, 326 profound hypoglycaemia counterregulatory responses, 134 hepatic autoregulation in, 9 increased cerebral blood flow in, 123 prolactin, changes during hypoglycaemia, 13, 14 prospective studies compared with retrospective studies, 51–52, 172 mild hypoglycaemia, 53, 54–55 severe hypoglycaemia, 57, 58, 59 children, 194 195, 286 type 2 diabetes, 248−249, 253, 254 protein snack, and nocturnal hypoglycaemia, 93 psychiatric disturbance/illness effect of hypoglycaemia, 295 as risk factor for death, 58, 269 treatment of, 281 psychological consequences of hypoglycaemia, 311
344 psychological factors, and risk of hypoglycaemia, 74 psychosocial manifestations of hypoglycaemia, 290, 310–312 reaction time, effect of hypoglycaemia, 38, 39 reactions to hypoglycaemia, 142, see also symptoms of hypoglycaemia rebound hyperglycaemia, 90–91 clinical approach, 92 recreational activities, 317 regional blood flow, changes during hypoglycaemia, 16–18 renin, changes during hypoglycaemia, 15 repaglinide, 249 retinopathy in diabetic pregnancy, 227, 232 effect of hypoglycaemia, 73, 279 reduction in prevalence, 122 and strict glycaemic control, 184 retrospective studies compared with prospective studies, 51–52, 172 mild hypoglycaemia, 52, 53 severe hypoglycaemia, 57, 58, 59, 66 in children, 196, 205 type 2 diabetes, 248−249, 252, 254 risk factors for hypoglycaemia, 63–75, 101–103, 161, 173–177 in children, 197–202 for severe hypoglycaemia, 63–75, 149, 161, 173–177 for sudden death, 272–275 road traffic accidents (RTAs), hypoglycaemia-induced, 319–320 rosiglitazone, 247 school examinations, 326–327 schools, emergency treatment arrangements, 328 Scottish Diabetes in Pregnancy Study, 219, 220 severe hypoglycaemia definitions, 51, 51, 172, 193, 220, 246, 286 effect of intensive insulin therapy, 63–67, 173–177, 287 in children and adolescents, 199–200, 208 emergency service involvement, 58, 256–257 fear of future episodes, 74, 311, 312 frequency, 57–60, 196 children, 192−193 pregnant women, 219, 220
INDEX hospital referrals, 58, 60, 115, 116 management of, 115–116, 208 neuropsychological deterioration due to, 287–290 rates (1992–2002), 121, 123, 198 risk factors, 63–75, 149, 161, 173–177 sulphonylurea therapy, 247 type 2 diabetes compared with type 1 diabetes, 256 insulin therapy, 253–256 oral antidiabetic agents, 247, 250 without warning symptoms, 65, 147, 162 and duration of insulin therapy, 69, 147 severe neuroglycopenia, blood glucose threshold, 142 shaking, see trembling (symptom) shift work, 325–326 single photon emission [computed] tomography (SPE[C]T) brain imaging study, 287 cerebral blood flow studies, 16, 297 skin, changes in blood flow during hypoglycaemia, 17 sleep, counterregulatory responses affected by, 73, 87–88, 203 sleep patterns counterregulatory response affected by, 88, 203 effect of caffeine, 110 sleepiness (symptom), 26, 36 see also drowsiness (symptom) smoking, risk of hypoglycaemia affected by, 75 socio-economic factors, risk of hypoglycaemia affected by, 74 somatostatin action of, 8, 14, 15 changes during hypoglycaemia, 13, 15 Somogyi phenomenon, 90–91, 206 speech difficulty (symptom), 26, 30, 31, 36, 196 children, 32 older people, 240 spleen, changes in blood flow during hypoglycaemia, 17, 18 sporting activities, 315–317 Stockholm Diabetes Intervention Study (SDIS), 64, 65, 66, 289 strenuous exercise, 313, 314, 315, 317 strict glycaemic control, see also intensive insulin therapy contraindications, 184 and fear of hypoglycaemia, 121, 310
INDEX and impaired awareness of hypoglycaemia, 147 mental performance affected by, 41 patients unsuitable for, 183–184 elderly, 182 young children, 185, 205 as risk factor for severe hypoglycaemia, 64, 67–68, 102, 172, 287 in children and adolescents, 197 risks, 171–190 side effects, 95, 171, 179–181 symptoms onset affected by, 41, 50, 150, 152, 155 Stroop test, effect of hypoglycaemia, 38 subcutaneous insulin injection, see also continuous subcutaneous insulin infusion (CSII) factors affecting absorption, 101–102 sudden death, 269–274, see also deaths due to hypoglycaemia and ‘Dead in bed’ Syndrome frequency, 272 possible mechanisms, 269 risk factors, 274–278 syndromes in non-diabetic young people, 270–271 suicide brain damage following attempt, 293 deaths due to, 271, 272 sulphonylureas, see also chlorpropamide; glibenclamide; gliclazide; glimepiride; glipizide; tolbutamide drug interactions, 247, 250 hypoglycaemic effects, 245, 247−248, 250, 252 pharmacokinetics, 249 supine posture, counterregulatory responses affected by, 87 sweating (symptom), 12, 18, 19, 26, 27, 30, 31, 36, 142, 143, 196 change with duration of insulin therapy, 148 children, 32, 32, 196 difficulty in interpreting, 38, 144 older people, 240 sympathetic nervous system, 10, 12 activation during hypoglycaemia, 10–12, 143 sympatho-adrenal responses to hypoglycaemia, 142, 143 deficient, 129, 132 and impaired awareness of hypoglycaemia, 151
345 symptoms of hypoglycaemia, 19, 25–37, 26, 142 blood glucose thresholds, 29, 50, 142, 149–152 effect of strict glycaemic control, 41, 50, 150, 178, 180 in older people, 240, 241 changes with duration of insulin therapy, 148 in children, 32–33, 196–197 classification, 33 correlation with blood glucose concentration, 28–29 detection of, 35–36 generation of, 34–35, 143 hierarchy affected by intensive insulin therapy, 41, 179, 180 identifying, 19, 25–29, 26 incorrect interpretation, 28, 37 individuality, 27–28 interpretation, 28, 36–37 in older people, 32, 239–240, 240 from perception to action, 33–37 in pregnant women, 221 scoring systems, 37 in type 2 diabetes, 242 systemic mediator theory, and counterregulatory response failure, 131–132 tachycardia (symptom), 16, 19, 26, 29, 31 target blood glucose concentration lower limit, 50, 114, 173, 192 recommended range(s), 185, 223 taxi drivers, 322, 326 tearfulness (symptom), in children, 32 196 temperature, changes during hypoglycaemia, 19–20 tenseness (symptom), 27, 29, 43, 314, 329 teratogenic effects due to hypoglycaemia, 233 terbutaline, and blood glucose levels, 94 therapeutic management (avoidance of hypoglycaemia), 181–184 target blood glucose ranges, 182, 225 thiazolidinediones, 247 thinking slowed (symptom), 26 31, 143 ‘tingling’ (symptom), 26 tiredness (symptom), 26, 27, 30, 31, 43, 312, 327 tolbutamide glucagon response to hypoglycaemia affected by, 128 pharmacokinetics, 249
346 train drivers, 323, 324 transient ischaemic attacks (TIAs), misinterpretation of symptoms, 33, 240 travel arrangements, 322–323 trembling (symptom), 18, 19, 26, 27, 29, 30, 31, 142, 143, 196 change with duration of insulin therapy, 148 children, 32, 32, 196, 196, 197 older people, 240 triglycerides, metabolic alterations, 3, 5 truck drivers, 320, 326 type 1 diabetes frequency of hypoglycaemia, 51–60 asymptomatic hypoglycaemia, 55–57 in children, 191–193 in diabetic pregnancy, 219–222 mild hypoglycaemia, 52–55 severe hypoglycaemia, 57–60, 122, 123, 198 type 2 diabetes frequency of hypoglycaemia, 246–256 incretin mimetics, 255 insulin therapy, 253–256 oral antidiabetic agents, 247–250 impaired awareness of hypoglycaemia, 245–246 moderators of hypoglycaemia, 245–246 pregnant women, 220 risk of severe hypoglycaemia, 173, 177, 247 types of insulin, see also insulin analogues; isophane (NPH) insulin risk of hypoglycaemia affected by, 72, 174 in children, 199–200, 208–209 unexplained deaths, 268–272, see also sudden death unhappiness during hypoglycaemia, 43, 312 United Kingdom Hypoglycaemia Study Group, 53–55, 252
INDEX United Kingdom Prospective Diabetes Study (UKPDS), 247 universities, emergency treatment arrangements, 326 university examinations, 326–327 unsteadiness, as symptom in older people, 240 vascular complications, reduction by intensive insulin therapy, 171 vaso-vagal episode, misinterpretation of symptoms, 240 vasopressin, changes during hypoglycaemia, 13, 14–15 ventricular tachycardia (VT), 274, 277–278 ventromedial nucleus of hypothalamus (VMH), role in counterregulation, 10, 13, 125, 133 verbal fluency, effect of hypoglycaemia, 38 Veterans Affairs Cooperative Study in type 2 Diabetes (VA CSDM), 253–254, 254 Vildagliptin, 255 visual disturbances (symptom), 26 31 children, 32 196 older people, 240 visual processing, effect of hypoglycaemia, 42 vocational driving licences, 320 restrictions, 324 warmness (symptom), 30, 31, 32 warning symptoms, see symptoms of hypoglycaemia weakness (symptom), 26, 29, 31 children, 196, 196 older people, 240 weight gain, and intensive insulin therapy, 171 working at heights, restrictions on, 324 worry about hypoglycaemia, 43–44, 83, 206, 313, see also fear of hypoglycaemia in children, 206 nocturnal hypoglycaemia, 73, 312