Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment

VASCULAR COMPLICATIONS OF DIABETES CURRENT ISSUES IN PATHOGENESIS AND TREATMENT

Editor Richard Donnelly MD, PhD, FRCP, FRACP Division of Vascular Medicine University of Nottingham The Medical School Derby DE22 3DT UK

Associate Edward Horton MD Editor Joslin Diabetes Center One Joslin Place Boston MA 02215 USA

VASCULAR COMPLICATIONS OF DIABETES CURRENT ISSUES IN PATHOGENESIS AND TREATMENT SECOND EDITION

Editor Richard Donnelly MD, PhD, FRCP, FRACP Division of Vascular Medicine University of Nottingham The Medical School Derby DE22 3DT UK

Associate Edward Horton MD Editor Joslin Diabetes Center One Joslin Place Boston MA 02215 USA

Supported by an Educational Grant by Eli Lilly & Co

Answers That Matter.

© 2005 by Blackwell Publishing Ltd Blackwell Publishing, Inc., 350 Main Street, Malden, Massachusetts 02148-5020, USA Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK Blackwell Publishing Asia Pty Ltd, 550 Swanston Street, Carlton, Victoria 3053, Australia The right of the Author to be identified as the Author of this Work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. 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 or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher. First published 2002 Second Edition 2005 Library of Congress Cataloging-in-Publication Data Vascular complications of diabetes: current issues in pathogenesis and treatment / edited by Richard Donnelly and Ed Horton.-- 2nd ed. p. ; cm. Includes bibliographical references and index. ISBN-13: 978-1-4051-2785-1 (alk. paper) ISBN-10: 1-4051-2785-6 (alk. paper) 1. Diabetic angiopathies. [DNLM: 1. Diabetes Complications. 2. Diabetic Retinopathy--etiology. 3. Protein Kinase C--adverse effects. 4. Vascular Diseases--etiology. WK 835 V3305 2005] I. Donnelly, Richard, 1960- II. Horton, Edward S. RC700.D5V375 2005 616.1'3--dc22 2005008488

ISBN-13: 978-1-4051-2785-1 ISBN-10: 1-4051-2785-6 A catalogue record for this title is available from the British Library Set in Branding Serif and Branding Sans Design and layout by Designers Collective Printed and bound in the USA by Walsworth Publishing Co., Marceline, Missouri Commissioning Editor: Alison Brown Development Editor: Claire Bonnett Production Controller: Kate Charman For further information on Blackwell Publishing, visit our website: http://www.blackwellpublishing.com The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards.

CONTENTS

List of contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .v Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vi

SECTION I MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES Chapter 1 The public health impact of the diabetes epidemic . . . . .3 Adrian R. Scott Chapter 2 Risk factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Adrian R. Scott Chapter 3 Diabetic nephropathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23 Adrian R. Scott Chapter 4 Coronary heart disease and diabetes . . . . . . . . . . . . . . . . . . . . . .35 Adrian R. Scott Chapter 5 Diabetes and cerebrovascular disease . . . . . . . . . . . . . . . . . . . . .45 Adrian R. Scott Chapter 6 Erectile dysfunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51 Adrian R. Scott Chapter 7 Evidence-based interventions to prevent or retard vascular complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59 Adrian R. Scott

SECTION II DIABETIC NEUROPATHIES Chapter 8 Classification and clinical features of neuropathy . . . . . . . . 79 Rayaz A. Malik Chapter 9 Pathophysiology of diabetic neuropathy . . . . . . . . . . . . . . . . . .85 Rayaz A. Malik Chapter 10 Epidemiology and natural history of DPN . . . . . . . . . . . . . . . . .91 Rayaz A. Malik Chapter 11 Detection/Screening/Assessment . . . . . . . . . . . . . . . . . . . . . . . . . .97 Rayaz A. Malik Chapter 12 Foot ulceration and Charcot arthropathy . . . . . . . . . . . . . . . . .105 Rayaz A. Malik

iii

iv

CONTENTS

Chapter 13 Treatments options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .113 Rayaz A. Malik Chapter 14 Management guidelines for diabetic peripheral neuropathy and foot ulceration . . . . . . . . . . . . . . . . . . . . . . . . . . .121 Rayaz A. Malik

SECTION III DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS Chapter 15 Diabetic retinopathy: epidemiology and risk factors . . . .129 Hean-Choon Chen Chapter 16 Classification and diagnosis of diabetic retinopathy . . . . .139 Hean-Choon Chen Chapter 17 Diabetic maculopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151 Hean-Choon Chen Chapter 18 Proliferative diabetic retinopathy . . . . . . . . . . . . . . . . . . . . . . . . .163 Hean-Choon Chen Chapter 19 Non-retinal diabetic ocular complications . . . . . . . . . . . . . . . .171 Hean-Choon Chen

SECTION IV MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION Chapter 20 Pathophysiology and potential targets for therapeutic intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .179 Richard Donnelly Chapter 21 Protein kinase C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .189 Richard Donnelly Chapter 22 Protein kinase C activation and vascular permeability . . .197 Richard Donnelly Chapter 23 Role of protein kinase C activation in cardiovascular and renal complications of diabetes . . . . . . . . . . . . . . . . . . . . . .205 Richard Donnelly Chapter 24 Experimental pharmacology using isoform-selective protein kinase C inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .213 Richard Donnelly Chapter 25 Clinical trials with ruboxistaurin . . . . . . . . . . . . . . . . . . . . . . . . . .221 Richard Donnelly Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .226

LIST OF CONTRIBUTORS

Hean-Choon Chen FRCS, FRCOpath Consultant Ophthalmologist Derbyshire Royal Infirmary Derby, UK

Richard Donnelly MD, PhD, FRCP, FRACP Professor of Vascular Medicine University of Nottingham and Honorary Consultant Physician The Medical School Derby, UK

Rayaz A. Malik MB.ChB, PhD, MRCP Senior Lecturer and Consultant Physician Academic Department of Medicine Manchester Royal Infirmary Manchester, UK

Adrian R. Scott MD, FRCP Consultant Physician Diabetes Centre Northern General Hospital Sheffield, UK

v

PREFACE

Diabetes-related cardiovascular complications often cause premature mortality, as well as disabilities such as blindness, foot ulceration and amputation. The health care and social care costs of managing these complications are enormous, but new treatments, devices and clinical management protocols are steadily improving the longer term outcomes for people with diabetes. This second edition has been revised and updated to reflect state of the art clinical practice. In particular, a new section on diabetic neuropathy covers important aspects of screening and detection, diagnosis and management. The book is aimed at healthcare professionals involved in the assessment and surveillance of patients with diabetes complications, and the section on protein kinase C (PKC) explains the basis of a major new pathway responsible for hyperglycaemia-induced vascular injury. Recent clinical trials have suggested that inhibition of PKC-β is an effective therapeutic intervention for improving the symptoms and outcomes from diabetes-related complications. Richard Donnelly

vi

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

SECTION I MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Adrian R. Scott MD, FRCP CHAPTER 1 THE PUBLIC HEALTH IMPACT OF THE DIABETES EPIDEMIC

INTRODUCTION The 21st century will see diabetes emerge as the world’s commonest chronic disease. Whilst the bulk of this will be type 2 diabetes (90%) the incidence has been rising in both types. The direct and indirect costs of diabetes and its complications, plus the associated reduction in quality and quantity of life, will have considerable economic consequences. This effect will be most noticeable in developing countries which are going to see a disproportionate increase in the prevalence of diabetes over the next few decades. It has been estimated that the worldwide prevalence of diabetes will double between 1990 and 2010. Epidemiological studies in the USA have shown that the number of people with known diabetes has increased from around 1.5 million in 1958 to 10.5 million in 1998. Most states in the US report a prevalence of over 8% and this fails to take into account those people with undiagnosed diabetes. Most screening studies indicate that at least 50% of people found to have diabetes were silently undiagnosed for sometime.

THE NATURAL HISTORY OF TYPE 1 DIABETES Although onset is predominantly in childhood or young adulthood, a significant proportion will be diagnosed over the age of 30 years. The peak ages for onset, however, are around puberty and between 4–6 years old. Life expectancy is reduced, though there is some evidence that this is improving. The British Diabetic Association Cohort study (1972–1993), a prospective follow-up of insulin-treated patients with diabetes diagnosed under the age of 30, showed increased mortality at all ages. Avoidable metabolic complications such as hypoglycaemia and diabetic ketoacidosis accounted for most of the excess mortality among those under 30 years but after 20 years of diabetes the impact of atherosclerotic macrovascular complications steadily increases. The prognosis is particularly disturbing for children diagnosed with type 1 diabetes under the age of 10 years; previous reports have indicated that 60% were dead within 40 years of diagnosis. With increasing duration of diabetes, the prevalence of retinopathy, nephropathy and neuropathy is highest in those with poor glycaemic control and lowest in those with good control. The Diabetes Control and Complications Trial (DCCT 1995) established quite clearly that good glycaemic control in type 1 diabetes can reduce the incidence and progression of microvascular complications but the risk of a vascular event increases with duration of diabetes and the presence of nephropathy.

3

4

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

The DCCT was under-powered, and the patients too young, to be sure if improved glycaemia reduced the risk of macrovascular complications, however, the trend was for good control to be associated with a reduction in vascular events.

EPIDEMIOLOGY OF TYPE 1 DIABETES The worldwide variation in incidence is considerable though the pattern of presentation is similar. The incidence is showing signs of increasing at all ages but most noticeably in the under 5s. In under 16s, Northern Europe (Finland, Scotland, Sweden) has the highest rates with up to 30–35 cases per 100,000 of the population aged <16 years per year. Japan, China and Korea have rates that are as low as 0.5–2 cases per 100,000 per year. It is tempting to think that this is due mainly to genetic differences but there are different incidences in genetically similar countries such as Norway and Iceland, suggesting that environmental factors have a very significant influence. Whilst the pathophysiology of islet cell destruction has been well defined, the trigger for this process remains uncertain. Despite the relatively sudden onset of symptoms, family studies have shown there is a long prodromal period of immune activation. Viruses and cow’s milk protein are currently the main contenders that may initiate this process in the genetically susceptible.

THE NATURAL HISTORY OF TYPE 2 DIABETES People with type 2 diabetes often have established complications at the time of diagnosis. In the UK Prospective Diabetes Study (UKPDS), for example, 36% of newly diagnosed patients had retinopathy, 12% neuropathy and 2% proteinuria at recruitment. This may well be an underestimate, because in UKPDS patients with established vascular disease or retinopathy requiring laser therapy were excluded. Using prospective studies which have studied the rate of progression of retinopathy, it is estimated that at diagnosis of type 2 diabetes patients will likely have had their diabetes for between 8 and 12 years and, prior to diabetes, impaired glucose tolerance for very much longer. Death from ischaemic heart disease, stroke or lower extremity ischaemia occurs in over 60% of patients.

EPIDEMIOLOGY OF TYPE 2 DIABETES Although there is good evidence that type 2 diabetes is a heterogeneous condition with a number of genetic sub-groups, the current view supports the idea that for the majority of people this is a metabolic disorder in the genetically susceptible, precipitated by lifestyle changes which have led to a sedentary lifestyle and obesity. Essentially it is a failure of adaptation to a new environment which has changed in the course of a few generations. Elliot Joslin

CHAPTER 1 • THE PUBLIC HEALTH IMPACT OF THE DIABETES EPIDEMIC

went to study the Pima Indians at the start of the 20th century because of their low prevalence of diabetes. By the end of the century they ranked alongside the Pacific Micronesians from Nauru for having the highest prevalence of type 2 diabetes in the world. The ‘thrifty gene’ hypothesis postulates that humans evolving in a harsh environment, where famine and high physical activity was the norm, may have developed fuel efficient systems, which, when faced with limitless supplies of food and a sedentary lifestyle, leads to the metabolic disturbances now characterized as the Metabolic Syndrome (central obesity, hypertension, hyperlipidaemia and glucose intolerance). Underlying this is insulin resistance, which partly relates to fat distribution – the greater the proportion of intra-abdominal fat compared to the total, the greater the degree of insulin resistance. The distribution of fat deposition is genetically determined and there is evidence that there are ethnic differences in body composition. This may explain insulin sensitivity studies which, despite matching for age and BMI, demonstrate significant differences between ethnic groups. BMI tables have been based on the weights and heights of white Europeans and recent studies from India and New Zealand suggest they are not applicable to all ethnic groups. For example, in Indian populations it is suggested that accumulation of intra-abdominal fat begins at a BMI of around 23 and this (rather than 25 in white populations) should be the cut-off between ‘normal’ and ‘overweight’. In part, this may explain the wide variation in prevalence of type 2 diabetes across the world. Nauruans and Pimas have already been mentioned, but other high risk groups include South Indians, Polynesians, Maori of New Zealand, native American Indians, Mexican Americans and African Americans, all of whom have a higher prevalence of diabetes than Whites. This is despite the epidemic of obesity which affects all ethnic groups in the developed world. Recent work has demonstrated defective mitochondrial function in the muscles of relatives of people with type 2 diabetes, though whether this is present in all ethnic groups has yet to be determined. There is also an inverse relationship with poverty, but this is insufficient to explain all the population differences in prevalence. Unlike infections or cancer which are either present or not, the prevalence of metabolic disturbances depends very much on the definition, which may change over time. This has certainly been true for the diagnosis of type 2 diabetes and the associated metabolic syndrome, where the dilemma has been to produce a workable definition that could be applied to large groups of people, in order to distinguish those at high risk of complications (Fig. 1.1). Recently, abnormalities of glucose tolerance have been re-defined by the American Diabetes Association (ADA) and World Health Organisation (WHO) (Table 1.1) and for diabetes the diagnostic cut-off for fasting plasma glucose (FPG) has been lowered from 7.8 to 7 mmol/l. For epidemiological studies and for routine

5

6

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Criteria for the diagnosis of diabetes mellitus Plasma venous glucose concentration (mmol/l) Diabetes Mellitus: Fasting or 2hr post glucose load

> 7.0 > 11.1

Impaired Glucose Tolerance (IGT): Fasting (if measured) and 2hr post glucose load

< 7.0 and > 7.8

Impaired Fasting Glycaemia (IFG): Fasting

> 6.1 and < 7.0

And (if measured) 2hr post glucose load

< 7.8

Normal Fasting Plasma Glucose (FPG)

< 5.6

For epidemiological or population screening purposes, the fasting or 2hr value after 75g oral glucose may be used alone. For clinical purposes, the diagnosis of diabetes should always be confirmed by repeating the test on another day unless there is unequivocal hyperglycaemia with acute metabolic decompensation or obvious symptoms.

Table 1.1 Criteria for the diagnosis of diabetes mellitus (WHO classification 1999). Note that the American Diabetes Association defines IFG as FPG 5.6–6.9 mmol/l. People with IFG and IGT are considered to have “pre-diabetes”. An OGTT (75g) may be indicated in people with IFG if considered at high risk of diabetes.

clinical practice the ADA recommended using fasting glucose testing alone, and the use of the two hour oral glucose tolerance test (OGTT) was not recommended. Subsequent investigations have shown that fasting and two hour glucose criteria do not identify the same group of individuals. The DECODE study, which combined the results of 13 prevalence studies in nine European countries, found that there was a distinct sex difference in the prevalence of diabetes, impaired fasting glucose (IFG) and impaired glucose tolerance (IGT). Undiagnosed diabetes and IFG were more common in men than in women at 30–69 years of age. IGT was higher in women than in men and was particularly high in the over 70s. In the USA, the NHANES III study (2000), confirmed that diagnosed diabetes is most prevalent in the middle-aged (6%) and elderly (11%) compared to only 1.5% of 18–44 year olds (Fig. 1.1). The incidence is increasing in childhood and is related to obesity.

CHAPTER 1 • THE PUBLIC HEALTH IMPACT OF THE DIABETES EPIDEMIC

Impaired fasting glucose Undiagnosed diabetes Diagnosed diabetes

Age-standardized per cent

25

20

15

10

5

0

NonNonMexican Hispanic Hispanic American white black

NonNonMexican Hispanic Hispanic American white black

Men

Women

Fig 1.1 Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in US adults: From the Third National Health and Nutrition Examination Survey 1988–1994. Diabetes Care 1998; 21: 518–524.

THE COST OF DIABETES There is considerable morbidity associated with diabetes and calculating the cost can at best be an imprecise estimate. Diabetes is the leading cause of blindness, end-stage renal failure and lower-extremity amputation. People with diabetes experience high rates of macrovascular complications at least twice that of those without diabetes. In the UK, estimates of the cost of diabetes were first attempted in 1989 using 1984 data and this suggested that 4–5% of all UK healthcare expenditure went on people with diabetes. More recent data suggest the figure is more like 8%, and that one third of total expenditure on diabetes is spent on those aged 0 to 24 years. Based on 1999 figures, it was estimated that it might cost £100 million, in England alone, to implement the findings of the UKPDS, in terms of intensive treatment of glycaemic control and blood pressure. The difficulties, however, are that it is difficult to cost episodes in patients with multiple pathologies and coding has been shown to under-

7

8

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

record secondary diagnoses such as diabetes. Most economic assessments concentrate on direct costs, though clearly indirect costs, such as time off work, and intangible costs, such as what might have been, will inflate the figures considerably. Surveys from the USA suggest that health care expenditure was over $11,000 per year per person with diabetes compared to $2,600 for people without diabetes. Over 60% was due to inpatient hospitalization. Nevertheless, there are effective strategies for the prevention or delay of complications associated with diabetes and both the DCCT in type 1 diabetes and the UKPDS in type 2 diabetes has demonstrated the effectiveness of intensification of treatment. An economic model has been used to analyse the costs of DCCT, enabling calculation of the costs of preventing end-stage complications. The model predicted that intensive therapy would result in approximately 15 years free from the first major complications of type 1 diabetes and additional years of life free from blindness, ESRF and lower-extremity amputation (Fig. 1.2). It was projected that intensive therapy would prolong life by about 5 years and the cost was approximately $29,000 per year of life gained. Similar economic modeling has looked at the cost-effectiveness of ‘comprehensive’ or intensive care of type 2 diabetes (Table 1.2). Such models are predicting the likelihood of complications developing in a diabetic population. One such model suggests that with standard care (non-intensive) over a life-time, 19% of subjects would become blind, 17% would develop ESRF and 16% would require lower-extremity amputation. With intensification of glucose control these figures could be reduced by up to 75% (but with no effect on cardiovascular outcome), with increased survival by 1.3 years. The average lifetime cost was $40,000 more than with standard care. Of course, these models are limited because they have only looked at the cost and benefits of intensification of glycaemic control and clearly there are many other interventions available to reduce macrovascular risk. These somewhat daunting economic assessments of diabetes care can be viewed from different perspectives. For politicians and public health specialists, it provides an incentive to invest now in primary prevention of type 2 diabetes, as treatment costs are unsustainable given the epidemic of diabetes that is sweeping the developed and developing worlds. For clinicians, the challenge is to develop cost-effective strategies and deliver high quality diabetes services that reach the majority rather than the minority. Examples of affordable interventions with proven benefits are: comprehensive eye-screening; ensuring everyone at high vascular risk receives low dose aspirin; and annual foot assessments, but these are not made available to all people with diabetes, even in the more affluent societies. The pharmaceutical industry must not forget that it remains part of the society in

(b)

Cumulative incidence of ESRD (%)

(a)

Cumulative incidence of blindness (%)

CHAPTER 1 • THE PUBLIC HEALTH IMPACT OF THE DIABETES EPIDEMIC

(c)

30 Standard care Comprehensive care 20

10

0 0

8

16

24

32

40

48

56

64

0

8

16

24

32

40

48

56

64

0

8

16

24

32

40

48

56

64

30

20

10

0

Cumulative incidence of IEA (%)

30

20

10

0 Years after diagnosis of diabetes

Fig. 1.2 For people receiving standard care, the model predicts sharply increasing cumulative incidence of complications, including blindness (a), ESRD (b), and lowerextremity amputation (LEA) (c) with increasing duration of type 2 diabetes. The model predicted a substantially lower incidence of these long-term complications with a program of comprehensive care. Eastman, et al. Diabetes Care 1998; 21 (Suppl 3): C19–C24.

9

10

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Predicted reduction in life time costs of end-stage complications through comprehensive care for type 2 diabetes Cost elements

Standard care

Present value costs (3% discount rate) General and diabetes-related medical care ($) 32,365 Eye disease ($) 3,128 Renal disease ($) 9,437 Neuropathy/lower-extremity amputations ($) 4,381 New coronary heart disease ($) 13,458 Total costs ($) 62,769 QALY (undiscounted) 16.04 QALY (discounted 3%) 11.43 Life-years (undiscounted) 17.05 Incremental cost/QALY gained —

Comprehensive care Difference

58,312 1,536 960

25,947 (1,592) (8,477)

1,469 14,414 76,922 18.03 12.30 18.37 —

(2,912) 956 13,922 1.9 0.87 1.32 16,002

Table 1.2 Predicted reduction in life time costs of end-stage complications through comprehensive care for type 2 diabetes. Data are averages per person per life time. Cost savings are indicated in parentheses.

which it operates and has a social responsibility. The challenge is to develop and market safe therapies which generate enough profit to encourage future shareholders to invest, but not so much that only the wealthy can afford them.

CHAPTER 1 • THE PUBLIC HEALTH IMPACT OF THE DIABETES EPIDEMIC

CURRENT ISSUES •

•

•

•

Population screening for type 2 diabetes is not widespread and may not be cost-effective but targeted opportunistic screening of high risk individuals (such as women with prior gestational diabetes, first degree relatives, high-risk ethnic groups, the obese) will identify 70% of those with undiagnosed diabetes. With the increase of obesity in childhood this may mean screening young people in their teens if they are from a family in which type 2 diabetes is common. Finnish and American randomized studies have demonstrated that interventions such as weight loss and exercise programs in patients with impaired glucose tolerance have a role in delaying or preventing the progression to frank diabetes by over 50%. Metformin used in this context reduces progression from IGT to diabetes by 30%. Studies of the early use of insulin sensitizers in IGT are ongoing. The epidemic of obesity affects all ages and consequently the emergence of type 2 diabetes in childhood is increasingly apparent. The prognosis is likely to be particularly bad in this age group and a high incidence of nephropathy and early onset cardiovascular disease is to be expected in South Indians, Maori and other indigenous populations. Effective prevention and treatment strategies for obesity are urgently required. Studies from the UK have shown that simple messages such as discouraging intake of high sugar carbonated drinks can reduce the development of excess weight gain and obesity in adolescents. Scandinavian countries have introduced bans on television advertising to children for ‘energy-rich/nutrition-poor’ (junk) foods and New Zealand is trying to introduce a ‘fat-tax’ which would tax snack foods and soft drinks.

FURTHER READING DCCT research group. Resource utilization and costs of care in the DCCT. Diabetes Care 1995; 18:1468–1478. Haffner SM, Stern MP, Hazuda HP, Pugh JA, Patterson JK. Hyperinsulinaemia in a population at high risk for non-insulin dependent diabetes mellitus. N Engl J Med 1986; 315: 220–224. Rubin RJ, Altman WM, Mendelson DN. Health care expenditures for people with diabetes mellitus, 1992. J Clin Endocrinol Metab 1994; 78: 809A–809F. The Worldwide Burden of Diabetes. Proceedings of a workshop. Phoenix, Arizona, USA. Diabetes Care 1998; 21: Suppl 3. Tuomilehto J, Linström J, Eriksson JG et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001; 344: 1343–50.

11

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

CHAPTER 2 RISK FACTORS

Adrian R. Scott MD, FRCP

INTRODUCTION Prior to the use of glycosylation products such as glycosylated haemoglobin and fructosamine in the late 1970s, estimates of glucose control relied on selfreported urine tests, random blood sugars measured in the outpatient setting and other surrogate measures such as frequency of hypoglycaemia, or measurement of 24 hour urinary glucose excretion. Despite these difficulties, the association between duration of diabetes, the degree of hyperglycaemia and the severity of microvascular and neuropathic complications had long been observed in both type 1 and type 2 diabetes. It was also clear that the relationship between glycaemic control and macrovascular disease was not straight forward since people with mild degrees of hyperglycaemia such as those with impaired glucose tolerance had twice the risk of developing coronary heart disease as those with normal glucose tolerance. In addition, the association of diabetes (particularly type 2) with multiple vascular risk factors such as hypertension and dyslipidaemia was apparent but it has taken until this last decade or so to realize that it is the interaction of these factors that so alters risk and that each must be viewed in this context, not in isolation. This chapter looks at the influence of hyperglycaemia and other factors on the development of microvascular and macrovascular complications.

HYPERGLYCAEMIA In a prospective personal series of 4,400 patients with diabetes, observed between 1947 and 1973, Pirart showed that poor glycaemic control was clearly related to a higher prevalence of retinopathy, nephropathy and neuropathy compared with patients with better control. With the discovery of glycosylated haemoglobin the association between long-term hyperglycaemia and complications was confirmed. Retinopathy and microalbuminuria are good markers of microvascular disease and indicative of a generalized vasculopathy. The numerous studies that have looked at the relationship between glycaemic control and both the onset and progression of microvascular complications have produced remarkably consistent results. For example, The Berlin Retinopathy Study was an observational report on children and adolescents with type 1 diabetes who were followed between 1977 and 1993. At that time Berlin was politically and geographically isolated and most young people with type 1 diabetes were followed up by a single centre. Glycosylated haemoglobin (HbA1C) was available from 1980 onwards and urine for microalbumin was tested from 1987 onwards. Data has been published on 346 patients (190 males) who were followed up to the age of 18–22 years. The rate of onset of

13

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

background retinopathy rose with increasing HbA1C from 0.7 events per 100 patient years in the group with a long-term HbA1C of < 7% to 7.3 events per 100 patient years when the long-term HbA1C was > 11%. The incidence of retinopathy increased steeply when the HbA1C was above 9% (Fig. 2.1) and was similar to the results seen in the Diabetes Control and Complications Trial (DCCT). Surprisingly, glycaemic control did not appear to influence the time to development of retinopathy, except in those with very poor control (HbA1C >13%). In all other groups the median time to onset of background diabetic retinopathy was approximately 12 years. Patients with microalbuminuria, however, developed retinopathy after a mean of 11.5 years compared to 14.7 years in those with normoalbuminuria. The chance of remaining free from background retinopathy after 12 years was < 25% in patients with microalbuminuria compared to 81% in patients without microalbuminuria. In the DCCT study 1,441 highly selected patients aged 13–39 years were randomly assigned to intensive or conventional treatment. Approximately half of those randomized had been selected as free of retinopathy and with normal urinary albumin excretion. The other half had mild to moderate retinopathy and urine albumin excretion <200 mg per 24 hours. This group

10 Rate of background retinopathy (per 100 patient years)

14

8 6 4 2 0 <7

7–8

8–9

9–10

10–11

>11

Glycated haemoglobin (%)

Fig. 2.1 Rate of development of background retinopathy per 100 patient years in different classes of HbA1C. Berlin Retinopathy Study. Diabetes Care 1994; 17(12): 1390–96.

CHAPTER 2 • RISK FACTORS

was similarly randomized into a secondary intervention arm. The intensively treated group sustained a 2% drop in HbA1C to 7%, but glycaemic control remained virtually unchanged in the conventional group (approximately 9% at baseline). There was a 76% adjusted mean risk reduction in the primary prevention arm for the development of retinopathy. In the secondary prevention arm the estimated risk reduction was 54% with intensive treatment. The Wisconsin Epidemiological Study of Diabetic Retinopathy (WESDR) followed a large population of people with diabetes who were living in Southern Wisconsin in the US from 1979–1980. There were around 1,200 type 1 diabetic patients originally diagnosed under the age of 30 and nearly 1,800 older onset patients who were predominately type 2 but around 800 of whom were treated with insulin. Approximately 1,300 of this study population were followed up at 10 years – the main reason for the dropout was death before 10 years. The incidence of retinopathy progression, progression to proliferative retinopathy and incidence of macula oedema increased from the lowest to the higher quartile of HbA1C. For a given level of HbA1C there were few differences in incidence or progression of retinopathy among the three groups (young type 1, older-onset patients on insulin, older-onset patients on tablets). In addition, there was no evidence of a threshold effect (Figs 2.2 and 2.3). The study group examined whether a change in glycaemic control was associated with a change in the risk of progression of retinopathy at 10 years. Using mathematical modelling, it was estimated that a 1.5% decrease in

100

Incidence (%)

80 60 40 Younger, P<.0001 Older, insulin P<.005 Older, no insulin P<.0001

20 0 6

8

10 12 Glycosylated haemoglobin (%)

14

Fig. 2.2 The 10-year incidence of retinopathy by quartile of HbA1C at baseline in younger-onset group taking insulin, the older group taking insulin, and the olderonset group not taking insulin in WESDR. Diabetes Care 1995; 18: 258–268.

15

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

60 Younger, P<.0001 Older, insulin P<.0005 Older, no insulin P<.0001 40 PDR (%)

16

20

0 6

8 10 Glycosylated haemoglobin (%)

12

Fig. 2.3 The 10-year progression to proliferative diabetic retinopathy by quartile of HbA1C at baseline in the younger-onset group taking insulin, and the older-onset group not taking insulin, and the older-onset group not taking insulin in the WESDR. Diabetes Care 1995; 18: 258–268.

HbA1C from baseline to 4-year follow-up would be expected to lead to a 33% decrease in the 10-year incidence of proliferative retinopathy in the younger age group. A similar fall in HbA1C produced a 24–40% decrease in incidence in the older age group. The 10-year incidence of proteinuria and renal failure was 28.3% and 7.1% in the younger group and 36.5% and 1.8% in the older onset group. Once again there was a relationship between HbA1C and the incidence of nephropathy (Fig. 2.4). The WESDR also indicated a relationship between hyperglycaemia and macrovascular disease. There was an increased risk of amputation in both younger and older groups and HbA1C was associated with increased risk of death. After correction for age and sex the hazard ratio of dying for the fourth quartile of HbA1C compared to the first quartile of HbA1C was 1.9. In the older group, for example, the 10-year survival in the lowest quartile of HbA1C was 62.8% compared with 41.7% of those in the highest quartile (Fig. 2.5). Interestingly, in WESDR 29% of younger onset patients and 43% of older onset patients did not manifest proliferative diabetic retinopathy or proteinuria despite being in the third or fourth quartile of hyperglycaemia. This raises the possibility that some patients are “protected” from complications or that others are more susceptible. A number of studies of patients with type 1 diabetes has suggested that as many as 20% of patients do not

CHAPTER 2 • RISK FACTORS

50

5.4–8.5 8.6–10.0 10.1–11.5 11.6–20.8

Incidence (%)

40

P<.0006

P<.0001

30

20

10

0 Younger

Older

Fig. 2.4 The incidence of gross proteinuria in people with insulin-dependent diabetes mellitus by quartile of HbA1C. Diabetes Care 1995; 18: 258–268.

Incidence (%)

15

5.4–8.5 8.6–10.0 10.1–11.5 11.6–20.8

P<.0006

10 P<.01 5

0 Younger

Older

Fig. 2.5 The 10-year incidence of lower extremity amputation by quartile of HbA1C at baseline, in the younger and older onset groups participating in the WESDR. Diabetes Care 1995; 18: 258–268.

17

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Estimated hazard ratios (95% Cl)

develop microvascular complications even as late as 30–40 years after the onset of the disease. On the other hand, a small minority may have severe retinopathy after only 5–7 years. Clustering of nephropathy, for example, has been observed in some families and the history of essential hypertension in a first-degree relative is associated with an increased risk of nephropathy in the family member with type 1 diabetes. In WESDR there were no obvious differences between people with type 1 and type 2 diabetes in the incidence of microvascular complications in relation to rising HbA1C. In the UKPDS the risk of each of the microvascular and macrovascular complications of type 2 diabetes was strongly associated with hyperglycaemia, as measured by HbA1C. There was no evidence of a threshold and there was a threefold increase over the range of <6% to >10% (Fig. 2.6). Despite all these studies demonstrating a clear relationship between hyperglycaemia, HbA1C and the development of microvascular complications, the reasons remain obscure. Although glycosylated haemoglobin reflects ‘recent’

Estimated hazard ratios (95% Cl)

18

3

1

0.4 40 45 50 55 60 65 70 2.5 3.0 3.5 4.0 4.5 5.0 0.9 1.0 1.1 1.2 1.3 Age (years) Low density lipoprotein High density lipoprotein cholesterol (mmol/l) cholesterol (mmol/l) 3

1

0.4 40 45 50 55 60 65 70 2.5 3.0 3.5 4.0 4.5 5.0 0.9 1.0 1.1 1.2 1.3 Haemoglobulin A1c (%) Systolic blood never excurrent smokers smokers smokers pressure (mmHg)

Fig. 2.6 Estimated hazard ratios for significant risk factors for coronary artery disease occurring in 335 out of 3,055 diabetic patients. BMJ 1998; 316: 823–828, with permission.

CHAPTER 2 • RISK FACTORS

glycaemic control, there are inevitably differences between individuals regarding the rates of formation and breakdown of glycated haemoglobin and no clear answers about how these may relate to risk of microvascular complications. A single HbA1C must be interpreted with caution since recent studies have shown that in an individual with stable glycaemic control the 95% confidence limits for an HbA1C of 7.0% are between 6.1 and 7.9%. The DCCT study data revealed that past glycaemia over 3–4 months contribute only 10% to the current HbA1C and the most recent 30 days contribute 50%. Monozygotic twin studies have shown that 62% of the population variance in HbA1C is genetically determined. Put simply, an HbA1C of 8.0% in a person with diabetes whose pre-diabetic HbA1C was 3.5% may represent a completely different risk of developing microvascular complications than a similar patient with the same HbA1C, whose prediabetic HbA1C was 5.8%.

HYPERTENSION, HYPERLIPIDAEMIA AND SMOKING Hypertension exacerbates the micro- and macrovascular complications of diabetes (Fig. 2.7) but it is important to differentiate between the hypertension associated with the two main types of diabetes. People with type 1 diabetes at diagnosis have similar blood pressures to those without diabetes and the development of hypertension increases with diabetes duration and is associated with the development of nephropathy. Microalbuminuria and proteinuria are manifestations of renal involvement. Untreated, the hypertension worsens, protein excretion rates increase and glomerular filtration rates fall. Clustering of nephropathy in families suggests that there may be a genetic predisposition to nephropathy and hypertension in some individuals with type 1 diabetes. Hypertension in people with type 2 diabetes, however, is much more common, may precede the diagnosis, and is present in between 30 and 50% at diagnosis. It is a major component of the metabolic syndrome that constitutes type 2 diabetes and appears to reflect insulin resistance. The effects on the cardiovascular system are more profound than similar blood pressure levels in a person without diabetes. For example, in the MRFIT study, rising systolic blood pressure was associated with increasing 10-year CHD mortality which was 3–5 times greater in those with diabetes. Epidemiological studies have demonstrated the continuous relationship between serum cholesterol and risk of atherosclerotic vascular disease, particularly CHD. This was confirmed by the Framingham study and, in the MRFIT study when over 300,000 men aged 35–57 years were screened, the relationship between cholesterol and death from CHD was independent of smoking and hypertension and continuous across the age range. There was also a strong relationship between CHD and cholesterol level in people with diabetes.

19

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

50 Adjusted incidence per 1000 person years (%)

20

Myocardial infarction Microvascular end points 40

30

20

10

0 110

120

130

140

150

160

170

Systolic BP

Fig. 2.7 Incidence rates (95% confidence interval) of myocardial infarction and microvascular end points by category of updated mean systolic blood pressure, adjusted for age, sex and ethnic group expressed for white men aged 50–54 years at diagnosis and mean duration of diabetes 10 years. BMJ 2000; 321: 412–419, with permission.

The lipid abnormalities associated with diabetes are both qualitative and quantitative. There are no quantitative differences between patients with type 1 diabetes and those without diabetes, though abnormalities may appear with the development of nephropathy or if glycaemic control is poor. HDL levels are often in the normal range but sub-fractions of HDL show significant differences from the normal population. For example, HDL2 levels have a strong negative correlation with CHD and in type 1 diabetes levels of this subfraction are low, in favour of HDL3, which does not have the same cardioprotective properties. In type 2 diabetes the spectrum of lipid abnormalities is broader and an essential element of the metabolic syndrome. HDL levels are low, and associated with hypertriglyceridaemia. Total and LDL-cholesterol levels are similar to non-diabetic levels, but again qualitative differences exist. LDL particles are small and dense and thought to be more atherogenic. Nevertheless, LDL levels correlate with the presence of clinical macrovascular disease in both type 1 and type 2 diabetes. Hypertriglyceridaemia as a risk factor CHD

CHAPTER 2 • RISK FACTORS

remains controversial, but accounts for some of the other lipid changes such as low HDL and the formation of small dense LDL. There are also correlations with plasminogen activator inhibitor I (PAI-I). Smoking is an independent risk factor for macrovascular disease and in the MRFIT study increased the 10-year risk of dying from CHD by 2.4 times in non-diabetics (from 10.1 to 23.9 per 1,000) and by 1.6 times in those with diabetes (from 44.5 to 68.7 per 1,000). Smoking has also been implicated in the progression of microvascular and other diabetic complications including retinopathy, nephropathy and necrobiosis lipoidica. Analysis of the MRFIT data suggested that stopping smoking was one of the most effective interventions at reducing mortality from macrovascular disease.

ETHNICITY

Rate of death from nephropathy

There appear to be definite ethnic variations in the prevalence of complications of type 2 diabetes. Compared to the WESDR population of nonHispanic whites, retinopathy is much more common in the Pima Indians of Arizona and the Mexican-Americans in San Antonio, Texas. Nephropathy and CHD is more common in indigenous peoples such as the Maori in New Zealand (Fig. 2.8) and South Asians, occurring a decade or so earlier, than Europeans. On the other hand, CHD is less frequent in the Pima Indians with type 2 diabetes than many non-diabetic white populations in the US.

0.08

0.06 0.04

0.02

0.00 0

1

2

3

4

5

6

7

Time (years)

Fig. 2.8 Risk of recorded death with nephropathy by Cox’s proportional hazards regression model after accounting for age, sex, and source of patient. Shown are Europeans with type 2 diabetes ( ■ ): Maori with type 2 diabetes ( ◆ ): and Pacific Islands people with type 2 diabetes ( ▲ ).

21

22

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

CONCLUSION Risk factors for the micro- and macrovascular complications of diabetes are similar in both type 1 and type 2 diabetes but significant differences exist in the prevalence and role of hypertension and hyperlipidaemia. Duration of diabetes is significantly correlated with complications in type 1 diabetes. An essential part of diabetes care is an annual structured risk assessment so that individual management plans can be developed to target these increased risks.

FURTHER READING Kannel WB, McGee DL. Diabetes and glucose tolerance as risk factors for cardiovascular disease: The Framingham Study. Diabetes Care 1979; 2: 120–6. Stamler J, Vaccaro O, Neaton JD, Wentworth D. for the Multiple Risk Factor Intervention Trial Research Group. Diabetes, other risk factors and 12 year cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care 1993; 16: 434–44. Yudkin JS. How can we best prolong life? Benefits of coronary risk factor reduction in nondiabetic and diabetic subjects. BMJ 1993; 306: 1313–18.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

CHAPTER 3 DIABETIC NEPHROPATHY

Adrian R. Scott MD, FRCP

INTRODUCTION Diabetes is now the commonest cause of end-stage renal failure (ESRF) in Europe and North America. This is mainly because of the increasing prevalence of diabetes and because people with diabetes are now accepted more readily onto renal replacement programs, having been excluded in the past. In Scandinavia and the USA, 30% of people requiring dialysis or transplantation have diabetes though there are significant variations across Europe. Between 20–30% of people with type 1 or type 2 diabetes will develop nephropathy but only a small percentage of people with type 2 diabetes will progress to ESRF, most dying of cardiovascular disease long before the need for renal dialysis. Nevertheless, the much higher prevalence of type 2 diabetes means that up to 50% of patients with diabetes requiring dialysis are from this group. There are considerable ethnic variations in the prevalence of nephropathy. For example, in the UK Anglo-Asians and Afro-Caribbeans have a much higher prevalence of nephropathy. In the USA Pima Indians with type 2 diabetes are particularly at risk, as are Maori in New Zealand.

CLINICAL PRESENTATION The earliest clinical manifestation of diabetic renal disease is the finding of small quantities of albuminuria (30–300mg/24 h or 20–200μg/min), often within 5–10 years of diagnosis, which increases progressively over a number of years. In cross-sectional studies, microalbuminuria is present in approximately 20–30% of insulin-treated adults and 10–30% of Caucasian adults with type 2 diabetes, and there is a clear relationship with glycaemic control (Fig. 3.1). The progression of nephropathy is best documented in type 1 diabetes (Fig. 3.2). A history of hypertension in a first-degree relative and differences in Na+–Li+ counter-transport suggest there is a component related to genetic predisposition. Poor glycaemic control may initiate functional changes within a year or two of the diagnosis of diabetes, including renal hypertrophy and hyperfiltration with an increase in renal blood flow and glomerular filtration rate (GFR). These changes are reversible and metabolically dependent. Recent studies suggest that even levels of albumin excretion rate above 10 μg/min are highly predictive of future microalbuminuria. Blood pressure rises progressively in parallel with increasing albuminuria and secondary lipid abnormalities also occur. The glomerular barrier loses its size selectivity and as macroproteinuria is reached GFR has already started to decline. Occasionally, nephrotic syndrome may result.

23

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

12

10

8 Odds ration

24

6

4

2

0 7

8

9

14 10 12 Hemoglobin A1 (%)

16

18

20

Fig. 3.1 Relation between mean hemoglobin A1: values and the risk of microalbuminuria in patients with IDDM. N Engl J Med 1995; 332: 1251–1255.

The rate of decline of renal function is linear but the time to ESRF varies considerably between individuals (Fig. 3.3). Albuminuria is a manifestation of a generalized vasculopathy and a high proportion of patients with nephropathy develop symptomatic coronary heart disease (CHD) in the early course of their renal disease. The associated anaemia and hypertension lead to left ventricular hypertrophy (LVH) and heart failure. Death from stroke, myocardial infarction or peripheral gangrene occurs in nearly 75% of patients, either before or during renal replacement therapy, and 50% of type 1 diabetics are dead within 10 years of the onset of proteinuria.

KEY DIAGNOSTIC FEATURES Although the diagnosis of nephropathy is based on the finding of Albustix positive albuminuria (equivalent to a urinary albumin excretion rate of more than 300 mg/day), it is important to remember that the renal disease is already well established at the stage of overt proteinuria with irreversible structural changes having occurred in the glomeruli. In type 1 diabetes,

CHAPTER 3 • DIABETIC NEPHROPATHY

Normal

Incipient nephropathy

Clinical nephropathy

UAER

<20 μg/min

>20 <200 μg/min

GFR

Stable 1% decline pa >40 years

Age related changes; more rapid loss when UAER approaches 200 μg /min or if blood pressure increases

Decline 10 ml/min/y (hypertensive), 1–4 ml/min/y (normotensive)

Blood pressure

Stable: higher in those progressing to incipient nephropathy

Initially stable, but higher than normal UAER controls Increases with increasing UAER

Most patients hypertensive (>140/85 mmHg) Increases with declining GFR

Kidneys remain large GBM thickening 54 nm pa Mesangial expansion ~ 4% pa

Kidneys shrink GBM 2–3 times normal, stable Nodules Global glomerulosclerosis Mesangial expansion ~ 7% pa

1%–2% (increase 20% pa) per annum

Pathology Large kidneys Tubular hypertrophy/ hyperplasia Glomerular enlargement Normal ultrastructure GBM thickening 20 nm pa

3%–4% per annum

<200 μg/min

pa = per anum; GBM = glomerular basement membrane

Fig. 3.2 Natural history of diabetic nephropathy: Oxford Textbook of Medicine 3rd edn. Oxford University Press, 1996.

patients over 12 years of age should be screened annually for microalbuminuria using the albumin/creatinine ratio (ACR) on the first morning urine sample (Fig. 3.4). Type 2 patients should be routinely screened for albuminuria and, if Albustix negative, annual screening undertaken for microalbuminuria, particularly if a positive result would alter or intensify their management. Classically, the finding of Albustix positive proteinuria on more than two occasions in a person with established diabetes (having first ruled out urinary tract infection) is indicative of diabetic nephropathy. Hypertension is

25

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

5 1 x 103 Creatinine

26

4 3 2 1 0 0

10

20 Months

30

40

Fig. 3.3 Decline of renal function in 16 patients with nephropathy. ABC of Diabetes 1982; 285: 627–629.

commonly associated (Fig. 3.5) and the absence of retinopathy should prompt consideration of an alternative diagnosis in a patient with type 1 diabetes. However, in type 2 diabetic patients only about 50% of those with nephropathy have associated retinopathy. Renal ultrasound can be helpful in excluding unrelated structural abnormalities such as hydronephrosis or polycystic kidneys, but asymetrical kidneys may be indicative of renal artery stenosis (RAS), especially in patients with peripheral vascular or aortic aneurysm disease. If renal artery dopplers are suggestive of RAS, arteriography or CT-angiography is justified but the contrast media used for these investigations can precipitate renal failure. Adequate hydration with intravenous saline is important, but early suggestions that oral treatment with acetylcysteine reduced the incidence of contrast-induced renal complications has not been confirmed by subsequent studies. Metformin should be stopped two days before the investigation because of the small risk of lactic acidosis. Haematuria is an unusual feature in diabetic nephropathy and these patients may have co-existent renal disease unrelated to their diabetes. Renal biopsy is rarely necessary but should be considered in these patients, since it may occasionally reveal potentially treatable glomerular disease, e.g. IgA nephropathy. The morphological changes seen in the diabetic kidney, e.g. diffuse and nodular glomerulosclerosis and arteriolohyalinosis (KimmelstielWilson kidney), which may be present in over 90% of kidneys after 10 years of

CHAPTER 3 • DIABETIC NEPHROPATHY

First morning urine sample

Albumin concentration <20 mg/l and or Albumin: creatinine < 2.5 mg/mmol in men < 3.5 mg/mmol in women

Albumin concentration ≥20 mg/l and or Albumin: creatinine ≥2.5 mg/mmol in men ≥ 3.5 mg/mmol in women

Retest to confirm if still abnormal

Normal No timed urine

Timed urine sample*

Retest in 1 year

AER > 20 μg/min (30 mg/24 h)

AER < 20 μg/min

Two other timed urine samples within 6–12 weeks

AER ≥ 20 μg/min (30 mg/24 h) in at least one sample

- Monitor as often as required -Monitor HbA1c, blood pressure, lipids, creatinine -Test for retinopathy, CHD, CVD, PVD, neuropathy -Commence ACE-inhibitor AER = albumin excretion rate CVD = cerebral vascular disease PVD = peripheral vascular disease

CHD = coronary heart disease HBA1c = glycated haemoglobin

* If timed urine sample difficult to obtain, monitoring should continue with the use of albumin/creatinine (A/C) ratio. No prospective study has however so far evaluated the validity of A/C ratio as a monitoring index

Fig. 3.4 Screening strategy and monitoring programme for microalbuminuria in type 1 diabetes. St Vincent Joint Force for Diabetes Report of the Renal Disease Subgroup, 1994, 2nd edn.

27

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

60 Cumulative incidence (%)

28

Hypertension Proteinuria

50 40 30 20 10 0 15

20

25

30

35 40 Age (years)

45

50

55

Fig. 3.5 Cumulative incidence of hypertension and persistent proteinuria in type 1 diabetes: Handbook of Diabetes, 2nd edn. From Williams G & Pickup JC, Blackwell Science, 1999.

type 1 diabetes, are not synonymous with diabetic nephropathy. Despite these common histological changes, the clinical syndrome of nephropathy with proteinuria and declining renal function only develops in about one-third of these patients, indicating that there is a poor correlation between renal morphology and function.

EVIDENCE-BASED PRACTICE Early detection and identification of individuals at higher risk (such as those with a family history of hypertension or a sibling with diabetes and nephropathy) should be the aim. Many of the early changes of diabetic renal disease, such as hyperfiltration and increasing albumin excretion, are readily reversible with improved glycaemic control. The Diabetes Control and Complications Trial (DCCT) showed a 39% reduction in the occurrence of microalbuminuria and a 54% reduction in albuminuria in the intensive therapy arm for both adults and adolescents with type 1 diabetes. Similarly, the UK Prospective Diabetes Study (UKPDS) showed a slowing of renal decline in the tight glycaemic control group with type 2 diabetes. Tight control of hypertension is essential to reduce the decline in renal function. Angiotensin-converting enzyme inhibitors (ACE-Is) are indicated in type 1 patients with persistent microalbuminuria or proteinuria, irrespec-

CHAPTER 3 • DIABETIC NEPHROPATHY

tive of initial blood pressure, but women of childbearing age must avoid pregnancy because of potential fetal toxicity. Renoprotection by ACE-Is is probably a class effect but evidence exists for the use of Captopril, Enalapril and Lisinopril. The sulphydril group present in Captopril has antioxidant properties but whether this is advantageous in clinical practice is unproven. To achieve a target BP of < 130/80 for patients with nephropathy may require several antihypertensive agents and for young people under 16 years BP targets may be set even lower (to achieve a BP < 90th centile for age). Type 2 patients with microalbuminuria or proteinuria are less likely to progress to ESRF but as with type 1 diabetes (Fig. 3.6) BP management is the mainstay of treatment (Fig. 3.7). The choice of antihypertensive drug is less important than the achieved BP but ACE-Is confer some advantages if the patient has established vascular disease, as shown in the diabetic sub-group of the HOPE study (MICRO-HOPE) where Ramipril reduced mortality, rates of

Albuminuria (mg/min)

Glomerular filtration rate (ml/min/1.73m2)

Mean arterial blood pressure (mmHg)

Start of antihypertensive treatment 125 115 105 95 ΔGFR = 0.94 (ml/min/mo)

105 95

ΔGFR = 0.29 (ml/min/mo)

85 75

ΔGFR = 0.10 (ml/min/mo)

65 55 1250 750 250 -2 6

-1

0

1

2 Years

3

4

5

Fig. 3.6 Effect of blood pressure control on progression of renal disease in 11 patients with type 1 diabetes. Parving H-H et al. BMJ 1987; 294: 1443–1447.

6

29

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

40 Less tight control (n = 390→161) Tight control (n = 758→325) Patients with events (%)

30

30

20

10

32% reduction in risk for diabetes related mortality P = 0.019

0 0

1

2

3 4 5 6 Years from randomization

7

8

9

Fig. 3.7 UK Prospective Diabetes Study Group (UKPDS): blood pressure control and mortality from macrovascular disease and renal failure. BMJ 1998; 317: 703–713.

myocardial infarction and episodes of heart failure. Angiotensin receptor antagonists (ARAs) achieve similar reductions in blood pressure and proteinuria as ACE-Is but combination therapy with both agents may produce superior falls in systolic and diastolic BP in comparison to either agent alone in type 2 diabetics. As yet, there is none of the long-term data that exists for ACE-Is, suggesting a specific renoprotective effect of ARAs over and above their effect on blood pressure. Most of the studies have been short term and placebo controlled and those that have compared ARAs with ACE-Is have not shown superior outcomes with ARAs. Indeed, to date there have been only three randomized double-blind studies lasting more than one year using ARAs as an anti-hypertensive treatment for people with diabetes. None showed a significant reduction in total or cardiovascular mortality. In the RENAAL study there was a statistical reduction in progression to end-stage renal failure in those that were treated with ARAs but the placebo group had higher BPs throughout the study which may account for the worse outcomes. Until there are more conclusive studies with direct comparisons with ACE-Is, ARAs should be reserved for patients intolerant of ACE-Is (e.g. cough). They have the same renal sideeffects as ACE-Is such as hyperkalaemia and must be avoided in RAS. Diet adjustment has a place in the management of patients with nephropathy prior to the need for renal replacement therapy, particularly with the aim of reducing vascular risk. Thus, a diet low in fat, high in antioxidants and low in salt (to reduce BP) must be balanced against potassium and phosphate intake where

CHAPTER 3 • DIABETIC NEPHROPATHY

the serum level may be elevated. Many people with diabetes mistakenly eat more protein than the general population, so modest reductions may be beneficial, though low protein diets for low GFR states are not well tolerated by many patients, despite evidence for their effects in slowing the deterioration.

MONITORING RENAL FUNCTION Serum creatinine is a poor indicator of renal function, rising only after there has been a severe reduction in GFR (Fig. 3.8). EDTA clearance measurements are impractical for routine tracking of renal function but can be useful if 24-hour urinary creatinine clearance values are inconsistent with the clinical picture. Otherwise, reciprocal-creatinine plots against time can helpfully indicate the rate of decline of renal function in individual patients. 1600

1400

Serum creatinine (μmol/l)

1200

1000

800

600

400

200

0 0

20

40

60 80 GFR (51Cr EDTA)

100

120

Fig. 3.8 Relationship between glomerular filtration rate (GFR) measured using chromium-51 edetic acid (51Cr EDTA GFR) and serum creatinine, in patients being investigated for renal disease. Values are means and ranges (unpublished data from Roberts B, Gabriel R, 1975. BMJ 1986; 293: 1119–1120).

31

32

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

After starting ACE-Is or AII blockers, urea and electrolytes should be monitored carefully (e.g. one week and one month after initiation) to detect hyperkalaemia or a drug-induced deterioration in renal function (usually associated with RAS). Metformin should be avoided completely if serum creatinine is >130μmol/l because of the potential but small risk of lactic acidosis. Occasional monitoring of haemoglobin, corrected calcium and fasting phosphate is indicated. Abnormal results should prompt further measurements of haematinics, especially ferritin, and indicators of bone metabolism such as alkaline phosphatase and parathyroid hormone. In general, increasing albuminuria is a sign of worsening nephropathy and a reduction in albumin secretion as measured by albumin creatinine ratio (ACR), correlates with beneficial effects on renal function. Spontaneous resolution of microalbuminuria may occur, independent of ACE-I use, in type 1 diabetes. Monitoring of microalbuminuria over time can be useful, and an increasing ACR may be an indication for more aggressive antihypertensive therapy. Arrangements for the involvement of a nephrologist in the care of a patient with diabetic nephropathy will vary from country to country but studies in the UK have suggested that many patients are never given the opportunity of assessment by a renal physician, or else they are referred at the point of requiring dialysis. The type of patient suitable for referral to a nephrologist might be: • patients with a serum creatinine > 150μmol/l; • type 1 diabetic patients with confirmed microalbuminuria (AER>30 μg/min); • younger patients with overt diabetic nephropathy (dipstick positive proteinuria) but a normal serum creatinine; • patients with nephropathy and resistant hypertension; • patients with asymmetrical kidneys on ultrasound or renal artery dopplers suggestive of renal artery stenosis; • patients with overt nephropathy but atypical features such as haematuria, absence of retinopathy, or nephrotic levels of proteinuria (>3g/24hr). Patients with diabetic nephropathy often become symptomatic at lower serum creatinine levels than those who are non-diabetic and thus require dialysis earlier. Choosing between haemodialysis and peritoneal dialysis (CAPD) will depend on a variety of factors including patient choice, but difficulties of vascular access and problems such as coexisting autonomic neuropathy makes CAPD the preferred choice for the majority of patients. Visually impaired patients can manage CAPD surprisingly well, but they may not detect their own foot ulcers. Symptomatic autonomic neuropathy is far more common in patients with diabetic nephropathy. Postural hypotension and impotence add to the concerns of the patient but one of the most distressing symptoms is gustatory sweating, which may be intolerable and is effectively untreatable.

CHAPTER 3 • DIABETIC NEPHROPATHY

Gastroparesis is fortunately rare. All of these patients will have other diabetic complications such as retinopathy and ischaemic or neuropathic feet which need regular inspection to detect and prevent deterioration. Anaemia may respond to iron infusion but there is some evidence in patients without diabetes that erythropoietin (EPO) given pre-dialysis may slow the progression of renal deterioration (Fig. 3.9).

80

40 20

Group I (untreated anaemic, overall) Group II (treated anaemic, overall) Group III (untreated non-anaemic, overall)

0 0

5

10

15 20 25 Months of folllow-up

30

35

P = 0.0003

60 P = 0.0024 P = 0.3111

Cumulative renal survival rate (%)

100

40

Fig. 3.9 Reversal of anaemia by erythropoietin can retard progression of chronic renal failure. Kuriyama S et al. Nephron 1997; 77: 176–185.

CURRENT ISSUES •

•

Studies investigating the renoprotective effects of various classes of drugs including thiazolidenediones, protein kinase C inhibitors and vasopeptide inhibitors such as Omapatrilat are ongoing. Only careful attention to the cluster of vascular risk factors will begin to reduce the high mortality from CHD in these patients. Coronary risk prediction charts, e.g. those based on the Framingham data, will probably underestimate the CHD risk and should not be used in this population unless modified to adjust for proteinuria. There is good evidence to suggest that patients with nephropathy should be treated as for secondary prevention, i.e. assume they have established vascular disease and treat them at lower thresholds with statins and aspirin (once BP is controlled). There is also some evidence that statins may reduce proteinuria and have ancillary anti-inflammatory effects in the kidney.

33

34

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

FURTHER READING Lewis EJ, Hunsicker LG, Bain RP, Rhode RD, for the Collaborative Study Group. The effects of angiotensin-converting enzyme inhibition on diabetic nephropathy. N Engl J Med 1993; 329: 1456–1462. Mogensen CE and Cooper ME. Diabetic renal disease: from recent studies to improved clinical practice. Diabetic Medicine 2004; 21: 4–17. Siebenhofer A, Plank J, Horvath K, Berghold A, Sutton AJ, Sommer R & Pieber TR. Angiotensin receptor blockers as anti-hypertensive treatment for patients with diabetes mellitus: meta-analysis of controlled double-blind randomised trials. Diabetic Medicine 2004; 21: 18–25. Viberti GC, Jarrett RJ, Mahmud U, Hill RD, Argyropoulos A, Keen H. Microalbuminuria as a predictor of clinical nephropathy in insulin-dependent diabetes mellitus. Lancet 1982; i: 1430–1432.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Adrian R. Scott MD, FRCP CHAPTER 4 CORONARY HEART DISEASE AND DIABETES

INTRODUCTION Traditionally diabetic vascular complications are divided into microvascular and macrovascular. The uniqueness of diabetic retinopathy, and to a lesser extent nephropathy, has led to microvascular disease sometimes overshadowing the more common and life-threatening coronary and cerebrovascular events. The fascination of doctors with unusual patterns of small vessel disease has meant that for many decades the true impact of vascular disease in people with diabetes has received less attention than it deserves. Both types of diabetes are associated with a many-fold increase in the risk of macrovascular disease, particularly coronary heart disease (CHD), stroke (CVA) and peripheral vascular disease (PVD) leading to gangrene and lower limb amputation. In type 1 diabetes this is strongly associated with nephropathy but many would now consider type 2 diabetes as a cardiovascular disease per se since hyperglycaemia is only one element of a syndrome characterized by insulin resistance, central obesity, hypertension and hyperlipidaemia, all of which contribute to atherogenesis.

EPIDEMIOLOGY OF CHD IN DIABETES In type 1 diabetes the risk of a vascular event increases with duration of diabetes and the presence of nephropathy. Many older studies failed to distinguish between the two major types of diabetes but the recent British Diabetic Association (BDA) cohort study followed insulin-treated patients diagnosed before the age of 30 years, and found an excess of deaths at all ages. Vascular disease was implicated from the third decade onwards (Fig. 4.1). With type 2 diabetes the increased risk of CHD is present from diagnosis and one cohort study found a history of MI in 16.5% of males and 9.7% of females at the time of diagnosis of diabetes. During a 10-year follow-up the age-adjusted incidence of first MI was 1.5-fold higher in diabetic men and up to 8.1-fold higher in diabetic women compared to age-matched non-diabetics. In the Framingham study, 3,000 non-diabetic women were followed for 24 years and no episode of MI prior to the menopause was documented. In diabetic women before the menopause, however, the morbidity and mortality from atherosclerotic events was equal to or greater than diabetic men. These differences in atherosclerosis between diabetics and non-diabetics are common across the world but significant differences also exist in the incidence of CHD events in diabetics from one country to another. This suggests that the risk of CHD can be modified and that environmental factors probably have a role. For example, a Central American with diabetes has less coronary atherosclerosis than a North American without diabetes. The incidence of CHD and

35

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

10 000 Mortality per 100 000 per year (log scale)

36

1000

100 10

Male general population Female general population Male cohort Female cohort

1 0

14 5–19 0–24 5–29 0–34 5–39 0–44 5–49 0–54 5–59 0–64 5–69 4 1 5 3 5 4 6 3 6 2 2

10–

Age (years)

Fig. 4.1 BDA Cohort Study (1972-1993): Cardiovascular disease mortality rates in insulin-treated diabetes diagnosed under age 30. Laing SP et al. Diabetic Medicine 1999; 16: 466–471.

PVD in the Japanese is very low and this is usually attributed to a diet high in carbohydrate, rich in fish and low in fat. Macrovascular complications among Japanese diabetics is one fifth as common as among Caucasian diabetics. Hawaiian Japanese diabetics, however, have an equivalent cardiovascular mortality to Caucasian diabetics, suggesting that these differences are modifiable.

AETIOLOGY Genetics and environment contribute to the cluster of (CHD) risk factors associated with diabetes but fetal nutrition would also appear to have a role. This is suggested by small-for-dates babies who are more likely to have CHD and/or type 2 diabetes in middle-age than normal birth weight babies, particularly if they also become obese (so exacerbating the hyperinsulinaemia/insulin resistance). CHD in people with diabetes is associated with the usual risk factors such as hypertension, smoking, obesity, elevated LDL-cholesterol, low HDL-cholesterol and renal disease, many of which are more common than amongst non-diabetics. The interactions are complex, however, and the Multiple Risk Factor Intervention Trial (MRFIT) study which followed over 300,000 men for seven years found that hypertension and hyperlipidaemia had a greater impact in the diabetic subgroup (Fig. 4.2). Diabetic men with a cholesterol of >7.3 mmol/l were nearly six times as likely to develop CHD over the period of follow-up than diabetic men with a cholesterol of <5.5 mmol/l. Thirty percent of smokers

CHAPTER 4 • CORONARY HEART DISEASE AND DIABETES

Ten-year CHD mortality (per 1000)

(a)

(b)

80

80

60

60

40

40

20

20

10

10 Diabetic Non-diabetic

0

0 110 120 130 140 150 160 Systolic blood pressure (mmHg)

4

5

6

7

Serum cholesterol (mmol/l)

Fig. 4.2 Multiple Risk Factor Intervention Trial: Effects of systolic blood pressure (a) and serum cholesterol concentration (b) on 10-year mortality from coronary heart disease (CHD) in 342,815 non-diabetic and 5,163 diabetic subjects aged 35–57 years who initially had not suffered a myocardial infarction. From Williams G & Pickup JC, Textbook of Diabetes 1999 2nd edn, Blackwell Science.

were dead at the end of the 7-year period. Diabetics also have metabolic abnormalities which encourage thrombosis and discourage fibrinolysis. Thus fibrinogen, von Willebrand factor, plasminogen activator inhibitor-1 (PAI-1) and plasma viscosity are all elevated and platelet function is abnormal.

PROGNOSIS In-hospital and one-year mortality has been shown to be two and fourfold greater amongst people with diabetes, especially women. All-cause mortality in men and women aged 61–75 years admitted to hospital with an acute MI in Southern Derbyshire, UK, confirmed this pattern, with the largest number of deaths occurring within the first month (Fig. 4.3). The reasons for this are not clear but sudden death, heart failure and reinfarction are all more common in diabetic patients following MI. There is no evidence that infarct size is greater but underlying coronary artery atherosclerosis is more severe and there is some evidence of a cardiomyopathy associated with diabetes which could predispose to a worse outcome post-MI. A raised admission blood glucose in patients with acute coronary syndrome is predictive of both in-patient and long-term mortality. A meta-analysis of 15 previous studies

37

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

50 Cumulative mortality (%)

38

40 30 20 10

Women with DM Men with DM

Women with no DM Men with no DM

0 10 day

30 day

3 mth After MI

6 mth

12 mth

Fig. 4.3 Mortality rates in men and women with and without diabetes aged 61-75 years in Southern Derbyshire between 1995 and 1998 (unpublished data).

reported that for people without diabetes an admission BG of >6 mmol/l increased hospital mortality nearly fourfold. For people with diabetes, an admission BG of >10 mmol/l increased mortality 1.7 times compared to people with diabetes with BG of <10 mmol/l. In a prospective study of admission BG in 336 patients with AMI (12% known DM), 1 year mortality was 19.3% if admission BG was <5.6 rising to 44% if BG was >11 mmol/l.

ASSESSING VASCULAR RISK IN PEOPLE WITH DIABETES Stratification of risk is helpful in prioritizing care. This ensures that those with the highest risk are treated with some urgency; resources are targeted according to need; and low risk individuals are not put at unnecessary disadvantage from long-term potential drug side-effects. All coronary risk prediction charts are based on the same Framingham data which takes into account age, gender, smoking and diabetes status, systolic blood pressure, and total cholesterol/HDL ratio. The presence or absence of left ventricular hypertrophy (LVH) and adjustment for microalbuminuria status can be included in the Framingham equations and results in an estimate of risk of a cardiovascular event (fatal or non-fatal) over a 10-year period. They are for primary prevention only, with patients categorized as high risk predicted to have over a 30% chance of an event over 10 years. To put this in context, a person with established vascular disease has a more than 40% risk over 10 years. The original Framingham cohort was largely white and contained only a relatively small number of patients with diabetes. Thus, coronary risk prediction charts should be used with caution as they will tend to under-estimate

CHAPTER 4 • CORONARY HEART DISEASE AND DIABETES

CHD risk in some groups, e.g. British Asians, New Zealand Maori, patients with nephropathy and where there is a family history of premature vascular disease suggestive of a primary hyperlipidaemia. Haffner and colleagues published evidence that the risk of death in patients with type 2 diabetes (but no prior MI) was similar to that of a non-diabetic with a past history of MI (Fig. 4.4). This suggests that, in terms of secondary prevention, patients with diabetes could be considered as having the same risk of an event as someone with established vascular disease. Despite such a high risk, screening for asymptomatic CHD remains controversial. A positive exercise test in this group has a low predictive value for significant CHD on angiography. The high prevalence of hypertension (and therefore increased left ventricular mass) can give abnormal ST-segment responses to exercise as well as false positive nuclear myocardial perfusion scanning.

EVIDENCE BASED PRACTICE The acute management of ST-segment elevation MI (STEMI) in a patient with diabetes is exactly the same as for a person without diabetes, except that all those with diabetics (and patients with unknown diabetes status but an admission serum glucose > 11 mmol/l) should receive a glucose and insulin infusion (so-called DIGAMI regimen) to maintain BG within the normal range for at least 24 hours. In the original Swedish DIGAMI studies using this i.v. regimen, patients were subsequently treated with a basal-bolus subcutaneous insulin

100

Survival (%)

90 80 70 60 50

Non-DM without MI DM without MI

Non-DM with MI DM with MI

40 0

1

2

3

4 Years

5

6

7

8

Fig. 4.4 Probability of death in 1,059 subjects with type 2 diabetes and 1,378 non-diabetic subjects with and without prior MI. Haffner S et al. N Engl J Med 1998; 339: 229–34.

39

40

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

regimen for at least three months and this was associated with a 29% reduction in mortality. The group which appeared to benefit most were the lower risk, younger patients on diet alone prior to the MI. Current evidence cannot distinguish between the benefits of early glucose and insulin i.v. infusion versus the benefits attributable to post-discharge subcutaneous insulin therapy in accounting for the overall reduction in mortality. DIGAMI2 has attempted to unravel the benefits of the individual components and although complete, the results have not yet been published. Effective interventions frequently show greatest benefits in the higher-risk patients and thrombolysis, beta-blockers and statins all demonstrate significantly larger absolute reductions in morbidity and mortality in the diabetic subgroups. Following STEMI treated with reperfusion therapy, there was an increased risk of adverse outcomes in patients with diabetes but they still appeared to derive a greater absolute benefit. The Fibrinolytic Therapy Trialists Group demonstrated that the use of fibrinolyic therapy saved 37 lives per 1,000 patients with diabetes, compared with 15 lives per 1,000 patients without diabetes. In trials studying the addition of GPIIb/IIIa inhibitors in patients with STEMI treated by thrombolysis, people with diabetes experienced a marked increase in bleeding rates which outweighed the small reduction in the composite end-points of death or reinfarction. Primary percutaneous transluminal angioplasty (PTCA) has been shown to improve outcomes among patients with STEMI when compared with fibrinolytic therapy. Patients with diabetes also benefit and do even better with primary stenting plus GPIIb/IIIa inhibition. Meta-analyses of beta-blocker trials show a 35% reduction in mortality when given at the time of the MI (orally or intravenously) in diabetics compared to a 13% reduction in non-diabetics. When given as secondary prevention post-MI, the corresponding results are –48% and –33% respectively. A similar pattern was seen with statins post-MI, e.g. in the Cholesterol And Recurrent Events (CARE) study using Pravastatin 40 mg and in the Scandanavian Simvastatin Survival Study (4S) which used 20–40mg simvastatin (Fig. 4.5). Reductions in mortality were greater in the diabetic subgroups, as were the reductions in duration and number of hospital admissions in the statin treated group. Angiotensin-converting enzyme inhibitors (ACE-Is) given post-MI reduce mortality by 22% in patients with a left ventricular ejection fraction of <40%. The Studies Of Left Ventricular Dysfunction (SOLVD) demonstrated similar benefits in diabetics. The metanalysis performed by the ACEInhibitor Myocardial Infarction collaborative group of ACE-I therapy started on the day of MI and continued for 4–6 weeks showed that five deaths were prevented per 1,000 treated patients and there was a reduced risk of heart failure. This analysis included the CONSENSUS II trial which found a

CHAPTER 4 • CORONARY HEART DISEASE AND DIABETES

1.1

Proportion without CHD event

1 0.9 0.8 0.7 0.6 0.5

DM, simvastatin DM, placebo

Non-DM, simvastatin Non-DM, placebo

0.4 0

1

2

3 Years

4

5

6

Fig. 4.5 Effects of simvastatin 20–40 mg on fatal and non-fatal cardiovascular events in patients with diabetes and non-diabetics with known CHD: Scandanavian Simvastatin Survival Study (4S).

negative interaction between aspirin and intravenous enalapril (an 11% increase in mortality in the enalapril group). However, in this study hypotension and severe cardiac failure at the time of randomization were not exclusion criteria. GISSI-3, using lisinopril, showed a 44% and 27% reduction in 6-week mortality in type 1 and type 2 diabetics respectively. The large Heart Outcomes Prevention Evaluation (HOPE) study randomized patients with CHD but clinically normal left ventricular (LV) function to ramipril 10 mg or placebo, with or without vitamin E. The diabetic subgroup (Micro-HOPE) of 3,577 patients were followed for a mean of 4.5 years and treatment with ramipril was associated with a 24% reduction in all-cause mortality, a 22% reduction in the rate of MI and a 33% reduction in the rate of stroke (Fig. 4.6). Vitamin E had no benefit.

Does improving glycaemic control reduce cardiovascular risk? The Diabetes Complications and Control Trial (DCCT) of conventional vs. intensive glycaemic control in type 1 diabetes was underpowered to answer this question as the cohort was relatively young and therefore the number of events was low. There was a trend towards a reduction in cardiovascular events in the intensively treated group. The UK Prospective Diabetes Study (UKPDS) trial

41

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

0.12 Cumulative mortality Kaplan-Meier rates

42

Ramipril Placebo

0.10 0.08

RRR = 37% (21-51) P = 0.0001

0.06 0.04 0.02 0.00 0

500

1000 Days of follow-up

1500

2000

Fig. 4.6 Micro-HOPE Study: Effects of Ramipril or placebo on cardiovascular death in 3,577 diabetics with normal left ventricular function. Lancet 2000; 355: 253–259.

also had a non-significant (16%) reduction in fatal and non-fatal MIs in the intensively treated group after 15 years. Again, this may be because of the relatively low event rate in a group of patients who were recruited at diagnosis and were relatively young. Additionally, the trial had as exclusion criteria diagnoses such as severe peripheral vascular disease and existing CHD. However, further analysis of the UKPDS data showed that irrespective of the treatment arm, the lower the HbA1C, the lower the event rate for both micro and macrovascular complications. There was a 21% decrease in all diabetes related end-points for each 1% decrement in HbA1C. The benefit was greater for microvascular complications but a significant reduction in macrovascular events was also seen. The previous fears raised by the University Group Diabetes Program (UGDP) in the 1970s that sulphonylureas were associated with increased cardiovascular mortality was not realized and there appeared to be no difference between insulin and sulphonylureas as long-term therapy, in this regard. Nevertheless, there remains uncertainty about their use in the acute setting (e.g. acute MI or angioplasty etc.) where there is some evidence that there is loss of protective ischaemic pre-conditioning, possibly due to their inhibitory effect on the ATP sensitive K+ channel. However, in the UKPDS, metformin used as monotherapy in the obese patient was associated with a significant reduction in CV deaths compared to insulin or sulphonylureas. There is no long-term data on thiazolidenedione derivatives or the newer sulphonylureas.

CHAPTER 4 • CORONARY HEART DISEASE AND DIABETES

Anti-platelet therapy There have been no specific trials of aspirin in acute MI in people with diabetes but secondary prevention studies demonstrate a 25% reduction in cardiovascular events and a 15% reduction in mortality with apparently similar results in people with diabetes. The optimal dose of aspirin remains uncertain but lower doses are associated with fewer haemorrhagic complications. Evidence of benefit of aspirin in people with diabetes for primary prevention must be extrapolated from trials in non-diabetics (mostly male) which have suggested that there is a reduction in non-fatal cardiovascular and cerebrovascular events. These trials have also shown that 75 mg is probably as effective as 500mg and is not associated with an increased risk of cerebral haemorrhage in patients with controlled hypertension. There was a non-significant reduction in fatal and non-fatal MI in the Early Treatment of Diabetic Retinopathy Study using 325 mg aspirin.

Lipid-lowering therapy Statins will reduce mortality, the need for revascularization and cardiovascular and cerebrovascular events, but the optimum dose and lipid targets remain uncertain. The British and Europeans recommend achieving a total cholesterol <5 mmol/l, and LDL <3 mmol/l, whilst the American Diabetes Association suggests an LDL target <2.6mmol/l, HDL >1.15mmol/l and triglycerides <2.3 mmol/l. Improved glycaemic control will enhance the lipid profile, though glitazones (more so rosiglitazone) cause a 12% increase in LDL (compensated in part by an increase in HDL). These LDL particles are less dense and theoretically less atherogenic though there are no long-term outcome studies yet completed. Fibrates have little effect on LDL but increase HDL and lower triglycerides. Only gemfibrozil has been shown to reduce cardiovascular mortality – which in the Veterans Affairs–High density Lipoprotein Intervention Trial (VA-HIT) study resulted in a 22% reduction in CHD deaths or non-fatal MIs. This study included 25% diabetics and so for patients intolerant of statins, gemfibrozil 600mg twice daily is a suitable alternative. Diabetic patients presenting with non-ST segment elevation (NSTE) acute coronary syndrome have a higher risk profile for subsequent events and some studies suggest they derive greater benefit when managed with rapid initiation of intensive management together with early angiography and revascularization. For most parts of the world, such aggressive reperfusion strategies remain beyond reach. Nevertheless, intensive multiple risk factor treatment in high risk patients (such as a patient with type 2 diabetes and microalbuminuria) has been shown to reduce cardiovascular events by as much as 53% (CI 27–76) over an 8-year period when compared to a less intensive strategy.

43

44

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

People with diabetes may not always be suitable for revascularization procedures as their atherosclerotic disease is often severe and widespread. Long-term survival following CABG is less good in diabetic patients. The Bypass Angioplasty Revascularization Investigation (BARI) trial suggested that 5-year outcomes from angioplasty were inferior to CABG in diabetic patients. Though with improved techniques, routine use of GPIIb/IIIa inhibitors, and drug-eluting stents, percutaneous intervention may become a more desirable option in diabetic patients with multivessel disease.

FURTHER READING Hales CN, Barker DJP, Clark PMS et al. Fetal and infant growth and impaired glucose tolerance at age 64. BMJ 1991; 303: 1019–22. Malmberg K, Ryden L, Efendic S et al. on behalf of DIGAMI Study Group. A randomised trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute MI. (Oigami Study): effects on mortality at 1 year. J Am Coll Cardiology 1995; 26: 57–65. Laing, S. P. Swerdlow, A. J. Slater, S. D. Bothat, J. l. Burden, A. C. Waugh, N. R. Smith, A. W. M. Hill, R.D. Bingley, P. J. Patterson, C. C. Qiao, Z. Keen, H. The British Diabetic Association Cohort Study, I: all-cause mortality in patients with insulin-treated diabetes mellitus; Diabet. Med. 1999; 16: 459–465.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Adrian R. Scott MD, FRCP CHAPTER 5 DIABETES AND CEREBROVASCULAR DISEASE

INTRODUCTION The great disabler of all the macrovascular complications – stroke – is, as one might expect, more frequent in people with diabetes and the outcome worse than in non-diabetics. The prevalence of cerebral infarcts, especially lacunar infarcts, is increased but the prevalence of subarachnoid haemorrhage, cerebral haemorrhage, and transient ischaemic attacks are decreased, despite hypertension being so common in the diabetic patient. The presence of diabetic nephropathy and coronary and peripheral vascular disease are risk factors for stroke in the diabetic patient. Afro-Caribbeans and Afro-Americans with diabetes are particularly at risk. A higher prevalence of stroke is found in the patient with both diagnosed and undiagnosed diabetes and glucose intolerance and, as with myocardial infarction (MI), most studies show that individuals with admission serum glucose of >6.6 mmol/l have a higher morbidity and mortality.

EPIDEMIOLOGY A number of large studies have confirmed the higher prevalence of stroke in the diabetic population. In the Framingham study the fourfold excess in male diabetics occurred in the 5th and 6th decades, whereas in females with diabetes the excess was a decade later. Most studies (usually of hospitalized patients) suggest a relative risk of stroke 2–3 times that of non-diabetics though the Swedish Gothenburg study put this excess as high as sixfold in men and 13-fold in women. Most studies have not distinguished between insulin requiring and non-insulin requiring diabetes, but for type 1 the excess may not be so great as with type 2. Certainly it is not as common as cardiovascular disease – Dekert’s long term mortality study (1976) of people with diabetes diagnosed before age 30 and followed up for more than 40 years showed a 10% incidence and 7% mortality from stroke. A similar UK study of diabetics dying before their 50th birthday found a similar mortality from stroke compared to 41% from coronary heart disease (CHD) and 19% from nephropathy. Diabetes mellitus is associated with higher mortality, worse functional outcome, more severe disability after stroke and a higher frequency of recurrent stroke. Short- and long-term mortality is increased and in one carefully matched Finnish study, 5-year mortality was 60% in the non-diabetic controls with stroke compared to 80% in those with diabetes.

45

46

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

AETIOLOGY Cerebral blood flow disturbances, impaired cerebrovascular reactivity, and damage to large and small extra- and intracranial cerebral vessels have been found in humans and animals with diabetes. Autopsy studies suggest that diabetic patients are susceptible to cerebral small-artery disease and lacunar infarction. These strokes result from vascular occlusion of small arteries at the base of the brain resulting in small deep arterial infarcts usually less than 15 mm in diameter and typically occur in hypertension and diabetes. Embolism from large vessel atheroma and heart (particularly post-infarct) is also more common. In one prospective study, carotid stenoses of >50% were present in 8.2% of diabetics compared with 0.7% of age-matched controls. However, only 28% of diabetics with an ischaemic cerebral event had a significant carotid stenosis suggesting that smaller vessel disease is more important. The precipitant for the occlusion is not clear but appears to be linked to excessive glycation and oxidation, endothelial dysfunction, increased platelet aggregation, impaired fibrinolysis and insulin resistance. Cerebrovascular blood flow has been shown to be abnormal in people with diabetes, both of auto-regulation and in response to vasodilators such as CO2. Endothelial dysfunction with failure to vasodilate in response to nitric oxide has been postulated; autonomic neuropathy may also be a factor. Blood glucose on admission correlates both with survival and degree of recovery. Several studies have demonstrated a worse outcome with a presenting blood glucose >6.6mmol/l. Whether hyperglycaemia adversely affects stroke outcome or primarily reflects stroke severity is not clear – animal studies of acute hyperglycaemia prior to cerebral ischaemia show more severe histological damage and a worse outcome but there is no evidence in humans that infarct size is larger. Hyperglycaemia might theoretically worsen stroke damage in a number of ways: the local hypoxia induced by acute cerebral ischaemia results in glucose being metabolized anaerobically causing lactic acid to accumulate. The resultant local acidosis damages vascular, glial and neuronal tissue. In addition, ischaemia causes accumulation of the neurotransmitters, glutamate and aspartate, in the extracellular tissues. Usually these neurotransmitters cause stimulation of a nerve at a post-receptor site and depolarization. When accumulation occurs hyperstimulation also occurs, followed by neuronal death, though glial and vascular tissue are spared. This neural toxicity may result from an increase in intracellular calcium following neuronal hyperstimulation.

CLINICAL PRESENTATION Strokes are common and the lack of challenging interventions has often meant that these patients are not always adequately assessed and hence

CHAPTER 5 • DIABETES AND CEREBROVASCULAR DISEASE

receive sub-optimal care. Not all acute neurological events are strokes and consideration must be given to the underlying cause – classically, hypoglycaemia may present with altered consciousness but also with focal neurology and if missed, permanent neurological sequelae may result. There are also reports in the literature of focal fits and neurological signs in association with hyperglycaemia but these preceded modern scanning technology so may have represented small strokes or transient ischaemic attacks (TIAs). However, it is important not to overlook the diagnosis of non-ketotic hyperosmolar states (HONK) as the dehydration and elevated viscosity may have led to arterial occlusion causing stroke, MI, or even peripheral gangrene. Silent ischaemia is relatively common and patients may present with a stroke and uncontrolled diabetes as a complication of an earlier painless MI.

EVIDENCE-BASED PRACTICE The reality is that acute stroke management in the person with diabetes is based on extrapolation of the data from non-diabetics as there are, as yet, no prospective studies of stroke management in diabetics. However, there is one proviso: these are high risk patients with a multi-system disorder, whom as a group do badly, both in terms of survival, and rehabilitation. Early interventions must be undertaken promptly as any delay is not likely to improve prognosis. The role of thrombolysis of acute stroke remains controversial but a systematic review of 12 controlled trials involving 3,435 patients assessed the use of intravenous thrombolytic therapy (with a number of agents) started within six hours of the onset of symptoms of ischaemic stroke. Thrombolysis reduced the proportion of patients who died or remained dependent on others at the end of trial follow-up, up to six months later (61.5% vs. 68% of control patients not given thrombolysis). Results were more impressive if treatment was started within three hours (56.6% vs. 70.7%). Alteplase seemed superior to streptokinase but overall there was an increased risk of symptomatic intracranial haemorrhage (9.6% vs. 2.6%). Overall, the risk of dying within two weeks was increased in those receiving thrombolytic therapy (20.9% vs. 11.9%) despite the improvement in the composite end-point of death or dependency. Whether people with diabetes benefit similarly from thrombolysis is not known. Use of anticoagulant therapy with unfractionated or low-molecular weight heparin for acute ischaemic stroke is associated with an increase in haemorrhagic stroke but with no positive benefit in terms of mortality or dependency. If CT scanning is not immediately available to rule out haemorrhage, administration of aspirin (orally or rectally) should, on balance, be given sooner rather than later. With the exception of immediate post-stroke hypertension management (about which little is known but for which avoidance of treatment is recommended for at least four weeks), correction of other co-morbidities

47

48

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

seems logical but lacks the confirmation of randomized-controlled trials. Thus, dehydration and hypoxia should be avoided, with administration of antibiotics if respiratory infection supervenes. The benefits of treatment of hyperglycaemia in this situation are unknown but, as it correlates with a worse outcome, a DIGAMI-style glucose and insulin infusion to maintain near-normoglycaemia would seem to be a cheap and easily implementable solution. Heparin prophylaxis is best avoided as there appears to be an increased risk of secondary cerebral haemorrhage but prevention of deep venous thromboses (DVTs) is achievable with elasticated thromboembolic stockings. Secondary prevention data is based entirely on general population studies with none having been conducted in people with diabetes alone. General management of vascular risk factors using targets similar to those for diabetic patients with CHD would seem logical. The dose of aspirin is uncertain and some authors have suggested that people with diabetes may need higher doses to achieve the same anti-platelet effects. The European Stroke Prevention Study showed that, in the general population, aspirin and sustained-release dipyridamole are equally effective secondary prevention in reducing the risk of stroke and/or death. Addition of dipyridamole is justified if the patient has a cerebrovascular incident whilst on aspirin as this study showed that the combination was significantly more effective than either alone. Clopidogrel is slightly superior to aspirin at the prevention of recurrent stroke but probably not sufficiently cost-effective to justify widespread use. In the Clopidogrel vs. Aspirin in Patients at Risk of Ischaemic Events (CAPRIE) study 7.2% of those treated with clopidogrel 75mg/day had an event compared with 7.7% of patients receiving aspirin 325mg/day. It may be more affordable if patients with true aspirin allergy are targeted or those who have gastrointestinal intolerance to aspirin but not clopidogrel. Reports of thrombotic thrombocytopenic purpura with clopidogrel are worrying but fortunately rare. Warfarin should be substituted for antiplatelet therapy if the cause of the stroke is attributable to atrial fibrillation or other emboli from the heart. It is probably safer to wait two weeks after the stroke before making this change. Primary prevention of stroke in people with diabetes is of interest because there are a number of studies where stroke prevention has been a significant secondary end-point. The UK Prospective Diabetes Study (UKPDS) of hypertension, where the target for tight blood pressure control was <150/85, showed a 44% reduction in strokes compared to the less tight blood pressure control group (target BP <180/105). Ramipril 10mg daily in the diabetic sub-group of the Heart Outcomes Prevention Evaluation (HOPE) study (Micro-HOPE) reduced the number of strokes from 6.1% to 4.2% (a 33% relative risk reduction) in a cohort of patients

CHAPTER 5 • DIABETES AND CEREBROVASCULAR DISEASE

most of whom had established CHD. The difference in blood pressure between the ramipril and placebo groups at the end of the study was only 2.5/1.0 – some would say this was insufficient to account for the difference in stroke rates, suggesting a different mode of vascular protection by angiotensin-converting enzyme inhibitors (ACE-Is) than simply lowering blood pressure. However, sub-groups studied with 24 hour blood pressure monitoring suggest that casual blood pressure readings underestimate the difference between the treated and untreated groups and most experts conclude that blood pressure lowering is more important than the type of antihypertensive drug used. In both CARE and LIPID (secondary prevention studies in patients with previous MI or angina) pravastatin 40mg reduced the risk of fatal and nonfatal cerebrovascular accidents by 31% and 20% respectively. The numbers with diabetes, however, were too small to analyse as a separate group. In summary, stroke prevention in people with diabetes is about aggressive management of all vascular risk factors but with an emphasis on tight blood pressure control and use of antiplatelet therapy, ACE-Is and statins. New drugs in development offer the possibility of limiting neuronal damage at the time of the acute event.

CURRENT ISSUES •

•

•

Correction of hyperglycaemia with a glucose and insulin infusion at the time of the acute stroke is the subject of a randomized controlled trial. Until this is reported, the DIGAMI study of glucose and insulin infusion in acute myocardial infarction provides sufficient evidence to suggest that stroke patients are likely to benefit in a similar way, and uncontrolled hyperglycaemia should not be neglected. The place of thrombolysis in the management of acute stroke has yet to be determined and should probably only be undertaken in the context of randomized controlled trials. Current evidence suggests it must be given within three hours of the onset of symptoms – a goal not deliverable in most countries. Neuroprotective therapy remains experimental but a number of agents are being investigated. They include clomethiazole, glycine antagonists, lubeluzole and magnesium. Such treatment, given promptly after stroke onset, aims to limit ischaemic damage by protecting damaged but potentially viable neural tissue. Animal studies have suggested a neuroprotective effect of lubelozole but in humans it has not been shown to reduce neurological disability or mortality.

49

50

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

FURTHER READING Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy – I: Prevention of death, myocardial infarction, and stroke by prolonged platelet therapy in various categories of patients. BMJ 1994; 308: 81–106. Counsell C, Sandercock P. Anticoagulant therapy compared to control in patients with acute presumed ischaemic stroke (Cochrane Review). The Cochrane Library, Issue 2, 1998. Diener HC for the European and Australian Lubelozole Ischaemic Stroke Study Group. Multinational randomised controlled trial of Lubelozole in acute ischaemic stroke. Cerebrovasc Dis 1998; 8: 172–181. Wardlaw JM, Warlow CP, Counsell C. Systemic review of evidence on thrombolytic therapy for acute ischaemic stroke. Lancet 1997; 350: 607–614.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

CHAPTER 6 ERECTILE DYSFUNCTION

Adrian R. Scott MD, FRCP

INTRODUCTION Diabetes may lead to sexual dysfunction in both men and women. For the purposes of this book it is not possible to consider diabetes-related sexual dysfunction in women though the causes are similar in both genders. For years the treatment of erectile dysfunction (ED) was complex, unsatisfactory and disliked by patients and/or their partners. Embarrassment lead to a conspiracy of silence: ‘I won’t ask if you won’t say’ – and this important and distressing complication was too often ignored. With the advent of oral therapies for ED and the publicity surrounding the launch of Viagra (sildenafil), there has been much more open discussion. Therefore, even though a third of diabetic men will not respond to sildenafil, the dialogue has begun and other treatment avenues can be explored. This chapter looks at the aetiology of erectile dysfunction in people with diabetes, the importance of a correct diagnosis, followed by a look at past, present and future treatments. Sexual function declines with age and it must be remembered that the range of normal in describing sexual activity is very wide. It is estimated that as many as 25% of men over 65 suffer ED and in the diabetic population this is much higher. Published studies vary, but quoted ranges are between 30 and 60% suffering from partial or complete ED, with ejaculatory abnormalities such as retrograde ejaculation occurring in a smaller proportion.

CLINICAL PRESENTATION The term ED is preferred to impotence as there are clearly grades of dysfunction and impotence has such negative connotations of failure. An erection insufficient for intercourse is the usual definition and the skill of the clinician is to allow the patient to acknowledge there is a problem without intimidating them with what could be seen as overly intrusive questions. Nevertheless, a clear description of the onset and type of problem is essential if the appropriate treatment is to be selected. Intermittent or occasional erectile failure is very common and usually due to anxiety, alcohol or sexual indifference. How the partner copes with this failure can often determine future function – annoyance, anger or even ridicule will heighten anxiety on future attempts, leading to ‘performance anxiety’ and recurrent failure. This is perhaps the commonest cause of ED in the non-diabetic population. People with diabetes are not immune to this type of failure and anxiety may be exacerbated by their awareness of ED as a complication of diabetes.

51

52

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Mechanical problems such as phimosis (caused by hyperglycaemia-induced balanitis) or Peyronie’s disease maybe concealed in the patient’s description and should be specifically asked for. ED can be a manifestation of depression (as well as leading to it) so questions regarding sleep pattern, self-worth, mood etc. are important. A loss of libido is not typical of ED caused by diabetes and may suggest depression, hypogonadism or chronic ill-health. Chronic alcohol dependency as well as acute binges are other causes. The natural history of sexual function in man is variable but changes and declines with age. Most organic causes of ED result in preserved libido but are associated with a relatively slow onset or loss of spontaneous erections, especially those present on waking. ED only with their partner but not in response to auto-erotic stimuli such as masturbation strongly suggests psychogenic causes. It is essential to explore the relationship (if there is one) as dysharmony will usually cause sexual tension and could lead to ED. Attitudes of the male to sex may need to be explored, as for many, an erect penis is confirmation of their manhood and successful place in society. Sexual failure can lead to shame or anger towards their partner and the mistaken belief that sex equals penetration leads them to withdraw from all physical contact, leaving them very isolated.

PATHOPHYSIOLOGY It is too simplistic to attribute impotence to autonomic neuropathy or macrovascular disease. There is a clear association with microvascular disease including retinopathy and renal disease and most patients have abnormal cardiovascular reflexes and/or peripheral neuropathy. However, the converse is not true – many patients have abnormal autonomic function but normal sexual function. It is worth considering the mechanisms of normal erection which, after the appropriate sexual stimulation, depends on relaxation of the smooth muscle of the corpus cavernosa allowing expansion of the lacunar spaces against the tunica albuginea, the mechanical compression of subtunical venules and the entrapment of blood in the corpus cavernosa. This increased arterial flow into the penis is achieved by dilatation of the penile arteries. Nervous control of erection is mediated by parasympathetic fibres from the 2nd and 3rd sacral nerve roots. Sympathetic activity fibres arise mainly from the lower thoracic (T11) down to the 2nd lumbar nerve root and are mainly associated with ejaculation and detumescence. These autonomic fibres travel with somatic fibres in the pudendal nerve which comes off the sciatic nerve. An essential part of the mechanism leading to penile erection is nonadrenergic, non-cholinergic relaxation of vascular and cavernous smooth muscle resulting in blood flow up to 100ml/min. A number of chemical pathways are involved but the most important is the release of nitric oxide (NO)

CHAPTER 6 • ERECTILE DYSFUNCTION

which stimulates the formation of cyclic GMP (cGMP) by guanylate cyclase (Fig. 6.1). CGMP relaxes smooth muscle by decreasing intracellular calcium and an erection ensues. Detumescence occurs when cGMP is broken down by phosphodiesterase type 5 (PDE5). Prostaglandins (PG) also play a role in erection and detumescence. Other chemicals include vasoactive intestinal polypeptide (VIP), and purinergic agonists e.g. adenosine, endothelin, and neuropeptide-Y. Diabetes has effects on the neural and vascular elements of erection, and penile biopsies from impotent diabetic men showed reduced number of immunoreactive nerves as well as VIP immunoreactivity. Diabetes causes inhibition of PG synthesis and cigarette smoking causes acute vasoconstriction of the penile arterial blood flow by inhibiting PGI2 synthesis. Hyperlipidaemia disrupts both PG and NO synthesis at a vascular endothelial level and is frequently part of the diabetic metabolic syndrome.

Nitrergic

Vipergic

NO

VIP Smooth muscle

cGMP GTP

VIP-R

NA AD β2-ADR

cAMP ↓ [Ca2+]

ATP

PGE

Hyperpolarization Na--Pump

EP-R

K--channel

K-

Endothelium

NO

Fig. 6.1 Regulation of penile smooth muscle relaxation: cGMP, cAMP and hyperpolarization. (NA, noradrenaline; AD, adrenaline; β2-ADR, β2-adrenergic receptor; EP-R, prostaglandin E receptor; VIP-R, vasoactive intestinal peptide receptor; NO, nitric oxide.

53

54

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

KEY DIAGNOSTIC FEATURES It is not always possible to differentiate between psychogenic and organic ED as there is often overlap, particularly in the man with diabetes. Also, it is easy to assume that all ED is a result of their diabetes which should be a diagnosis arrived at by excluding other causes (Table 6.1). The multiple co-morbidities in people with diabetes mean they are frequently on drugs that may interfere with sexual function. Antihypertensives are a particular problem but unless there is a clear history associating the onset of ED with drug therapy it is usually an unrewarding task to stop one agent and try another. The UK Prospective Diabetes Study (UKPDS) showed that over a third of patients need three or more drugs to control hypertension and ultimately the patient may have to face the difficult choice between well-controlled hypertension and erectile function. Frequently however, the ED is a result of the diabetes and the presence of microvascular complications increases this possibility. The onset is gradual, with the patient observing a reduction in the quality of erections and spontaneous early morning erections become increasingly scarce. The sensation of ejaculation diminishes and retrograde ejaculation may occur. Libido remains normal unless depression supervenes.

EVIDENCE-BASED PRACTICE Although the diagnosis is usually made on the history, a careful physical examination is important, particularly of the genitals looking for evidence of hypogonadism or phimosis. A few simple blood tests should be done in all patients but more sophisticated investigations such as autonomic function tests and penile vascular studies are rarely necessary outside of the research setting (Table 6.2). Sex hormone binding globulin may be helpful in interpreting a borderline low testosterone. Involvement of the partner at an early stage should be encouraged as their cooperation is essential if treatment is to be successful. Options for treatment include, of course, no therapy and many couples may opt for this having modified their sexual lifestyle to accommodate the male partner’s ED. For others, just the open discussion and the restoration of physical contact without sex is enough. The treatment options are as follows: • Selective inhibitors of phosphodiesterase 5 (PDE5) – the enzyme responsible for breaking down cGMP – include sildenafil, vardenafil and tadalafil. They help to restore natural erectile function but will only work in the presence of sexual stimulation. An overall response rate of 80% has been reported (Fig. 6.2) but in diabetic men up to a third gain little benefit even with the larger dose. It is taken about an hour before sexual activity starting with a low dose and adjusting according to response. Headache, flushing, dyspepsia, visual disturbances and nasal congestion have all been

CHAPTER 6 • ERECTILE DYSFUNCTION

Causes of erectile dysfunction • • • • • • • • • • • •

Macrovascular disease Microvascular disease Surgical or traumatic damage to pelvic vasculature Neurological disorders, e.g. MS Spinal cord damage/disease Autonomic neuropathy Hypogonadism Psychogenic Previous priapism Peyronie’s disease Drugs: antihypertensives, antidepressants, tranquillizers, anti-androgens etc. Chronic liver/renal disease

Table 6.1 Causes of erectile dysfuntion.

Minimum investigations in the assessment of erectile dysfunction in a patient with diabetes • • • • •

Physical examination Testosterone Prolactin Gonadotrophins Sex hormone binding globulin

Table 6.2 Minimum investigations in the assessment of erectile dysfunction in a patient with diabetes.

reported. PDE5 inhibitors may potentiate the hypotensive effects of nitrates and use in these patients is contraindicated. Early fears of sudden deaths in patients taking sildenafil (the first of the PDE5 inhibitors to reach the market) have not been borne out but it must be remembered that intercourse can be very physical and in patients with a high prevalence of coronary heart disease may be potentially harmful. There have been no head to head studies comparing the efficacy of the three agents and the main difference between them is the longer duration of action of tadalafil (up to 24 hours after administration). • Sub-lingual or intra-nasal apomorphine is an alternative to PDE5 inhibitors though is less effective and the evidence base very limited. First discovered as a side-effect when used for the treatment of Parkinson’s disease, the main limitation is nausea when first used (though this subsides with time). In a study of 854 patients (16% with diabetes) the 4 mg dose resulted in a rapid onset erection in 50% of patients.

55

56

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Mean score Almost always or always

5

Most times

4

About haf the time

3

Baseline 24 weeks

Much less than half the time 2 Almost never or never

1 Placebo

25 mg

50 mg

100mg

Fig. 6.2 Effectiveness of sildenafil over 24-week period in 216 men with erectile dysfunction. Br J Urology 1996; 78: 257–261.

• Intracavernosal injection therapy has been superseded by oral agents but nevertheless remains a useful second-line therapy if PDE5s are unsuccessful or contraindicated. Unlicensed drugs include papaverine (a non-specific PDE inhibitor) which can be used alone or with phentolamine (an alphablocker), and VIP. However, the only licensed product is Prostaglandin E1 (alprostadil) which has an overall response rate of 70–80%. Erection occurs within 5–20min, it is rapidly metabolized and the incidence of priapism low, but the invasive nature of the treatment is unacceptable to many men and there is always a high drop-out rate. Pain may occur at the injection site and repeated injections into the same site has been reported to cause areas of fibrosis. Use should be limited to once a week and careful instructions given to the patient to immediately report to the hospital if the erection lasts for more than six hours. Repeated hip movements or going for a brisk walk can cause detumesence probably by diverting blood to the legs but if priapism persists penile aspiration is required. • Transurethral drug application using a narrow pellet of synthetic PGE1 reportedly has a success rate of 60–70%. Erection is normally achieved within 5–10 min and lasts for up to an hour but local pain is a common problem leading to discontinuation of this form of therapy. There is a low risk of priapism. • Vacuum constriction devices were first invented in the late 19th century to enhance penile function. However, it was not until the 1980s that they

CHAPTER 6 • ERECTILE DYSFUNCTION

began to be used successfully in the treatment of ED. There are three components to the device: • A cylinder with one open end into which the penis is inserted; • A vacuum pump (hand or battery operated); and • A constriction ring. Creating a vacuum in the tube causes blood to be drawn into the penis producing an erection – the ring is slipped off the cylinder onto the base of the penis to maintain rigidity. This is sufficient for penetration in the majority of patients. The ring should not be left on for more than 30min, and correct training is essential if successful and safe use of this device is to be achieved. Couples complain about the lack of spontaneity with vacuum devices. • Surgery has a very limited place in the treatment of ED but several penile prosthetic implants are available. The high cost and risk of infection limit their use. • Penile constriction rings (as used with vacuum devices) can sometimes be helpful in those men with partial ED by increasing penile engorgement. • Yohimbe has been the subject of a number of double-blind trials with conflicting results but is not licensed for use in the UK. It has only modest effects at best. Other oral agents under development include derivatives of yohimbine and delquamine which are centrally acting adrenoceptor antagonists. • Transcutaneous nitroglycerine has not been confirmed to be of use in ED. • Phentolamine mesylate, an oral alpha-adrenoceptor blocker, has a faster onset of action than sildenafil. At doses of 40mg and 80mg about a half of those treated managed an erection sufficient for penetrative sex on 75% of attempts.

CURRENT ISSUES •

Combination therapies may have to be used if only a partial response is seen with one treatment alone, though good quality randomized trials of combination therapy have not been undertaken.

FURTHER READING: Burnet AL. Nitric oxide in the penis: physiology and pathology. J Urol 1997; 157: 320–324. Maggi M et al. Erectile dysfunction: from biochemical pharmacology to advances in medical therapy. European J of Endocrinology 2000; 143 (2): 143–154. Porst H. Current perspectives on intracavernosal pharmacotherapy for erectile dysfunction. Int J Impotence Research 2000; 12 Suppl. 4: S91–S100. Rajfer J. Opportunities and challenges in oral therapy. Int J Impotence Research 2000; 12 Suppl. 4: S59–S61.

57

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Adrian R. Scott MD, FRCP CHAPTER 7 EVIDENCE-BASED INTERVENTIONS TO PREVENT OR RETARD VASCULAR COMPLICATIONS

INTRODUCTION It was not until very late in the 20th century that physicians began to see diabetes (particularly type 2 diabetes) as a vascular disorder and for reasons that seem unclear now, these patients were excluded (along with women) from many of the ground-breaking trials to prevent vascular complications by the treatment of hypertension, hyperlipidaemia etc. With the exception of treatment of hyperglycaemia, there have been very few primary or secondary prevention studies that only include patients with diabetes. Much of our information on interventions that prevent or retard the vascular complications of diabetes are derived from either sub-group analyses of larger trials, or extrapolation of data from the non-diabetic population to those with diabetes. In general, however, where an intervention has been shown to reduce risk in the general population, the benefit to those with diabetes has been even greater. A word of caution: throughout this section I have emphasized that relative risk reduction has to be seen in the context of the base-line risk. Furthermore, the benefits of intensive treatment, whether it be glycaemic, lipids or hypertension control, must be viewed against both the cost and potential for harm of this approach, versus the outcomes of a less intensive approach. This is best summed up by an analysis of the UK Prospective Diabetes Study (UKPDS) of conventional vs. intensive control of blood glucose and hypertension (Table 7.1). When viewed in terms of the events prevented by more intensive therapy of hyperglycaemia, one view is that the results are quite disappointing. Others would argue that these benefits were achieved by actually only a small mean difference in HbA1C (1%) and that interventions that achieve even tighter control are likely to show even greater benefits. Whilst clinicians have to try and make sense of these dilemmas, the challenge is to help people with diabetes make informed choices about their therapy – some will view a weight gain of 5 kg (the mean likely to occur with intensive insulin therapy) unacceptable if the health gain is as follows: out of 100 newly diagnosed patients with intensified therapy (mean HbA1C 7%), 41 will develop a diabetes-related end-point (DREP) over 10 years. But if glycaemic control is less tight (mean HbA1C 7.9%) 46 will develop a DREP. Unfortunately the person opting for insulin therapy (and 5 kg weight gain) cannot assume that they will be one of the 5 out of the 100 who will benefit.

59

60

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Various ways of presenting identical data on the effects of intensified vs. conventional therapy on ‘any diabetes-related end-point’ Intensified therapy Conventional therapy 100 patients over 10 years 100 patients over 10 years median HbA1C 7.9% median HbA1C 7%

Patients with at least one end-point No. of patients (%) 41 (41%)

46 (46%)

Patients without any end-point No. of patients (%) 59 (59%)

54 (54%) Differences

Benefit of intensified vs. conventional therapy Decrease in patients with end-point No. of patients Absolute risk reduction Relative risk

5 5% 11%

Increase in patients without end-point No. of patients Absolute increase Relative increase

5 5% 9%

Lack of benefit of intensified vs. conventional therapy Patients with end-point despite intensified therapy No. of patients Absolute per cent Relative per cent (41 of 46)

41 41% 89%

All patients who do not benefit No. of patients (%)

95 (95%)

Table 7.1 Various ways of presenting identical data from the UKPDS on the effects of intensified vs. conventional therapy on ‘any diabetes-related end-point’. I Mulhauser and M Berger. Diabetic Medicine 2000; 17: 823–829.

Despite the wealth of information available to clinicians, we are unable to quantify the benefit of multiple proven interventions against no treatment, but we can be fairly confident that both quality and quantity of life are improved in the short to medium term. What uncertainty must not do is lead to inaction or sloppy clinical practice. For example, debate over

CHAPTER 7 • EVIDENCE-BASED INTERVENTIONS

which is the most appropriate blood pressure target should not prevent a pragmatic approach that aims to achieve a sustainable reduction in blood pressure without causing unacceptable side-effects. Using the latest lipidlowering agent is for nought, if an annual assessment of the foot to identify at-risk features is omitted, since this simple intervention has been demonstrated to reduce the risk of lower limb amputation.

EVIDENCE BASED INTERVENTIONS Macrovascular disease – Acute coronary syndromes (ACS), primary and secondary prevention of coronary heart disease (CHD) Intensification of multiple therapeutic strategies has been shown to reduce hospital mortality in diabetic patients, though other studies have shown that high risk patients, such as those with diabetes, are less likely to receive these evidence based interventions. In a recent study from Germany, hospital mortality fell from 29% to 17% in people with diabetes following STEMI, after a change in policy resulted in greater use of glucose/insulin infusion, GP IIb/IIIa inhibitors, and earlier angiography. In another study from Denmark, intensive risk factor management in type 2 diabetes reduced the risk of cardiovascular and microvascular events by about 50% compared to a group treated less aggressively.

Thrombolysis / anticoagulation Several large trials have proved the efficacy of thrombolysis for acute myocardial infarction (AMI) with a variety of agents such as streptokinase, urokinase, tissue plasminogen activator (TPA) or similar. No trials have been conducted in diabetics alone. The earlier the treatment is given the better the outcome and the absolute benefit at the 5th week is 30 lives saved per 1,000 patients treated before the 6th hour and 20 lives saved per 1,000 patients treated between the 6th and 12th hour. The mortality benefit persists at one year and beyond. The maximum reduction in mortality was achieved in those patients treated during the first hour after the onset of symptoms. The benefit of thrombolysis is significant regardless of age, gender, previous MI or diabetes. In fact, although the relative risk reduction of mortality is 20% in most sub-groups, this means that the number of lives saved is greatest in those at higher risk. So for example, the number of lives saved per 1,000 patients treated is 49 in the presence of bundle branch block, and 37 in the case of AMI compared to only 8 in the case of inferior myocardial infarction. In the presence of diabetes 37 lives are saved compared to only 15 in the absence of diabetes. In the WARIS study warfarin given post-AMI reduced mortality by 24%, recurrent AMI by 34% and CVAs by 55% at the cost of an annual severe bleeding rate of 0.6%. Similar results were found in the ASPECT study but there was only a 10% (NS) reduction in mortality. There have been no subanalyses in people with diabetes.

61

62

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Aspirin, thienopyridines, and IIb/IIIa glycoprotein receptor inhibitors Aspirin alone given as a dose of at least 150mg soon after the onset of symptoms of ACS and continued for at least a month reduced cardiovascular mortality by 20% compared to placebo. This represents a benefit of 25 lives saved per 1,000 treated patients. Reduction in re-infarction and non-fatal CVAs are also reduced by almost 50%. Continuing the aspirin beyond the first month doubles the initial benefit by preventing an additional 40 deaths, MIs or CVAs per 1,000 treated patients during the first four years. The beneficial effects of aspirin and thrombolysis are additive and the streptokinase-aspirin combination has been shown to be the most effective regime with a 38% reduction of mortality. The benefits of aspirin in diabetics appear to be similar to those without diabetes though the optimum dose of aspirin remains to be determined. In the general population, aspirin for primary prevention has been shown to reduce the number of non-fatal MIs and TIAs but showed no effect on cardiovascular mortality. In some studies there was a suggestion of an increase in the number of haemorraghic strokes. However, this was not confirmed in the HOT study which suggested that, provided hypertension is well controlled, the addition of aspirin is beneficial. In practice, the treatment of 1,000 high-risk subjects by aspirin for one year would prevent about three coronary ischaemic events. In people with diabetes, aspirin is indicated if the 10-year CHD risk exceeds 15%. Thienopyridines (ticlopidine and its derivative clopidogrel) do not inhibit cyclo-oxygenase like aspirin, but inhibit ADP-dependent binding of fibrinogen to IIb/IIIa glycoprotein receptors. Unfortunately, ticlopidine can cause neutropenia and thrombocytopaenia so has not been fully evaluated. Clopidogrel, on the other hand, both alone and in combination with aspirin, has been shown to be superior to aspirin alone at reducing cardiovascular and cerebrovascular events. However, the benefits are small (an 8% difference compared to aspirin) and at today’s prices it is not cost-effective. In the CURE study (Clopidogrel in Unstable angina to prevent Recurrent ischaemic Events) over 12,000 patients (22% with diabetes) who presented within 24 hours of onset of symptoms of ACS, were randomized to receive aspirin 75–325mg/day and either clopidogrel or placebo. There was a 20% reduction in events (9.28% vs. 11.47%) in the clopidogrel group and an 8% reduction in mortality. Thus, if clopidogrel has a place, it is in secondary prevention in those most at risk; this may include people with diabetes. IIb/IIIa glycoprotein receptor inhibitors are powerful antiplatelet agents which if given to high risk patients with unstable angina or ACS, in conjunction with aspirin, reduce mortality and risk of MI. Studies in patients with diabetes suggest similar benefit. These agents are also useful following coronary angioplasty at reducing the risk of death and infarction.

CHAPTER 7 • EVIDENCE-BASED INTERVENTIONS

Beta blockers Beta-blockers inhibit sympathetic nervous system activity, reducing heart rate and decreasing myocardial oxygen consumption, so reducing ischaemia. They also have an anti-arrhythmic effect and early intravenous administration has been shown to limit the size of myocardial damage. Many of the studies that were performed on beta blockade at the time of the infarct, and post infarction, were done in the pre-thrombolysis era. In combination with thrombolysis, intravenous beta-blockers, given early after the onset of symptoms, have been shown to significantly decrease recurrent angina and non-fatal recurrent MI during the first six days. Controversy remains over the mode of administration since the GUSTO-1 study showed that although overall mortality on the 30th day was significantly lower in patients treated with beta-blockers, i.v. administration worsened the prognosis compared to oral beta-blockers introduced after haemodynamic stabilization. Intravenous beta-blockers were associated with an increase in incidence of heart failure, shock and the need for ventricular pacing. A meta-analysis of data from 29 trials involving 29,000 patients on betablockers during MI showed a reduction in mortality of 13%. Pooled data from six trials conducted in patients with diabetes showed a 35% reduction in mortality. The benefits of beta-blockers given post MI (usually after a few days) have been analysed in another meta-analysis of some 24,000 patients and a 23% reduction in mortality was demonstrated. Diabetics had a greater benefit, however, of up to 48%. Not all beta-blockers are the same, however, and only timolol, propranolol and metoprolol have been proven to be of value in these situations. Despite this, atenolol is the most popular beta-blocker in the UK. For primary prevention of CHD events, beta-blockers appear to have no advantages over angiotensin-converting enzyme inhibitors (ACE-Is) according to the UKPDS which compared atenolol with captopril in the treatment of hypertension, though patients on atenolol gained more weight (3.4 vs. 1.6kg).

Angiotensin-converting enzyme inhibitors (ACE-Is) Studies have been conducted using ACE-Is acutely, at the time of MI or post MI. A meta-analysis of 15 acute trials with more than 100,000 patients showed a reduction in mortality of some 6%. This included the Consensus-2 trial using intravenous enalapril, in which there was an increased mortality of 11% in those treated with an ACE-I and a negative interaction with aspirin use was observed. In GISSI-3 patients who were hypotensive or in severe congestive cardiac failure were excluded and acute use of lisinopril was associated with a 6-week reduction in mortality of 44% in patients with type 1 diabetes and 27% in those with type 2 diabetes.

63

64

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

ACE-Is given post MI in patients with reduced left ventricular (LV) function show a reduction in mortality of 22% in patients with an LV ejection fraction <40%. Similar reductions in mortality have been observed in patients with diabetes. The benefits of ACE-Is observed in patients with LV dysfunction have subsequently been confirmed in high risk patients with normal LV function. In the HOPE study, 9,297 patients over the age of 55 years (mean age 66) with established vascular disease were randomized to receive ramipril 10mg or placebo. 38% (n=3,578) were diabetic (Micro-HOPE) and were followed for a mean of 4.5 years. Treatment with ramipril was associated with a 24% reduction in all-cause mortality, a 22% reduction in the rate of MI and a 33% reduction in the rate of stroke (Fig. 7.1).

Treatment of hyperglycaemia The DIGAMI study involved patients with an AMI and an admission blood glucose >11mmol/l irrespective of their diabetes status. Patients were given a glucose/insulin infusion to maintain blood glucose control between 7 and 9mmol/l followed by three months of four times daily subcutaneous insulin, or assigned to the control group who received their usual treatment of diabetes (which might have included insulin if it were clinically indicated). A significant reduction in mortality of 29% was seen in the insulin/glucose infusion group compared to controls. The greatest benefit was seen in those at relatively low risk i.e. younger patients not previously treated by insulin. They experienced a relative risk reduction of 52% at one year. Some authors have suggested that this increased benefit was related to discontinuation of oral anti-diabetic agents such as sulphonylureas but this seems unlikely as the UKPDS study showed no difference in mortality between those treated with sulphonylureas and insulin. The picture has not become clearer since a follow-up study (DIGAMI 2) that included a total of 1,253 patients with suspected AMI from 48 hospitals in Europe. At the time of writing, this study had not been published but the results were presented at the 2004 meeting of the European Association for the Study of Diabetes. Patients were randomized to one of three groups: (1) acute insulin-glucose infusion followed by insulin-based long-term glucose control; (2) insulinglucose infusion followed by standard glucose control (i.e. no insulin); or (3) routine metabolic management according to local practice. Mortality was surprisingly low in all three groups (18.4%) but there were no significant differences between them. 41% of the control group received insulin therapy, which may be why no mortality benefit of glucose-insulin infusion was demonstrated. Further analysis of UKPDS demonstrates that for each 1% decrement in HbA1C, there was a 21% decrease in all diabetes related end-points. The benefit was greater for microvascular complications but a significant reduction in macrovascular events was also seen.

CHAPTER 7 • EVIDENCE-BASED INTERVENTIONS

0.16

Myocardial infarction

Kaplan-M

0.12

0..08 0.04

0 0 0.08

500

1000

1500

2000

500

1000

1500

2000

1500

2000

Stroke

Kaplan-M

0.06

0..04 0.02

0 0 0.12

Cardiovascular death

Kaplan-M

0.09

0..06 0.03

0 0

500

1000

Duration of follow-up (days)

Fig. 7.1 Effects of ramipril on cardiovascular outcomes in people with diabetes – MICRO-HOPE. Lancet 2000; 355: 253–259.

65

66

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Revascularization Urgent revascularization is usually undertaken in refractory unstable angina that fails to settle despite optimum medical therapy. In 90–95% medical therapy stabilizes the situation. However, early intervention, with angiography and angioplasty where appropriate, has now been shown to be superior to medical treatment alone. One study compared thrombolysis within 30 minutes of admission with primary angioplasty within 90 minutes. After six months of follow-up, patients in the angioplasty group fared better on all measures: the mortality rate was 6.2%, compared with 7.1% for TPA; the rate of recurrent ACS was 5.3%, compared with 10.6% for TPA; and the stroke rate was 2.2%, compared with 4% for TPA. Length of time in hospital also was shorter for the angioplasty group: 4.5 days compared with 6 days for those who received TPA. These differences, though small, represent significant improvements in outcome. When available and performed by experienced operators at high-volume centers, primary percutaneous coronary intervention (PPCI) saves 20 lives and results in 60 fewer events for every 1,000 patients treated. ‘Time is muscle’ and most hospitals do not have these facilities, so unless transfer and urgent PPCI can be achieved within 90 minutes, thrombolysis remains the treatment of choice. Surgery and angioplasty have similar results – vein grafting only improves prognosis in patients with triple vessel disease or LV dysfunction. There is no evidence that diabetics need an alternative strategy except that the Bypass Angioplasty Revascularization Investigation (BARI) study suggested that people with diabetes did less well with angioplasty – this remains to be confirmed. The BARI 2 Diabetes (BARI 2D) trial has not yet reported but will compare whether attenuation of insulin resistance can arrest or retard progression of coronary artery disease compared with treatment targeted to the same level of glycaemic control with an insulin-providing approach. It is designed also to determine whether early revascularization reduces mortality and morbidity in patients with type 2 diabetes whose cardiac symptoms are mild and stable. The role of other antiplatelet agents such as clopidogrel, or angioplasty followed by drug-eluting stents remains uncertain. Comparisons of restenosis after angioplasty have shown slightly higher rates in people with diabetes compared to non-diabetics. This difference is probably explained by the older ages of those with diabetes in these studies.

Statins Statins have transformed the management of hyperlipidaemia reducing both total mortality, coronary events and need for revascularization. What is clear is that this is not due to cholesterol lowering alone. They have significant antiinflammatory effects reducing C-reactive protein (CRP) and probably stabilize atheromatous plaques.

CHAPTER 7 • EVIDENCE-BASED INTERVENTIONS

Several clinical trials have now looked at the effect of initiating a statin within hours and days of a coronary event and the findings suggest there may be significant benefits. The RECIFE (Reduction of Cholesterol in Ischaemia and Function of the Endothelium) trial looked at the effect of rapidly lowering cholesterol on endothelial function days after a coronary event. Patients admitted with MI or unstable angina were randomized to placebo or pravastatin 40mg within 10 days. Endothelial function was assessed by measuring flow-mediated dilatation of the brachial artery, which increased by 42% with pravastatin compared with placebo. More excitingly, other short-term studies with pravastatin 40mg and atorvastatin 80mg initiated within four days have shown significant reductions in coronary events in as little as 16 weeks. Larger trials are on-going which will help to determine if these preliminary findings are confirmed and if this is a class effect. Not all statins have been demonstrated to lower CRP, and smooth muscle cell proliferation (which may help stabilize plaques) is inhibited by some statins in vivo but not by pravastatin. Potential but unproven interventions such as use of antioxidants, treatment of homocysteinaemia (with folic acid) or hypertriglyceridaemia remain untested. In the HOPE study vitamin E conferred no benefit.

Peripheral vascular disease The Wisconsin Epidemiological Study of Diabetic Retinopathy (WESDR) included nearly 3,000 people with diabetes (1,200 type 1 patients diagnosed under 30 years) and showed an association between rising HbA1C and risk of lower limb amputation. However, neither DCCT nor UKPDS were powered to look at PVD or an amputation as an end-point – DCCT patients were young and few had macrovascular events; UKPDS patients with significant peripheral vascular disease at diagnosis were excluded from the study. There was a non-significant (16%) reduction in fatal and non-fatal MIs in the intensively treated group after 15 years and there was no difference between the intensive and conventional treatment groups in the proportion of patients who had evidence of peripheral vascular disease by Doppler blood pressure or absent peripheral pulses. No controlled trials have been undertaken with diabetic patients to assess the effects of risk factor modification on the regression of vascular disease, however, smoking cessation is associated with improvement of symptoms in non-diabetics. Among patients with symptomatic PVD, continued smoking is associated with worsening claudication, limb-threatening ischaemia and amputation, and the need for revascularization. Patency rates are lower following revascularization in patients who continue to smoke and survival is reduced. The effects of management of hyperlipidaemia and hypertension on the progression of PVD has not been fully evaluated but an analysis of the Scandinavian Simvastin Survival Study (4S) study showed that the incidence of ‘new or worsening claudication’ was significantly lower in

67

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

4.0 Percentage (%)

68

Placebo Simvastatin RR = 0.62 P<0.008

3.0

2.0

1.0

0 0

1

2

3

4

5

Years

Fig. 7.2 Effect of simvastatin on incidence of ‘new or worsening’ intermittent claudication in the Scandinavian Simvastatin Survival (4S) study. Am J Cardiol 1998; 81: 333–335.

CHD patients treated with simvastatin (Fig. 7.2). In the Physician’s Health study, low doses of aspirin resulted in a 54% reduction in the risk of peripheral arterial surgery, compared with placebo, but no study has shown any improvement in claudication symptoms with antiplatelet agents. Drug treatment for claudication has generally been disappointing but a new agent, cilostazol, (a phosphodiesterase III inhibitor with antiplatelet, antithrombotic and vasodilatory effects), looks promising. Phase III studies involving just over 2,000 patients (approximately 25% with diabetes) using doses of between 50 and 100mg twice daily, showed improvement in maximal walking distance in most but not all studies. Pain-free walking distance and quality of life were improved and cilostazol was significantly better than both placebo and oxpentifylline 400mg tid (Fig. 7.3). Peripheral vascular disease is a contributory factor in most diabetic foot ulcers, and evidence exists that a combination of a dedicated multidisciplinary foot service and the utilization of specialist footwear can reduce both ulceration and amputation rates.

Cerebrovascular disease In UKPDS each 1% reduction in HbA1C was associated with a 37% decrease in risk for microvascular complications and a 21% decrease in the risk of any diabetes related end-point or death. The association with glycaemia was less steep for stroke and heart failure (Fig. 7.4), for which hypertension is a much more important contributory factor but nevertheless improved glycaemia reduced

Maximal walking distance mean % change from baseline

CHAPTER 7 • EVIDENCE-BASED INTERVENTIONS

70 60 50

(Mean ± 95% confidence interval) Cilostazol 100 mg bid Oxpentifylline 400 mg tid Placebo

40 30 20 10 0 Baseline

4

8

12

16

20

24

Weeks of treatment

Fig. 7.3 Maximal walking distance in patients with claudication treated with cilostazol or oxpentifylline or placebo. Circulation 1998; 98 (Suppl 1): 1012, Abstract 58.

the incidence of stroke. Unfortunately, acute stroke management in the person with diabetes is based on extrapolation of the data from non-diabetics as there are, as yet, no prospective studies of stroke management in diabetics. Trials have shown that thrombolytic therapy started within six hours of the onset of symptoms of ischaemic stroke reduces the proportion of patients who die or remain dependent on others, up to six months later, (61.5% vs. 68% of control patients not given thrombolysis). Results were more impressive if treatment was started within three hours (56.6% vs. 70.7%). Alteplase seemed superior to streptokinase but overall there was an increased risk of symptomatic intracranial haemorrhage (9.6% vs 2.6%). Overall, the risk of dying within two weeks was increased in those receiving thrombolytic therapy (20.9% vs. 11.9%) despite the improvement in the composite end-point of death or dependency. Whether people with diabetes benefit similarly from thrombolysis is not known. Use of anticoagulant therapy with unfractionated or low-molecular weight heparin for acute ischaemic stroke is associated with an increase in haemorrhagic stroke but with no positive benefit in terms of mortality or dependency. Hyperglycaemia on admission is associated with a worse outcome and although the benefits of treatment of hyperglycaemia in this situation are unknown, a DIGAMI-style glucose and insulin infusion would seem sensible (trials are ongoing).

69

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Hazard ratio

10

Fatal and non-fatal myocardial infarction

1

P<0.035

14% decrease per 1% reduction in HbA1c

12% decrease per 1% reduction in HbA1c

Microvascular end points

Hazard ratio

10

Cataract extraction

P<0.0001

1

P<0.0001

37% decrease per 1% reduction in HbA1c

0.5

10

Fatal and non-fatal stroke

P<0.0001

0.5

Hazard ratio

70

19% decrease per 1% reduction in HbA1c

Amputation or death from peripheral vascular disease

Heart failure

P<0.0001

1

P<0.021

43% decrease per 1% reduction in HbA1c

0.5 5

6

7

8

9

16% decrease per 1% reduction in HbA1c

10

5

6

7

8

9

10

Updated mean HbA1c concentration

Fig. 7.4 Hazard ratios (with 95% confidence intervals) showing association between mean HbA1C and various micro- and macrovascular complications in the UKPDS trial. BMJ 2000; 321: 405–441, with permission.

Secondary prevention data is based entirely on general population studies with none having been conducted in people with diabetes alone. Targets similar for those of diabetic patients with CHD are appropriate. The dose of aspirin required is uncertain. The European Stroke Prevention Study showed that, in the general population, aspirin and sustained-release dipyridamole are an equally effective secondary prevention in reducing the risk of stroke

CHAPTER 7 • EVIDENCE-BASED INTERVENTIONS

and/or death. However, doubts exist because of the relatively low dose of aspirin used and the exclusion of one centre from the study because of scientific fraud. In the Clopidogrel versus Aspirin in Patients at Risk of Ischaemic Events (CAPRIE) study, 7.2% of those treated with clopidogrel 75mg/day had an event compared with 7.7% of patients receiving aspirin 325mg/day. So, if the patient has a cerebrovascular incident whilst on aspirin, addition of dipyridamole may be justified; where patients have a true aspirin allergy clopidogrel is probably the most appropriate choice. There are a number of studies where stroke prevention has been a significant secondary end-point. The UKPDS hypertension study, in which the target for tight blood pressure control was <150/85, showed a 44% reduction in strokes compared to the less tight blood pressure control group (target blood pressure <180/105) (Fig. 7.5). Ramipril 10mg daily in the diabetic sub-group (Micro-HOPE) of the HOPE study reduced the number of strokes from 6.1% to 4.2% (a 33% relative risk reduction) in a cohort of patients most of whom had established CHD. The difference in blood pressure between the ramipril and placebo groups at the end of the study was only 2.5/1.0. Some commentators say this is insufficient to account for the difference in stroke rates, but as these were based on casual readings rather than 24-hour profiles it remains uncertain whether the benefit is due to blood pressure lowering or a different mode of vascular protection by ACE-Is. In both CARE and LIPID (secondary prevention studies in patients with previous MI or angina) pravastatin 40mg reduced the risk of fatal and nonfatal cerebrovascular accidents by 31% and 20% respectively. The numbers with diabetes, however, were too small to analyse as a separate group.

Nephropathy The DCCT showed a 39% reduction in the occurrence of microalbuminuria and a 54% reduction in albuminuria in the intensive therapy arm for both adults and adolescents with type 1 diabetes. Similarly, the UKPDS showed a slowing of renal decline in the tight glycaemic control group with type 2 diabetes (Fig. 7.6). Type 2 patients with microalbuminuria or proteinuria are less likely to progress to ESRF but, as with type 1 diabetes, blood pressure management is the mainstay of treatment to reduce the decline in renal function. ACE-Is are indicated in type 1 patients with persistent microalbuminuria or proteinuria, irrespective of initial blood pressure (Fig. 7.7) and are first line agents in type 2 diabetes if microalbuminuria or proteinuria is present, though lowering blood pressure is the priority. Renoprotection by ACE-Is is probably a class effect. Angiotensin receptor antagonists (ARAs) achieve similar reductions in blood pressure and proteinuria as ACE-Is. There have been only three randomized double-blind studies lasting more than one year using ARAs as

71

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Hazard ratio

10

1 0.5

Fatal and non-fatal myocardial infarction

Hazard ratio

1

Fatal and non-fatal stroke P<0.0001

P<0.0001

12% decrease per 10mmHg reduction in systolic blood pressure

Microvascular end points 10

P<0.0001

19% decrease per 10mmHg reduction in systolic blood pressure

Cataract extraction P<0.041

13% decrease per 10mmHg reduction in systolic blood pressure

0.5

10 Hazard ratio

72

Amputation or death from peripheral vascular disease P<0.0001

Heart failure P<0.0028 12% decrease per 10mmHg reduction in systolic blood pressure

1

16% decrease per 10mmHg reduction in systolic blood pressure

0.5 110 120 130 140 150 160 170

110 120 130 140 150 160 170

Updated mean systolic blood pressure (mmHg)

Fig. 7.5 Hazard ratios (with 95% confidence intervals) showing association between mean systolic blood pressure and various micro- and macrovascular complications in the UKPDS study. BMJ 2000; 321: 413–417, with permission.

anti-hypertensive treatment for people with diabetes. None showed a significant reduction in total or cardiovascular mortality. In the RENAAL study there was a statistical reduction in progression to end-stage renal failure in those that were treated with ARAs but the placebo group had higher blood pressure throughout the study which may account for the worse outcomes.

CHAPTER 7 • EVIDENCE-BASED INTERVENTIONS

P = 0.0052

Conventional therapy Intensive therapy

Patients (%)

3 P = 0.034

2

1

P = 0.54

P = 0.14

3

6

0

P = 0.099

74% risk reduction

4

9

12

15

Years

Fig. 7.6 Effect of glycaemic control on creatinine increase in UKPDS. Lancet 1998; 352: 837–853.

0.5

Proportion with event

0.4 0.3 0.2 0.1 0 Baseline n at risk: Placebo 114 Captopril 111

3

6

9

12 15 Month

18

21

24

113 109

108 104

98 96

91 90

78 85

69 77

40 41

86 89

Fig. 7.7 Effect of captopril on progression of microalbuminuria in normotensive patients with type 1 diabetes. Diabetologia 1996; 39: 587–593.

To achieve a target blood pressure of <130/80 for patients with nephropathy may require several anti-hypertensive agents and for young people blood pressure targets may be set even lower to achieve a blood pressure < 90th centile for age.

73

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

Retinopathy The DCCT demonstrated beyond doubt that good glycaemic control retarded the onset of microvascular complications and delayed progression of those complications already present. Almost a half of patients in the intensive treatment arm achieved an HbA1C of 6.1% or less at least once during the study, giving a mean of approximately 7% for the duration of the study. From five years into the study there was a 50% reduction in the cumulative incidence of retinopathy in the primary prevention cohort and a 54% reduction in progression in the secondary prevention arm compared to the conventional treatment group (mean HbA1C 8.9%) (Figs. 7.8 & 7.9). Intensive therapy reduced the risk of pre-proliferative and proliferative retinopathy by 47% and the need for photocoagulation by 56%. In the UKPDS, hypertensive patients were assigned to either tight or moderate blood pressure control over nine years. The group randomized to tight blood pressure control (mean 144/82 mmHg) experienced a 47% reduction in the risk of losing three lines of vision compared to the group with standard

60 50 40 Patients (%)

74

Conventional 30 P<0.001

20 10

Intensive 0 0

1

Conventional Intensive

2

3

375 342

4 5 Year of study 220 202

6

7

79 78

8

9

52 49

Fig. 7.8 Cumulative incidence of sustained change in retinopathy in patients with IDDM receiving intensive or conventional therapy in the primary prevention arm of the DCCT. Year of study refers to sample sizes at different years of the study. NEJM 1993; 329: 977–986.

CHAPTER 7 • EVIDENCE-BASED INTERVENTIONS

blood pressure control (mean BP 154/87 mmHg). There was also a 34% reduction in the risk of progression of retinopathy status. The EURODIAB Controlled Trial of Lisinopril in Insulin-Dependent Diabetes Mellitus (EUCLID), also comparing tight with moderate blood pressure control, produced similar results, i.e. a significantly reduced risk of blindness and reduced rate of progression of retinopathy in those patients randomized to more intensively controlled blood pressure. Use of an ACE-I may reduce the progression of retinopathy, independently of blood pressure reduction, especially in normotensive type 1 diabetics.

Neuropathy In the primary prevention cohort of the DCCT, intensive therapy delayed the appearance of neuropathy at 5 years by 69%. Ten per cent of the conventional treatment group developed neuropathy compared to 3% in the conventional group. Progression (as judged by clinical findings and nerve conduction studies) was reduced by 59%.

60 50 Conventional Patients (%)

40 30

P<0.001

20

Intensive

10 0 0

1

Conventional Intensive

2

3

348 354

4 5 Year of study 324 335

6

7

128 136

8

9

79 93

Fig. 7.9 Cumulative incidence of sustained change in retinopathy in patients with IDDM receiving intensive or conventional therapy in the secondary prevention arm of the DCCT. Year of study refers to sample sizes at different years of the study. NEJM 1993; 329: 977–986.

75

76

SECTION I • MICRO- AND MACROVASCULAR COMPLICATIONS OF DIABETES

In UKPDS, however, no differences were observed between treatment groups. Aldose reductase inhibitors have not been found to give sustained or clinically relevant improvements in nerve function.

CONCLUSION The last two decades have seen the publication of a large number of well designed, randomized controlled trials which have identified interventions that can prevent or retard the progression of the micro- and macrovascular complications of diabetes. The combination of intensified glycaemic control and aggressive treatment of hypertension and hyperlipidaemia offers the person with diabetes the chance to extend life-expectancy and improve quality of life. Health organizations, pharmaceutical companies and politicians must work closely together to devise strategies aimed at preventing type 2 diabetes and setting up health surveillance systems that identify patients early in their disease, prior to the onset of complications, so that these interventions can have the maximum benefit. Nevertheless, these treatments require great commitment from both patient and health professionals and are, for most of the world’s diabetic population at present, unaffordable.

FURTHER READING Klein R. Hyperglycaemia and microvascular and macrovascular disease in diabetes. Diabetes Care 1995; 18(2) 258–268. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin dependent diabetes mellitus. NEJM 1993; 329: 977–986. UK Prospective Diabetes Study (UKPDS) Group Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). The Lancet 1998; 352: 837–853.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

SECTION II DIABETIC NEUROPATHIES

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Rayaz A. Malik MB.ChB, PhD, MRCP CHAPTER 8 CLASSIFICATION AND CLINICAL FEATURES OF NEUROPATHY

INTRODUCTION Diabetic neuropathy may simply be defined as the presence of symptoms and/or signs of peripheral nerve dysfunction in people with diabetes after the exclusion of other causes. It can only be diagnosed with a careful clinical examination, and absence of symptoms cannot be equated with absence of neuropathy. Diabetic neuropathy should not be diagnosed on one symptom, sign or test alone: a minimum of two abnormalities (from symptoms, signs, nerve conduction abnormalities, quantitative sensory tests or quantitative autonomic tests) is recommended.

CLASSIFICATION The preferred clinical classification is outlined in Table 8.1.

Generalized symmetrical polyneuropathies Chronic sensorimotor neuropathy is the commonest manifestation of the diabetic neuropathies. Positive painful symptoms tend to be intermittent and more pronounced at night. Patients may also experience ‘negative’ symptoms such as numbness and unsteadiness due to disturbed proprioception and abnormal muscle sensory function resulting in repetitive minor trauma, falls or Charcot’s neuroarthropathy. Neurological examination demonstrates a symmetrical sensory loss to all modalities in a stocking distribution which in severe cases may extend to the knees and hands. This is accompanied by reduced or absent ankle

Classification of diabetic neuropathy Generalized symmetrical polyneuropathies Sensorimotor (chronic) Acute sensory Hyperglycemic neuropathy Autonomic neuropathy Focal and multifocal neuropathies Cranial Thoracolumbar radiculoneuropathy Focal limb Proximal motor (amyotrophy) Superimposed chronic inflammatory demyelinating neuropathy (CIDP)

Table 8.1 Classification of diabetic neuropathy.

79

80

SECTION II • DIABETIC NEUROPATHIES

and knee reflexes, respectively. Autonomic dysfunction may manifest as dry skin. Motor weakness is unusual, although small muscle wasting can be seen in the feet and hands in severe neuropathy. The ‘at-risk’ foot for neuropathic ulceration might also have a high arch (pes cavus) and clawing of the toes. Acute Sensory neuropathy is chararacterized by severe sensory symptoms (hyperaesthesiae, dysaesthesia), allodynia on sensory testing, preserved strength and occasionally reduced ankle reflexes. Pain is the outstanding complaint in all patients which may be accompanied by severe weight loss and depression. Relatively rapid alterations in glycaemic control as a result of ketoacidosis, weight loss and eating disorders or a sudden improvement in blood sugars by either insulin or oral hypoglycemic agents (insulin neuritis) may precipitate an acute or sub-acute onset of debilitating symptoms which resolve in less than a year. The pathogenesis is unclear although epineurial arterio-venous shunts and proliferating “new vessels” have been observed in such patients suggesting that endoneurial ischemia precipitated by the sudden improvement of glycaemic control may cause this condition. Hyperglycaemic neuropathy: Rapidly reversible distal uncomfortable sensory symptoms may be accompanied by abnormalities of nerve conduction in patients with recently diagnosed or transiently poorly controlled diabetes with recovery following restoration of euglycaemia.

Autonomic neuropathy Diabetic autonomic neuropathy is common though rarely severely symptomatic and results in significant morbidity and mortality. The cardiovascular, gastrointestinal, urogenital, sudomotor and pupillomotor systems may be involved alone or in combination. The most common dysautonomic conditions together with their symptoms are listed in Table 8.2.

Staging The agreed clinical stages of diabetic neuropathy are shown in Table 8.3.

Focal and multifocal neuropathies Entrapment neuropathies occur in older type 2 diabetic patients. Carpal tunnel syndrome is the most common entrapment neuropathy encountered in diabetic patients as a result of median nerve compression under the transverse carpal ligament. Whilst an electrophysiological deficit can be demonstrated in 20–30% of patients it is symptomatic in only 5.8%. Painful paraesthesiae of the fingers may progress to a deep-seated ache, which radiates up the forearm and arm with nocturnal exacerbation. Demyelination occurs and treatment options include wrist splints, injections of cortisone into the carpal tunnel and surgical decompression by sectioning the transverse carpal liga-

CHAPTER 8 • CLASSIFICATION AND CLINICAL FEATURES OF NEUROPATHY

Clinical features of autonomic neuropathies System

Feature

Symptoms

Cardiovascular

Orthostatic hypotension

Dizziness, lightheadedness, presyncope and syncope. Early fatigue/weakness on exercise

Exercise intolerance Gastrointestinal

Hypomotility

Hypermotility Urogenital

Bladder dysfunction Sexual dysfunction

Dysphagia and heartburn, early satiety, nausea, vomiting, belching, bloating, constipation Diaorrhea (nocturnal), incontinence Frequency, urgency, nocturia, urinary retention/ incontinence Erectile dysfunction, ejaculatory failure, retrograde ejaculation, vaginal dryness

Sudomotor

Sweating abnormalities

Anhydrosis, heat intolerance, gustatory sweating.

Pupillomotor

Visual abnormalities

Visual blurring, impaired adaptation to light

Autonomic afferent

Reduced visceral sensation

Silent myocardial ischemia, abdominal crises

Table 8.2 Clinical features of autonomic neuropathies.

Staging severity of diabetic neuropathy N0 =

No objective evidence of diabetic neuropathy

N1 =

Asymptomatic polyneuropathy N1a no symptoms or signs but neuropathic test abnormalities N1b test abnormalities1 + neuropathy impairment on neurological exam

N2 =

Symptomatic neuropathy N2a symptoms, signs and test abnormality N2b N2a + significant ankle dorsiflexor weakness

N3 =

Disabling polyneuropathy

1nerve

conduction, QST or autonomic test abnormalities.

Table 8.3 Staging severity of diabetic neuropathy.

81

82

SECTION II • DIABETIC NEUROPATHIES

ment. These provide variable degrees of pain relief but do not necessarily benefit muscle wasting or sensory loss. Ulnar neuropathy occurs in 2.1% of patients as a result of ulnar nerve compression in the cubital tunnel. Painful paraesthesiae in the 4th and 5th digits may be associated with hypothenar and interosseous muscle wasting. Both demyelination and axonal degeneration occur and management is conservative as the results of surgery are poor. Other entrapment neuropathies involve the radial, common peroneal, lateral femoral cutaneous and very rarely sciatic and obturator nerves. Cranial neuropathies are extremely rare (0.05%) and occur in older individuals with a long duration of diabetes. For the ocular neuropathies cranial nerves III (3.3%), IV (3.3%) and VI (2.1%) are affected. Oculomotor nerve palsy presents with acute diplopia, ptosis and pupillary sparing which resolves over approximately 2.5 months but can recur in 25% of patients. Subperineurial microfasciculation and demyelination have been shown to occur. 6–48% of patients with idiopathic facial neuropathy or Bell’s palsy have diabetes. Acute onset weakness of facial muscles, widening of the palpebral fissure and secondary corneal irritation may be accompanied by a disturbance in taste and hyperacusis. Concomitant hypertension and severity of paralysis determine the degree of recovery at one year. There are reports of an increased frequency of trigeminal neuralgia and hearing loss as a result of VIII nerve involvement. Diabetic amyotrophy occurs in patients with type 2 diabetes aged 50–60 years and presents with severe pain and uni- or bilateral muscle weakness and atrophy in the proximal thigh muscles. Electrophysiology shows spontaneous fibrillation, alterations in the amplitude of the motor unit potential, a reduction in femoral nerve conduction velocity and an attenuated compound muscle action potential of the quadriceps muscle. Recent reports have demonstrated centro-fascicular degeneration of the intermediate cutaneous nerve of the thigh with an inflammatory infiltrate and occlusion of epineurial blood vessels with features of a necrotizing vasculitis (arteriolar, venular and capillary wall infiltration with inflammatory cells) with haemosiderin deposition (Fig. 8.1). Based on the observations of a vasculitis in a proportion of patients, immunosuppressive therapy has been recommended using initial intravenous, followed by high dose oral corticosteroids, or intravenous immunoglobulin. Diabetic truncal radiculoneuropathy affects middle-aged to elderly diabetic patients. Pain occurs in a girdle like distribution over the lower thoracic or abdominal wall and is of an aching or burning quality, superimposed with lancinating stabs in a unilateral and rarely bilateral distribution. Profound weight loss may accompany the onset of symptoms. Clinical examination demonstrates heterogeneous neurological findings ranging from no abnormality to sensory loss and hyperaesthesia in a complete dermatomal pattern. Electromyography demonstrates denervation potentials in the intercostal, anterior abdominal wall, and paraspinal muscles. The natural history is for

CHAPTER 8 • CLASSIFICATION AND CLINICAL FEATURES OF NEUROPATHY

Fig. 8.1 Light microscopic H&E section showing epineurial arteriolar and venular wall infiltration with inflammatory cells and haemorrhage.

spontaneous resolution within 4–6 months. An improvement in glycaemic control has been advocated, as has immunosuppressive therapy with corticosteroids, or intravenous immunoglobulin.

CHRONIC INFLAMMATORY DEMYELINATING POLYNEUROPATHY (CIDP) Diabetic patients occasionally develop clinical and electrodiagnostic features suggestive of CIDP. Thus if an unusually severe and progressive polyneuropathy develops in diabetic patients one must consider CIDP. Current electrodiagnostic and pathological criteria appear to be insufficient for defining many cases of CIDP and therefore certainly should not be relied on to differentiate from diabetic polyneuropathy. The presence of increased numbers of macrophages suggestive of a macrophage-associated demyelination may be helpful as this is a characteristic feature of CIDP not observed in diabetic polyneuropathy. Treatment requires long-term immunomodulatory therapy with combinations of corticosteroids, azathioprine, plasmapharesis and intravenous immune globulin which produces relatively rapid and substantial improvement in neurological deficits and electrophysiology.

83

84

SECTION II • DIABETIC NEUROPATHIES

FURTHER READING Boulton AJM, Malik RA, Arezzo JC, Sosenko JM. Diabetic somatic neuropathies: technical review. Diabetes Care 2004; 27: 1458–1486. Vinik AI, Maser RE, Mitchell BD, Freeman R. Diabetic autonomic neuropathy. Diabetes Care 2003; 26: 1553–1579.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Rayaz A. Malik MB.ChB, PhD, MRCP CHAPTER 9 PATHOPHYSIOLOGY OF DIABETIC NEUROPATHY Studies in animal models and cultured cells provide a conceptual framework for the pathogenesis of experimental diabetic neuropathy. However, translational studies in diabetic patients continue to generate much debate and controversy over the cause(s) and treatment of human diabetic neuropathy.

Hyperglycaemia Longitudinal data from the Rochester cohort supports the contention that the duration and severity of exposure to hyperglycaemia are related to the severity rather than the onset of neuropathy. Similarly, in a study of newly diagnosed patients with type 2 diabetes followed from base-line for a period of 10 years, the overall severity but not the development of neuropathy was related to the degree of hyperglycaemia. Although recent studies in patients with impaired glucose tolerance (IGT) suggest that even minor degrees of glucose dysmetabolism may lead to the development of neuropathy. In patients with IGT sural nerve amplitude and myelinated fiber density do not differ from those with normal glucose tolerance. Although, of 121 patients with a painful neuropathy and electrodiagnostic evidence of axonal injury and epidermal nerve fibre abnormalities, 25% had impaired glucose tolerance. Intensive insulin therapy or pancreatic transplantation improves neuropathy in patients with type 1 diabetes. This is not proven in type 2 diabetes as both the VA Cooperative Study on type 2 Diabetes Mellitus (VACSDM) and Steno-2 study failed to demonstrate an improvement in somatic neuropathy.

Polyol Pathway Animal models of diabetes consistently demonstrate an association between increased flux through the polyol pathway and a reduction in nerve conduction velocity which can be ameliorated with aldose reductase inhibitors (ARIs). In humans the situation is not clear and in one of the earliest studies, sorbitol and fructose levels were increased in only one third of the sural nerve biopsies studied and could not be related to clinical, neurophysiological, or pathological severity of neuropathy. A significant increase in glucose, fructose and sorbitol was observed in post-mortem sciatic nerves from diabetic patients and in nerves obtained at amputation, though the actual concentrations differed markedly in these two studies. In a recent study of patients with normal glucose tolerance, IGT and type 2 diabetes only, the diabetic patients demonstrated an elevation in nerve sorbitol, suggestive of a glycaemic threshold for activation of this pathway. Linear regression analysis has demonstrated a significant inverse correlation between nerve sorbitol and myelinated fibre density. It also appears that those at greatest risk of

85

86

SECTION II • DIABETIC NEUROPATHIES

developing the complications are those with a higher set-point for aldose reductase activity. Furthermore, polymorphisms in the promoter region of the aldose reductase gene leading to a highly significant decrease in the frequency of the Z+2 allele have been demonstrated in patients with overt neuropathy compared to those without neuropathy. A meta-analysis of all randomized controlled trials of ARIs was disappointing. It identified 19 trials, testing four different ARIs for 4–208 weeks (median 24 weeks) and demonstrated a small but statistically significant reduction in decline of median (0.66 m/s 95% CI 0.18–1.14 m/s) and peroneal (0.53 m/s 95% CI 0.021.04 m/s) motor nerve conduction velocity (NCV) without benefit in sensory nerves. More recent trial data has been more encouraging and seems to show that the degree of aldose reductase inhibition may determine the improvement observed. Thus zenarestat demonstrated a dose-dependent increment in sural nerve sorbitol suppression accompanied by significant improvement in NCV, and in doses producing >80% sorbitol suppression there was a significant increase in the density of small-diameter myelinated fibres of the sural nerve. More recently, fidarestat, a potent ARI significantly improved median nerve F-wave conduction velocity (FCV) and minimal latency as well as symptoms of numbness, spontaneous pain, paraesthesiae and hyperaesthesiae.

Myoinositol Myoinositol deficiency has been proposed to play an important role in the pathogenesis of diabetic neuropathy on the basis of data in experimental diabetes. This has not been borne out in human nerve biopsies with no study to date showing a reduction in myoinositol levels.

Glycation Hyperglycaemia results in the formation of advanced glycation end-products (AGEs), which in turn act on specific receptors (RAGE), inducing monocytes and endothelial cells to increase the production of cytokines and adhesion molecules. Glycation has also recently been shown to have an effect on matrix metalloproteinases (MMPs), in particular MMP-2 which degrades type IV collagen, membrane type 1 MMP, tissue inhibitors of MMPs (TIMP)-1 and -2, and transforming growth factor β (TGF-β‚). Glycation also prevents epidermal growth factor auto-phosphorylation and activates extra cellular signal-regulated kinases (ERKs). In experimental diabetes these changes can be prevented by AGE inhibitors such as the nucleophilic compounds pyridoxamine, tenilsetam, 2, 3-diaminophenazone or aminoguanidine. Alternatively the administration of recombinant RAGE hinders the AGE-RAGE interaction. Sural nerves obtained from diabetic and non-diabetic amputation specimens demonstrate significantly elevated pentosidine

CHAPTER 9 • PATHOPHYSIOLOGY OF DIABETIC NEUROPATHY

levels in both cytoskeletal and myelin protein. Enhanced staining for carboxymethyllysine has been demonstrated in the perineurium, endothelial cells and pericytes of endoneurial microvessels as well as myelinated and unmyelinated fibres of sural nerves of patients with type 2 diabetes. Pyrraline, an advanced glycation end product is also increased in postmortem samples of optic nerve from diabetic patients. Intervention trials have focused on nephropathy and no trial data is currently available for human diabetic neuropathy.

Oxidative stress There is emerging evidence that single-nucleotide polymorphisms of the genes for mitochondrial (SOD2) and extracellular (SOD3) superoxide dismutases may confer an increased risk for the development of neuropathy. Alpha-lipoic acid (LA), a powerful antioxidant that scavenges hydroxyl radicals, superoxide and peroxyl radicals and regenerates glutathione has been shown (ALADIN-II study) to improve symptomatic neuropathy and, over two years sural SNCV, SNAP and tibial MNCV. In ALADIN-III 509 diabetic patients were randomized to LA intravenously for three weeks, followed by oral treatment, compared to placebo. Whilst the total symptom score remained unchanged, an improvement in the neuropathy impairment score was observed after three weeks of i.v. therapy, which was maintained until the end of the study. Most recently the SYDNEY study has demonstrated a significant improvement in the neuropathy symptom score, neuropathy impairment score and one attribute of nerve conduction after daily i.v. with racemic LA for five days/week for 14 treatments.

Vascular factors Large vessel revascularization studies have shown an improvement in NCV in one but not another study. Although in the latter study at long-term follow up a prevention of worsening of peroneal NCV was shown. In a doubleblind placebo controlled clinical trial of trandalopril over 12 months, peroneal motor NCV, M-wave amplitude F-wave latency and sural nerve amplitude improved significantly. Although in the Appropriate Blood Pressure Control in Diabetes (ABPCD) trial the effects of intensive versus moderate blood pressure control with either nisoldipine or enalapril were assessed and surprisingly failed to prevent progression of not only diabetic neuropathy but also nephropathy and retinopathy.

Protein kinase C-β 1, 2-diacylglycerol (DAG) induced activation pf protein kinase C (PKC). In particular, PKC-β, has been proposed to play a major role in diabetic neuropathy. Although in nerves from diabetic animals a fall in DAG levels and a consistent

87

88

SECTION II • DIABETIC NEUROPATHIES

pattern of change in PKC activity has not been observed, inhibition of PKC-β corrects reduced nerve blood flow and NCV. Based on the findings of a phase II clinical trial demonstrating some benefit in diabetic patients with neuropathy, multi-centre randomized, double-blind, placebo controlled trials are underway, and are due to complete in 2006.

Hydroxymethylglutaryl CoA reductase inhibitors An increasing body of evidence suggests that conventional risk factors for macrovascular disease such as deranged lipids are important in the pathogenesis and progression of human diabetic neuropathy. Recent studies show that hydroxymethylglutaryl CoA reductase inhibitors may enhance endothelial cell nitric oxide bioavailability and prevent AGE-induced NF-kb induced protein-1 activation and up-regulation of vascular endothelial growth factor (VEGF) mRNA, thus ameliorating experimental diabetic neuropathy. Thus simvastatin has shown a trend towards slower progression of neuropathy measured by vibration perception threshold but no change in clinical neuropathy. As a caution, recent case studies suggest a link between statin use and an increased risk of peripheral neuropathy.

GROWTH FACTORS Neurotrophins Neurotrophins promote the survival of specific neuronal populations by inducing morphological differentiation, enhancing nerve regeneration, stimulating neurotransmitter expression, and altering the physiological characteristics of neurones. Studies in experimental models of diabetes demonstrate a depletion of nerve growth factor (NGF) and NT3 which if corrected ameliorates NCV deficits. Although initially studies in the skin of patients with diabetic neuropathy demonstrated a depletion of NGF, subsequent studies have shown a significant increase in skin NGF mRNA and NT3 concentrations with normal sciatic nerve ciliary neurotrophic factor levels. In situ hybridization studies have also demonstrated an increased expression of trkA, the high-affinity receptor for NGF, and trkC, the receptor for NT3, in the skin of diabetic patients. Despite these apparently contradictory findings, a phase II clinical trial of recombinant human NGF in 250 patients with diabetic neuropathy demonstrated a significant improvement in the sensory component of the neurological examination and two quantitative sensory tests. However, a phase III trial in 1,019 diabetic patients with sensory polyneuropathy failed to demonstrate a significant benefit. Recently a randomized, double-blind, placebo controlled study of brainderived neurotrophic factor (rhBDNF) in 30 diabetic patients demonstrated no significant improvement in nerve conduction and quantitative sensory and autonomic function tests, including the cutaneous axon-reflex.

CHAPTER 9 • PATHOPHYSIOLOGY OF DIABETIC NEUROPATHY

Insulin like growth factors (IGFs) In cultured Schwann cells and the STZ-diabetic rat, IGF-1 demonstrates a protective effect via PI 3-kinase, in preventing glucose-mediated neuronal and Schwann cell apoptosis. Both the STZ-diabetic and BB/W rat develop severe hyperglycaemia and a deficiency in circulating IGF-I levels with neuroaxonal dystrophy (NAD) in the nerve terminals of the prevertebral sympathetic ganglia and the distal portions of noradrenergic ileal mesenteric nerves. In contrast the Zucker Diabetic Fatty (ZDF) rat, an animal model of type 2 diabetes, also develops severe hyperglycaemia but maintains normal levels of plasma IGF-I and does not demonstrate NAD in sympathetic ganglia and ileal mesenteric nerves. However, IGF-I and IGF-I receptor mRNA levels have not been shown to differ in the sural nerve of diabetic patients compared with control subjects.

C-Peptide In experimental studies C-peptide has demonstrated effects on Na(+)/K(+)ATPase activity, endothelial nitric oxide synthase, expression of neurotrophic factors and regulation of molecules controlling degeneration of the nodal apparatus in Type 1 diabetic nerves, as well as DNA binding of transcription factors and modulation of apoptotic phenomena. These findings have recently been effectively translated into benefits in patients with Type 1 diabetes with the demonstration of a significant improvement in sural sensory nerve conduction velocity and vibration perception but without a benefit in either cold or heat perception after 12 weeks of daily subcutaneous C-peptide treatment.

Vascular endothelial growth factor (VEGF) VEGF was originally discovered as an endothelial-specific growth factor with a predominant role in angiogenesis. However, recent observations indicate that VEGF also has direct effects on neurones and glial cells stimulating their growth, survival and axonal outgrowth. Thus with its potential for a dual impact on both the vasculature and neurones it could represent an important therapeutic intervention in diabetic neuropathy. Although immunohistochemistry of sciatic nerves and dorsal root ganglia from STZ-diabetic rats demonstrates intense VEGF staining in cell bodies and nerve fibers with no or very little VEGF expression in controls. Intramuscular gene transfer of plasmid DNA encoding VEGF1 or VEGF-2 in the STZ-diabetic rat and alloxan diabetic rabbit results in restoration of nerve vascularity, blood flow and both large and small fibre dysfunction. Thus there is an intrinsic capacity to up-regulate VEGF but this appears insufficient and may require exogenous delivery possibly via gene therapy. A phase I/II, double-blind, placebo controlled study to evaluate the safety and impact of phVEGF165 gene transfer in patients with diabetic neuropathy is currently underway and will involve 192 patients over a period of four years.

89

90

SECTION II • DIABETIC NEUROPATHIES

IMMUNE MECHANISMS Studies suggest that sera from type 2 diabetic patients with neuropathy contain an autoimmune immunoglobulin that induces complement-independent, calcium-dependent apoptosis in neuronal cells. The expression of these cytotoxic factors has been related to the severity of neuropathy and the type of neuronal cell killed. Thus it has been suggested that such toxic factors may contribute to diabetic neuropathy by acting in concert with hyperglycaemia to damage sensory/autonomic neurones.

CURRENT ISSUES • Translational studies in diabetic patients continue to generate much debate and controversy over the cause(s) and treatment of human diabetic neuropathy. • Recent studies in patients with impaired glucose tolerance (IGT) suggest that even minor degrees of glucose dysmetabolism may lead to the development of neuropathy. • A meta-analysis of 19 randomized controlled trials of ARIs was disappointing as it demonstrated a small but statistically significant reduction in decline of motor NCV without benefit in sensory nerves. • Single-nucleotide polymorphisms of the genes for mitochondrial (SOD2) and extracellular (SOD3) superoxide dismutases may confer an increased risk for the development of neuropathy. • 1, 2-diacylglycerol (DAG) induced activation pf protein kinase C (PKC), in particular PKC-β has been proposed to play a major role in diabetic neuropathy. • Conventional risk factors for macrovascular disease such as deranged lipids and hypertension are important in the pathogenesis of diabetic neuropathy. • In cultured Schwann cells and the STZ-diabetic rat, IGF-1 prevents glucosemediated neuronal and Schwann cell apoptosis. • VEGF was originally discovered as an endothelial-specific growth factor.

FURTHER READING Dyck PJ, Davies JL, Wilson DM, Service FJ, Melton LJ 3rd, O’Brien PC. Risk factors for severity of diabetic polyneuropathy: intensive longitudinal assessment of the Rochester Diabetic Neuropathy Study cohort. Diabetes Care 1999; 22: 1479–1486. Eichberg J. Protein kinase C changes in diabetes: is the concept relevant to neuropathy? Int Rev Neurobiol 2002; 50: 61–82. Gaede P, Vedel P, Larsen N, Jensen GV, Parving HH, Pedersen O. Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med 2003; 348: 383–393.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Rayaz A. Malik MB.ChB, PhD, MRCP CHAPTER 10 EPIDEMIOLOGY AND NATURAL HISTORY OF DPN

INTRODUCTION Epidemiological studies of diabetic peripheral neuropathy (DPN) have produced widely varying figures on incidence and prevalence. There are several reasons for this and include differing criteria to define neuropathy and the use of different populations i.e. clinic versus hospital patients. Furthermore, the clinical complexity of DPN due to the relative involvement of sensory motor and autonomic fibres has led to the use of multiple assessments to define DPN. These have commonly included varying combinations of positive and negative symptoms, neurological deficits derived from a clinical examination, quantitative sensory tests and electrophysiology. To attempt to draw some valid comparisons between different studies each of the end points utilized will be evaluated separately. The incidence or prevalence will be defined and the natural history and risk factors for the end point considered.

POSITIVE SENSORY SYMPTOMS (PAINFUL NEUROPATHY) Positive sensory symptoms arise spontaneously or as a response to stimuli and they may be divided into painful and non-painful categories. Table 10.1 presents the long list of positive sensory symptoms, which have been utilized in different studies and provides an insight into the difficulties of how different symptoms may be recorded and interpreted.

Descriptions of positive neuropathic sensory symptoms Non-painful

Painful

thick stiff asleep prickling tingling

prickling tingling knife-like electric shock-like squeezing constricting hurting burning freezing throbbing allodynia* hyperalgaesia

* Allodynia: the perception of pain from a non-noxious stimulus.

Table 10.1 Descriptions of positive neuropathic sensory symptoms. J Neurolog Sci 2001; 189: 3–5.

91

92

SECTION II • DIABETIC NEUROPATHIES

A prevalence of 27% was demonstrated in the only population-based study to date and it employed questionnaires to define symptoms according to pain, tingling, numbness, and the inability to feel hot or cold. Other clinic or hospital based studies utilizing similar symptom profiles have reported a prevalence of painful symptoms which varies from 3–30%, although their focus was more heavily on pain. Data on the natural history of painful neuropathy are conflicting, with one study showing a decrease in the intensity of symptoms with worsening of quantitative measures of nerve function, whilst another study found an improvement of pain with improvement of sensory function in individuals treated with continuous subcutaneous insulin infusions (CSII). Of course CSII has been shown to have a direct effect on painful symptoms via decreasing glucose flux. The data on remission of symptoms over time are inconsistent, with one study showing an overall reduction in the severity of pain scores but without any full remissions, whilst the majority of other studies have observed remissions. Despite the large number of studies in patients with painful neuropathy the risk factors remain ill defined. Thus in a population based study the degree of hyperglycaemia, hypertension and diabetes duration were deemed important, but in a clinic based study, symptoms were related to diabetes duration but not HbA1C.

NEGATIVE SENSORY SYMPTOMS (HYPOAESTHETIC NEUROPATHY) Since it is well recognized that patients tend to underestimate their degree of insensitivity, there has been more reliance on quantitative sensory testing than on symptomatology in studies of hypoesthesia. The prevalence of hypoesthesia may vary greatly according to the criteria used to define abnormality. In a study that utilized three different quantitative sensory measurements in the same individuals, the prevalence of abnormalities varied from 8–34%, highlighting the impact of different tests on the outcome data generated. The site chosen to measure the deficit is also an important factor as the presence of hypoesthesia increases substantially as measurements become more distal. Whilst vibration perception thresholds may be elevated in children with diabetes and also in adults at diagnosis of type 2 diabetes, age requires careful consideration since normal values increase markedly with age, especially after the 6th decade. In patients with type 1 diabetes followed from diagnosis, abnormalities of thermal thresholds and electrophysiology preceded any abnormality of vibration threshold in the first five years after diagnosis. Early in the course of type 2 diabetes, small but statistically significant increases of vibration and thermal thresholds have been observed over two years. A study that utilized the 10 g monofilament apparently showed no

CHAPTER 10 • EPIDEMIOLOGY AND NATURAL HISTORY OF DPN

progression of neuropathy in the first few years of diabetes. This simply reflects that the 10 g monofilament is neither sensitive enough to detect early loss of sensation nor to detect a small change in progressive loss of sensation. This study highlights the inappropriate use of the 10g monofilament and shows that it is useful only to detect those at risk of foot ulceration i.e. those with severe neuropathy. Progression appears to be more rapid once decreased sensation appears and has been consistently related to diabetes duration, degree of hyperglycaemia, and height.

COMBINED ASSESSMENTS Studies which have utilized various combinations of positive and negative symptoms, quantitative sensory testing, abnormalities of the neurological exam and electrophysiology have generally produced higher estimates of the prevalence of neuropathy. In one of the largest and earliest epidemiological studies the prevalence of neuropathy was greater than 40% after 25 years of known diabetes duration. The EURODIAB IDDM Complications Study provided an overall prevalence for neuropathy of 28%. However, among the 27 centres included in the study, the prevalence ranged from less than 20% in several centres to over 50% in two centres. This highlights the variability in interpreting examination findings in different countries and also that the patients chosen for study in the different centres came from differing clinical populations. In the Diabetes Control and Complications Trial (DCCT) ‘clinically detectable’ neuropathy was found in 39% of the participants who had type 1 diabetes. In the Epidemiology of Diabetes Complications (EDC) Study, a prospective study of patients with type 1 diabetes, the overall prevalence of DPN at base-line was 37%. In the San Luis Valley Diabetes Study, a population based study of type 2 diabetic patients, there was an overall prevalence of 28%. The natural history of DPN is difficult to ascertain from such studies. Pirart’s study revealed strong univariate associations between neuropathy and the duration of diabetes as well as degree of hyperglycaemia. The EURODIAB IDDM study has employed multivariate modeling to identify associations of neuropathy with age, duration of diabetes, HbA1C and severe ketoacidosis. Furthermore, the prevalence of neuropathy was related to elevated diastolic blood pressure and triglycerides and decreased HDL-cholesterol. In the DCCT impairment of nerve conduction was found to be associated with age, diabetes duration HbA1C, male gender and Cpeptide deficiency. In the EDC Study DPN at base-line was related to age, diabetes duration, HbA1, HDL-cholesterol, hypertension and cigarette smoking. The SLVDS found that DPN was related to age, diabetes duration, HbA1C and insulin use. Several studies have also observed associations of DPN with retinopathy and nephropathy. In a case-control study, DPN was associated with lifetime cigarette smoking in type 1 but not type 2 diabetes.

93

94

SECTION II • DIABETIC NEUROPATHIES

CONCLUSION Despite the seeming inconsistencies of findings in epidemiological studies of diabetic neuropathy, one thing is clear: DPN is very common. At least one manifestation of DPN is present in well over 20% of individuals with diabetes. However, the prevalence of painful neuropathy appears to be appreciably lower. Regarding natural history, some degree of abnormality occurs virtually from the diagnosis of Type 1 and particularly type 2 diabetes. Once DPN is detectable, especially with an abnormality in vibration sense, there is a tendency toward rapid progression. DPN is associated with the degree of hyperglycaemia, diabetes duration, height, conventional cardiovascular risk factors (including lipid and blood pressure indices), and other complications of diabetes. Other risk factors such as alcohol consumption and cigarette smoking have been less consistent in their associations with DPN. The definitive risk factors that have been identified have biological plausibility for their involvement in the pathogenesis of DPN (Table 10.2). Duration and degree of hyperglycaemia indicate the extent of overall exposure to hyperglycaemia. Height, as a proxy for nerve length entertains the hypothesis that the longer nerves are more susceptible to the metabolic and vascular consequence of diabetes. Blood pressure and lipid indices would appear to strengthen the assertion that vascular abnormalities contribute to the development and progression of DPN and the relation with other complications suggests they have common (vascular) pathogenetic pathways. Thus epidemiologic studies provide important clues for future mechanistic research and also provide support for mechanisms already defined in the pathogenesis of DPN. Future epidemiological studies should pay particular attention to the endpoints and methodology utilized. Risk factors and their biological basis for DPN Risk factor

Biological basis

Age, duration of diabetes, HbA1C severe ketoacidosis Height

Measures of total exposure to primary inducer of multiple pathogenetic pathways. Proxy for longer nerves being more susceptible to metabolic/vascular mechanisms Inducer of vascular dysfunction Inducer of vascular dysfunction Common vascular basis for all complications May aggravate neuropathy May aggravate vascular dysfunction

Dyslipidaemia Hypertension Nephropathy and Retinopathy *Alcohol consumption *Cigarette smoking

* Less consistent associations with DPN.

Table 10.2. Risk factors and their biological basis for DPN.

CHAPTER 10 • EPIDEMIOLOGY AND NATURAL HISTORY OF DPN

CURRENT ISSUES • • •

•

Epidemiological studies of diabetic peripheral neuropathy (DPN) have produced widely varying figures on incidence and prevalence. The prevalence of painful symptoms varies from 3–30%. The 10 g monofilament is neither sensitive enough to detect early loss of sensation nor to detect a small change in progressive loss of sensation – and should only be used to detect those at risk of foot ulceration. The EURODIAB IDDM study has identified associations between neuropathy and age, duration of diabetes, HbA1C, diastolic blood pressure, triglycerides and decreased HDL-cholesterol.

FURTHER READING EURODIAB IDDM Study Group. Prevalence of diabetic peripheral neuropathy and its relation to glycaemic control and potential risk factors: the EURODIAB IDDM compliction study. Diabetologia 1996; 39: 1377–1386. The EURODIAB Prospective Complications Study (PCS) Group. Cardiovascular risk factors predict diabetic peripheral neuropathy in type 1 subjects in Europe. Diabetologia 1999; 42: A50–181. Witte DR, Tesfaye S, Chaturvedi N, Eaton SE, Kempler P, Fuller JH, EURODIAB Prospective Complications Study Group. Risk factors for cardiac autonomic neurpathy in type 1 diabetes mellitus. Diabetologia 2005; 48: 164–171.

95

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Rayaz A. Malik MB.ChB, PhD, MRCP CHAPTER 11 DETECTION/SCREENING/ASSESSMENT

SYMPTOMS Patients often have difficulty in describing the symptoms of neuropathy. Therefore when physicians record symptoms in clinical practice they must avoid ‘interpreting’ or ‘translating’ what patients report, rather they should record the patient’s description verbatim. For simple screening, a number of questionnaires are available and include the Neuropathy Symptom Score (NSS), Michigan Neuropathy Screening Instrument (MNSI), and more recently the NSS +4. Where a response to treatment is being assessed, visual analogue or verbal descriptive scales are used, but few have been validated for diabetic neuropathy.

Signs Composite scores which quantify the degree of neurological deficit in the lower limbs were pioneered by Dyck and co-workers who first described the Neuropathy Disability Score (NDS) and later the Neuropathy Impairment Score (NIS). A modified NDS has been used in several large studies (Table 11.1) and has been shown to predict foot ulceration in a large prospective community study. A clinical scoring system which documents and monitors neuropathy in the clinic has been validated by the Toronto group.

CLINICAL SCREENING Whilst the simple hand held screening devices are less sensitive than the more sophisticated QST devices, they are cheap, portable and easy to use. Currently the most widely and sometimes perhaps inappropriately used device is the Semmes-Weinstein monofilament. The lack of perception of pressure to gentle pressure applied to the handle sufficient to buckle the nylon filament defines an abnormal response. Although filaments of many different sizes are available, the 10 g or 5.07 (10 x log of the force generated to deform the filament) monofilament should be used. In identifying feet at risk of ulceration it has a sensitivity of 86–100%. The commonest algorithm recommends four sites per foot (hallux and metatarsal heads 1, 3, and 5). Although a recent study suggests that there is little advantage gained from multiple site assessment. Additionally a recent study has shown that filaments manufactured by a number of companies actually buckle at 8 g rather than the validated 10 g monofilament. The graduated Rydel-seiffer tuning fork is widely used in Europe, particularly in Germany. It allows the assessor to define a threshold on a 0–8 scale and has been shown to correlate well

97

98

SECTION II • DIABETIC NEUROPATHIES

Risk factors and their biological basis for DPN Right

Left

Vibration perception threshold 128 Hz tuning fork; apex of big toe: normal = can distinguish vibrating/ not vibrating Temperature perception on dorsum of the foot Use tuning fork with beaker of ice/warm water

Normal = 0 Abnormal = 1

Pin-prick apply pin proximal to big toe nail just enough to deform the skin; trial pair = sharp, blunt ; normal = can distinguish sharp / not sharp Achilles reflex

Present = 0 Present with reinforcement = 1 Absent = 2 NDS Total out of 10

Table 11.1 Neuropathy Disability Score.

with other QST measures. The recently reported Neuropen combines the assessment of pain using a neurotip with pressure using a 10 g monofilament.

QUANTITATIVE SENSORY TESTING (QST) QST may be defined as any procedure requiring a power source where the intensity and characteristics of the stimuli are well controlled and where the detection threshold is determined in parametric units that can be compared to established normal values. QST measures the three major modalities of vibration, thermal and pain thresholds and has been shown to detect sub-clinical neuropathy, track its progression and predict those patients at risk of foot ulceration. Despite the strengths and weaknesses of QST (Table 11.2) it has been used as a primary efficacy measure in a number of completed and on-going clinical trials of DPN.

Vibration perception threshold (VPT) VPT reflects the activation of mechanoreceptors (i.e. Pacinian and Meissner corpuscles), conduction in peripheral large diameter myelinated axons, and transmission through the dorsal column spinal pathways. Vibratory thresholds

CHAPTER 11 • DETECTION/SCREENING/ASSESSMENT

Strengths and limitations of QST Strengths • Accurate control of stimulus characteristics • Ability to assess multiple modalities • Use of well established psychophysical procedures to enhance sensitivity • Measures function over a wide range of intensity and hence neuropathic severity • Measures sensation at multiple anatomical sites • Data from large, age-matched normal groups available for comparison Limitations • Semi-objective measure affected by the subject’s motivation and cooperation • Affected by age, gender, body mass, history of smoking and alcohol consumption • Expectancy and subject bias • Affected by any change along the entire neuroaxis from nerve to cortex. • Chronic liver/renal disease

Table 11.2 Strengths and limitations of QST.

have been shown to detect not only sub-clinical neuropathy in children and adolescents with type 1 diabetes but also provide a strong indication of risk for future ulceration. In a 4-year prospective study patients with base-line VPT >25 volts with the biothesiometer were seven times more likely to develop foot ulcers. Recently in 187 type 2 diabetic patients multivariate logistic regression has shown that an elevated VPT score was the strongest predictor of foot ulceration (relative risk of 25.4). In a large prospective study of 1,035 type 1 and type 2 diabetic patients, each one unit increase in vibration threshold (voltage scale) at base-line increased the hazard of foot ulceration by 5.6% over one year.

Thermal thresholds Separate cold and warm thermoreceptors conduct thermal energy in thinly myelinated Aδ or unmyelinated C fibres, respectively. Additionally ‘pain’ can be driven principally by high intensity stimulation of warm thermoreceptors. Thermal thresholds are abnormal in patients with sub-clinical neuropathy, increase with progression of neuropathy and also predict foot ulceration. Generally, there is a high correlation between elevated thermal and vibration thresholds, but when these measures are dissociated it suggests a predominant small or large fibre neuropathy.

Electrophysiology Multiple consensus panels have recommended electrophysiology in the evaluation of diabetic peripheral neuropathy (DPN), and in its use as a primary measure of therapeutic efficacy in multicentre clinical intervention trials. The

99

100

SECTION II • DIABETIC NEUROPATHIES

speed of both sensory and motor conduction, amplitude of the propagating neural signal, density and synchrony of muscle fibres activated by maximal nerve stimulation, and the integrity of neuromuscular transmission are assessed. When used together they are objective, non-invasive and highly reliable. However, the standard procedure of maximal nerve conduction velocity (NCV) has major limitations as it reflects the integrity of only a small subset of large diameter and heavily myelinated axons. It also misses subtle but important early alterations such as a reduction in Na+/K+ adenosine triphosphatase activity, which would primarily diminish the ability of neurones to rapidly re-establish transmembrane ion gradients and can only be assessed by defining refractory cycles and axonal recovery. A number of principal factors influence the speed of NCV (Table 11.3). In early diabetic neuropathy the slowing of NCV is thought to be related to an alteration in nodal ion distribution, a diminished length constant of large diameter axons due to an altered cross-sectional volume and demyelination which is further augmented at later stages by Wallerian degeneration. NCV provides a sensitive, but nonspecific index on the onset of DPN and can be valuable in detecting sub-clinical deficits. It diminishes by approximately 0.5 metres/sec/year. In a study of 133 patients with newly diagnosed type 2 diabetes, NCV deteriorated by 3.9 and 3.0 metres/sec in the sural and peroneal nerves respectively over 10 years. In the Diabetic Control and Complications Trial (DCCT) the sural and peroneal nerve velocities in the conventionally treated group diminished by 2.8 and 2.7 metres/sec, respectively over the 5-year study period. The changes in NCV relate to glycaemic control as in the DCCT the incidence of abnormal NCV was 40.2% in the conventionally treated group and only 16.5% in the group receiving intensive insulin therapy. In a study of 45 type 1 diabetic patients a regression analysis has shown that a 1% change in HbA1 is associated with a 1.3 metres/sec change in maximal nerve conduction. The peak amplitude of either the sensory response (SNAP) or the compound muscle action potential (CMAP) reflects the number of responding

Principle factors governing electrophysiological alterations in diabetic neuropathy • Integrity and degree of myelination of the largest diameter fibres • Mean cross-sectional diameter of the responding axons • Representative internodal distance in the segment under study • Micro-environment at the nodes • Distribution of ion channels

Table 11.3 Principle factors governing electrophysiological alterations in diabetic neuropathy.

CHAPTER 11 • DETECTION/SCREENING/ASSESSMENT

fibres and the synchrony of their activity. There is a strong correlation (r = 0.74; p <0.001) between myelinated fibre density and whole nerve sural amplitude in DPN and a change of 1.0 μV in sural SNAP amplitude is associated with a decrease of approximately 150 myelinated fibres/mm2, while a loss of 200 fibres/mm2 is associated with an approximate 1.0 mV reduction in the mean amplitude of the CMAP from the ulnar, peroneal and tibial nerves. The total area of the SNAP and CMAP may reflect the contribution of slower conducting fibres, but it is severely limited by variability. F-waves reflect the antidromic conduction of the compound neural volley to the ventral spinal cord, the activation of a sub-population of spinal motor neurons, the orthodromic conduction of the newly established volley and the postsynaptic activation of a portion of the muscle fibres in the innervated muscle. A subtle change affecting each node may not be detected in measures focused on an isolated distal segment, but may accumulate and become evident in the long latency F-wave response. However, changes limited to the distal segment of the axon, including possible therapeutic benefits, may be poorly represented in F-wave measures. The distribution of conduction velocities allows an assessment of the activity in small diameter axons. The recent fusion of a collision technique with an analysis of the distribution of velocities has shown that it is highly sensitive for detecting sub-clinical neuropathy as 58% of patients displayed an abnormality in the distribution of conduction velocities compared to only 11% using standard procedures. In addition to measuring the speed of conduction or the size of the activated signal, the excitability (magnitude and nature of the current necessary to establish the electrophysiological (EP) response) can be an important parameter in assessing neuropathy. At the cellular level, changes in excitability may be related to alterations in intracellular levels of cyclic adenosine monophosphate (cAMP) that have been reported with DPN.

INVASIVE ASSESSMENT Nerve biopsy/exposure Sural nerve biopsy is established as a technique to aid in the diagnosis of atypical neuropathies and more recently as a measure of therapeutic efficacy. However, this is an invasive procedure with recognized sequelae which include persistent pain at the biopsy site, cold intolerance, unpleasant though mild mechanically elicited sensory symptoms and sensory deficits in the sural distribution, particularly in diabetic patients. Thus, with the widespread availability of accurate QST and EP techniques, biopsies are rarely required for the routine diagnosis of DPN. The use of morphological measures of neuropathy from biopsies as endpoints in trials of potential pharmacological therapies for DPN is a more controversial area. Concerns have been expressed regarding the

101

102

SECTION II • DIABETIC NEUROPATHIES

invasive nature of having two nerve biopsies on separate occasions. Furthermore, although myelinated fibre density relates well to alterations in electrophysiology and clinical neurological deficits, the validity of assessing axonal atrophy and axo-glial dysjunction has been questioned. Despite these drawbacks nerve biopsy has been employed to help determine the aetiopathogenesis of neuropathy. In particular it has given unprecedented insights into the development of diabetic amyotrophy. A number of studies have investigated alterations in sural nerve function and structure in vivo without actually performing a biopsy. This has included using microelectrodes to measure endoneurial oxygen tension and epineurial vessel photography with fluorescein angiography to study the structure and function of the epineurial blood vessels. Similarly microlight-guide spectrophotometry has been employed to measure blood flow and oxygen saturation in the sural nerve.

SKIN BIOPSY This technique, although still invasive, only requires a 3 mm skin biopsy and enables immunohistological staining with PGP 9.5 to quantify small nerve fibre morphology. It has been shown to be highly sensitive, as epidermal nerve fibres are abnormal even in patients with impaired glucose tolerance (IGT) who have entirely normal electrophysiology and QST. It may also enable the assessment of small fibre regeneration and hence has been advocated as a measure to evaluate therapeutic efficacy in clinical trials of diabetic neuropathy.

NON-INVASIVE ASSESSMENT Magnetic resonance imaging (MRI) MRI has been used to assess involvement of the spinal cord in neuropathy. In an exploratory study patients with DPN had a lower cross-sectional cord area than healthy controls in the cervical and thoracic regions. More recently alterations have been shown in patients with sub-clinical neuropathy preferentially affecting the spinothalamic tracts and thalamus.

Corneal confocal microscopy The technique of corneal confocal microscopy is a completely non-invasive and reiterative technique that assesses nerve structure in vivo without the need for biopsy (Fig. 11.1). This technique accurately defines the extent of corneal nerve damage and repair, which correlates with a range of measures of neuropathic severity including NDS, QST and electrophysiology. This may act as a structural surrogate measure of neuropathic severity and may be utilized as a primary efficacy measure of therapeutic efficacy in clinical trials of human diabetic neuropathy.

CHAPTER 11 • DETECTION/SCREENING/ASSESSMENT

Fig. 11.1 CCFM image of normal cornea demonstrating normal corneal nerve fibre density and branching with normal tortuosity.

CURRENT ISSUES •

• •

•

•

•

•

Patients often have difficulty in describing the symptoms of neuropathy. Therefore when physicians record symptoms they must record the patient's description verbatim. Composite scores which quantify the degree of neurological deficit in the lower limbs were pioneered by Dyck and co-workers. Hand held screening devices are less sensitive than the more sophisticated QST devices but they are cheap and portable and hence widely used. QST measures vibration, thermal and pain thresholds and has been shown to detect sub-clinical neuropathy, track its progression and predict those patients at risk of foot ulceration. Multiple consensus panels have recommended electrophysiology in the evaluation of DPN, and in its use as a primary measure of therapeutic efficacy in multicentre clinical intervention trials. Skin biopsy enables quantification of small nerve fibre morphology and has been shown to be highly sensitive, as epidermal nerve fibres are abnormal in patients with IGT who have entirely normal electrophysiology and QST. Corneal confocal microscopy may act as a structural surrogate measure of neuropathic severity and may be utilized as a primary efficacy measure of therapeutic efficacy in clinical trials of human diabetic neuropathy.

103

104

SECTION II • DIABETIC NEUROPATHIES

FURTHER READING Boulton AJM, Malik RA, Arezzo JC, Sosenko JM. Diabetic somatic neuropathies: technical review. Diabetes Care 2004; 27: 1458–1486. Shy ME, Frohman EM, So YT Arezzo JC, Cornblath DC, Giuliani MJ and the subcommittee of the American Academy of Neurology. Quantitative sensory testing. Neurology 2003; 602: 898–906.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Rayaz A. Malik MB.ChB, PhD, MRCP CHAPTER 12 FOOT ULCERATION AND CHARCOT ARTHROPATHY The late sequelae of diabetic peripheral neuropathy (DPN) are recognized to be foot ulceration (cumulative lifetime incidence ~15%) and Charcot’s neuroarthropathy.

FOOT ULCERATION Causation Rothman’s model of causation defines a combination of neuropathy, trauma, and foot deformity to be the commonest pathways to foot ulceration and faulty healing may eventually lead to amputation (Fig. 12.1). Both large and small fibre somatic nerve damage lead to the insensate foot, but it per se does not ulcerate spontaneously. It is the combination of neuropathy with either extrinsic factors (e.g. ill-fitting shoes, foreign body in shoe) or intrinsic factors (e.g. high foot pressures or plantar callus) that results in ulceration. Long periods of high pressure due to insensitivity leads to tissue ischaemia, necrosis and ulceration. Additionally autonomic dysfunction results in blunted pressure-induced vasodilation (PIV) where cutaneous blood flow in response to locally applied pressure is impaired in diabetic patients. Once ulceration has occurred, removing pressure has been shown to have a striking effect on

Neuropathy

Neuropathy

Trauma

Faulty healing

Neuropathy

Neuropathy

Trauma

Trauma

Amputation

Ulceration

Ulceration

Fig. 12.1 Pathways to ulceration and amputation in the diabetic foot.

105

106

SECTION II • DIABETIC NEUROPATHIES

wound biology aiding healing. Wound biopsies from 10 patients with neuropathic ulceration who underwent pressure relief for 20 days with a total contact cast were compared with 10 patients without intervention. Biopsies from patients with pressure relief showed a shift towards a reparative pattern with angiogenesis and granulation and less hyperkeratosis, fibrosis, inflammation, and cellular debris.

Classification Foot ulceration classification systems are important to risk stratify patients and to implement timely and appropriate intervention to aid healing and predict outcomes. The oldest and probably the most widely used classification is the Wagner classification, which if used rigorously expedites timely management, particularly when local amputation can limit progression to further higher level amputation. Newer classifications have evolved recently which take into account a number of additional features which may predict outcomes. Thus the UTDWCS (University of Texas diabetes wound classification system) takes into account not just severity of the wound but other key factors which limit healing i.e. presence of infection, ischaemia or both (Table 12.1). The Nottingham based SAD classification system has been validated recently and showed that ulcer area, depth and presence of arteriopathy but not neuropathy contribute independently to predicting outcomes. Recently the International Working Group of the Diabetic Foot (IWGDF) has developed a new classification system for research purposes (PEDIS) which proposes that all foot ulcers should be classified according to five categories: perfusion, extent/size, depth/tissue loss, infection and sensation. It needs to be validated before it can be formally introduced into clinical and research practice.

UTDWCS (University of Texas diabetes wound classification system) 0

1

2

3

A

Pre- or postulcerative lesion (epithelialized)

Superficial, not involving tendon, capsule or bone

Penetrates to tendon or capsule

Penetrates to Bone

B

Infection

Infection

Infection

Infection

C

Ischaemia

Ischaemia

Ischaemia

Ischaemia

D

Infection and Ischaemia

Infection and Ischaemia

Infection and Ischaemia

Infection and Ischaemia

Table 12.1 UTDWCS (University of Texas diabetes wound classification system).

CHAPTER 12 • FOOT ULCERATION AND CHARCOT ARTHROPATHY

PREVENTION Education Although it is generally believed that education and preventative foot care should reduce the risk of ulceration in high risk individuals, there are few data to support this contention. A recent study of home daily foot temperature monitoring with advice on a reduction in activity and regular contact with the study nurse when foot temperatures were elevated (>4 0F vs. opposite foot), demonstrated a significant reduction in diabetic foot complications (enhanced therapy group 2% vs. standard therapy group 20%, seven ulcers and two Charcot fractures vs. one ulcer).

Healing In a recent study from Germany 120 diabetic patients with neuropathic foot ulcers were grouped according to the initial ulcer size (<100 mm, 100–150 mm, 150–200 mm, >200 mm) and their healing velocity evaluated. The reduction in wound radius and the time needed for healing were related to the initial ulcer area. In 203 patients those with a reduction in ulcer area greater than the 4-week median (53%) had a 12-week healing rate of 58%, whereas those with a reduction in ulcer area less than the 4-week median had a healing rate of only 9%. Thus the percent change in foot ulcer area after four weeks of observation appears to be a robust predictor of healing at 12 weeks.

Revascularization When associated with significant ischemia, diabetic foot ulcers require arterial revascularization to achieve wound healing. Typically distal by-passes have been shown to be successful in only those centres where there is considerable experience and expertise in this procedure and therefore this limits widespread use. More recently below-knee stent-supported angioplasty for critical limb ischaemia has been shown to be highly effective with an improvement in ankle brachial indices comparable to tibial bypass. This was associated with healing of amputation sites and ulceration in 96% of patients who underwent successful intervention.

Hyperbaric oxygen Recently 18 diabetic patients with ischaemic, non-healing lower-extremity ulcers were assigned to receive 100% oxygen for 90 minutes each day (treatment group) or air (control group) for a total of 30 treatments. Healing was achieved in five out of eight ulcers in the treatment group compared to one out of eight ulcers in the control group.

107

108

SECTION II • DIABETIC NEUROPATHIES

Recombinant platelet derived growth factor The healing of wounds is a complex procedure involving multiple growth factors. One such factor, platelet derived growth factor (PDGF) is active in all stages of the healing process promoting the chemotactic recruitment and proliferation of cells involved in wound repair, and has been tested in several Phase II/III studies (Table 12.2). Becaplermin (0.01% Regranex gel) is the only growth factor licensed for use in wound healing.

CHARCOT NEUROARTHROPATHY Charcot neuroarthropathy (CN) is a rare and disabling condition affecting the bones and joints of the foot of patients with both somatic and autonomic neuropathy but in whom the peripheral circulation is intact. Patients will present with a warm swollen foot often without history of trauma. Radiologically, bony fragmentation, fracture, and dislocation occur (Fig. 12.2). This progresses to a foot deformity with the development of a bony prominence and eventual ulceration which is difficult to manage as standard off-loading is made difficult due to the deformity (Fig. 12.3). A study of randomly selected neuropathic patients reported radiological evidence of CN in 16% of patients, suggesting a key role of neuropathy in the pathogenesis of this condition. Other factors which may predispose to CN include an alteration in bone architecture and strength. In a study of 16 patients with acute CN a significant reduction in calcaneal stiffness was observed in the Charcot and non-Charcot foot and a significantly lower foot, femoral neck and lumbar bone mineral density (BMD) in comparison with a control group. However, it is important to remember that osteoporosis is more prevalent in

Analysis of % complete wound closure at 20 weeks in PDGF studies Placebo 30 μg/g 100 μg/g Study 1 (II-118) Steed et al. J Vasc Surg 1995; 21: 71–78.

25

48

Study 2 (III-382) Wieman et al. Diabetes Care 1998; 21: 822–827.

35

36

Study 3 (III-172) D’Hemercourt et al. Wounds 1998; 10: 69–75.

36

44

Study 4 (III-250) Smiell et al. Wound Rep Regen 1999; 7: 335–346.

32

36

Study 5 (IIIb-134) Embil et al. Wound Rep Regen 2000; 8: 162–168.

50

57

Table 12.2 Analysis of % complete wound closure at 20 weeks in PDGF studies.

CHAPTER 12 • FOOT ULCERATION AND CHARCOT ARTHROPATHY

Fig. 12.2 X-ray showing a mid foot Charcot in a diabetic patient.

Fig. 12.3 Foot of diabetic patient with evidence of neuropathy.

109

110

SECTION II • DIABETIC NEUROPATHIES

type 1 diabetes but not in type 2 diabetes. Thus calcaneal bone density does not differ in men and is higher in women with type 2 diabetes. And in the Health, Aging, and Body Composition Study, elderly (70–79yrs) type 2 diabetic patients had a total hip BMD which was 4–5% higher.

TREATMENT Bisphosphonates In a double blind randomized controlled trial study the effect of one infusion of 90mg pamidronate, a bisphosphonate, was assessed in the management of acute diabetic Charcot neuroarthropathy in 39 diabetic patients over 12 months. There was a fall in temperature of the affected foot in both groups after two weeks with a further non-significant reduction in temperature in the active group at four weeks. However, an improvement in symptoms and a reduction in bone turnover (urinary deoxypyridinoline, bone-specific alkaline phosphatase) were observed at four weeks.

Combined magnetic field therapy In a study of 31 patients with acute Charcot, 20 received magnetic field therapy (MFT) which resulted in a statistically significant reduction in time to consolidation (23.8 weeks in controls vs. 11 weeks in the MFT group). Additionally, less destruction of the bony architecture with a decrease in the amount of residual deformity was noted in the study group.

CURRENT ISSUES • Neuropathy, trauma, and foot deformity are the commonest pathways to foot ulceration. • Foot ulceration classification systems are important to risk stratify patients and implement appropriate intervention to expedite healing. • Few data support the belief that education and preventative foot care reduce the risk of ulceration in high risk individuals. • Change in foot ulcer area after four weeks of observation appears to be a robust predictor of healing at 12 weeks. • Charcot neuroarthropathy is a rare but disabling condition affecting the bones and joints of the foot in patients with neuropathy but intact circulation. • Bisphosphonates have been advocated in the treatment of Charcot joints, but the evidence of benefit is limited.

CHAPTER 12 • FOOT ULCERATION AND CHARCOT ARTHROPATHY

FURTHER READING Abidia A, Laden G, Kuhan G, Johnson BF, Wilkinson AR, Renwick PM, Masson EA, McCollum PT. The role of hyperbaric oxygen therapy in ischaemic diabetic lower extremity ulcers: a double-blind randomised-controlled trial. Eur J Vasc Endovasc Surg 2003; 25: 513–518. Boulton AJ. The diabetic foot: from art to science. The 18th Camillo Golgi lecture. Diabetologia 2004; 47: 1343–1353. Feiring AJ, Wesolowski AA, Lade S. Primary stent-supported angioplasty for treatment of below-knee critical limb ischemia and severe claudication: early and one-year outcomes. J Am Coll Cardiol 2004; 44: 2307–2314. Hanft JR, Goggin JP, Landsman A, Surprenant M. The role of combined magnetic field bone growth stimulation as an adjunct in the treatment of neuroarthropathy/Charcot joint: an expanded pilot study. J Foot Ankle Surg 1998; 37: 510–515. Jirkovska A, Kasalicky P, Boucek P, Hosova J, Skibova J. Calcaneal ultrasonometry in patients with Charcot osteoarthropathy and its relationship with densitometry in the lumbar spine and femoral neck and with markers of bone turnover. Diabet Med 2001; 18: 495–500. Jude EB, Selby PL, Burgess J, Lilleystone P, Mawer EB, Page SR, Donohoe M, Foster AV, Edmonds ME, Boulton AJ. Bisphosphonates in the treatment of Charcot neuroarthropathy: a double-blind randomised controlled trial. Diabetologia 2001; 44: 2032–2037. Koitka A, Abraham P, Bouhanick B, Sigaudo-Roussel D, Demiot C, Saumet JL. Impaired pressure-induced vasodilation at the foot in young adults with type 1 diabetes. Diabetes 2004; 53: 721–725. Lavery LA, Higgins KR, Lanctot DR, Constantinides GP, Zamorano RG, Armstrong DG, Athanasiou KA, Agrawal CM. Home monitoring of foot skin temperatures to prevent ulceration. Diabetes Care 2004; 27: 2642–2647. Nagai MK, Embil JM. Becaplermin: recombinant platelet derived growth factor, a new treatment for healing diabetic foot ulcers. Expert Opin Biol Ther 2002; 2: 211–218. Piaggesi A, Viacava P, Rizzo L, Naccarato G, Baccetti F, Romanelli M, Zampa V, Del Prato S. Semiquantitative analysis of the histopathological features of the neuropathic foot ulcer: effects of pressure relief. Diabetes Care 2003; 26: 3123–3128. Schaper NC. Diabetic foot ulcer classification system for research purposes: a progress report on criteria for including patients in research studies. Diabetes Metab Res Rev 2004; 20: S90–95. Sheehan P, Jones P, Caselli A, Giurini JM, Veves A. Percent change in wound area of diabetic foot ulcers over a 4-week period is a robust predictor of complete healing in a 12week prospective trial. Diabetes Care 2003; 26: 1879–1882. Treece KA, Macfarlane RM, Pound N, Game FL, Jeffcoate WJ. Validation of a system of foot ulcer classification in diabetes mellitus. Diabet Med 2004; 21: 987–991. Valk GD, Kriegsman DM, Assendelft WJ. Patient education for preventing diabetic foot ulceration: a systematic review. Endocrinol Metab Clin N Amer 2003; 31:633–658. Zimny S, Schatz H, Pfohl M. The effects of ulcer size on the wound radius reductions and healing times in neuropathic diabetic foot ulcers. Exp Clin Endocrinol Diabetes 2004; 112: 191–194.

111

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

CHAPTER 13 TREATMENTS OPTIONS

Rayaz A. Malik MB.ChB, PhD, MRCP

Treatments that may prevent the onset or modify the natural history of diabetic peripheral neuropathy (DPN) by targeting known pathogenetic mechanisms will be discussed first. These treatments are mostly experimental or in phase II or early phase III clinical trials.

Near normoglycemia The most reliable method of achieving and maintaining near normal glycaemia is by pancreatic or islet cell transplantation. In the Minneapolis series only modest improvements in measures of neuropathy were seen after several years of normoglycaemia. However, the pancreatic transplant was often in combination with a renal transplant and therefore the recipients generally had a long duration of diabetes and severe neuropathy at baseline. The Diabetic Control and Complications Trial (DCCT) demonstrated a significant benefit in both electrophysiology and the degree of clinical neuropathy. Three much smaller but long-term interventional studies have confirmed that maintaining near-normoglycaemia prevents the development and retards the progression of DPN. The Stockholm Diabetes Intervention Study, over 7.5 and 10 years, the Oslo Study over eight years in Type 1 diabetes and the Kumamato Study over six years in type 2 diabetes. Thus achieving near-normoglycaemia should be the aim both in prevention and in the first step of management of DPN.

Aldose reductase inhibitors (ARIs) Polyol pathway (Fig. 13.1) hyperactivity has been postulated as a mechanism which links hyperglycaemia to neuropathy.

Aldose reductase Glucose

NADPH

Fig. 13.1 Polylol pathway.

Sorbitol

NADP+

Sorbitol dehydrogenase

NAD+

Fructose

NADH

113

114

SECTION II • DIABETIC NEUROPATHIES

However, despite 25 years of clinical trials with ARIs in diabetic neuropathy, only epalrestat is currently available in Japan. Most of the early trials can be summarized as: • Too small. Drug effect inadequate to inhibit nerve sorbitol accumulation; • Too few. Inadequate numbers of subjects randomized; • Too short. Many trials only lasted weeks or months for a chronic disease of many years duration; • Too late. Targeting a pathogenetic mechanism is not likely to be effective when the complication is well established; • Too toxic. A number of ARIs were withdrawn in phase III due to hepatic or renal toxicity.

Antioxidants Increased free radical production and a reduced ability to neutralize free radicals due to nicotinamide adenine dinucleotide (NADH) depletion, partly the result of non-enzymatic glycation and polyol pathway hyperactivity supports the role of oxidative stress in the pathogenesis of neuropathy. Studies with the antioxidant alpha-lipoic acid (LA) have provided evidence of potential efficacy with this agent for both symptoms and signs. Two large North American/European clinical trials of the efficacy of LA are in progress and should report in 2005.

Gamma-linolenic acid (GLA) GLA can prevent abnormalities in essential fatty acid and prostanoid metabolism. GLA treatment for one year in a randomized trial resulted in improvement in electrophysiology and deficits. A LA/GLA conjugate has produced impressive improvements in electrophysiological and neurochemical abnormalities in experimental diabetic neuropathy but this has not been tested in diabetic patients.

Inhibitors of glycation Non-enzymatic glycation occurs in diabetes and may be an important initiating factor for nerve demyelination and may also interfere with axonal transport. Advanced glycation end products (AGE) can also absorb nitric oxide (NO) and impair nerve blood flow. Studies of aminoguanidine, an inhibitor of advanced glycation have shown no benefit in diabetic nephropathy and few data are available on this or other inhibitors of AGE formation in human diabetic neuropathy.

Protein Kinase C-β (PKC-β) inhibition Intracellular hyperglycaemia increases diacylglycerol levels and activates PKC-β formation, which alters expression of endothelial nitric oxide synthetase and vascular endothelial growth factor (VEGF). These changes are

CHAPTER 13 • TREATMENTS OPTIONS

associated with abnormalities in vascular function and PKC-β inhibitors have been shown to correct endothelial dependent relaxation, nerve perfusion deficits and conduction velocity in animal models. Preliminary phase II data suggest that treatment with a PKC-β inhibitor might ameliorate symptoms and neurological deficits in DPN. Phase III multicentre clinical trials are currently in progress and will report in 2006.

Vasodilators Treatment with angiotensin-converting enzyme inhibitors (ACE-Is) has been shown to improve electrophysiological measures of nerve function in mild neuropathy. The short-acting vasodilator isosorbide dinitrate has been shown to improve painful symptoms, but its effect on deficits and electrophysiology are unknown.

Neutrophins Neurotrophic factors are important in the growth and development of neurones. These include insulin-like growth factor I and II and the neurotrophin family which includes nerve growth factor (NGF) and brain derived neurotrophic Factor (BDNF). NGF is reduced in experimental diabetes and treatment with NGF corrects some aspects of sensory neuropathy related to small fibre dysfunction in diabetic rats. A promising phase II study showed an improvement in human diabetic neuropathy. However this was not confirmed in a phase III study and therefore the clinical development of NGF was halted. Similarly, a small study with BDNF has also shown no benefit.

SYMPTOMATIC MANAGEMENT Control of hyperglycaemia A number of small open label uncontrolled studies have suggested that achieving stable near-normoglycaemic control is helpful in the management of painful neuropathic symptoms. In an early study of continuous subcutaneous insulin infusion (CSII) for four months in patients with painful neuropathy, relief of neuropathic symptoms was associated with an overall improvement of glycaemic control and a reduction in blood glucose flux. More recently patients with painful neuropathy were compared to those with painless neuropathy. Those with painful symptoms had poorer control, more excursions to hyper- and hypoglycaemic levels and greater blood glucose flux. Thus, it may be that the stability of glycaemic control is equally important to the level of overall normoglycaemia.

Pharmacotherapy A large number of therapeutic agents have been used in the management of painful symptoms. There is little evidence to support the use of non-steroidal

115

116

SECTION II • DIABETIC NEUROPATHIES

anti-inflammatory drugs in symptomatic neuropathy. Moreover, these agents should be used with caution in neuropathic diabetic patients, many of whom may have renal impairment, a contraindication to non-steroidal drugs.

Tricyclic drugs Several small randomized clinical trials have supported the use of these agents in the management of neuropathic pain. Putative mechanisms by which these drugs relieve pain include inhibition of norepinephrine and/or serotonin reuptake at synapses of central descending pain pathways. More recently, the antagonism of n-methyl-d-aspartate receptors has been shown to alleviate hyperalgaesia and allodynia. Most experience has been achieved with amitriptyline and imipramine. The dosage of either one of these two drugs required for symptomatic relief is similar, being 25–150 mg daily. In order to avoid undue drowsiness, the dose can be taken once a day in the evening. Desipramine is also a useful drug that may be better tolerated than amitriptyline. The usefulness of these agents was confirmed in a systematic review performed by McQuay and colleagues. The major problem remains the frequency of side-effects which are predictable. Although drowsiness and lethargy are common, it is the anticholinergic sideeffects, particularly dry mouth, that are the most troublesome.

Selective serotonin-reuptake inhibitors (SSRIs) These agents inhibit pre-synaptic reuptake of serotonin but not norepinephrine. Studies suggest that treatment with paroxetine but not fluoxetine is associated with significant pain relief. Similarly citalopram 40 mg/day is efficacious but less effective than imipramine. Their main advantage is lesser side-effects.

Anticonvulsants Anticonvulsants have been used in the management of neuropathic pain for many years. In total there are 13 published trials of anticonvulsants of which only six are randomized controlled trials which show that, for those diabetic patients with painful neuropathy given anticonvulsants, the true risk of a reduction in pain relief is 2.3 times greater than with placebo (Fig 13.2). Only limited evidence exists for the efficacy of phenytoin and carbamazepine for diabetic neuropathy. Agranulocytosis has been reported with carbamazepine and must be monitored by obtaining blood counts at baseline and thereafter periodically to screen for bone marrow suppression and leucopenia. Oxcarbazepine is a second generation antiepileptic drug and like carbamazepine, it blocks sodium channels, reducing hyperexcitability of peripheral nerves. It has been effectively used in trigeminal neuralgia. There are currently five randomized-controlled trials underway in Japan evaluating its efficacy in diabetic neuropathy. Gabapentin is now widely used for neuropathic

CHAPTER 13 • TREATMENTS OPTIONS

Rull 1969 Carbamazepine

1.87 (1.35, 2.84)

Wilton 1974 Carbamazepine

2.67 (1.48, 5.08)

Chadda 1978 Phenytoin

2.80 (1.66, 5.05)

Backonja 1998 Gabapentin

1.84 (1.30, 2.67)

Kochar 2002 Valproate

3.43 (1.66, 7.93)

Rosenstock 2004 Pregabalin

2.76 (1.50, 5.25)

combined (fixed)

2.31 (1.87, 2.85)

1

2

5

10

Fig. 13.2 Meta-analysis of six placebo controlled trials of anticonvulsants for pain in painful diabetic neuropathy. The dotted vertical line shows the combined relative risk estimate with the 95% confidence interval (CI). Here we can say, with 95% confidence, that for those given anticonvulsants the true population risk of a reduction in pain relief is 2.3 times greater than with placebo.

symptoms based on the results of a large controlled trial in symptomatic neuropathy where significant pain relief together with reduced sleep disturbance was reported using dosages of 900–3,600 mg daily. In a recent review of all the trials of gabapentin for neuropathic pain, it was concluded that dosages of 1,800–3,600 mg per day of this agent were effective. The side-effect profile also seems superior to that of the tricyclic drugs, though somnolesence is a problem with higher doses. Pregabelin has been shown to be more efficacious than gabapentin with even lesser side effects. Lamotrigine has at least two antinociceptive properties and in a randomized placebo controlled study was shown to have efficacy in patients with neuropathic pain.

Anti-arrhythmics Mexilitine is a class 1B antiarrhythmic agent and a structural analogue of lignocaine. Its efficacy in neuropathic pain has been confirmed in controlled trials using doses of up to 450 mg daily which are lower than those usually used for the treatment of cardiac arrhythmias. However, regular ECG monitoring is necessary and the long-term use of mexilitine cannot be recommended.

117

118

SECTION II • DIABETIC NEUROPATHIES

Other agents Tramadol is an opioid-like centrally acting, synthetic non-narcotic analgesic which was shown to produce significant symptomatic relief in a randomized controlled trial of patients with painful diabetic neuropathy initially over six weeks which was maintained for at least six months. However, side-effects are relatively common and are similar to other opioid-like drugs. Similarly, two randomized trials have confirmed the efficacy of controlled release oxycodone for neuropathic pain in diabetes.

TOPICAL AND PHYSICAL TREATMENT Topical nitrate A recent placebo controlled study suggested that the local application to the feet of isosorbide dinitrate spray was effective in relieving overall pain and burning discomfort in painful diabetic neuropathy and has been confirmed by applying GTN patches to both lower limbs.

Capsaicin This alkaloid depletes tissue of substance P and results in degeneration of epidermal nerve fibres. Therefore it is not surprising that several controlled studies combined in a meta-analysis seem to provide some evidence of efficacy in diabetic neuropathic pain. However, true blinding of these studies has been questioned because of the local hyperalgesia experienced when applying the active drug. Furthermore, with the overt epidermal nerve fibre damage it causes one must question the ethical basis of promoting denervation in a foot which already has or will have sensory loss.

Acupuncture A number of unmasked studies support the use of acupuncture and, in the most recent published report, benefits of acupuncture lasted for up to six months and reduced the use of other analgesics.

OTHER PHYSICAL THERAPIES A number of other physical therapies have been proposed, though few have been rigorously tested in controlled clinical trials. Efficacy has been shown with percutaneous nerve stimulation and static magnetic field therapy, with no benefit with electrical socks.

CHAPTER 13 • TREATMENTS OPTIONS

Electrical spinal cord stimulation A case series of patients with severe painful neuropathy unresponsive to conventional therapy suggested efficacy of using an implanted spinal cord stimulator. This approach can only be recommended in very resistant cases as it is invasive, expensive, and unproven in controlled studies.

CURRENT ISSUES • • •

• • • •

Maintaining near-normoglycaemia prevents the development and retards the progression of DPN. Despite 25 years of clinical trials with ARIs in diabetic neuropathy, only epalrestat is currently available in Japan. Increased free radical production and a reduced ability to neutralize free radicals due to NADH leads to oxidative stress which is involved in the pathogenesis of neuropathy. Preliminary phase II data suggest that treatment with a PKC-β inhibitor might ameliorate symptoms and neurological deficits in DPN. Treatment with ACE-Is has been shown to improve nerve function in mild neuropathy. A large number of therapeutic agents have been used in the management of painful symptoms. Of 13 published trials of anticonvulsants only six are randomized controlled trials and show a risk reduction in pain relief 2.3 times greater than placebo.

FURTHER READING Backonja M, Glanzman RL. Gabapentin dosing for neuropathic pain: evidence from randomized placebo controlled clinical trials. Clin Ther 2003; 25: 81–104. Boulton, AJM. Treatment of symptomatic diabetic neuropathy. Diabet Metab Res Rev 2003; 19: S16–S21. McQuay H, Carroll D, Jadad AR, et al. Anticonvulsant drugs for the management of pain: a systematic review. BMJ 1995; 311: 1047–1052.

119

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Rayaz A. Malik MB.ChB, PhD, MRCP CHAPTER 14 MANAGEMENT GUIDELINES FOR DIABETIC PERIPHERAL NEUROPATHY AND FOOT ULCERATION

DIABETIC NEUROPATHY In 1995 the Neurodiab, a subgroup of the European Association for the Study of Diabetes, suggested that there was a need for guidelines for the outpatient management of patients with diabetic neuropathy. Such guidelines were developed from an international consensus meeting attended by diabetologists, neurologists, primary care physicians, podiatrists and diabetes specialist nurses. In brief, these guidelines agreed on a practical means of assessment (Table 14.1), classification (Table 14.2) and management (Table 14.3) of diabetic patients presenting in an outpatient setting. More recently in 2004 and 2003 respectively, two detailed technical reviews have been published on somatic diabetic peripheral neuropathy (DPN)and autonomic (AN) diabetic neuropathy. On behalf of the American Diabetes Association a consensus report has emerged from these two lengthy reviews. It is agreed that the diabetic neuropathies are heterogeneous, affect different parts of the nervous system and may present with diverse clinical manifestations. A simple clinical classification is proposed (Table 14.4). DPN is considered to be a diagnosis of exclusion. The early recognition and appropriate management of neuropathy is thought to be important for a number of reasons: 1) non-diabetic neuropathies may be present in patients with diabetes; 2) a number of treatment options exist for symptomatic diabetic neuropathy; 3) up to 50% of DPN may be asymptomatic and patients are at risk of insensate injury to their feet; 4) AN may involve every system in the body; 5) AN causes substantial morbidity and increased mortality particularly if cardiovascular autonomic neuropathy is present. The diagnostic methods employed in diabetic neuropathy are varied and show considerable heterogeneity for both specificity and sensitivity. The American Academy of Neurology recently proposed a consensus statement on the use of quantitative sensory testing (QST) in the diagnosis of diabetic neuropathy. This assessment evaluated the clinical utility, efficacy, and safety of QST. Using a search strategy on MEDLINE, Current Contents, and their personal files, the authors identified 350 articles. No adequately powered class I studies and only a small number of class II and III studies demonstrated that QST probably identified small or large fibre sensory abnormalities in patients with diabetic neuropathy. They concluded that QST was a potentially useful tool for measuring sensory impairment for clinical and research studies but should not be the sole criteria used to diagnose pathology.

121

122

SECTION II • DIABETIC NEUROPATHIES

Stages of diabetic peripheral neuropathy Stage of neuropathy

Characteristics

No neuropathy

No symptoms or signs

Clinical neuropathy Chronic pain

Acute pain

Painless with complete/partial sensory loss Late complications

Burning, shooting, stabbing pains ± pins and needles; increased at night; absent sensation to several modalities; reduced/absent reflexes Severe symptoms as above (hyperaesthesia common); may follow initiation of insulin in poorly controlled diabetes; signs minor or absent Numbness/deadness of feet or no symptoms; painless injury; reduced/absent sensation; reduced thermal sensitivity; absent reflexes Foot lesions; neuropathic deformity; non-traumatic amputation

1. Types of diabetic neuropathy: frequent, sensorimotor symmetrical neuropathy (mostly chronic, sensory loss or pain), autonomic neuropathy (history of impotence and possibly other autonomic abnormalities); rare, mononeuropathy (motor involvement, acute onset, may be painful), diabetic amyotrophy (weakness/wasting usually of proximal lower limb muscles). 2. Staging does not imply automatic progression to the next stage. The aim is to prevent, or at least delay, progression to the next stage.

Table 14.1 Stage of diabetic peripheral neuropathy.

Neurological tests Pin prick test

Use a disposable instrument, e.g. a disposable dressmaker’s pin. Do not use a hypodermic needle. Ask ‘Is it painful?’ not ‘Can you feel it?’

Light touch

Use a consistent method, ideally a cotton wisp

Vibration test

Use a 128 Hz tuning fork, initially on the big toe

Ankle reflex

Compare the ankle reflex with the knee reflex

Pressure perception

Absence of sensation in the foot to a 10 g monofilament may be used to assess the risk of foot ulceration

Table 14.2 Neurological tests.

Treatment should be directed at underlying pathogenetic mechanisms, although the benefits of this approach are limited at present (Table 14.5). Effective symptomatic treatments are available for the manifestations of DPN but treatments should take into account number needed to treat (NNT) and needed to harm (NNH) (Table 14.6).

CHAPTER 14 • MANAGEMENT GUIDELINES FOR DPN AND FOOT ULCERATION

Management of the stages of neuropathy Stage

Key elements

Referral

No clinical neuropathy (Stage 0/1)

Education; glycaemic controla Annual assessment

Chiropodist/podiatrist/ diabetes specialist nurse

If disabled, treatment with tricyclic drugs; glycaemic control Simple analgesics/ tricyclic drugs/NSAIDs/ opiates; glycaemic control Education, especially footcare; glycaemic control Early referral Emergency referral if lesions present; otherwise referral within 4 weeks

Diabetologist/neurologist

Clinical neuropathy (Stage 2) Chronic painful

Acute painful

Painless/loss of sensation Diabetic amyotrophy Late complications (Stage 3)

Diabetologist/neurologist

Appropriate member of footcare team according to needs Neurologist/diabetologist Diabetologist/neurologist/ chiropodist/podiatrist/ diabetes specialist nurse

aIf,

in future, specific therapies become available for the treatment of early neuropathy, the guidelines will require amendment to provide advice on the separate diagnosis and management of people with no neuropathy (Stage 0) and those with subclinical disease (Stage 1). Guidelines for the Diagnosis and Outpatient Management of Diabetic Peripheral Neuropathy, Diabetic Medicine 1998; 15: 508–514.

Table 14.3 Management of the stages of neuropathy.

Classification of diabetic neuropathy Generalized symmetric polyneuropathies: • acute sensory • chronic sensorimotor • autonomic Focal and multifocal neuropathies: • cranial • truncal • focal limb • proximal motor (amyotrophy) • (co-existing chronic inflammatory demyelinating polyneuropathy [CIDP] ) Note: Clinicians should be alert for treatable neuropathies occurring in diabetic patients including CIDP, monoclonal gammopathy, vitamin B12 deficiency etc.

Table 14.4 Classification of diabetic neuropathy.

123

124

SECTION II • DIABETIC NEUROPATHIES

Treatment of diabetic neuropathy based on the putative pathogenic mechanisms Abnormality

Compound

Aim of treatment

Polyol pathway? Aldose reductase Nerve sorbitol ↓ inhibitors Sorbinil Tolrestat Ponalrestat Zopolrestat Zenarestat Lidorestat Fidarestat AS-3201

myo-Inositol ↓ Oxidative stress Nerve hypoxia

Epalrestat Myo-inositol α-Lipoic acid

Vasodilators ACE inhibitors Prostaglandin analogs phVEGF165 gene transfer Protein kinase C PKC-β inhibitor (ruboxistaurin) C-peptide ↓ C-peptide Neurotrophism ↓ Nerve growth factor (NGF) BDNF LCFA metabolism ↓ GLA synthesis ↓ NEG

Nerve myo-inositol Oxygen free radicals ↓ NBF ↓

Status of RCTs

Withdrawn (AE) Withdrawn (AE) Ineffective Withdrawn (marginal effects) Withdrawn (AE) Withdrawn (AE) Effective in RCTs, trials ongoing Effective in RCTs, trials ongoing Marketed in Japan Equivocal Effective in RCTs, trials ongoing Effective in 1 RCT Effective in 1 RCT

Angiogenesis?

RCTs ongoing

NBF ↓

RCTs ongoing

NBF Nerve regeneration, growth Nerve regeneration, growth Acetyl-L-carnitine LCFA accumulation ↓ γ-Linolenic acid EFA metabolism (GLA) Aminoguanidine AGE accumulation ↓

Studies ongoing Ineffective Ineffective Ineffective Withdrawn Withdrawn

BDNF: brain-derived neurotrophic factor; NEG: non-enzymatic glycation; AGE: advanced glycation end products; EFA: essential fatty acids; LCFA: long-chain fatty acids; AE: adverse events; NBF: nerve blood flow; RCTs: randomized controlled trials.

Table 14.5 Treatment of diabetic neuropathy based on the putative pathogenetic mechanisms. Copyright © 2005 American Diabetes Association. From Diabetes Care 2005; 28: 956–962. Reprinted with permission from The American Diabetes Association.

CHAPTER 14 • MANAGEMENT GUIDELINES FOR DPN AND FOOT ULCERATION

Oral symptomatic therapy of painful neuropathy Drug class

Drug

Daily dose (mg)

NNT

NNH

Side effects

Tricyclics

Amitriptyline Imipramine

25–150 25–150

2.4 (2.0–3.0)

2.7 (2.1–3.9)

++++ ++++

SSRIs

Paroxetine Citalopram

40 40

ND ND

ND ND

+++ +++

Anticonvulsants

Gabapentin Pregabalin Carbamazepine

900–1800 150–600 200–400

3.7 (24–8.3)

2.7 (2.2–3.4)

3.3 (2.0–9.4)

1.9 (1.4–2.8)

++ ++ +++

Opioids

Tramadol Oxycodone CR*

50–400 10–60

3.4 (2.3–6.4)

7.8

+++ ++++

NNT = Numbers needed to treat to achieve pain relief in 1 patient NNH = Numbers needed to treat to harm 1 patient CI = Confidence Interval ND = Not determined

Table 14.6 Oral symptomatic therapy of painful neuropathy. Copyright © 2005 American Diabetes Association. From Diabetes Care 2005; 28: 956–962. Reprinted with permission from The American Diabetes Association.

FOOT ULCERATION It is agreed that diabetic foot ulcers pose a great burden on both the patient and the health care systems of many countries. A multifactorial approach with the input of many different specialists is advocated in diagnosing and treating diabetic patients with foot ulceration. An International Consensus on the Diabetic Foot, resulting in a worldwide network of professionals involved in the management of diabetic patients with foot problems, has emerged. Prophylactic foot care has been shown to decrease patient morbidity, utilization of expensive resources, and risk for amputation and premature death. In 2005 the Diabetes Committee of the American Orthopaedic Foot and Ankle Society has recently developed guidelines for the implementation of such prophylactic foot care. Screening and treatment algorithms are based on an evaluation of neuropathy, skin integrity, ulcer history, deformity and vascular insufficiency. Treatment includes paring of calluses, debridement of infected or nonviable tissue, dressings, and off-loading with patient education and provision of appropriate orthoses/footwear. Timely specialty assistance and cross referral to a vascular surgeon, orthopaedic surgeon, podiatrist, diabetologist, infectious disease consultant and radiologist is encouraged.

125

126

SECTION II • DIABETIC NEUROPATHIES

CURRENT ISSUES •

•

•

• •

One of the first sets of international guidelines for the outpatient management of diabetic neuropathy was published from an international consensus meeting in 1997. Recently two detailed technical reviews on somatic and autonomic diabetic neuropathy have been synthesized into a consensus report on behalf of the American Diabetes Association. The American Academy of Neurology recently proposed that QST was a potentially useful tool for measuring sensory impairment for clinical and research studies but should not be the sole criteria used to diagnose pathology. Treatments aimed at underlying pathogenetic mechanisms of DPN are limited but moderately effective symptomatic treatments are available. A multifactorial approach with the input of many different specialists is advocated in diagnosing and treating diabetic patients with foot ulceration.

FURTHER READING Boulton AJ, Gries FA, Jervell JA. Guidelines for the diagnosis and outpatient management of diabetic peripheral neuropathy. Diabet Med 1998;15: 508–14. Boulton AJ, Malik RA, Arezzo JC, Sosenko JM. Diabetic somatic neuropathies. Diabetes Care 2004; 27: 1458–1486. Boulton AJM, Vinik AI, Arezzo JC, Bril V, Feldman EL, Freeman R, Malik RA, Maser RE, Sosenko JM, Ziegler D. Position statement: diabetic neuropathies Diabetes Care 2005 (In press). Schaper NC, Apelqvist J, Bakker K. The international consensus and practical guidelines on the management and prevention of the diabetic foot. Curr Diab Rep 2003; 3: 475–479. Shy ME, Frohman EM, So YT, Arezzo JC, Cornblath DR, Giuliani MJ, Kincaid JC, Ochoa JL, Parry GJ, Weimer LH; Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Quantitative sensory testing: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology 2003; 60: 898–904. Pinzur MS, Slovenkai MP, Trepman E, Shields NN. Guidelines for diabetic foot care: recommendations endorsed by the Diabetes Committee of the American Orthopaedic Foot and Ankle Society. Foot Ankle Int 2005; 26: 113–119. Vinik AI, Maser RE, Mitchell BD, Freeman R. Diabetic autonomic neuropathy. Diabetes Care 2003; 26: 1553–1579.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

SECTION III DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Hean-Choon Chen FRCS, FRCOphth CHAPTER 15 DIABETIC RETINOPATHY: EPIDEMIOLOGY AND RISK FACTORS

INTRODUCTION The annual rate of blind registration due to diabetic retinopathy appears to be falling in some European countries, but diabetes remains the leading cause of blindness in the working age group, accounting for approximately 12% of all cases of registrable blindness among those <65 years old. The incidence of visual impairment during a 4-year observation period has been best documented by the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR), a large population-based study conducted in 10 counties in southern Wisconsin, USA (Table 15.1). Approximately 10% of all diabetic patients in primary care were examined and followed-up. The subjects were divided into two groups: those with an age of onset of diabetes below 30 years and those with an age of onset above 30 years. This latter group was further subdivided into those treated with insulin and those not taking insulin. Subjects were re-examined at several time points (four and ten years) to document the incidence and prevalence of diabetic retinopathy and the associated risk factors. In the WESDR, visual impairment was subdivided into four categories ranging from none to blind and data at base-line was compared with that four years later (Table 15.1). For example, in the 832 younger onset (i.e. <30 years) subjects with no visual impairment at baseline, 2.8%, 1.4% and 0.5% had progressed to mild impairment, moderate impairment and blindness by 4 years (Table 15.1).

BLINDNESS RATES IN DIABETIC PATIENTS The annual incidence of blindness from diabetic retinopathy varies between 0.02 and 1%, and the prevalence of blindness in diabetic patients is approximately 2%. Diabetic retinopathy accounts for between 8 and 12% of all registrable blindness (defined as a Snellen visual acuity of 6/60 or worse in the better eye).

PREVALENCE OF DIABETIC RETINOPATHY Summary of findings from the WESDR: Onset of diabetes before age 30 years (Fig. 15.1) • The prevalence of any retinopathy is 2% in those within two years of diagnosis of diabetes and 98% in those with disease duration >15 years. • The prevalence of proliferative retinopathy is 0% in those within five years of diagnosis of diabetes, rising to 67% in those with disease duration >35 years.

129

130

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Onset of diabetes after age 30 years (Fig. 15.2) • The prevalence of any retinopathy is 29% in those within five years of diagnosis of diabetes and 78% in those with disease duration >15 years. • The prevalence of proliferative retinopathy is 2% in those within five years of onset of diabetes, rising to 16% in those with disease duration >15 years. The higher prevalence of retinopathy in those diagnosed with diabetes over the age of 30 is partly due to the difficulty in determining the precise date of onset of disease. The prevalence of retinopathy at the time of diagnosis of type 2 diabetes has been reported to be as high as 38%.

Four-year incidence of visual impairment in participants surviving and completing the follow-up examination Visual impairment at baseline

Number of participants

None (%)

Mild (%)

Moderate (%) Blind (%)

832 26 10 20

95.3 26.9 10.0 0

2.8 42.3 10.0 0

1.4 15.4 30.0 0

0.5 15.4 50.0 100.0

Older-onset, taking insulin † None 423 Mild 27 Moderate 15 Blind 8

83.9 29.6 6.7 0

10.6 22.6 13.3 0

4.3 40.7 26.7 0

1.2 7.4 53.3 100.0

Older onset, not taking insulin § None 454 Mild 29 Moderate 7 Blind 4

91.0 20.7 0 0

5.5 31.0 0 0

2.9 31.0 28.6 0

0.7 17.2 71.4 100.0

Younger onset* None Mild Moderate Blind

* In three persons the baseline and/or follow-up impairment level could not be determined. † In 12 persons, the baseline and/or follow-up impairment level could not be determined.

Table 15.1 Four-year incidence of visual impairment in participants surviving and completing the follow-up examination. Derived from the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR). The levels of visual acuity were classified as: no impairment (better than 20/40), mild impairment (20/40–20/63), moderate impairment (20/80–20/160) and blind (≥20/200). Ophthalmology 1988; 95: 1340–1348.

CHAPTER 15 • EPIDEMIOLOGY AND RISK FACTORS

100 Any retinopathy

Percentage

80 60 40 Proliferative retinopathy 20 0 0

5

10

15

20

25

30

35

40

45

50

Duration (years)

Fig. 15.1 Frequency of any retinopathy and prolifereative retinopathy by duration of diabetes in type 1 diabetes. Derived from the Wisconsin Epidemiologic Study of Diabetic Retinopathy.

100 Receiving insulin Not receiving insulin

Percentage

80

Any retinopathy

60

40 Proliferative retinopathy

20

0 0

5

10

15

20

25

30

Duration (years)

Fig. 15.2 Frequency of any retinopathy and prolifereative retinopathy by duration of diabetes in patients with type 2 diabetes, according to insulin status. Derived from the Wisconsin Epidemiologic Study of Diabetic Retinopathy.

131

132

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

INCIDENCE OF DIABETIC RETINOPATHY Summary of findings from the WESDR: Onset of diabetes before age 30 years • The incidence of any retinopathy developing for the first time over a 4year and 10-year period is 59% and 90%, respectively. • The incidence of proliferative retinopathy over a 4-year and 10-year period is 11% and 30%, respectively. Onset of diabetes after age 30 years • The incidence of any retinopathy developing for the first time over a 4year and 10 year period is 34–47% and 67–80%, respectively. • The incidence of proliferative retinopathy over a 4-year and 10-year period is 2–7% and 10–23%, respectively.

RISK FACTORS FOR DIABETIC RETINOPATHY Hyperglycaemia Chronic hyperglycaemia is the major risk factor in the development of diabetic retinopathy. The US Diabetes Control and Complications Trial (DCCT) evaluated the rates of development of microvascular complications in two groups of type 1 diabetics assigned either to strict or standard glycaemic control over a mean follow-up period of 6.5 years. Those who had no retinopathy at entry into the study and who were assigned to tight control showed a 76% reduction in the mean risk of developing retinopathy. In addition, among those subjects with mild retinopathy at base-line, tight control slowed the progression of retinopathy by 54% and reduced the likelihood of severe or proliferative retinopathy by 47%. The UK Prospective Diabetes Study (UKPDS) similarly examined the effect of tight versus standard glycaemic control in newly diagnosed patients with type 2 diabetes. Over 10 years, HbA1C levels averaged 11% lower in the tight control group, which resulted in a 25% risk reduction in microvascular complications, including the need for panretinal laser photocoagulation. The WESDR examined the 4-year incidence and progression of retinopathy in relation to HbA1C and showed that patients in the highest quartile of HbA1C experienced the greatest incidence and progression of retinopathy (Fig. 15.3). It is estimated that by reducing HbA1C from 11% to 9% the rate of progression to proliferative disease would be halved.

Duration of diabetes Duration of diabetes is a reliable predictor of the presence of retinopathy, but the severity of the retinopathy is primarily influenced by other risk factors, especially glycaemic control.

CHAPTER 15 • EPIDEMIOLOGY AND RISK FACTORS

100

Progression (%)

80 60 40 20 0 0

5

10

15

20

25

30

0

5

10

15

20

25

30

0

5

25

30

100

Progression (%)

80 60 40 20 0 100

Progression (%)

80 60 40 20 0 10 15 20 Duration at baseline (years)

Fig. 15.3 Four-year progression of retinopathy by quartile of HbA1C and duration of diabetes at baseline in persons with (top) younger-onset diabetes; (centre) olderonset diabetes, taking insulin; and (bottom) older-onset diabetes, not taking insulin. Triangles indicate first quartile; diamonds, fourth quartile. Derived from Wisconsin Epidemiologic Study of Diabetic Retinopathy.

133

134

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Blood pressure High blood pressure (BP) is a major independent risk factor in the development of retinopathy in both type 1 and type 2 diabetes. Even among relatively normotensive individuals (BP <140/90 mmHg), those with BP levels above the 90th percentile show a higher prevalence of retinopathy and higher rate of progression. Hypertension is roughly twice as common in diabetic compared with non-diabetic patients. For example, 17–24% of patients with type 1 diabetes and 38% of patients with type 2 diabetes have BP >160/90 mmHg. Because of the link between BP and diabetic renal complications, it has been difficult to establish from epidemiological work that hypertension is an independent risk factor for retinopathy, but several intervention studies have been conclusive. In the UKPDS, hypertensive patients were assigned to either tight or moderate BP control over nine years. The group randomized to tight BP control (mean 144/82 mmHg) experienced a 47% reduction in the risk of losing three lines of vision compared to the group with standard BP control (mean 154/87 mmHg). There was also a 34% reduction in the risk of progression of retinopathy status. The EURODIAB Controlled Trial of Lisinopril in Insulin-Dependent Diabetes Mellitus (EUCLID), also comparing tight with moderate blood pressure control, produced similar results, i.e. a significantly reduced risk of blindness and reduced rate of progression of retinopathy in those patients randomized to more intensively controlled BP. Use of an angiotensin-converting enzyme inhibitor (ACE-I) may be particularly effective in reducing the progression of retinopathy, independently of BP reduction, especially in normotensive type 1 diabetics.

Age Diabetic retinopathy is rare under 10 years of age; only one case was seen in the first decade of life in the WESDR. Puberty, with its attendant hormonal changes, brings about accelerated changes in retinopathy status; the highest 4-year incidence in the WESDR was among patients aged 10–12 years at the base-line examination. In those with onset of diabetes below 21 years, females developed retinopathy approximately two years ahead of males, once again suggesting a pubertal influence. There is some evidence, however, to suggest that the pre-pubertal years are not as influential on the course of retinopathy as those after puberty.

Nephropathy Patients with diabetic nephropathy are much more likely to have associated microvascular disease in the eye; up to 96% of those with nephropathy also have retinopathy. In type 2 diabetes, microalbuminuria is an independent pre-

CHAPTER 15 • EPIDEMIOLOGY AND RISK FACTORS

dictor of retinopathy. This suggests similar pathogenic mechanisms, but a significant proportion of patients with advanced retinopathy show no clinical evidence of nephropathy. Similarly, although unusual, advanced renal disease has been described in the absence of retinopathy. It therefore seems likely that each form of microvascular disease has overlapping but not identical pathogenic mechanisms. Beyond its predictive value, nephropathy may also signify the presence of a more severe form of retinopathy, one that is more resistant to the effects of laser therapy.

Plasma lipids The role of plasma lipids in the development of diabetic retinopathy is unclear. In the WESDR, base-line serum cholesterol level had no relationship with retinopathy status five years later in type 1 diabetics. Other studies, however, have suggested an association between higher levels of total serum cholesterol, declining ratios of high density lipoprotein (HDL), cholesterol/total cholesterol and more severe retinopathy. As part of the Sorbinil Retinopathy Trial in type 1 diabetes, total serum cholesterol was found to have a marginal effect on the rate of progression of retinopathy. In a small case-control study of type 2 diabetic patients by Dodson and Gibson, patients with maculopathy were followed-up over seven years and compared with a matched group of patients without maculopathy. The group with maculopathy was found to have higher serum cholesterol levels (6.6 vs. 5.9 mmol/l), higher levels of the HDL2 subfraction (0.46 vs. 0.32 mmol/l) and a higher prevalence of hyperlipidaemia (54% vs. 35%). The Early Treatment Diabetic Retinopathy Study (ETDRS) group reported that patients with persistently poor vision had higher blood cholesterol levels. They also found that patients with high total cholesterol and LDL-cholesterol had twice the number of retinal hard exudates compared with those with normal lipid levels. The risk of losing vision was related to the extent of hard exudates, even after adjusting for macular oedema. These patients also developed more hard exudates during the course of the study. This finding is supported by the WESDR, which also reported an association between higher cholesterol levels and the extent of retinal hard exudation. There is some evidence indicating that the correction of dyslipidaemia can significantly reduce macular exudation.

Genetic factors It seems likely that there is a component of inherited susceptibility to the various microvascular complications of diabetes, because some individuals appear not to develop retinopathy despite poor glycaemic control. Familial clustering of diabetic retinopathy was investigated among 372 subjects with type 1 diabetes in the DCCT. The presence and severity of retinopathy was

135

136

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

assessed in first-degree diabetic relatives of subjects in the trial. The severity of retinopathy in parents of children probands was found to be linked with the severity of retinopathy in their children, with the link strongest between mother and child. However, no relationship could be established between siblings. Although this observation suggests that the severity of diabetic retinopathy may be influenced by familial factors, the difficulty remains in differentiating environmental from genetic influences.

Pregnancy Diabetic retinopathy often gets worse during pregnancy, but it has been difficult to demonstrate that the progression of retinopathy is related to the pregnancy and would not otherwise have occurred. The issue has recently been clarified by two case-control studies. Moloney and Drury studied the progression of retinopathy prospectively in two groups of diabetic women who were matched apart from pregnancy status. The pregnant group demonstrated both increased prevalence and severity of retinopathy status compared to the control group. This finding is further supported by a larger study where the rate of progression of retinopathy was also found to be greater in the pregnant group after accounting for the influence of duration of diabetes, glycaemic control and hypertension. The Diabetes in Early Pregnancy Study reported that the risk of progression of retinopathy is greatest in women with moderate to severe retinopathy at the start of the pregnancy. For example, 29% of subjects with moderate retinopathy at base-line developed proliferative features, compared with 6% of those with minimal retinopathy at base-line. The study also found that women with the highest glycosylated haemoglobin levels at the start of pregnancy, and those who experienced the greatest improvement in glycaemic control during pregnancy, had the highest risk of progression of retinopathy status. This may be one major reason why retinopathy progresses during pregnancy, since retinopathy status can temporarily deteriorate after institution of tight glycaemic control. Aiming for improved glycaemic control during pregnancy features strongly in the antenatal care of diabetic patients. Although current pregnancy may adversely affect retinopathy status, it has been reported that in the long-term parous women have a lower prevalence of any retinopathy and of severe retinopathy. This may be the consequence of the institution of better control during pregnancy continuing into the post-partum period. Ophthalmic supervision is therefore important during pregnancy particularly when retinopathy status at the start of pregnancy is moderately advanced. Those women with poor glycaemic control at the start of pregnancy also require close supervision.

CHAPTER 15 • EPIDEMIOLOGY AND RISK FACTORS

Ocular factors It has been observed that patients with asymmetrical diabetic retinopathy have a higher ophthalmic arterial perfusion pressure in the more severely affected eye. Ocular perfusion pressure is associated with the incidence of retinopathy. Higher intraocular pressure (therefore lower perfusion pressure) may offer some protection against retinopathy. Other factors found to be associated with a reduced risk of retinopathy include myopia, extensive chorioretinal scarring (perhaps simulating the effects of panretinal laser treatment) and amblyopia.

CURRENT ISSUES • The incidence of blindness from diabetic retinopathy appears to have been diminishing over the last two decades, but diabetic eye disease remains the leading cause of registrable blindness in the working age group (accounting for 12% of all cases). • The principal risk factor in the development of diabetic retinopathy is that of hyperglycaemia; it has now been definitively shown that tight glycaemic control reduces the incidence and progression of retinopathy. Sudden improvements in glycaemic control may, however, worsen retinopathy during the first year but any adverse effect is usually transient. • Intensive BP control reduces the risk of progression of retinopathy. Even small reductions in BP translate into large clinical benefits.

FURTHER READING Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes. N Engl J Med 1993; 329: 977–986. Klein, R, Klein BEK, Moss SE, Davis MD, DeMets KL. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. II. Prevalence and risk of diabetic retinopathy when age at diagnosis is less than 30 years. Arch Ophthalmol 1984; 102: 520–526. Klein, R, Klein BEK, Moss SE, Davis MD, DeMets KL. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. III. Prevalence and risk of diabetic retinopathy when age at diagnosis is 30 or more years. Arch Ophthalmol 1984; 102: 527–532. UK Prospective Diabetes Study (UKPDS) Group. 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 1998; 352: 837–853. UK Prospective Diabetes Study Group. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. BMJ 1998; 317: 703–13.

137

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Hean-Choon Chen FRCS, FRCOphth CHAPTER 16 CLASSIFICATION AND DIAGNOSIS OF DIABETIC RETINOPATHY

INTRODUCTION The clinical features of diabetic retinopathy result from pathological changes within the retinal vasculature which are classified into two principal stages, background retinopathy and proliferative retinopathy. Background retinopathy has also been referred to as non-proliferative retinopathy to differentiate it from proliferative retinopathy although the proliferative changes occur amongst the ‘background’ of non-proliferative features. The terms most commonly used to denote the various stages of diabetic retinopathy are: background retinopathy, preproliferative retinopathy (a subclassification within background retinopathy) and proliferative retinopathy.

BACKGROUND DIABETIC RETINOPATHY Background retinopathy is common and ranges from the very earliest signs of retinopathy to the severe changes seen just before the development of proliferative retinopathy. For practical purposes, background retinopathy can be subclassified into two subgroups of patients: those with sight-threatening disease and those without. The patterns of retinopathy most likely to reduce vision are those with macular involvement (maculopathy) and preproliferative disease (Table 16.1). Maculopathy denotes the lesions of background disease within the macular area, whereas preproliferative retinopathy has additional features which differentiate it from the lesions of less severe background retinopathy. The three basic lesions which make up background retinopathy are microaneurysms, retinal haemorrhages and exudates. Haemorrhages and exudates have a limited lifespan, and, to a lesser extent, this also applies to microaneurysms.

Stages of diabetic retinopathy Background

Non-sight threatening

Microaneurysms, haemorrhages, exudates

Maculopathy

Cotton wool spots, large intraretinal haemorrhages, intraretinal vascular abnormalities (IRMA)

Pre-proliferative Venous abnormalities (beading, duplication, loops) Proliferative

New vessels, fibrovascular tissue, retinal detachment, vitreous haemorrhage

Table 16.1 Stages of diabetic retinopathy.

139

140

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Microaneurysms These are the very earliest clinically detectable lesions of diabetic retinopathy (Fig. 16.1). They appear as small, round, red dots and may be found in any part of the retina although they predominate in the posterior pole of the eye. They are not associated with any visible blood vessels and represent localized dilatations of retinal capillaries. The number of microaneurysms increases with increasing severity of retinopathy. A microaneurysm indicates a localized area in the microvascular circulation where the blood-retinal barrier is deficient and may therefore be associated with abnormal vascular leakage. The pathogenesis of microaneurysms is unclear but they may represent outpouchings of capillaries at areas of relative weakness where there is pericyte loss. Pericytes are cells which partly enclose retinal capillaries and may be considered the smooth muscle equivalent of the microvasculature; pericyte numbers diminish early in the development of diabetic retinopathy. Microaneurysms may also represent a localized response to surrounding hypoxia, i.e a limited proliferative process, as they tend to predominate in areas where there is closure of surrounding capillary beds.

Haemorrhages Haemorrhages co-exist with microaneurysms but are more variable in their appearance (Fig. 16.1). At their smallest, they may be difficult to differentiate from microaneurysms. A haemorrhage, unlike a microaneurysm, is not nec-

Fig. 16.1 Moderate background diabetic retinopathy with microaneurysms, haemorrhages and exudates.

CHAPTER 16 • CLASSIFICATION AND DIAGNOSIS

essarily round and may take on a variety of outlines; the phrase ‘dot and blot’ is an apt description. Haemorrhages can occur within the retina, where they remain confined by the retina, or they can occur on the retinal surface (flameshaped haemorrhage) where they spread out over the superficial nerve fibre layer taking on a characteristic flame appearance. This latter form of haemorrhage is less obviously a feature of diabetic retinopathy and may suggest the co-existence of hypertensive vessel damage. Haemorrhages probably occur from rupture of microaneurysms or other weak-walled vascular abnormalities. Small intraretinal haemorrhages occur early in diabetic retinopathy and their numbers increase with increasing severity. In more advanced disease, large dark blot intraretinal haemorrhages suggest severe retinal ischaemia with arteriolar occlusion, a feature of preproliferative disease.

Exudates These are usually small collections of lipoprotein which have accumulated within the retina from abnormal vascular leakage, and are therefore found in the vicinity of microaneurysms (Fig. 16.1). They are usually reflective and may appear to have a rigid, multifaceted contour, ranging in colour from white to yellow. They were previously referred to as ‘hard’ exudates to differentiate them from soft exudates (now called cotton wool spots); however, this separation is now redundant since it is well established that cotton-wool spots are not the products of exudation. Like microaneurysms, exudates are most frequently detected in the posterior pole and may be distributed in the form of a whole or partial ring appearance (Fig. 16.2). Such ‘circinate’ ring arrangements usually have microaneurysms in the centre, which are responsible for the vascular leakage that gives rise to the exudates at the margins. The number of exudates may paradoxically increase as the degree of extravascular fluid diminishes due to precipitation of lipids and proteins, analogous to a saline solution depositing salt upon drying. There may, therefore, be a transient increase in the number of exudates following laser treatment as the macula becomes drier.

What is diabetic maculopathy? The term macula refers to the important centre of the retina. It measures approximately 5 mm in diameter and is the area centred upon the fovea with a radius that extends to the temporal margin of the optic disc.The fovea itself is about the same size as the optic disc (1.5–1.7 mm in diameter), with its centre (foveola) recognizable in normal eyes by the foveolar reflex. More practically, the macula can be considered as the area within the major temporal vascular arcades.

141

142

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Fig. 16.2 Diabetic maculopathy with a circinate exudate ring.

Diabetic maculopathy can be defined as any retinopathy lesion located within the macula. However, the term maculopathy is usually reserved for sight-threatening lesions close to the centre of the macula. The Early Treatment Diabetic Retinopathy Study (ETDRS) group produced the following list of criteria, any one of which is sufficient to diagnose clinically significant macular oedema (CSMO) requiring laser treatment: • Thickening of the retina located less than 0.5 mm from the centre of the macula. • Exudates (with thickening of adjacent retina) located less than 0.5 mm from the centre of the macula. • An area of retinal thickening 1 disc diameter in size located less than 1 disc diameter from the centre of the macula. For practical purposes, sight-threatening maculopathy is any retinopathy lesion within 1⁄2 a disc diameter from the centre of the macula; this simplified definition will assist the non-ophthalmologist in identifying what may be sightthreatening but this would not necessarily be an indication for laser treatment.

PREPROLIFERATIVE DIABETIC RETINOPATHY Although classified as a sub-category of background retinopathy, preproliferative retinopathy is a sight-threatening condition that is usually considered separately from background disease. It also differs from background

CHAPTER 16 • CLASSIFICATION AND DIAGNOSIS

retinopathy in having four new features, i.e. cotton wool spots, retinal venous abnormalities, large blot intraretinal haemorrhages and intraretinal microvascular abnormalities.

Cotton wool spots Cotton wool spots appear as pale cream patches of variable sizes (Fig. 16.3). They do not have clearly defined outlines and are most frequently seen in the posterior pole. A cotton wool spot is an area of infarction in the nerve fibre layer, and the appearance is due to swollen nerve axons with impaired axoplasmic flow. It therefore represents an area of localized retinal ischaemia and suggests the presence of arteriolar occlusion. Cotton wool spots persist for a long time, ranging from 8–17 months. Five or more cotton wool spots are generally required to suggest preproliferative disease.

Intraretinal microvascular abnormalities (IRMA) These usually appear as irregular loops of vessels within the retina which may straddle normal vessels (Fig. 16.3). IRMA occur adjacent to areas of capillary bed closure and their origin is unclear. Unlike ‘new vessels’, IRMA do not always leak fluorescein, although some leakage may occur at their growing tips. At least two different theories exist as to what abnormal vasculature are presently classified as IRMA: shunt vessels and intraretinal new vessels.

Fig. 16.3 Early preproliferative retinopathy with intraretinal retinal microvascular abnormalities associated with cotton wool spots.

143

144

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Venous abnormalities Various abnormalities occur in the retinal veins in response to the hypoxic environment (Fig. 16.4). These take the form of: • beading, e.g. the ‘string-of-sausages’ appearance; • reduplication of veins, whereby the vein appears to divide into two parallel channels over a short segment; and • venous loops, where the vein makes a sudden deviation in the form of a loop. Venous abnormalities, particularly those of beading and reduplication, are strong indicators of hypoxia and suggest that new vessel development is imminent.

Deep retinal haemorrhages These are large dark haemorrhages within the retina representing haemorrhagic infarction secondary to retinal arteriolar occlusion.

PROLIFERATIVE DIABETIC RETINOPATHY Ischaemia within the retina due to widespread closure of capillary beds leads to newly formed blood vessels appearing on the retinal surface, or overlying the optic disc. These vessels extend in the plane between the retina and the vitreous and are accompanied by a supporting network of fibroglial proliferation.

Fig. 16.4 Severe preproliferative retinopathy venous abnormalitits.

CHAPTER 16 • CLASSIFICATION AND DIAGNOSIS

New vessels developing from the vasculature of the optic disc are called ‘disc new vessels’ (NVD) (Fig. 16.5) whilst those developing on the surface of the retina are called ‘new vessels elsewhere’ (NVE) (Fig. 16.6). It is thought that NVD represents severe generalized ischaemia of the retina, whereas NVE are a response to local ischaemia in the quadrant of the retina where they occur. New vessels usually arise from a vein and have a haphazard growth pattern. As they grow, the combination of new vessels and supporting fibroglial tissue becomes adherent to both the retinal and posterior vitreous surfaces, inducing the vitreous to detach from the retina. The subsequent traction may cause haemorrhage either because the fragile new vessels break or because they are avulsed from their point of origin on the main retinal vessel. If bleeding is confined to the space between the retina and the vitreous, a preretinal or retrohyaloid haemorrhage is clearly visible on ophthalmoscopy (the so-called ‘boat shaped’ haemorrhage with a fluid level appearance). Depending on whether the haemorrhage obscures the macula, vision may be severely affected or minimally compromised. If the haemorrhage is to break through into the main body of the vitreous, the view of the retina may be variably obscured, likewise the patient’s vision. At worse, no view may be possible of the retina. The second outcome of neovascular traction is a retinal detachment. A tractional retinal detachment usually occurs slowly, and may remain stable for years assuming laser treatment has been applied to control the neovascular process. A tractional retinal detachment affects vision in two ways. Firstly, if it directly affects the fovea, vision will be reduced; if extrafoveal traction exists,

Fig. 16.5 Proliferative retinopathy with disc new vessels (NVD).

145

146

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Fig. 16.6 Proliferative retinopathy with retinal new vessels (NVE).

tension induced retinal folds may secondarily affect the fovea, producing visual distortion. Secondly, a stable tractional detachment may suddenly become unstable if a full thickness hole occurs in the retina, leading to a rhegmatogenous retinal detachment which may spread to involve the fovea. The proliferative process may not be confined to the posterior segment of the eye. Iris neovascularization is a feared complication because of the risk of neovascular or thrombotic glaucoma; a form of glaucoma which is difficult to manage once established. As with retinal neovascularization, fibrous tissue eventually develops which can occlude the trabecular meshwork and the anterior chamber angle leading to uncontrolled neovascular glaucoma or secondary angle closure glaucoma, resulting in a painful, red blind eye.

INTERNATIONAL CLINICAL DIABETIC RETINOPATHY DISEASE SEVERITY SCALE The International Council of Ophthalmology has produced a new classification for diabetic retinopathy, in an attempt to standardize terminology. The classification comprises five stages from no retinopathy to proliferative retinopathy (Table 16.2). The Council has also proposed a classification for diabetic maculopathy based on whether it is absent or present; the latter is then subclassified into three grades of severity (Table 16.3).

CHAPTER 16 • CLASSIFICATION AND DIAGNOSIS

International clinical diabetic retinopathy disease classification scale Disease severity level

Findings observable upon dilated ophthalmoscopy

No apparent retinopathy

No abnormalities

Mild non-proliferative diabetic retinopathy

Microaneurysms only

Moderate non-proliferative diabetic retinopathy

More than just microaneurysms but less than severe NPDR

Severe non-proliferative diabetic retinopathy

Any of the following: • More than 20 intraretinal haemorrhages in each of 4 quadrants • Definite venous beading in 2+ quadrants • Prominent IRMA in 1+ quadrant And no signs of proliferative retinopathy

Proliferative diabetic retinopathy

One or more of the following: • Neovascularization • Vitreous/preretinal haemorrhage

Table 16.2 International clinical diabetic retinopathy disease classification scale.

International clinical classification of the severity of diabetic macular oedema Classification

Findings observable upon dilated ophthalmoscopy*

Diabetic macular oedema present

• Mild diabetic macular oedema Some retinal thickening or hard exudates in posterior pole but distant from the macula • Moderate diabetic macular oedema Retinal thickening or hard exudates approaching the centre of the macula but not involving the centre • Severe diabetic macular oedema Retinal thickening or hard exudates involving the centre of the macula

Table 16.3 International clinical classification of the severity of diabetic macular oedema.

Diagnosis of diabetic retinopathy It is essential that patients with diabetes are regularly examined for the presence of symptomless retinopathy. Retinopathy screening is presently performed by health care professionals from a variety of disciplines, including optometrists, nurses, medical photographers, general practitioners and diabetologists using a variety of techniques.

147

148

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Screening for diabetic retinopathy Diabetic retinopathy is a disease which fulfills all the necessary criteria for a screening programme: Those at risk form an identifiable population; it has a recognized disease pattern; and laser treatment, if performed early, is effective in preventing loss of vision, particularly in proliferative disease. Advanced disease, when diagnosed late, is less amenable to treatment and much more costly, both economically and in terms of the patient’s quality of life. In the UK and many other countries, however, there is no national strategy for the screening of diabetic retinopathy. In any one year in the UK, the proportion of diabetic patients who receive retinopathy screening varies from 38%–85%, and from 14%–97% between different primary care practices. A variety of methods are in use in the UK at present, e.g. selected optometrists accredited to perform the screening using slit-lamp biomicroscopy, and schemes that are based on retinal photography, either fixed-site or via a mobile unit.

Techniques for screening Direct ophthalmoscopy using a hand-held ophthalmoscope is used to a varying extent, and with varying degrees of success, by general practitioners, optometrists and diabetologists. It is technically difficult, allowing only a twodimensional view of the retina. Therefore retinal oedema cannot be accurately diagnosed using this technique. The peripheral parts of the retina are difficult to examine using this technique. It is a form of examination that has largely been abandoned by ophthalmologists, who now mainly examine the retina by slit-lamp biomicroscopy. Slit-lamp biomicroscopy provides a much wider three-dimensional view of the retina using a 78 or 90 dioptre lens. Although an effective technique, it is very skill-dependent and the operator requires extensive training. Both methods of ophthalmoscopy have to be performed with mydriasis (dilated pupils) and both have the disadvantage of not providing a hard record for qualitative assessment and for monitoring signs of progression of disease. Retinal photography is a technique that is more easily acquired, and the image can be interpreted later by another health professional, e.g. diabetologist, ophthalmologist or a specially trained grader. The number of ungradeable photographs ranges from 3.7% to 20%, the failure rate being lower with mydriasis. Increasingly, digital photography is supplanting the analogue techniques of slides and Polaroid photography. It has major advantages in its ease of image acquisition, data storage and there is also the option of electronic data transfer. The computerized interpretation of images is a real possibility and the screening process may eventually become entirely electronic.

CHAPTER 16 • CLASSIFICATION AND DIAGNOSIS

The British Diabetic Association (BDA) has proposed that a screening test for diabetic retinopathy should have at least 80% sensitivity and 95% specificity. In a wide-ranging review of multiple studies examining the effectiveness of direct ophthalmoscopy, indirect ophthalmoscopy and retinal photography, it has been reported that retinal photography with a dilated pupil is the most effective. Most of the studies which employed retinal photography had sensitivity levels of over 80%, and mydriatic photographs gave an even higher level of sensitivity.

Proposed UK national screening scheme Screening for retinopathy in the UK is likely to change as part of the “Preservation of sight in diabetes: a risk reduction programme” initiative. A nationwide screening programme will be established by 2005/2006 with clear aims: (1) to reduce the rate of avoidable visual loss by early detection of sightthreatening retinopathy so that it can be treated promptly; and (2) detection of any retinopathy, so that the diabetic patient can be made aware that changes have begun to occur in their eyes, and attempts can be made to improve glycaemic and blood pressure control. It is anticipated that screening will be offered to all diabetic patients by 2007. The main features of this screening program will include: • Target population: all diabetic patients, types 1 and 2, over 12 years of age, or post-puberty. • Frequency: annually, initially, but after a few screening rounds those deemed to be at low risk can probably be screened less frequently whilst those with more severe disease may be screened more frequently. • Technique: the Steering Committee has recommended mydriatic digital photography (two fields, macular and nasal) with visual acuity measurement (visual acuity alone will not lead to referral to an assessment clinic unless accompanied by retinopathy, although the general practitioner will be informed if visual acuity falls below 6/12 in either eye). • Examination: grading of photographs by specially trained graders using a standardized grading scheme employing reference images, with opinions from ophthalmologists and/or diabetologists if required. • Positive patients: referral to special assessment clinics at convenient ophthalmology departments.

149

150

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

CURRENT ISSUES •

•

•

•

•

The classification of diabetic retinopathy into background and proliferative stages is well established and assists in the management of sight-threatening retinopathy. There has been a recent attempt by the International Council of Ophthalmology to standardize the terminology employed in the classification of diabetic retinopathy. Screening for diabetic retinopathy in many countries, including the UK, is not sufficiently well established or co-ordinated to achieve the standards required by the St Vincent declaration of 1989. The most sensitive screening technique for diabetic retinopathy is mydriatic fundal photography and the least effective is direct ophthalmoscopy. Successful retinopathy screening is principally technique-dependent and less personnel-dependent.

FURTHER READING Bachmann MO, Nelson SJ. Impact of diabetic retinopathy screening on a British district population: case detection and blindness prevention in an evidence-based model. J Epidemiol Community Health 1998; 52: 45–52. Bagga P, Verma D, Walton C, Masson EA, Hepburn DA. Survey of diabetic retinopathy screening services in England and Wales. Diabetic Medicine 1998; 15: 780–782. Garvican L, Clowes J, Gillow T. Preservation of sight in diabetes: developing a national risk reduction programme. Diabetic Medicine 2000; 17: 627–634. Hart PM, Harding S. Is it time for a national screening programme for sight threatening retinopathy? Eye 1999; 13: 129–130. Hutchinson A, McIntosh A, Peters J et al. Effectiveness of screening and monitoring tests for diabetic retinopathy—a systematic review. Diabetic Medicine 2000; 17: 495–506.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Hean-Choon Chen FRCS, FRCOphth CHAPTER 17 DIABETIC MACULOPATHY

INTRODUCTION Diabetic maculopathy is the commonest cause of visual loss in patients with diabetes and it can be defined as the presence of sight-threatening lesions within the macula. These lesions commonly consist of microaneurysms, haemorrhages and exudates. The cause of visual loss is a consequence of either or both of the following: • leaking blood vessels leading to exudate formation and accumulation of extracellular fluid causing macular oedema (exudative diabetic maculopathy or diabetic macular oedema); and • capillary closure giving rise to macular ischaemia (ischaemic diabetic maculopathy). Loss of vision from exudation and oedema is the commoner of the two mechanisms and is fortunately responsive to laser treatment, at least in part. There is no treatment for macular ischaemia.

EPIDEMIOLOGY The Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR) reported an overall prevalence rate of 10% for macular oedema; predictably, the prevalence increases with increasing severity of overall retinopathy status, ranging from approximately 2% in those with mild background retinopathy to 20–37% in those with moderate to severe background disease, and in those with proliferative retinopathy the prevalence of macular oedema was approximately 70% (Table 17.1). The prevalence of maculopathy also increases with duration of diabetes, and is higher in type 2 compared with type 1 diabetes (Figs 17.1 & 17.2).

DEFINITIONS Exudative maculopathy The Early Treatment Diabetic Retinopathy Study (ETDRS) group produced a list of criteria to denote clinically significant macular oedema, i.e. maculopathy for which laser treatment is indicated (see chapter 16).

Ischaemic maculopathy The normal fovea possesses a central avascular area known as the foveolar avascular zone (FAZ); this exists so as to provide the centre of the fovea with the least possible impedance to incident light. In the normal eye, the FAZ varies significantly in size with an average of approximately 0.5–0.6 mm. In

151

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Relationship of Diabetic Retinopathy with Macular Oedema* and Duration of Diabetes (Wisconsin, HAS-1, 1980-82) Younger onset

Older onset

Duration 10+ years Retinopathy status

% Macular No. oedema P

Non-proliferative Mild 172 Moderate to severe 128 Proliferative 85

Duration 0–14 years

Duration 15+ years

% Macular No. oedema P

% Macular No. oedema

P

1.7

–

152

2.6

–

126

6.3

–

20.3 69.7

<0.001 –

60 26

36.7 73.1

<0.001 –

87 35

63.2 74.3

<0.001

*Percentage with macular oedema = (number of persons with macular oedema status 2 or 3 in either or both eyes/number of persons with macular oedema status 0 in both eyes + number of persons with macular oedema status 2 or 3 in either or both eyes) × 100.

Table 17.1 Relationship of diabetic retinopathy with macular oedema* and duration of diabetes (Wisconsin, HAS-1, 1980–82). The Wisconsin Epidemiologic Study of Diabetic Retinopathy. IV. Diabetic macular oedema. Ophthalmology 1984; 91: 1464–1474.

Per cent with macula oedema

152

40 30 20 10 0 0

5

10

15 20 Duration (years)

25

30

35

Fig. 17.1 Frequency of macular oedema by duration of diabetes in years for insulintaking early onset persons. Ophthalmology 1984; 91: 1464–1474.

CHAPTER 17 • DIABETIC MACULOPATHY

Per cent with macula oedema

40 Insulin taking Non-insulin taking 30 20 10 0 0

5

10

15 20 Duration (years)

25

30

Fig. 17.2 Frequency of macular oedema by duration of diabetes for insulin- and non-insulin-taking older onset persons. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. IV. Diabetic macular oedema. Ophthalmology 1984; 91: 1464–1474.

ischaemic maculopathy, there is a gradual increase in the size of the FAZ. Although there is no defining measurement of FAZ for a diagnosis of ischaemic maculopathy, when the diameter of FAZ exceeds 1 mm, vision is usually compromised (Fig 17.3).

DIAGNOSIS The diagnosis of diabetic maculopathy is made clinically and, when necessary, with the aid of fluorescein angiography. More recently, a new imaging technique called optical coherence tomography (OCT) has provided a means of objectively assessing macular thickening non-invasively. Changes in visual function may raise the suspicion of macular disease: • reduced visual acuity; • reduced contrast sensitivity; and • colour vision defects, usually along the blue-yellow (tritan-like) axis, may occur from an early stage in the disease and may even predate clinically visible lesions. Clinically, the presence of microaneurysms and exudates indicates the presence of pathological vascular leakage, although these changes alone may not reduce visual acuity. The retinal pigment epithelial ‘pump’ and surrounding competent

153

154

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

capillaries are able to remove extravasated fluid and extracellular fluid accumulates when the rate of leakage exceeds the capacity for fluid removal, giving rise to retinal thickening. Vision is affected when this occurs at the centre of the macula. However, exudates in large numbers, and especially at the centre of the macula, can also affect vision in the absence of retinal thickening (probably due to direct photoreceptor damage). When large numbers of exudates are present, macular oedema is usually also present. It is not unusual to observe a paradoxical increase in the number of exudates as oedema dissipates, since lipoproteins are precipitated within the retina as the water component of oedema is removed. The diagnosis of macular thickening requires stereoscopic views of the macula. This is possible with slit-lamp biomicroscopy using either a noncontact 78 or 90 dioptre lens, or with a fundus contact lens. The contact lenses provide better stereopsis and may be better at detecting subtle degrees of retinal thickening. Fluorescein angiography is usually not necessary in the diagnosis of macular oedema. However, when there is an unexplained loss of vision, i.e. when it cannot be attributed to vascular leakage, fluorescein angiography is helpful. The diagnosis of macular ischaemia can only be definitively made with fluorescein angiography.

Fluorescein angiography Fluorescein angiography outlines the retinal circulation, illuminating the otherwise invisible microvasculature. It is a photographic investigation technique that uses an adapted fundus camera. Fluorescein, a vegetable dye extract, is able to absorb light with a wavelength of approximately 490 nm (blue light) and in response emits light at a longer wavelength of 530 nm (green light). Fluorescein is injected into an antecubital vein (usually 5 ml of

Fig. 17.3 This fluorescein angiogram of the posterior pole of the eye demonstrates the presence of ischaemic maculopathy with an enlarged foveal avascular zone.

CHAPTER 17 • DIABETIC MACULOPATHY

a 10% solution), where it becomes 70–85% plasma protein bound, and photographs are taken of the fundus as it traverses the retinal circulation. The adapted fundus camera possesses two filters to ensure that blue light enters the eye and yellow-green light enters the camera, where it is captured on film. Photographs are usually taken from about 10 seconds after injection and thereafter at one to two second intervals. Late pictures may be taken after a few minutes to determine the possible presence of late leakage. Fluorescein angiography provides assistance in the diagnosis of: • Capillary closure or non-perfusion, especially in the macula, where capillaries exist as a monolayer with an increased melanin background, providing greater contrast. This is particularly helpful in the diagnosis of ischaemic maculopathy, where, in good quality fluorescein angiographic pictures, FAZ is clearly delineated. • Vascular leakage, although fluorescein angiography is seldom required to make this diagnosis (Fig. 17.4). Cystoid macular oedema takes on a characteristic petalloid appearance (Fig. 17.5). • Subtle neovascularization, which may not be immediately obvious on clinical examination. New vessels do not possess a normal blood-retinal barrier and are hyperpermeable, therefore giving rise to extensive fluorescein leakage. Several other retinopathy lesions can also be highlighted, e.g. microaneurysms and intraretinal neovascular abnormalities (IRMA).

Optical coherence tomography This is a relatively novel non-invasive, non-contact imaging technique based on interferometry whereby high-resolution, cross-sectional images of the retina are obtained. It is similar to ultrasound except that light is used in place

Fig. 17.4 Fluorescein angiogram demonstrating diffuse macular leakage.

155

156

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

Fig. 17.5 Fluorescein angiogram demonstrating cystoid macular oedema with its characteristic petal-like leakage pattern.

of sound; ‘echoes’ of light are produced at junctions of tissue layers with different density. It allows for objective measurements of retinal, in particular macular, thickness. The technology is expensive but is becoming more widely available. It is in certain circumstances a substitute for fluorescein angiography, which is a more invasive technique and non-quantitative.

PATHOGENESIS Exudative maculopathy or macular oedema is the consequence of a breakdown of the blood-retinal barrier (BRB), more specifically the inner BRB, i.e. the retinal capillary endothelium. This is usually at sites of microaneurysms although other retinopathy lesions also signify breaches in the BRB such as dilated capillaries, IRMA and retinal neovascularization. There is however, also some evidence suggesting a breakdown in the outer BRB, i.e. the retinal pigment epithelium (RPE), as well. The RPE forms a barrier between the neural retina and the choriocapillaris, the capillary network of the choroid, which is highly permeable to macromolecules (unlike retinal capillaries). The normally tight junctions of the RPE, therefore, maintain a diffusion gradient for water from retina to choroid because of the higher oncotic pressure within the choroid. It has also recently been demonstrated that a taut, thickened posterior hyaloid (vitreous), which is still attached to the retina, can give rise to traction upon the macula producing intraretinal cystic spaces, and possibly a tractional detachment. These changes in the posterior hyaloid usually occur in response to ischaemic disease and therefore this form of macular oedema more frequently occurs in severe forms of retinopathy.

CHAPTER 17 • DIABETIC MACULOPATHY

TREATMENT Ischaemic maculopathy is not amenable to treatment. Exudative maculopathy may, for treatment purposes, be subclassified into two varieties: • Focal oedema. This is a localised area of leakage, usually from a cluster of microaneurysms which exist in the centre of an area of oedema. The area affected is usually circular and its periphery often delineated by a ring of exudates, the so-called ‘circinate of exudates’. • Diffuse oedema. Leakage appears to occur from a large area of dilated capillaries, sometimes with little evidence of exudate formation. This is possibly due to capillary beds dilating to compensate for surrounding areas of capillary occlusion, or there may be an autoregulatory increase in local retinal blood flow. Prolonged diffuse oedema can lead to the formation of cystic spaces in the centre of the macula, known as cystoid macular oedema (CMO), and is more likely to be associated with systemic problems such as renal failure and uncontrolled hypertension. Both forms of oedema can coexist. There will usually be areas of focal leakage in patients with diffuse oedema but an area of focal oedema frequently exists in isolation. The visual deficit is usually more severe in cases of diffuse oedema, which is unfortunately also more refractory to treatment, particularly in the presence of cystoid oedema.

Laser therapy Laser therapy is effective in both focal and diffuse macular oedema. In focal oedema, the treatment is applied directly at the area(s) of leakage, for example the microaneurysms at the centre of a ring of exudates. An attempt may be made to coagulate the microaneurysm. The ETDRS protocol included direct treatment of microaneurysms greater than 40μm in diameter. The technique employed to do this is to initially ‘whiten’ the underlying retinal pigment epithelium using a 100μm spot size. The aim of this step is to ensure that the whitened underlying tissue will no longer absorb the energy from subsequent laser shots. The spot size should then be reduced to 50μm and aimed at the microaneurysm, with the aim of turning it white or a darker shade of red. In diffuse oedema the method used is known as ‘grid macular treatment’. It involves applying laser burns with a spot size of between 100μm and 200μm, aiming for a blanching effect of the retinal pigment epithelium, i.e. a grey-white effect. The spots are applied to cover the part of the macula that is thickened, sparing the fovea; treatment should commence approximately 500μm from the centre of the macula and be spaced about one burn-width apart. If oedema persists, retreatment with laser burns applied closer to the centre of the macula can be applied; the ETDRS protocol allowed for treatment from 300μm from the centre of the macula, unless there was perifoveal

157

158

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

capillary dropout. Treatment over the papillo-macular bundle is possible since the nerve fibre layer is spared, as the level of laser energy used affects only the pigment epithelium and its immediate neighbours. The ETDRS protocol advocated treating not just the areas that were oedematous but also any adjacent areas of capillary non-perfusion. Treatment should be applied using an appropriate contact lens to provide good visibility of the macula, e.g. the wide-angle Volk Transequator. The type of thermal laser wavelength used should be either green (e.g. argon and double-frequency YAG lasers) or yellow (dye laser); blue light laser should be avoided in treatment of the macula because the absorption of blue light by abundant macular xanthophyll pigment can cause greater degrees of collateral damage to the retina. The red light of the krypton laser will not be absorbed by haemoglobin and therefore will not be useful in closing microaneurysms. At the commencement of treatment, it is prudent to assess the energy requirement away from areas close to the fovea, perhaps with a trial shot a short distance away. An initial energy setting of approximately 80–100 mW should be considered with a pulse duration of 0.1 seconds; these settings may then be modified according to the response. Following treatment, an appropriate review interval will be between two and four months. It often takes this period of time for any effect on oedema resolution to become obvious.

Surgical Procedures If a taut posterior hyaloid is visible (possibly with the aid of OCT), pars plana vitrectomy has been found to be effective in resolving macular oedema. Features supporting this particular treatment option may include oedema resistant to grid laser treatment and the presence of cystoid oedema. Vitrectomy for macular oedema even in the absence of overt posterior hyaloid traction has been reported to produce a beneficial effect although this is not widely carried out in the absence of significant clinical evidence. The removal of large subfoveal exudates has also been reported with some success.

THE EFFECTIVENESS OF LASER TREATMENT In treating focal leakage, successful closure of the leaking points, i.e. the microaneurysms, implies that leakage stops. However, part of the success is likely to be due to those factors which lead to the resolution of oedema following treatment for diffuse leakage. Why laser treatment works in the treatment of diffuse leakage is unclear. Because the laser is targeted at the retinal pigment epithelium, the actual leaking points, i.e. the retinal capillaries, are not directly treated. Various hypotheses have been formulated to explain the therapeutic benefits of laser:

CHAPTER 17 • DIABETIC MACULOPATHY

• Laser-damaged pigment epithelial cells regenerate and the new cells that fill the gaps created by thermal necrosis may create a more effective outer BRB. • The laser also damages and causes loss of retinal photoreceptor cells, which are the major consumers of oxygen within the neural retina. This may relieve hypoxia and reduce the autoregulatory increment in blood flow through dilated capillary beds, thereby reducing leakage. • The effect of the laser on the pigment epithelium may release a diffusible factor that induces retinal capillary repair, leading to a restoration of the inner BRB. Laser treatment of diabetic maculopathy is not always successful; it is frequently only able to maintain current levels of vision, or slow the rate of progression, rather than lead to an improvement in vision. If macular oedema coexists with a degree of ischaemia, it is reasonable to apply laser treatment although the prognosis is poorer compared with eyes without ischaemia. The diagnosis of ischaemia may be difficult in the presence of significant leakage since this obscures the view of the retinal microcirculation on fluorescein angiograms.

Complications of macular laser treatment • Accidental damage to the foveola. • Rupture of Bruch’s membrane with use of high laser energy, which may lead to the formation of a choroidal neovascular membrane. • Development of paracentral scotomata; these may be subtle and detectable only with more specialized perimetric techniques, such as blue-on-yellow perimetry. • Submacular fibrosis from laser scars which appear to spread and merge.

WHEN SHOULD LASER TREATMENT BE APPLIED? If clinically significant macular oedema (CSMO) exists, laser treatment should be applied soon, i.e. within weeks, and this is especially so if the centre of the macula is involved or directly threatened. When macular oedema coexists with proliferative retinopathy, the ETDRS group has clearly demonstrated that scatter photocoagulation leads to a worsening of the macular situation. It is recommended that whenever possible the macula should be treated first, or simultaneously, with the scatter treatment initially confined to the nasal quadrants. Intraocular surgery, for example cataract extraction, has been shown to worsen maculopathy and, if the view allows it, macular laser treatment should be applied before the patient undergoes cataract surgery. If maculopathy is present but has not yet reached the severity of CSMO, the patient should be

159

160

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

closely observed in the postoperative period as the level of maculopathy is very likely to increase in severity. A significant proportion of patients who develop macular oedema after cataract surgery will experience a spontaneous regression in their oedema, usually over several months after surgery, particularly if the underlying retinopathy status is mild. However, if oedema was present preoperatively, any worsening is likely to be persistent.

MICROPULSED DIODE LASER This is a relatively novel approach using a diode laser with a wavelength of 810 nm (invisible, infrared wavelength) and the laser applied in a ‘micropulse’ fashion, i.e. frequent, short (microsecond) pulses delivering subthreshold laser energy (an energy level that does not produce an immediate, clinically evident effect). The energy is delivered in ‘envelopes’, each consisting of multiple short pulses. The advantage over conventional macular laser treatment is the specific targeting of the laser on the retinal pigment epithelium, therefore inducing less damage to surrounding tissues with fewer complications.

TREATMENTS UNDER CONSIDERATION The intravitreal injection of triamcinolone (a long-acting steroid) has been demonstrated to have some effect on reducing macular oedema. However, not all patients respond and the effect may be relatively short-lived. The efficacy of a protein kinase C inhibitor in reducing macular oedema is being examined.

CURRENT DEVELOPMENT • The ETDRS group has recommended laser treatment for those patients with features fulfilling the ‘clinically significant macular oedema’ criteria. • Laser treatment is most likely to maintain current levels of vision, or retard the rate of progression of visual loss, rather than produce an improvement in vision. • Laser treatment should be applied using a green or yellow wavelength, and may be applied in a focal and/or grid fashion for localized areas of leakage or diffuse leakage, respectively. • Pars plana vitrectomy is effective in reducing diffuse macular oedema if a taut thickened posterior hyaloid is present. There is also evidence that persistent diabetic cystoid macular oedema may respond to pars plana vitrectomy, even if there is no clinical evidence of vitreous traction upon the macula. • There is some evidence that intravitreal triamcinolone can reduce macular oedema with an improvement in vision, at least in the shorter term.

CHAPTER 17 • DIABETIC MACULOPATHY

FURTHER READING Early Treatment Diabetic Retinopathy Study Research Group: Photocoagulation for diabetic macular oedema: Early Treatment Diabetic Retinopathy Study report number 1. Arch Ophthalmol 1985; 103: 1796–1806. Ikeda T, Sato K, Katano T, Hayashi Y. Vitrectomy for cystoid macular oedema with attached posterior hyaloid membrane in patients with diabetes. Br J Ophthalmol 1999; 83: 12–14. Klein R, Klein BEK, Moss SE, Davis MD, DeMets DL. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. IV. Diabetic macular oedema. Ophthalmology 1984; 91: 1464–1474. Lewis H, Abrams GW, Blumenkranz MS, Campo R. Vitrectomy for diabetic macular traction and oedema associated with posterior hyaloidal traction. Ophthalmology 1992; 99: 753–759. Moorman CM, Hamilton AM. Clinical applications of the MicroPulse diode laser. Eye 1999; 13: 145–150.

161

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Hean-Choon Chen FRCS, FRCOphth CHAPTER 18 PROLIFERATIVE DIABETIC RETINOPATHY

INTRODUCTION The development of newly formed blood vessels either on the retina or on the optic disc, with or without fibrous tissue, signifies the presence of proliferative retinopathy. New vessels develop in response to retinal ischaemia, and the presence of neovasularization carries a high risk of significant loss of vision due to various complications: • Vitreous haemorrhage. • Traction upon the fovea. • Retinal detachment. • Neovascular glaucoma.

EPIDEMIOLOGY The Wisconsin Epidemiological Study of Diabetic Retinopathy (WESDR) showed that the prevalence of proliferative retinopathy in type 1 diabetic patients is 0% in the first five years of diagnosis, increasing to 67% in those with diabetes of >35 years duration. In type 2 diabetes, the corresponding prevalence rates are 2% rising to 16% among those with diabetes for >15 years. Because type 2 diabetes is much more common, the absolute numbers of patients with proliferative eye disease is similar for the two types of diabetes. In the WESDR, approximately 43% of those examined with proliferative disease had type 1 diabetes whilst 42% had insulin-treated type 2 diabetes. The severity of proliferative disease is generally greater in type 1 diabetes, and the WESDR also showed that among patients with severe background retinopathy the 4-year incidence of proliferative retinopathy is around 40–50%. In the Diabetic Retinopathy Study (DRS), the risk of severe visual loss (worse than 5/200) in eyes with proliferative disease was 16% over two years. The risk was highest in patients with new vessels on the optic disc and those with vitreous or pre-retinal haemorrhages. The appearance of new vessels on the retina did not further increase the risk of severe visual loss in eyes with established new vessels on the disc.

DEFINITIONS Pathological newly formed blood vessels in the eye can be classified into three groups: • New vessels on the retina, commonly called new vessels elsewhere (NVE), usually defined as being greater than one disc diameter from the optic disc (Fig. 18.1).

163

164

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

• New vessels on the optic disc (NVD), i.e. on the disc or within one disc diameter from the optic disc (Fig. 18.2). • New vessels on the iris (Fig. 18.3).

DIAGNOSIS AND NATURAL HISTORY The diagnosis of proliferative disease is made clinically after a thorough examination of the retina with mydriasis, ideally using slit-lamp biomicroscopy with a non-contact 78 or 90 dioptre lens. The normal sequence of examination would be to firstly examine the posterior pole where neovascularization is common, arising from the optic disc or the larger vessels in the posterior pole, followed by a sequential examination of each quadrant of the retina, extending out to the periphery.

Fig. 18.1 New vessels elsewhere (NVE).

Fig. 18.2 New vessels on the optic disc (NVD).

CHAPTER 18 • PROLIFERATIVE DIABETIC RETINOPATHY

New vessels are most frequently found within 45 degrees of the optic disc, and NVD is present in approximately 70% of cases of proliferative disease, often in combination with NVE. Disc new vessels usually appear as fine vessels across the optic disc cup. New vessels are distinguished from normal vessels by their growth pattern and by being in a more superficial plane. They usually appear to emanate from a localized area of a retinal vein, and the size of these vessels can vary considerably. Retinal new vessels can sometimes be difficult to differentiate from intraretinal microvascular abnormalities (IRMA). NVE are on the retinal surface and will therefore grow over retinal structures such as other blood vessels and they also leak fluorescein profusely (Fig. 18.4). New vessels eventually acquire a varying degree of enveloping fibroglial

Fig. 18.3 Iris new vessels.

Fig. 18.4 Fluorescein angiogram revealing extensive fluorescein leakage from retinal new vessels.

165

166

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

tissue, and in the early stages the vitreous remains adherent to both the fibrovascular tuft and the surrounding retina. Eventually, however, a localized vitreous detachment is induced around the area of fibrovascular adhesion, with the fibrovascular tissue still attached to the vitreous (the vitreous detachment is therefore incomplete). Contraction of the posterior vitreous can lead to a vitreous haemorrhage, which can be subhyaloid (preretinal, i.e. between the retina and the posterior vitreous surface), or into the body of the vitreous. The process of vitreous detachment usually occurs gradually, over months or even years, during which time multiple vitreous haemorrhages can occur. Evidence of traction may be visible in the form of retinal tractional lines or striae, or a localized area of retinal detachment. Occasionally, a rhegmatogenous retinal detachment develops if the tractional forces produce a tear in the retina. In the presence of a vitreous haemorrhage, fundal examination may be difficult. Consideration should be given to whether there is another cause for the vitreous haemorrhage, e.g. a retinal tear, or if there is a retinal detachment. An ultrasound examination may be helpful. Blood in the preretinal space retains its red colour but blood within the vitreous eventually takes on a yellowish-grey appearance (ochre membrane). The possibility of haemolytic or ghost-cell glaucoma needs to be considered, especially after a vitreous haemorrhage has been present for a while.

PATHOGENESIS Retinal hypoxia due to capillary and arteriolar closure is the primary pathophysiological stimulus inducing new vessel formation. Hypoxia induces the local production of diffusible growth factors, e.g. vascular permeability factor (VPF), which initiate formation of new endothelial cells from existing blood vessels.

TREATMENT The principal form of treatment for proliferative retinopathy remains that of panretinal laser photocoagulation or ablation. It is still unclear how panretinal laser treatment works but various hypotheses include: • Ablation of ischaemic inner retinal tissue (the outer retina is avascular as it derives its oxygen by diffusion from the choroid and is therefore not directly affected by diabetes). Since most of the laser energy is absorbed by the retinal pigment epithelium, a heavy burn is required for the heat energy to diffuse inwards to destroy the inner retina. • Ablation of oxygen-consuming photoreceptors which lie adjacent to the retinal pigment epithelium, allowing more oxygen to diffuse further into the ischaemic inner retina.

CHAPTER 18 • PROLIFERATIVE DIABETIC RETINOPATHY

• Destruction of the retinal pigment epithelium may release some sort of new vessel inhibiting factor. Panretinal laser treatment is indicated for NVD and vitreous haemorrhage (Fig. 18.5). It should also be considered for patients with isolated NVE and those with severe preproliferative features.

Fig. 18.5 Diffuse intragel vitreous haemorrhage.

LASER TREATMENT TECHNIQUES The argon laser is the most widely used, but panretinal laser treatment can be applied with a krypton laser, gas laser, diode laser or the double-frequency YAG laser. For slit-lamp delivery, a wide-angle contact lens such as the Volk Quadraspheric lens provides the best view. The spot size setting is usually 200–500 μm, spaced 0.5–1.5 burn-widths apart, with a pulse duration of 0.05–0.1 seconds. The power should be set to deliver a medium intensity burn (creamy-white effect), e.g. an initial power setting for the argon laser of 200 mW for slit-lamp delivery (depending on spot size and duration) or approximately 300 mW for delivery through the indirect ophthalmoscope. The number of laser shots may vary from 1,200 to several thousand, according to severity of disease and response, and two or three treatment sessions are usually undertaken at weekly intervals. It is useful at the start of treatment to delineate the macula by a line of laser burns linking the superior and inferior temporal vascular arcades. Treatment should then begin 2–3 disc diameters from the centre of the macula. It is advisable to treat the inferior retina in the first session because any subsequent vitreous haemorrhage may obscure its view. Direct laser treatment of new vessels is rarely indicated.

167

168

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

COMPLICATIONS OF LASER TREATMENT • Macular oedema, which is usually transient. • Constriction of the visual field with implications for night vision and driving. • Retinal haemorrhage. • Rupture of Bruch’s membrane and development of a choroidal neovascular membrane. • Suprachoroidal effusion. • Uveitis. • Fibrous tissue contraction precipitating a tractional retinal detachment. Risk of these complications is minimized by using lower energy levels and applying the treatment over several sessions.

INDICATIONS FOR VITRECTOMY Pars plana vitrectomy techniques have improved over recent years and the indications for a vitrectomy in the context of proliferative diabetic retinopathy have widened but the main ones remain: (1) persistent severe vitreous haemorrhage; and (2) tractional macular retinal detachment. The timing of a vitrectomy depends on a variety of factors, but the Diabetic Retinopathy Vitrectomy Study reported that in type 1 diabetic patients early intervention (within six months of a dense vitreous haemorrhage) produced better visual results compared with deferring surgery for a year. In patients with type 2 diabetes, however, deferment did not alter outcome. Other factors which may influence the timing of vitrectomy include: • Severity of haemorrhage and prior state of the retina (the more severe the retinopathy status, the sooner surgery should be considered). • Extent of panretinal photocoagulation before haemorrhage. • Visual potential. • Presence of tractional macular detachment. • Extensive neovascularization, refractory to the effects of laser.

Additional indications for vitrectomy • Traction on the optic disc or peripapillary retina. • Macular traction or distortion, if macular function is otherwise good. • Significant premacular haemorrhage, which if left untreated may produce significant fibrosis of the overlying posterior hyaloid with subsequent tractional sequelae. • Significant fibrous proliferation anterior to the macula, reducing vision.

CHAPTER 18 • PROLIFERATIVE DIABETIC RETINOPATHY

Complications of vitrectomy • • • • • • •

Retinal detachment. Cataract. Endophthalmitis. Elevated intraocular pressure; usually transient. Corneal epithelial defects. Persistent vitreous haemorrhage. Recurrent vitreous haemorrhage. This may result from residual neovascularization (or subsequent neovascularization) at the vitreous base, or from fibrovascular ingrowth through sclerotomy sites. Visual improvement following vitrectomy has been reported in 59–83% of patients, with greater than 80% retaining a clear vitreous cavity. However, the complications following vitrectomy can be considerable and rates of no light perception have been reported at over 20%.

CURRENT ISSUES • •

• •

Laser photocoagulation remains the mainstay of treatment for proliferative diabetic retinopathy. Since the primary pathogenic mechanism of proliferative retinopathy is ischaemia-induced formation of growth factors, treatments that block angiogenic pathways, e.g. VPF release and action, are likely to provide effective prevention or lessen severity. Pharmacological vitreolysis may be a possible future technique for separating the vitreous from the retina. The indications for vitrectomy are expanding as surgical techniques improve.

FURTHER READING Early Treatment Diabetic Retinopathy Study Research Group. Early photocoagulation for diabetic retinopathy. ETDRS report number 9. Ophthalmology 1991; 98: 766–785. Flynn HW Jr, Chew EY, Simons BD, Barton FB, Remaley NA, Ferris FL 3rd. Pars plana vitrectomy in the Early Treatment Diabetic Retinopathy Study. ETDRS report number 17. The Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology 1992; 99: 1351–1357. The Diabetic Retinopathy Vitrectomy Study Research Group. Early vitrectomy for severe vitreous haemorrhage in diabetic retinopathy. Arch Ophthalmol 1990; 108: 958–964.

169

170

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

The Diabetic Retinopathy Research Study Group. Four risk factors for severe visual loss in diabetic retinopathy: the third report from the Diabetic Retinopathy Study. Arch Ophthalmol 1979; 97: 654–655. The Early Treatment Diabetic Retinopathy Study Research Group. Techniques for scatter and local photocoagulation treatment of diabetic retinopathy: Early Treatment Diabetic Retinopathy Study Report no. 3. Int Ophthalmol Clin 1987; 27: 254–264.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Hean-Choon Chen FRCS, FRCOphth CHAPTER 19 NON-RETINAL DIABETIC OCULAR COMPLICATIONS

INTRODUCTION Whilst retinopathy is the principal ocular complication of diabetes, other associated ocular pathologies can also give rise to significant visual deficit. The following conditions have been linked with diabetes: • Cataract formation. • Glaucoma. • Uveitis. • Retinal vascular occlusion. • Ocular nerve palsies..

CATARACT It is not always easy to demonstrate the exact cause of a cataract because it is a common problem, particularly in people of more advanced age. However, it is unusual for a young, otherwise healthy lens to develop a cataract and therefore a cataract in a young diabetic patient is likely to be secondary to the presence of chronic hyperglycaemia. There is evidence of an inverse association between glycaemic control and lens clarity and, like retinopathy, rapid improvement in glycaemic control may also adversely affect the lens. The lens opacities which occur in diabetes frequently take the form of a cortical cataract, with white dots or specks developing predominantly in the anterior and posterior subcapsular regions. This form of cataract is similar to the sugar cataracts in experimental diabetic animals and is more likely to be seen in poorly controlled type 1 diabetic patients. Sugar cataracts can develop fairly rapidly, and improved control retards their progression. It is also recognized that a nuclear-sclerotic type of cataract, which is usually age related, can develop at an earlier age in patients with diabetes. The polyol pathway has been implicated in cataract formation. Historically, it has been thought that high levels of intralenticular sorbitol, converted from glucose, raises lenticular cellular osmotic pressure; and lens cell membranes and the capsule of the lens are relatively impermeable to sorbitol. The resulting increased intake of water causes lens cells to swell, disrupting their function and causing rupture with loss of lens clarity. However, this osmotic hypothesis may not be as important a pathogenic mechanism as previously assumed and other effects of increased aldose reductase activity, such as oxidative stress, may play a greater role in cataract formation. Glycation of lens proteins is a pathological mechanism that leads to crosslinking and also cataract formation, since the orientation of proteins is crucial

171

172

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

to lens clarity. Accumulation of glycation end-products also accentuates the oxidative damage to lens proteins, especially with the lens exposed to light. A nuclear sclerotic cataract is more likely to develop following a pars plana vitrectomy for the complications of proliferative retinopathy. The majority of patients undergoing cataract surgery will have a good visual outcome. However, the complication rate is higher in diabetic compared with non-diabetic patients. The complications of cataract extraction in patients with diabetes include: • Increased risk of cystoid macular oedema, which is more likely to be persistent. It is therefore important to apply macular laser treatment for preexisting maculopathy, if possible, before contemplating cataract surgery. • Increased risk of post-operative uveitis, especially in the presence of active proliferative retinopathy. It is therefore desirable to treat proliferative disease with panretinal laser treatment before undertaking cataract surgery, but this may not be possible because of a dense cataract, in which case laser treatment should be undertaken in the operating room with the indirect laser after cataract removal. There is some evidence that use of heparin surface-modified intraocular lenses reduces the degree of postoperative uveitis and lens surface deposition of giant cells. • Increased incidence of iris neovascularization. • A larger than usual anterior capsular opening should be created since the capsulotomy is more likely to contract postoperatively in patients with retinopathy. • Increased risk of endophthalmitis. Cataract surgery may be more difficult to perform in patients with diabetes because pupillary dilatation may be limited, there may be posterior synechiae present and the presence of iris neovascularization increases the risk of a haemorrhage peri-operatively.

GLAUCOMA There is conflicting evidence as to whether the incidence of chronic open angle glaucoma is more common in diabetic patients. On balance, there is probably a slightly increased risk. This is in addition to the higher risk of angle-closure glaucoma in neovascular glaucoma, whereby neovascularization develops in the anterior chamber angle leading to its occlusion. Neovascular glaucoma is a difficult condition to manage. Panretinal laser treatment should be applied if iris neovascularization is present in order to prevent neovascular glaucoma. The relatively recent introduction of diode laser cycloablation has added to the armementarium of treatment options. Following a trabeculectomy, diabetic patients have an increased incidence of late-onset endophthalmitis.

CHAPTER 19 • NON-RETINAL DIABETIC OCULAR COMPLICATIONS

RETINAL VASCULAR ABNORMALITIES Several types of vascular abnormality are more common in patients with diabetes. These include: • Central retinal vein occlusion (CRVO) (Fig. 19.1); • Branch retinal vein occlusion (BRVO) (Fig. 19.2); and • Ocular ischaemic syndrome. These conditions can be mistaken for diabetic retinopathy, but are distinguished by the characteristic distribution of various vascular lesions, principally haemorrhages, exudates and cotton wool spots and varying degrees of retinal oedema. In the venous occlusive diseases, the retinal veins are usually more tortuous, either generally (CRVO) or segmentally (BRVO). In CRVO, there may be varying degrees of swelling of the optic disc. It is usually possible to determine

Fig. 19.1 Central retinal vein occlusion (CRVO).

Fig. 19.2 Branch retinal vein occlusion (BRVO).

173

174

SECTION III • DIABETIC RETINOPATHY AND ASSOCIATED OPHTHALMIC DISORDERS

the presence of a collateral circulation either at the optic disc (CRVO) or at the boundaries of the area affected by a BRVO. These can sometimes be mistaken for IRMA, or neovascularization, but can usually be differentiated from each other clinically or with the aid of fluorescein angiography (Fig. 19.3). The ocular ischaemic syndrome usually results from occlusive carotid disease and produces a combination of retinal signs which may be difficult to differentiate from diabetic retinopathy, although there may be a preponderance of cotton wool spots and vision may deteriorate rapidly in a way that is disproportionate to the degree of retinopathy. Fluorescein angiography will demonstrate that the arm-to-retina time (i.e. the time taken from injection of fluorescein to its first appearance in the retinal vessels) is significantly reduced. Carotid Doppler examination will be helpful in diagnosing abnormal flow in the carotid arteries.

UVEITIS Uveitis in a diabetic patient may be due to ischaemia of the anterior segment, especially in the presence of iris neovascularization. A breakdown in the blood-aqueous barrier reflects increasing severity of retinopathy, giving rise to varying degrees of anterior chamber flare.

EXTERNAL OCULAR MUSCLE PALSIES There is an increased incidence of ischaemia-related nerve palsies affecting the function of the external ocular muscles. The most commonly affected nerve is the 6th cranial nerve (abducens) giving rise to a failure of abduction and therefore producing horizontal diplopia. The 3rd and 4th cranial nerves are less frequently affected. In the majority of cases, the function of the nerve recovers fully.

Fig. 19.3 Fluorescein angiogram in branch retinal vein occlusion demonstrating considerable collateral vessel development.

CHAPTER 19 • NON-RETINAL DIABETIC OCULAR COMPLICATIONS

CURRENT ISSUES • •

•

Cataract development is the most common ocular complication in diabetes after retinopathy. Cataract surgery in diabetic patients with retinopathy has a higher incidence of postoperative complications, including cystoid macular oedema, uveitis and endophthalmitis. The use of heparin surfacemodified intraocular lenses may lessen the severity of postoperative uveitis. Panretinal laser treatment in patients with active proliferative retinopathy performed before cataract surgery, or at the time or surgery using the indirect ophthalmoscope, may reduce the risk of postoperative complications.

FURTHER READING Chew EY, Benson WE, Remaley NA et al. Results after lens extraction in patients with diabetic retinopathy: early treatment diabetic retinopathy study report number 25. Arch Ophthalmol 1999; 117: 1600–1606. Dowler JG, Sehmi KS, Hykin PG, Hamilton AM. The natural history of macular oedema after cataract surgery in diabetes. Ophthalmology 1999; 106: 663–668. Ellis JD, Evans JM, Ruta DA et al. DARTS/MEMO collaboration. Diabetes Audit and Research in Tayside Study. Medicines Monitoring Unit. Glaucoma incidence in an unselected cohort of diabetic patients: is diabetes mellitus a risk factor for glaucoma? Br J Ophthalmol 2000; 84: 1218–1224. Trocme SD, Li H. Effect of heparin-surface-modified intraocular lenses on postoperative inflammation after phacoemulsification: a randomized trial in a United States patient population. Heparin-Surface-Modified Lens Study Group. Ophthalmology 2000; 107: 1031–1037. West JA, Dowler JG, Hamilton AM, Boyd SR, Hykin PG. Panretinal photocoagulation during cataract extraction in eyes with active proliferative diabetic eye disease. Eye 1999; 13: 170–173.

175

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

SECTION IV MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Richard Donnelly MD, PhD, FRCP, FRACP CHAPTER 20 PATHOPHYSIOLOGY AND POTENTIAL TARGETS FOR THERAPEUTIC INTERVENTION

INTRODUCTION Sustained exposure of vascular tissues to hyperglycaemia plays a major role in the pathogenesis of diabetic angiopathy. There is a particularly strong relationship between serum glucose levels and microvascular disease, but macrovascular events are also related to glycaemic exposure, even in the nondiabetic population (Table 20.1). There is some debate about which marker of glycaemic control has the strongest clinical relationship with vascular disease. While fasting plasma glucose levels and HbA1C relate closely to the incidence of diabetic retinopathy, recent evidence suggests that postprandial hyperglycaemia may be of greater prognostic value in regard to macrovascular events (Figs 20.1 & 20.2). There is also a relationship between HbA1C and the incidence of heart failure (Table 20.2). Vascular tissues exposed to high circulating levels of glucose develop a number of clinical and biochemical abnormalities, including structural and functional alterations to endothelial cells, vascular smooth muscle cells, glomeruli and mesangial cells, and cardiomyocytes. Recent studies have provided clearer insights into the underlying biochemistry and the mechanisms by which hyperglycaemia causes vascular disease. Four principal pathways are involved (Table 20.3).

ADVANCED GLYCOSYLATION END-PRODUCTS A major glucose-dependent pathway of cardiovascular damage is the process of advanced glycation. This pathway involves a spontaneous, non-enzymatic reaction between glucose and NH2-groups on tissue proteins, lipids and nucleic acids, particularly the long-lived structural proteins such as collagen, leading to the formation of advanced glycation (glycosylation) end-products (AGEs). This process involves a series of biochemical reactions (known classically as the Maillard reaction), initially resulting in the formation of Amadori (early glycation) products, the most well known example being HbA1C, followed by a series of reactions generating a range of intermediates, e.g. 3deoxyglucosone, and ultimately forming in a variety of AGEs (Fig. 20.3). Advanced glycation occurs over a period of weeks, and the early steps in the Maillard reaction are glucose concentration-dependent. The formation of AGEs is catalyzed by transitional metals and is inhibited by antioxidant compounds such as ascorbic acid (vitamin C). If oxidation accompanies glycation, the resultant products are also known as glycoxidation products. The AGEs pentosidine and Nε-[carboxymethyl]-lysine (CML) are examples of glycoxidation products (Fig. 20.3).

179

180

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

Age-adjusted death rates: all-cause, CVD, IHD & non-cardiovascular Glycated haemoglobin (%)

Cause of death

<5 5-.5.4 5.5-6.9 ≥7 (n=1204) (n-1606) (n=1611) (n=81)

Selfreported diabetes (n=160)

χ2 (linear trend) P value

All causes (n=135) Age adjusted rate/100 1.65 (18) Relative risk 1.00

2.33 (35) 1.41

3.43 (61) 2.07

4.35 (5) 2.64

5.92 (16) 3.59

40.8 <0.001

Cardiovascular disease (n=60) Age adjusted rate/100 0.50 (5) Relative risk 1.00

1.27 (19) 2.53

1.24 (22) 2.46

2.54 (3) 5.04

4.11 (11) 8.16

31.8 <0.001

Ischaemic heart disease (n=42) Age adjusted rate/100 0.31 (3) Relative risk 1.00

0.86 (13) 2.74

0.87 (15) 2.77

1.63 (2) 5.20

3.43 (9) 10.91

29.0 <0.001

Non-cardiovascular disease (n=75) Age adjusted rate/100 1.15 (13) Relative risk 1.00

1.06 (16) 0.92

2.19 (39) 1.91

1.81 (2) 1.58

1.82 (5) 1.58

11.8 <0.001

BMJ 2001; 322: 15–18. Number of persons with macular oedema status 2 or 3 in either or both eyes) × 100.

Table 20.1 Data from the Epic-Norfolk study showing the relationship between HbA1C and macrovascular events even in the non-diabetic population. Adapted from BMJ 2002; 322: 15–18.

Number of events, persons, person-years & age-adjusted rates per 1000 P-Y of heart failure hospitalization and/or death by HbA1C All (n=48 858)

HbA1C

n/N

P-Y

Rate per 103 P-Y (95% CI)

<7 7 to <8 8 to <9 9 to <10 ≥ 10

145/10,631 197/10,692 181/9,238 172/7,354 240/10,943

21,963 23,417 20,808 16,576 23,594

4.5 (2.9–7.0) 5.8 (3.8–8.9) 6.3 (4.1–9.7) 8.3 (5.5–12.6) 9.2 (6.2–13.8)

Table 20.2 There is a relationship between HbA1C and heart failure. These data show hospitalizations and/or deals from heart failure at different HbA1C values. Adapted from Circulation 2001; 103: 2668.

CHAPTER 20 • PATHOPHYSIOLOGY AND THERAPEUTIC INTERVENTION

80

Adjusted incidence per 1000 person years (%)

Myocardial infarction Microvascular endpoints

60

40

20

10 5

6

7

8

9

10

11

Updated mean HbA1c concentration (%)

Fig. 20.1 Data from the UK Prospective Diabetes Study (UKPDS 35) showing the relationship between mean updated HbA1C concentration and the adjusted incidence (per 1,000 person years) of acute myocardial infarction and microvascular endpoints. Adapted from BMJ 2000; 321: 405–412.

Major pathways that have been implicated in the pathogenesis of hyperglycaemia-induced vascular injury • Non-enzymatic glycation formation of advanced glycosylation end products (AGEs) • Polyol pathway: aldose reductase-mediated changes in sorbitol and myoinositol • Protein kinase C (PKC) activation: formation of diaclyglycerol and PKC activation • Redox potential alterations: changes in free radicals and oxidation state

Table 20.3 Major pathways of glycaemic vascular dysfunction.

181

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

1.3

1.21

1.21 1.2 Relative risk

182

1.16 1.13 1.10

1.1 1.07 1.05

1.0

1.00

1.00

0

1

2

2-hour-PG mmol/L

3

0

5

10

15

20

SBP mm/Hg

Fig. 20.2 Data from the DECODE study showing that 2-hour post-challenge glucose concentration is as powerful as systolic BP in determining all-cause mortality.

The formation of reactive intermediate products during Amadori rearrangement is a crucial step in the Maillard reaction. These compounds are known as α-dicarbonyls or oxoaldehydes and include such products as 3deoxyglucosone (3-DG) and methylglyoxal (MGO). 3-DG is formed by nonoxidative rearrangement and hydrolysis of Amadori adducts and from fructose-3-phosphate, which is a product of the polyol pathway (Fig. 20.3). MGO is also formed from non-oxidative mechanisms in anaerobic glycolysis and from oxidative decomposition of polyunsaturated fatty acids.

Carbonyl stress MGO, 3-DG and glyoxal (collectively the α-oxoaldehydes) are formed from various steps in the glycation process via degradation of glucose or Schiff’s bases in early glycation, or from Amadori products such as fructosamine in the intermediate stages of glycation (Fig. 20.4). Thus, α-oxoaldehydes appear to be important biochemical mediators through which glucose goes on to form AGEs by the classical Maillard reaction, the polyol pathway, as well as from in vivo factors such as the catabolism of ketone bodies and threonine.

α-oxoaldehydes

ion

Via catabolism of ketone bodies or threonine

MGO

rad at

Via fragmentation of triose phosphate

Amadori product

e.g. Fructosamine

Intermediate glycation product

Schiff's base

Fructose-3phosphate

Early glycation product

Fructose

Lipid peroxidation

Glyoxal

Glucose

Amino group of proteins

Sorbitol

Deg

In vivo sources of α-oxoaldehydes

Polyol pathway

CHAPTER 20 • PATHOPHYSIOLOGY AND THERAPEUTIC INTERVENTION

Classic rearrangement

3-DG

CML CEL PYRRALINE GOLD MOLD DOLD Advanced glycated end-products

α-oxoaldehydes

Oxidative pathway

Non-oxidative pathway

CML PENTOSIDINE PYRRALINE Advanced glycated end-products

Fig. 20.3 Pathways of AGE formation incorporating the polyol pathway and AGE formation via α-oxoaldehydes. AGEs are biochemically heterogeneous but associated with a number of structural and functional abnormalities particularly involving cross-linking of long-lived structural tissue proteins, e.g. collagen. Diabetologia 2001; 44: 129–146.

The accumulation of reactive dicarbonyl precursors is known as carbonyl stress, i.e. the accumulation of α-oxoaldehyde intermediates which may progress to form oxidative AGEs such as CML and pentosidine or non-oxidative AGEs derived from 3-DG or MGO. The process of carbonyl stress has been observed in both diabetes and end-stage renal failure, and has been implicated in the accelerated vascular damage characteristic of both conditions.

183

184

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

Pentosidine Arginine N N N H

N

Lysine

CML

COOH CH2 NH

NH

(CH2)2

O

CH

C

MOLD Lysine

N

N

Lysine

Fig. 20.4 The structure of three major AGEs.

Structure of AGEs Pentosidine and CML are the best characterized AGEs, and are often referred to as glycoxidation products (Fig. 20.4). Antioxidants reduce the formation of these two AGEs. In addition, lipid oxidation also plays a role in CML formation, and therefore CML appears to be a general biomarker of oxidative stress resulting from both carbohydrate and lipid oxidation reactions.

CHAPTER 20 • PATHOPHYSIOLOGY AND THERAPEUTIC INTERVENTION

Glycation of haemoglobin forms HbA1C, which has been described as an Amadori product but HbA1C is not an AGE. HbA1C is an indicator of glycaemic control for the preceding 6–12 weeks, whereas advanced glycation reflects a process that can occur over a much longer time period.

AGE and cross-link formation Although AGEs have a variety of chemical structures, a common consequence of AGE formation is covalent cross-linking (Table 20.4). Structural tissue proteins, e.g. collagen, are particularly affected by the cross-link process, which mainly involves lysine residues. AGE-induced cross-link formation leads to increased stiffness of the tissue protein matrix, in part due to increased matrix protein production accompanied by decreased matrix degradation. This process has a major effect on tissue remodelling (Table 20.4). Increased AGE-mediated cross-link formation occurs with advancing age and is markedly accelerated in diabetes. The physiological consequences of cross-link formation include sclerosis of renal glomeruli, thickening of the capillary basement membrane and atherosclerosis progression.

AGE receptors Several AGE receptors have been identified, including macrophage scavenger receptor types 1 and 2, receptor for AGE (RAGE), oligosaccaharyl transferase48 (AGE-R1), 80K-H phosphoprotein (AGE-R2) and galectin-3 (AGE-R3).

Harmful effects of AGEs Possible roles of AGEs and age receptors in complications of diabetes Diabetes atherosclerosis Vascular tissue AGE accumulation → protein cross-linking → oxidative damage ↑ Vascular matrix → thickening and narrowing of lumen ↑ Endothelial cell permeability and procoagulant activity → thrombosis Mononuclear cell chemotaxis/activation → cytokine and growth factor release: T-cell stimulation → interferon-γ production ↑ Macrophage uptake of AGE–LDL → atheroma Diabetic kidney disease ↑ Mesangial matrix secretion ↑ Basement membrane deposition ↑ Vascular permeability ↑ Growth factor secretion Glomerular hypertrophy → glomerulosclerosis

Table 20.4 Harmful effects of AGEs.

185

186

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

These receptors are expressed on a wide range of cells including vascular smooth muscle cells, monocytes, macrophages, endothelial cells, podocytes and microglia. In particular, expression of some of these receptors is increased in diabetes, e.g. galectin-3. Similarly, RAGE expression is increased in the blood vessels and kidneys of diabetic patients. The best characterized AGE receptor is RAGE, which is a multiligand member of the immunoglobulin superfamily. Experimental studies have shown that AGE binding to the RAGE receptor on macrophages leads to oxidative stress and activation of the transcription factor NF-κB. In addition, animal studies using a soluble RAGE which blocks the RAGE receptor have shown suppression of vascular lesion formation as well as improved vascular permeability and dysfunction.

Aminoguanidine Studies involving inhibition of AGE formation using the hydrizine derivative, aminoguanidine, have provided more direct evidence supporting the role of AGEs in diabetic vascular injury. Aminoguanidine treatment has been shown to decrease AGE formation in rat tissues, including the aorta, kidney and retina. Aminoguanidine does not inhibit the formation of early glycated products but acts at a site beyond this step (one that is not yet clearly defined), leading to reduced AGE formation. Experimental studies have shown that aminoguanidine treatment of diabetic rats reduces micro-aneurysm formation and pericyte loss. In addition, beneficial effects on peripheral nerve function and structure have also been reported. Under normal conditions, advanced glycated haemoglobin (Hb-AGE) accounts for 0.42% of circulating haemoglobin, but this increases to 0.75% in diabetic subjects. In an early clinical trial, aminoguanadine 1,200mg/day reduced levels of Hb-AGE by 28% after one month in subjects with diabetes for an average duration of 20 years. A related compound, pimagedine, has also been shown in clinical studies to retard the progression of diabetic nephropathy in optimally treated type 1 diabetic subjects. The side-effect profile of aminoguanidine makes it unsuitable for widespread clinical use. Early studies showed changes in the blood film resembling pernicious anaemia, but of greater concern is the development of antinuclear antibodies and clinical features of vasculitus and glomerulonephritis. New, more potent inhibitors of advanced glycation have been developed recently, including ALT-462 and ALT-486, which are approximately fiveand 20-fold more potent than aminoguanidine in vitro. In addition, the cross-link breaker, PTB, is a thiazolium compound which cleaves di-ketone bridges between adjacent carbonyl groups. Daily intraperitoneal injections of PTB in STZ-diabetic rats have been shown to ameliorate AGE-induced

CHAPTER 20 • PATHOPHYSIOLOGY AND THERAPEUTIC INTERVENTION

cross-links. Pyridoxamine (also known as pyridorin) is another pharmacological inhibitor of the conversion of Amadori intermediates to AGEs. Similarly, the compound OPB-9195 reduces AGE accumulation in glomeruli of experimental diabetic rats, in part through suppression of TGF-β and VEGF expression.

THE POLYOL PATHWAY In this metabolic pathway, glucose is reduced to sorbitol by the enzyme aldose reductase (Fig. 20.5). Under physiological conditions of normoglycaemia, the polyol pathway plays a very minor role in glucose disposal, but in diabetic states vascular cells undergoing insulin-dependent glucose uptake produce increased amounts of sorbitol which is slowly metabolized to fructose. The accumulation of sorbitol inside vascular cells results in osmotic

Vascular cell membrane GLUT-1 Aldose reductase Sorbitol

G-3-P

Fructose

GAP

Phosphatidic acid

Pyrovate

DAG Unesterified fatty acids

Lactate

Polyol pathway

Glycolysis

PKC activation de novo synthesis of DAG and PKC activation

Fig. 20.5 Glucose enters vascular cells via the GLUT-1 transporter, and most glucose undergoes glycolysis with a minor component (<5%) entering the polyol pathway. Under conditions of hyperglycaemia and increased glycolysis, glyceraldehyde-3phosphate (GAP) is converted to phosphatidic acid, which in turn results in increased de novo synthesis of diacylglycerol (DAG). Newly synthesized DAG is rich in palmitate, which preferentially activates the PKC-β isoforms in vascular cells. Unesterified free fatty acids augment DAG-mediated PKC activation.

187

188

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

stress, decreased myoinositol content, abnormal phosphoinositide metabolism, and decreased Na+/K+-ATPase activity. The polyol pathway and formation of sorbitol have been implicated in diabetic complications, including retinopathy, neuropathy, cataracts, nephropathy and corneal disease. In experimental animals, inhibition of aldose reductase leads to reduced neuropathy, and prevention of myoinositol depletion and reduced ATPase activity in the kidney. Other experimental studies have shown that aldose reductase inhibition produces significant reductions in proteinuria and other markers of diabetic nephropathy in animal models. Unfortunately, clinical studies using aldose reductase inhibitors over the past 20 years have been disappointing and this therapeutic approach has been largely abandoned as an option for preventing diabetic vascular dysfunction.

CURRENT ISSUES •

•

•

Defining the biochemical pathways involved in AGE formation, and the mechanisms of AGE-induced protein cross-linking, will lead to the identification of new pharmacological approaches that mimic the profile of aminoguanadine but have acceptable safety characteristics for routine clinical practice. A number of compounds are in development. Experimental studies have clearly suggested that oxidative stress plays an important part in the vascular damage associated with AGE formation and PKC activation, but so far clinical trials of vitamin C and vitamin E supplementation have been disappointing in terms of reducing cardiovascular events. Aldose reductase inhibitors have been largely disappointing in clinical trials, perhaps because only 5% of glucose transported into vascular cells, even under conditions of hyperglycaemia, passes through the polyol pathway.

FURTHER READING Idris I, Gray S, Donnelly R. Protein kinase C activation: isozyme-specific effects on metabolism and cardiovascular complications in diabetes. Diabetologia 2001; 44: 659–673. Singh R, Barden A, Mori T, Beilin L. Advanced glycation end products: A review. Diabetologia 2001; 44: 129–146. Tsuchida K, Makita Z, Yamagishi S et al. Suppression of transforming growth factor β and vascular endothelial growth factor in diabetic nephropathy in rats by a novel AGE inhibitor, OPB-9195. Diabetologia 1999; 42: 579–588.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

CHAPTER 21 PROTEIN KINASE C

Richard Donnelly MD, PhD, FRCP, FRACP

INTRODUCTION It is now recognized that activation of protein kinase C (PKC) under conditions of hyperglycaemia is one of the principal mechanisms of vascular damage in patients with diabetes. Glucose is transported into vascular cells by GLUT-1 transporters and then metabolized, mostly via glycolysis (<5% is metabolized by the aldose reductase/polyol pathway, even under conditions of hyperglycaemia). GLUT-1 expression in vascular cells is up-regulated by high extracellular glucose concentrations and other local factors involved in diabetic angiopathy, e.g. hypoxia. The increase in glycolysis results in increased de novo synthesis of diacylglycerol (DAG), which is the main endogenous activator of a ubiquitous intracellular enzyme known as PKC. Several studies have shown both in animals and humans with diabetes that there is a widespread increase in DAG levels and PKC activity in different types of vascular cell (Table 21.1).

PROTEIN KINASE C (PKC): A MULTIFUNCTIONAL FAMILY OF ISOENZYMES It has long been recognized that adding and removing phosphate groups is one of the most important physiological mechanisms by which the activity

Diabetes-related activation of DAG-PKC pathway in vascular cells and tissues

Cells in culture Aortic Endothelial Aortic VSM Retinal Endothelial Retinal Pericytes Renal mesangial Renal glomerular

Species

DAG Content

PKC Activity

Isoforms Activated

Rat, bovine Rat, human Bovine Bovine Rat Rat

↑ ↑ ↑ NM ↑ ↑

↑ ↑ ↑ ↑ ↑ ↑

β -β -βII, -δ -βII, -δ -α, βI -α, βI

↑ ↑ ↑ ↑ NM

↑ ↑ ↑ ↑ ↑

-β, -ε -βII, δ -β -δ, -βI, -α -βII

Tissues Heart Rat, human Retina Rat, canine Aorta Rat, canine Glomeruli Rat, mouse Monocytes Human ↑ = increased; NM = not measured

Table 21.1 Diabetes-related activation of DAG–PKC pathway in vascular cells and tissues.

189

190

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

of cellular proteins (e.g. enzymes and receptors) is regulated. For example, key metabolic enzymes such as glycogen synthase are switched on and off by intracellular kinases (enzymes that add phosphate groups) and phosphatases (enzymes that remove phosphate groups), which are themselves regulated by other biochemical signals, e.g. hormones and growth factors. Intracellular kinases are broadly divided into two different types: those that phosphorylate proteins at tyrosine residues (known as tyrosine kinases) and those that phosphorylate serine and threonine sites (known as serine/threonine kinases). There are two major serine/threonine kinases that are widely distributed in all tissues: cyclic-AMP-dependent protein kinase (also known as protein kinase A) and PKC. PKC was first described over 20 years ago as a single, proteolytically activated kinase, and cancer biologists were the first to take a keen interest in this enzyme because early studies showed that tumour-promoting substances known as phorbol esters caused prolonged activation of PKC. Since then, however, it has become clear that PKC plays an important regulatory role in a variety of cellular responses, in addition to cell growth and differentiation, and that PKC is involved in gene expression, secretion of hormones and post receptor signalling. Thus, PKC phosphorylates (and thereby regulates) a large number of intracellular substrates, including proteins such as the insulin receptor and key metabolic enzymes involved in glucose transport and utilization. Although PKC was first described as a single enzyme, molecular and genetic studies over the last 10 years have shown that PKC is in fact a family of structurally and functionally related proteins which are derived from multiple genes (at least three) and from alternative splicing of single mRNA transcripts. Twelve isoenzymes of PKC have so far been cloned and characterized. They are classified into three groups according to their structural homologies (Table 21.2). Individual isoforms have different patterns of tissue distribution, substrate specificity and cofactor requirements. For example, the group A (classical) PKC isoforms (PKC-α, -βI and -βII) require the presence of both calcium and phospholipid for enzyme activation, whereas the group B (novel) PKC isoforms are calcium-independent and group C (atypical) PKC isoforms are both calcium- and phospholipid-independent (Table 21.2). The brain and liver contain virtually all PKCs, but most other tissues express only certain PKC isoforms. The different patterns of tissue expression reflect a complex multifunctional role for this family of kinases, but specific functions related to individual isoenzymes are incompletely understood. Activation and translocation of PKC in vascular cells correlates with circulating glucose concentrations, as illustrated in a recent clinical study using monocytes (Fig. 21.1).

CHAPTER 21 • PROTEIN KINASE C

EFFECTS OF DIABETES ON DAG-PKC ACTIVATION IN VASCULAR TISSUES Several studies have clearly demonstrated increased tissue levels of DAG and isoform-selective activation of PKC in a range of vascular cell types under conditions of clinical or experimental diabetes (Table 21.1). Increased intracellular

Protein kinase C isoforms Type

Isoform

Novel (n)

Ca** independent

α β γ δ ε η θ

Atypical (a)

Ca** independent

μ ζ ι⁄λ

Conventional (c) Ca** and Phospholipid-dependent

Distribution Widespread Widespread Brain Widespread Brain, hematopoietic tissue Heart, skin, lung Hematopoietictissue, skeletal muscle, brain Lung, epithelial cells Widespread Kidney, brain, lung

Table 21.2 Protein kinase isoforms.

Membrane PKC activity (pmol.min-1.mg protein-1)

150

100

50

0 0

10

20

30

Plasma glucose (mmol/l)

Fig. 21.1 In a recent clinical study PKC activity in the membrane sub-fraction of circulating monocytes was measured in 19 patients with diabetes (●) and 14 nondiabetic control subjects (●) and showed a linear correlation with circulating plasma glucose levels (r2 = 0.4, p = 0.0001). Adapted from Ceolotto, et al. Diabetes 1999; 48: 1316–1322.

191

192

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

release of DAG in response to high circulating glucose concentrations is the primary step leading to activation and translocation of PKC. Various species of DAG (varying in fatty acid composition) are generated from four principal sources (Fig. 21.2): (1) classical receptor-mediated, phospholipase C-catalyzed hydrolysis of inositol phospholipids; (2) via the release of DAG from phospholipase D-mediated hydrolysis of phosphatidylcholine (PC); (3) the release of free fatty acids (FFAs) from precursor lipids by the action of phospholipase A2; and (4) de novo synthesis of DAG from phosphatidic acid (PA). This latter pathway is mainly responsible for hyperglycaemia-induced DAG formation in a range of cardiovascular tissues, but high glucose levels also increase the turnover of PC. The excess DAGs that accumulate in diabetic vascular tissues are particularly rich in the FFA palmitate which suggests that pathways 2 and 4 are the principal sources of hyperglycaemia-induced DAG formation.

Hyperglycaemia

G-3-P

(4) de novo synthesis

Phosphatidic acid PLD (2)

Glycolysis Phosphoinositides

Phospharidylcholine

PLA2(3)

PLC (1)

Diacylglycerols

Monoacylglycerols

Lyso = PC -FFAs

Triacylglycerols

Fig. 21.2 Four principal pathways are involved in the generation of diacylglycerols in vascular tissues, but under conditions of hyperglycaemia de novo synthesis (4) and hydrolysis of phosphatidylcholine (2) are particularly important. See text for further details.

CHAPTER 21 • PROTEIN KINASE C

Experimental studies have also shown that DAG-mediated activation of PKC is augmented by specific FFAs of varying chain lengths. For example, unesterified fatty acids and their CoA esters (especially arachidonic, oleic, linoleic and linolenic acids) appear to activate PKC synergistically with DAG (Fig. 21.3), and it has been suggested that cis-unsaturated fatty acids act as ‘PKC enhancer’ molecules. Thus in diabetes increased FFA levels, particularly in the postprandial state, may enhance hyperglycaemia-induced PKC activation, independently of (and in addition to fuelling) de novo synthesis of DAG. There is evidence that different species of DAG preferentially activate one or more PKC isoforms in different tissues, and George King’s Group at the Joslyn Diabetes Centre in Boston, USA, first made the important observation that PKC isoforms are differentially up-regulated in different tissues in

Hyperglycaemia

VSM contractility

Cardiomyopathy Vit E

PA Na--K--ATPase activity

DAG FFAs

+ Isozymeselective PKC activation (esp. PKC-β)

Endothelial dysfunction and activation

Monocyte activation

Extracellular matrix production

Angiogenesis

Vascular permeability

Fig. 21.3 Hyperglycaemia-induced accumulation of diacylglycerol (DAG) is ameliorated, in part, by vitamin E supplementation, which activates DAG kinase. Free fatty acids augment DAG-induced activation of specific PKC isoforms, especially PKC-β, which in turn leads to a number of important pathophysiological mechanisms involved in the structural and functional abnormalities associated with diabetic cardiovascular disease.

193

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

response to hyperglycaemia. In particular, they showed that increased activity of PKC-β is the dominant PKC response in macrovascular and renal tissues, including vascular smooth muscle and endothelial cells, as well as the retina. Furthermore, PKC-βII seems to be the main PKC isoform activated in vascular tissues in response to high glucose levels, whereas in glomerular cells PKCβI is the predominant isoform activated by hyperglycaemia (Fig. 21.4).

ACTIVATION OF PKC-β The pathophysiological consequences of PKC activation in vascular tissues will be addressed in detail in the next two chapters (Fig. 21.3), but it seems clear that hyperglycaemia-induced formation of certain species of DAG leads to preferential activation of PKC-βII in vascular tissues, including the retina, and preferential activation of PKC-βI in glomerular and mesangial cells within the kidney. Activation and translocation of these isoforms from the cytosol to the plasma membrane correlates with plasma glucose levels (Fig. 21.1) and leads to a number of undesirable pathophysiological changes involving membrane transport, gene transcription and local vasoactive hormone secretion/responsiveness (Fig. 21.3).

VITAMIN E It has been shown that the accumulation of DAG in vascular tissues in hyperglycaemic states is ameliorated, in part, by D-α-tocopherol (vitamin E), which

Aorta PKC isoforms (% control)

194

PKC-α

Heart PKC-β11

PKC-α

PKC-β11

200

100

0

C D Memb

C D Cyto

C D Memb

C D Cyto

C D Memb

C D Cyto

C D Memb

C D Cyto

Fig. 21.4 In diabetic animal models it was shown that individual PKC isoforms are differentially up-regulated under conditions of hyperglycaemia. In particular, in aorta and heart, PKC-βII was increased to a greater extent than PKC-α in the cellular membrane fraction. Adapted from Inoguchi, et al. Proc Natl Acad Sci 1992; 89: 11059–11063.

CHAPTER 21 • PROTEIN KINASE C

activates DAG kinase and promotes the conversion of DAG to PA (Fig. 21.3). Several experimental studies have shown that glucose-induced PKC activation is attenuated by vitamin E therapy, and that the functional consequences of PKC activation in the kidney and retina are reversed. This raises the possibility that vitamin E has therapeutic benefits via reducing the DAG-PKC pathway in diabetic vascular tissues.

CURRENT ISSUES • Hyperglycaemia-induced activation of PKC, especially PKC-β, appears to be a major pathway in the development of structural and functional abnormalities of vascular tissues in diabetes. Reducing the accumulation of DAG using vitamin E supplementation, combined with selective PKC isoenzyme inhibition, provides a logical therapeutic approach to ameliorating and reversing diabetic microangiopathy, especially in the eyes and kidneys. • Numerous protein substrates are phosphorylated and thereby regulated in response to PKC activation, which in turn results in changes in cell growth and differentiation; contractile function; matrix production; vascular permeability; and neovascularization. • Different species of DAG (varying in fatty acid composition) seem to activate different PKC isoforms in various tissues, and there is particular interest in the clinical relationships between meal-related increases in glucose and triglyceride levels, PKC activation and diabetic vascular disease.

FURTHER READING Hug H & Sarre TF. Protein kinase C isoenzymes: divergence in signal transduction? Biochem J 1993; 291: 329–343. Inoguchi T, Battan R, Handler E et al. Preferential elevation of protein kinase C isoform βII and diacylglycerol levels in the aorta and heart of diabetic rats. Differential reversibility to glycaemic control by islet transplantation. Proc Natl Acad Sci USA 1992; 89: 11059–11063. Newton AC. Protein kinase C: structure, function and regulation; mini review. J Biol Chem 1995; 270: 28495–28498. Xia P, Inoguchi T, Kern TS et al. Characterization of the mechanism for the chronic activation of diacylglycerol- protein kinase C pathway in diabetes and hypergalactosaemia. Diabetes 1994; 43: 1122–1129.

195

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Richard Donnelly MD, PhD, FRCP, FRACP CHAPTER 22 PROTEIN KINASE C ACTIVATION AND VASCULAR PERMEABILITY

INTRODUCTION The vascular endothelium is a multifunctional barrier between the intravascular and tissue compartments; it is much more than an inert lining of blood vessels. Endothelial cells have antiadhesive and anticoagulant properties, modulate the effects of vasoconstrictor agonists, and through tight intercellular junctions control the permeability to large circulating molecules (Fig. 22.1). Leakage of macromolecules through the endothelial barrier is an early feature of diabetic microvascular disease and responsible for the increase in urinary albumin excretion rate (UAE) and the typical exudative changes in diabetic retinopathy. More importantly, increased endothelial permeability — as indicated clinically by a raised UAE — confers a substantial increase in cardiovascular risk. Studies such

Glucose

Hypoxia

VEGF

Cytokines Hormones

Monolayer

Signal transduction

Ca2+ NO PKC

Cytoskeletal proteins Shape change/rounding-up of EC Loss of tight junctions Permeability to macromolecules

Fig. 22.1 Endothelial monolayer with intercellular junctions. Cellular, hormonal and physical factors regulate endothelial permeability via signal transduction pathways that involve PKC, nitric oxide and calcium. These various stimuli lead to shape change and reduced intercellular communication, which in turn creates increased permeability of the monolayer to large molecules.

197

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

as the WHO Multinational Study of Vascular Disease in diabetes showed a clear relationship between proteinuria and reduced survival in both type 1 and type 2 diabetes (Fig. 22.2). Thus, endothelial barrier dysfunction is an early hallmark of widespread microvascular damage, but is also indicative of increased morbidity and mortality from macrovascular complications. Metabolic and haemodynamic abnormalities are responsible for the increases in endothelial permeability in patients with diabetes, but high glucose levels, in particular, via activation of protein kinase C (PKC), increase vascular permeability (Fig. 22.1).

MECHANISMS OF INCREASED ENDOTHELIAL PERMEABILITY The transport of fluid and solute across the endothelial barrier is governed by filtration pressure (i.e. ‘Starling forces’) and the local generation of cellderived mediators that influence endothelial barrier function. Several morphological and functional abnormalities of endothelial cells are associated with increases in vascular permeability (Fig. 22.1).

Intercellular gaps Adjacent endothelial cells form junctional complexes consisting of tight junctions and adherence junctions which are the sites of diffusional transport of solutes from the vascular to the interstitial space. The increase in transendothelial permeability in response to pro-inflammatory mediators such as

1.0 0.9 Survival probability

198

0.8 0.7

None

0.6

Light

0.5 Heavy

0.4 0.3 0.2 0.1 0.0

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Years of follow-up from base-line

Fig. 22.2 Survival according to the degree of proteinuria (non-, slight or heavy) at base-line among patients with type 2 diabetes. Reproduced with permission from Diabetic Medicine 1995; 12: 149–155.

CHAPTER 22 • PKC ACTIVATION AND VASCULAR PERMEABILITY

histamine and thrombin can occur via contraction or retraction of cells and the resultant formation of interendothelial cell gaps. ‘Rounding up’ of endothelial cells is a characteristic morphological change associated with widening of the intercellular junctions and increased trans-endothelial albumin flux. Intracellular contractile proteins such as F-actin in the microfilaments are responsible for the shape change of endothelial cells in response to inflammatory mediators such as histamine and thrombin.

Endothelial cell contraction vs. retraction The characteristic shape change of endothelial cells in response to inflammatory stimuli involves contraction of microfilaments within the cytoskeleton. In particular, phosphorylation of a key enzyme, myosin light chain kinase (MLCK), regulates the intracellular actin-myosin contractile mechanism. PKC plays an important role in phosphorylating MLCK and other acting-regulating proteins such as vinculin and talin which are important for maintaining cell-cell and cell-matrix contacts. Connexin-43 is another protein involved in tight junctions which is phosphorylated by PKC (Fig. 22.3).

A R PLC

G

Connexin 43

PIP2 DAG Ins (1,4,5)P3 PKC ER Ca2+

C

J

PKG

L-Arg NOS cGMP NO

GC GTP

Fig. 22.3 Signal transduction of agonist-induced increases in endothelial permeability. Nitric oxide and PKC are important, e.g. via PKC-mediated phosphorylation of connexion. Adapted from Am J Physiol 1997; 273: H2442–2451.

199

200

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

Activation of PKC by phorbol esters causes reorganization of actin and vinculin and disruption of junctional complexes in epithelial cells, consistent with the notion that PKC-dependent phosphorylation of actin-binding proteins is a critical signalling event responsible for shape change and loss of endothelial barrier function (Fig. 22.4). PKC also phosphorylates caldesmon and vimentin, two other cytoskeletal proteins in endothelial cells that are important for cell shape change.

HIGH-GLUCOSE INDUCED HYPERPERMEABILITY Several mechanisms are involved in hyperglycaemia induced hyperpermeability: • Increased formation of vascular permeability factor (VPF) (Fig. 22.5). • Advanced glycation end-product (AGE) mediated oxidative stress. • Intracellular calcium release and activation of nitric oxide synthase (NOS) (Fig. 22.3). • PKC activation and phosphorylation of intracellular contractile proteins, e.g. MLCK, causing shape change and rounding-up of endothelial cells. The biochemical mechanisms of glucose-induced hyperpermeability are not clearly understood, but activation of PKC plays an important role in each of the above pathways. For example, PKC inhibition blocks glucose-induced over-expression of VPF (Fig. 22.6).

PKC ACTIVATION AND ENDOTHELIAL PERMEABILITY Activation and translocation of PKC in endothelial cells has been associated with the hyperpermeability responses to a number of circulating factors, including thrombin, histamine and H2O2. In addition, non-specific PKC inhibitors such as H-7 block increases in trans-endothelial permeability in response to thrombin and H2O2. Thus, PKC activation appears to be a critical intracellular signalling event mediating the increase in endothelial permeability associated with a range of circulating factors (apart from hyperglycaemia). An increase in intracellular calcium concentration seems to be an important trigger in endothelial cell permeability, in part via activation of calciumdependent PKC isoforms. PKC-α and PKC-β isoforms are the predominant calcium-dependent PKC isoforms in endothelial cells, and elegant work using antisense oligonucleotides to PKC-β1 in human microvascular endothelial cells has shown that this isoform plays a critical role in phorbol ester-induced hyperpermeability (Fig. 22.7). Similarly, in bovine pulmonary microvascular endothelial cells, H2O2 increased the trans-endothelial permeability to albumin in parallel with increased translocation of PKC-β to the plasma membrane, suggesting that PKC-β activation mediates the H2O2-

CHAPTER 22 • PKC ACTIVATION AND VASCULAR PERMEABILITY

Albumin flux (% of control)

150 140 130 120 n.s 110 100 0 0

10

20

30

40

Glucose (mM)

520 mmol Mannose

170

Albumin flux (% of control)

160 150 140 130 120 110 100 0

0

10-11

10-10

10-9

10-8

10-7

10-6

TPA (M)

Fig. 22.4 Data showing the effects of different concentrations of glucose (upper panel) and the phorbol ester TPA (lower panel) on albumin flux across aortic endothelial monolayers. High glucose and phorbol ester-induced PKC activation are associated with dose-dependent increases in endothelial permeability. Adapted from Hempel et al. Circulation Research 1997; 81: 363–371.

201

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

3000

2500

5 mm glu 20 mm glu

2000 VPF (pg/ml)

202

1500

1000

500

0 3

6

12

24

Time (h)

Fig. 22.5 Vascular permeability factor (VPF) (also known as vascular endothelial growth factor, VEGF) plays an important role in retinal and glomerular protein leakage. Hyperglycaemia increases VPF peptide production by human vascular smooth muscle cells. VPF concentration in the culture media overlying human vascular smooth muscle (VSM) cells was determined using a specific ELISA after incubation of the cell mono-layers with control (glucose 5 mmol/l, purple bars) or high glucose medium (glucose 20 mmol/l, orange bars) for up to 24 hours. Reproduced from Williams et al. Diabetes 1997; 46: 1497–1503.

induced permeability response. Thus, several experimental studies indicate that the calcium requirement for the increase in endothelial permeability may be related to a requirement for activation of calcium-dependent PKC isoforms, particularly PKC-β. The importance of PKC in vascular permeability has been emphasized in other clinical conditions apart from diabetes, e.g. pre-eclampsia. Serum from pre-eclamptic women increased endothelial permeability in vitro in parallel with increased translocation of classic PKC isoforms. Furthermore, the hyperpermeability response to serum from pre-eclamptic women was attenuated using a non-specific PKC inhibitor.

CHAPTER 22 • PKC ACTIVATION AND VASCULAR PERMEABILITY

VPF/GAPDH ratio

1.5

1.0

0.5

0.0

NG

HG

HG+H-7

HG-CC

Albumin clearance rate (x10-2 μl/min)

Fig. 22.6 Hyperglycaemia stimulates increased gene expression for VPF, and this pathway is PKC-dependent. These data show VPF mRNA expression (relative to a housekeeping gene, GAPDH) under conditions of normal glucose (NG), high glucose (HG) and high glucose in combination with the non-specific PKC inhibitors, H-7 and chelerythrine chloride (CC). Reproduced with permission from Diabetes 1997; 46: 1497–1503.

20

Control PKC β1-AS

15

10

5

0 0

2 20 PMA concentration (nM)

Fig. 22.7 Human microvascular endothelial cells were transfected with retroviral vectors containing the antisense oligonucleotide for PKC-β1. Knockout of PKC-b1 attenuated the increase in albumin permeability induced by the phorbol ester PMA (an exogenous, non-specific PKC activator), confirming that PKC-β1 is a critical PKC isoform involved in PKC-dependent hyperpermeability responses. Adapted from J Cell Physiol 1995; 166: 249–255.

203

204

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

CURRENT ISSUES • A rise in intracellular calcium, combined with activation of calciumdependent PKC isoforms, especially PKC-α and PKC-β, serves as a common biochemical signalling pathway in the hyperpermeability responses of endothelial cells to a variety of circulating factors, especially high-glucose and VPF. • Hyperglycaemia stimulates the gene expression for VPF, a pathway that is blocked by PKC inhibitors, and VPF has been strongly implicated in the pathogenesis of exudative diabetic retinopathy. • Restoring endothelial barrier function in patients with diabetes and microvascular complications, e.g. with blood pressure lowering therapy, seems to improve cardiovascular outcome, but novel therapeutic approaches such as PKC inhibition would augment conventional blood pressure and glucose-lowering strategies for ameliorating microangiopathy.

FURTHER READING Bonnardel-Phu E, Wautier JL, Schmidt AM et al. Acute modulation of albumin microvascular leakage by advanced glycation end products in microcirculation of diabetic rats in vivo. Diabetes 1999; 48: 2052–2058. Huang Q & Yuan Y. Interaction of PKC and NOS in signal transduction of microvascular hyperpermeability. Am J Physiol 1997; 273: H2442-H2451. Haller H, Hempel A, Homuth V et al. Endothelial-cell permeability and protein kinase C in pre-eclampsia. Lancet 1998; 351: 945–949. Hempel A et al. High glucose concentrations increase endothelial cell permeability via activation of protein kinase C-α. Circ Res 1997; 81: 363–371. Hinder F et al. Nitric oxide and endothelial permeability. J Appl Phsiol 1997; 83: 1941–1944. Kuroki T et al. High glucose induces alteration of gap junction permeability and phosphorylation of connexin-43 in cultured aortic smooth muscle cells. Diabetes 1998; 47: 931–936. Lum H et al. Mechanisms of increased endothelial permeability. Can J Physiol Pharmacol 1996; 74: 787–800. Nagpala PG et al. PKC-β1 over expression augments phorbol ester-induced increase in endothelial permeability. J Cell Physiol 1995; 166: 249–255. Siflinger-Birnboim A et al. Activation of protein kinase C pathway contributes to hydrogen peroxide-induced increase in endothelial permeability. Laboratory Invest 1992; 67: 24–30. Williams B et al. Glucose-induced protein kinase C activation regulates vascular permeability factor mRNA expression and peptide production by human vascular smooth muscle cells in vitro. Diabetes 1997; 46: 1497–1503.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

Richard Donnelly MD, PhD, FRCP, FRACP CHAPTER 23 ROLE OF PROTEIN KINASE C ACTIVATION IN CARDIOVASCULAR AND RENAL COMPLICATIONS OF DIABETES

INTRODUCTION In clinical practice, metabolic and haemodynamic abnormalities contribute to the initiation and progression of diabetic vascular complications but hyperglycaemia is particularly important in the pathogenesis of microangiopathy. Hyperglycaemia-induced de-novo synthesis of diacylglycerol (DAG) and isoform-selective activation of protein kinase C (PKC), especially PKC-β, plays a major role in the development of structural and functional abnormalities in vascular tissues, particularly the retina, peripheral nerves, glomerulus and vascular endothelium. Individual isoforms of PKC vary in their tissue distribution, cofactor requirements for activation, and substrates, but PKC-mediated phosphorylation of various intracellular enzymes, receptors and transcription factors adversely affects tissue function under conditions of sustained hyperglycaemia. The mechanisms by which PKC activation in vascular tissues causes structural and functional abnormalities associated with diabetic microangiopathy are not completely understood, but several key pathways have been identified in experimental models of diabetic retinopathy, nephropathy and endothelial dysfunction (Fig. 23.1).

PKC-MEDIATED CHANGES IN GENE EXPRESSION Experimental evidence showing coactivation of PKC and mitogen-activated protein (MAP) kinases in vascular cells grown in high glucose conditions strongly suggests that these two biochemical pathways are linked. Indeed, there is now good evidence showing that activation of PKC leads to activation of MAP kinases, which in turn phosphorylate and regulate transcription factors leading to up-regulation of mRNA and protein production for a range of key intracellular proteins. In particular, PKC-mediated increases in vascular permeability factor (VPF) expression in the eye play an important role in the permeability and neovascularization changes seen in diabetic maculopathy and proliferative retinopathy. In addition, increased mesangial expression of transforming growth factor-β (TGF-β) leads to increases in glomerular matrix protein production, particularly fibronectin and type IV collagen. PKC activation also influences the expression of genes coding for other growth factors and metalloproteinase enzymes involved in the degradation of matrix proteins. Thus, PKC activation, in part via MAP kinase activation, leads to up-regulation of a cluster of genes involved in the formation of structural proteins, growth factors and peptides such as VPF and TGF-β which play a pivotal role in the development of endothelial hyperpermeability, angiogenesis and matrix deposition.

205

206

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

ENDOTHELIAL AND VASCULAR SMOOTH MUSCLE FUNCTION It is well established that PKC enzyme activity in endothelial and vascular smooth muscle (VSM) cells increases in relationship to the ambient glucose concentration (Fig. 23.2). Furthermore, glucose-stimulated PKC activation involves multiple PKC isoforms, especially the conventional PKC isoforms e.g. PKC-β, which in turn affect vascular tone, local blood flow and the formation of platelet-derived and endothelial-derived vasodilator molecules, including nitric oxide.

Hyperglycaemia de novo synthesis

↑ DAG and PKC activation, esp. PKC-β MAP-kinase

Membrane transport

• ↑ PLA2 activity • ↑ Archidonic acid release • ↑ PGE2 production • ↓ Na+-K-+-ATPase

Gene expression

• ↑ VPF- permeability and angiogenesis • ↑ TGF-β - matrix production • ↑ Growth factors, e.g. βFGF • ↑ VSM hyperplasia • ↑ Inter-cellular adhesion molecules (esp. ICAM-1) •↑ Endothelin-1

Endothelial and vascular smooth muscle function • ↑ Reactive oxygen species • Impaired plateletmediated vasodilation • NOS regulation • Renin - angiotensin responses • Angiotensin II and endothelin responsiveness • Cardiomyopathy

Fig. 23.1 Hyperglycaemia, via increased de-novo synthesis of DAG and activation of several PKC isoforms, especially PKC-β, leads to various unwanted pathological effects which can be broadly grouped under three headings: increased gene expression; changes in endothelial and vascular smooth muscle (VSM) function; and altered membrane transport.

CHAPTER 23 • PKC ACTIVATION IN CARDIOVASCULAR AND RENAL COMPLICATIONS

Formation of reactive oxygen species Local formation of reactive oxygen species (ROS) by endothelial and VSM cells initiates a host of unwanted effects, and free radical generation has been implicated in various pathological changes associated with diabetic vascular disease. Formation of ROS involves activation of intracellular enzymes, e.g. NADPH oxidase and xanthine oxidase. However, recent evidence has shown that high glucose levels, via activation of PKC, stimulate ROS formation by VSM and endothelial cells, a pathway that is blocked by PKC inhibitors (Fig. 23.3). Thus, oxidative stress, a well established feature of diabetic vascular disease, is associated with hyperglycaemia via a PKC-dependent pathway in both endothelial and smooth muscle cells. In addition, ROS-induced apoptosis of VSM cells is also PKC-dependent.

Nitric oxide synthase The local generation of nitric oxide (NO) from the oxidation of L-arginine is catalysed by a family of nitric oxide synthases (NOS). Release of NO mediates the vasodilator responses to a number of important molecules, including acetylcholine, bradykinin and substance P. NO-mediated vasodilation is impaired in patients with type 1 and type 2 diabetes. This is likely to reflect both reduced NO availability and reduced NO responsiveness of VSM.

PKC peptide phosphorylation (pmol/min/mg protein)

50 40 30 20 10 0

5

10

15 Glucose (mM)

20

25

Fig. 23.2 Vascular smooth muscle cells exposed to high glucose concentrations show proportional increases in PKC enzyme activity and translocation of PKC to the plasma membrane. Data from Diabetes 1992; 41: 1464–1472.

207

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

Three separate genes encode the known isoforms of NOS: endothelial NOS (eNOS) and neuronal NOS (nNOS) catalyze the constitutive production of NO via a calcium-dependent pathway in blood vessels and neural tissues, respectively, while the third NOS isoform, inducible NOS (iNOS), is located in macrophages and catalyzes NO formation in inflammatory cells.

(a)

14 1x10-2 12

ROS production

10 8 6 4 2 0 (b)

14 1x10-2

P<0.05

12 ROS production

P<0.05

10 8 6 4 2

X GF

A A

+

PM PM

X

Hi

gh

G

Hi

+

gh

GF

G

X GF

nt Co

tro l+

ro l

0

Co n

208

Fig. 23.3 Experimental data showing the effects of PKC inhibitors (GFX or calphostin C, CAL) on high glucose or phorbol ester (PMA) induced increases in ROS production in cultured aortic vascular smooth muscle cells (a), or endothelial cells (b). Adapted from Diabetes 2000; 49: 1939–1945.

CHAPTER 23 • PKC ACTIVATION IN CARDIOVASCULAR AND RENAL COMPLICATIONS

There is good evidence that specific PKC isoforms regulate iNOS expression in macrophages and VSM cells. However, it would seem that different PKC isoforms can regulate iNOS in both a positive and negative manner. In addition, experimental evidence has shown that elevated glucose levels impair endothelial-dependent relaxation via PKC activation.

Impaired platelet-mediated vasodilation Activated normal platelets produce local vasodilation via release of plateletderived adenosine diphosphate (ADP), which in turn stimulates the release of endothelium-derived NO. NO causes VSM relaxation and inhibits platelet aggregation and excessive thrombus formation. Several studies have shown that platelets from patients with diabetes lack the ability to produce NO-dependent vasodilation. Furthermore, this defect can be reproduced experimentally by exposing normal human platelets to high glucose concentrations. Experimental evidence would suggest that this defect is PKC-dependent. For example, platelets from patients with diabetes that were treated with a PKC inhibitor, calphostin-C, produced normal vasodilation, while untreated platelets from the same patients lacked the ability to cause vasorelaxation (Fig. 23.4a). Similarly, normal platelets incubated in high glucose conditions lost their ability to cause vasorelaxation, but co-incubation with calphostin-C prevented glucose-mediated impairment of platelet-mediated vasodilation (Fig. 23.4b).

Monocyte adhesion Binding of circulating moncytes to endothelial cells is an important early event in atheroma formation and one that is enhanced in diabetes. A recent clinical study has shown that membrane-associated PKC activity is increased in monocytes from diabetic patients and that total PKC activity, and expression of the glucose-sensitive PKC-β isoform, decreased by 40% under normoglycaemic conditions. Because increased PKC activity in monocytes enhances their adhesion to the vascular wall, increases fibrinogen binding and promotes differentiation into macrophages, augmentation of this signal transduction pathway could well account for the accelerated progression of atheroma in patients with diabetes. The adhesion molecules most involved in monocyte-endothelial cell interactions are intercellular adhesion molecule-1 (ICAM-1), vascular adhesion molecule-1 (VCAM-1) and E-selectin. High glucose levels up-regulate ICAM-1 protein and mRNA expression via a PKC-dependent pathway.

MEMBRANE TRANSPORT AND GLOMERULOPATHY Expansion of the glomerular mesangium is an early feature of diabetic nephropathy, and isoform-specific translocation of PKC has been identified in glomerular cells. As well as increasing the synthesis of extracellular matrix

209

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

components, PKC-mediated phosphorylation of the glomerular sodiumpotassium ATPase (Na+/K+-ATPase) affects cellular adhesion, vascular permeability and sodium-hydrogen transport. The activity of key membrane transporters, e.g. Na+/K+- ATPase and calcium-ATPase, is reduced in diabetes, in part via PKC activation.

(a)

Change in vessel diameter (%)

210

(b)

40 35 30 25 20 15 10 5 0 -5

DM-Plts DM-Plts+Calp-C

1x108 Plts/ml

Ctrl-Plts Gluc-Plts Gluc-Plts+Calp-C

1x108 Plts/ml

Fig. 23.4 (A) Platelets from diabetic patients were incubated with or without the PKC inhibitor, calphostin C, prior to being perfused through a normal rabbit carotid artery. Untreated diabetic platelets induce no vasodilator response, but platelet-mediated vasodilation was restored by treating diabetic platelets with a PKC inhibitor. (B) similarly, platelets from non-diabetic subjects were incubated under normal or high glucose conditions. Treatment with high glucose impaired platelet-mediated vasodilation, but this was restored by treatment with the PKC inhibitor. Adapted from Brit J Pharmacol 1999; 127: 903–908.

CHAPTER 23 • PKC ACTIVATION IN CARDIOVASCULAR AND RENAL COMPLICATIONS

CURRENT ISSUES •

•

•

•

Hyperglycaemia, via isoform-selective PKC activation, affects a variety of pathophysiological pathways involved in the structural and functional abnormalities associated with diabetic vascular complications. These mechanisms are not completely understood, but PKC activation influences gene expression, endothelial and VSM function and various cell membrane transporters. PKC-mediated stimulation of gene expression, particularly VPF and TGF-β formation, involves activation of the MAP kinase pathway and interaction with specific binding sites and transcription factors. Some of the therapeutic protection conferred by angiotensinconverting enzyme inhibitors (ACE-Is) and experimental compounds such as aminoguanadine may, in part, be mediated via reductions in tissue PKC activities. Activation of the calcium-dependent PKC isoforms, especially PKC-β, plays the most important role in mediating the diverse unwanted effects of PKC activation in vascular tissues.

FURTHER READING Chan NN, Vallance P, Colhoun HM. Nitric oxide and vascular responses in type 1 diabetes. Diabetologia 2000; 43: 137–147. Idris I, Gray S, Donnnelly R. Protein kinase C activation: isozime-specific effects on metabolism and cardiovascular complications in diabetes. Diabetologia 2001; 44: 659–673. Inoguchi T, Li P, Umeda F et al. High glucose level and free fatty acid stimulate reactive oxygen species production through PKC-dependent activation of NADPH oxidase in cultured vascular cells. Diabetes 2000; 49: 1939–1945. Oskarsson HJ, Hofmeyer TG, Coppey L, Yorek MA. Effect of protein kinase C and phospholipase A2 inhibitors on the impaired ability of human diabetic platelets to cause vasodilation. Br J Pharmacol 1999; 127: 903–908. Tesfamariam B, Brown ML, Cohen RA. Elevated glucose impairs endothelium-dependent relaxation by activating protein kinase C. J Clin Invest 1991; 87: 1643–1648. Williams B, Schrier RW. Characterization of glucose-induced in situ protein kinase C activity in cultured vascular smooth muscle cells. Diabetes 1992; 41: 1464–1472.

211

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

CHAPTER 24 Richard Donnelly MD, PhD, FRCP, FRACP EXPERIMENTAL PHARMACOLOGY USING ISOFORM-SELECTIVE PROTEIN KINASE C INHIBITORS INTRODUCTION Because protein kinase C (PKC) plays a fundamental role in normal pathways of cellular signal transduction, conventional thinking would suggest that a pharmacological inhibitor of PKC is unlikely to be a feasible option for clinical drug development. More recently, however, realization that PKC is a family of multifunctional isoenzymes with different patterns of tissue distribution, regulation and biochemical structure has led to renewed interest in the therapeutic potential of isoform-selective blockade of PKC activation, e.g. using antisense oligonucleotides or macrocyclic bis-indolylmaleimide compounds.

DISCOVERY OF RUBOXISTAURIN (LY333531) — AN ORALLY ACTIVE AND HIGHLY SPECIFIC PKC-β INHIBITOR First and second generation PKC inhibitors, such as the staurosporine-like compounds and isoquinolinesulphonamides (e.g. GF109203X), block the catalytic domain of PKC which carries a high degree of sequence homologue with other kinases, and are therefore non-specific for PKC. George King’s group at the Joslyn Diabetes Centre in Boston, USA, collaborating with scientists at Eli Lilly, undertook an extensive screening program to identify and optimise a PKC-β-selective inhibitor. The drug discovery programme was initiated on the basis of numerous experimental studies showing that hyperglycaemia-induced PKC activation involves predominantly PKC-β in retinal, glomerular and vascular tissues. Ruboxistaurin is a bisindolylmaleimide compound that was discovered in the early 1990s (Fig. 24.1). The selectivity for PKC-β relative to other PKC isoforms, and indeed other intracellular kinases, was clearly documented (Table 24.1). Ruboxistaurin inhibits PKC-βI and PKC-βII with half-maximal inhibitory constants (IC50) of 4.7 and 5.9 nmol/l, respectively, whereas for other PKC isoenzymes (except PKC-η) the IC50 was 250 nmol/l or greater (Table 24.1). Thus, the compound has a unique profile of isoform selectivity. The pharmacological effects of ruboxistaurin have been evaluated successfully in numerous experimental models of diabetic microangiopathy, and this compound is now progressing into the late stages of full clinical development.

EFFECTS OF RUBOXISTAURIN IN DIABETIC RETINOPATHY Hyperglycaemia-induced PKC activation increases retinal vascular permeability factor (VPF) gene expression and peptide synthesis, which markedly increases retinal endothelial permeability and also plays an important role in

213 213

214

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

H N

O

O

N

N

O N(CH3 )2

Fig. 24.1 Structure of ruboxistaurin, a macrocyclic bis-indolylmaleimide, which is an orally active, selective PKC-β inhibitor.

Tabulated IC 50 values for ruboxistaurin Kinase

IC50 (nM) ruboxistaurin Staurosporine

PKC-α PKC-β1 PKC-β2 PKC-γ PKC-δ PKC-ε PKC-ζ PKC-η Cyclic AMP kinase Ca2+-calmodulin kinase Casein kinase Src tyrosine kinase

360 4.7 5.9 300 250 600 >105 52 >105 8 × 103 >105 >105

45 23 19 110 28 18 >1.5 × 103 5 100 4 1.4 × 104 1

Table 24.1 Tabulated IC50 values (i.e. concentrations in nM required to achieve 50% inhibition of enzyme activity) for ruboxistaurin and the non-specific PKC inhibitor, staurosporine, with respect to each PKC isoform and related intracellular kinases. Adapted from Science 1996; 472: 728–731.

CHAPTER 24 • EXPERIMENTAL PHARMACOLOGY USING ISOFORM-SELECTIVE PKC INHIBITORS

new vessel formation. Thus, blocking VPF-mediated retinal permeability is a prime target for therapeutic amelioration of diabetic maculopathy. Studies in rats have clearly shown that intravitreal administration of VPF increases vitreous fluorescein leakage, and that pretreatment of these animals for one week with ruboxistaurin 25 mg/kg/day via oral administrattion ameliorated VPF and phorbol ester-induced vitreous fluorescein leakage (Fig. 24.2). Furthermore, whereas control rats showed a two-fold increase in vitreous fluorescein leakage after intravitreal VPF administration, rats pretreated with the PKC-β inhibitor showed no difference in basal vitreous fluorescein leakage but there was a 96% reduction in VPF-induced vitreous fluorescein leakage (Fig. 24.2). Increased retinal permeability is a hallmark of neovascularization within the diabetic eye, as well as being a sight-threatening pathological entity even in the absence of new vessel formation. These experimental data have shown that oral administration of ruboxistaurin is well tolerated and considerably

P = 0.015

P = 0.043

Vitreous fluorescein leakage (arbitrary units)

20

10

0 0 0

2 0

0 25

2 VPF (ng/eye) 25 PKC-β inhibitor (mg/kg rat/day)

Fig. 24.2 Oral administration of the PKC-β inhibitor, ruboxistaurin, to normal rats prevents the increase in vitreous fluorescein leakage following intravitreal injection of VPF. Adapted from Diabetes 1997; 46: 1473–1480.

215

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

attenuates VPF-mediated retinal permeability. Furthermore, diabetes is characterized by an increase in retinal mean circulation time (MCT), and oral treatment with ruboxistaurin for two weeks in STZ-diabetic rats reduced retinal MCT, as measured by video fluorescein angiography (Fig. 24.3). This experimental data has now been confirmed in phase II clinical trials in which ruboxistaurin administration for one month produced significant improvements in retinal blood flow and MCT among 27 diabetic patients (chapter 25). Larger, multicentre clinical trials are in progress.

(a)

(b) 0.5

*

4

§

§ 2

0

0

(c) 20 *

15 10

+ +

+ +

+ +

0.3 0.2 0.1 1.0

10

+ +

1.5

5 0

* 0.4

0 1.0 10 0 0.1 Dose of ruboxistaurin (mg/kg) (d) 2

0.1

MCT (s)

GFR (ml/min)

+

Filtration fraction (GFR/RPF)

6

Urinary AER (mg/day)

216

§ 1

§

§

0.5

0

0

10

0

0

0.1

1.0

10

Dose of ruboxistaurin (mg/kg)

Fig. 24.3 Effect of oral dosing with ruboxistaurin on renal and retinal vascular function in non-diabetic (●) and STZ-diabetic (●) rats. Untreated diabetic animals show increases in glomerular filtration rate (GFR), renal filtration fraction (GFR corrected for renal plasma flow, RPF), urinary albumin excretion rate (AER) and retinal mean circulation time (MCT). Oral treatment with ruboxistaurin 0.1–10 mg/kg/day ameliorated these renal and retinal haemodynamic abnormalities. Science 1996; 272: 728–731.

CHAPTER 24 • EXPERIMENTAL PHARMACOLOGY USING ISOFORM-SELECTIVE PKC INHIBITORS

Further experimental studies have shown that diabetes-induced reductions in Na+/K+-ATPase and Ca2± -ATPase in the retina are mediated, in part, via PKC-β activation. Oral administration of ruboxistaurin normalizes Na+/K+-ATPase activity in retinal microvessels (Fig. 24.4).

PKC-β INHIBITION AND EXPERIMENTAL NEPHROPATHY The early stages of diabetic renal disease are characterized by glomerular hyperfiltration, mesangial expansion and microalbuminuria. Hyperglycaemiainduced de novo synthesis of DAG, coupled with activation of PKC, especially PKC-β, affects the structural and functional changes in the kidney via several different mechanisms involving various phosphorylation substrates of PKC. For example, mesangial expansion has been attributed, in part, to PKC-mediated increases in transforming growth factor-β (TGFβ) gene expression, activation of cytosolic phospholipase A2 and inhibition of Na+/K+-ATPase activity.

40

#

35

Na+/K+-ATPase activity

30 25 *

20 15 10 5 0 Normal

Diabetes

Diabetes + ruboxistaurin

Fig. 24.4 Oral treatment with the PKC-β inhibitor ruboxistaurin, reverses diabetesrelated reductions in Na+/K+-ATPase activity in retinal microvessels. Adapted from Diabetes 1998; 47: 464–469.

217 217

218

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

Experimental studies with ruboxistaurin have shown that, following oral administration for eight weeks to STZ-diabetic and non-diabetic rats, urinary albumin excretion rate (AER) and glomerular hyperfiltration were significantly reduced (Fig. 24.3). Interestingly, higher doses of the PKC-β inhibitor (1–10 mg/kg/day) were required to inhibit diabetes-mediated PKC activation in the kidney compared with the retina (0.1 mg/kg/day). In addition, treatment with ruboxistaurin had no significant effect on GFR and filtration fraction in non-diabetic animals (Fig. 24.3). Among diabetic rats, however, the dose-response curve for ruboxistaurin in normalizing GFR paralleled its inhibitory effect on PKC activity.

Renal protection with aminoguanidine and angiotensinconverting enzyme inhibition (ACE-I) involves normalization of glomerular PKC activity In experimental models of diabetic renal disease, e.g. the STZ-diabetic rat, it is well established that ACE-Is and aminoguanidine retard the structural and functional abnormalities characteristic of diabetic nephropathy, particularly with respect to reducing urinary AER. The exact mechanisms by which these therapeutic interventions work is not entirely clear, but recent work by George Jerums and colleagues has shown that glomerular PKC activity levels are normalized in STZ-diabetic rats during experimental treatment with aminoguanidine and the ACE-I, ramipril. Thus, diabetes-related increases in glomerular PKC activity may serve as an important common pathway by which metabolic and haemodynamic factors contribute to the initiation and progression of diabetic renal disease. Existing renoprotective agents, e.g. ACE-Is, may slow the progression of nephropathy, in part, by normalizing diabetes-induced increases in glomerular PKC activity.

EFFECTS OF RUBOXISTAURIN IN EXPERIMENTAL DIABETIC NEUROPATHY Various pathways have been implicated in the pathogenesis of diabetic neuropathy, including increased polyol pathway activity, enhanced non-enzymatic glycation and PKC activation. In addition, neural ischaemia is thought to play an important role in diabetic nerve injury, in part via PKC activation which impairs vasodilation and increases vasoconstrictor pathways in the endoneurial microvasculature. In experimental STZ-diabetic rats, motor nerve conduction velocity and sciatic nerve blood flow are reduced. Treatment with ruboxistaurin ameliorated these abnormalities via mechanisms attributable to prevention of neural ischaemia (Fig. 24.5).

CHAPTER 24 • EXPERIMENTAL PHARMACOLOGY USING ISOFORM-SELECTIVE PKC INHIBITORS

Velocity (m/s)

65

control level

60

55

0.0

0.1

0.3

1.0

10.0

25.0

Dose (mg/kg)

Fig. 24.5 Ruboxistaurin improves sciatic nerve conduction velocity in experimental models of peripheral neuropathy. A dose-related effect is illustrated.

CLINICAL IMPLICATIONS OF AN ORALLY ACTIVE PKC-β INHIBITOR, RUBOXISTAURIN Extensive experimental studies have shown that ruboxistaurin selectively inhibits PKC-β in retinal, neural, renal and vascular tissues following oral administration without any significant adverse effects. The encouraging tolerability profile of ruboxistaurin is no doubt attributable to its pharmacological specificity for PKC-βI and PKC-βII. The animal studies have convincingly shown that, following chronic oral treatment, ruboxistaurin ameliorates the early increases in retinal blood flow, glomerular filtration rate and renal and retinal permeability. This data opens the possibility of a new and exciting pathway for therapeutic intervention in the earliest stages of diabetic microvascular disease. In particular, such an approach would be unique in offering protection against the development and progression of retinopathy and nephropathy via a mechanism that is independent of (and complementary to) glucose or blood pressure reduction. Thus, in clinical practice, PKC-β inhibition would be used as an adjunct to all existing therapies for the prevention of diabetic vascular complications. Large multicentre clinical trials are on-going not only in diabetic retinopathy and renal disease but also in patients with other diabetic complications, e.g. erectile dysfunction and diabetic neuropathy.

219

220

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

CURRENT ISSUES • Ruboxistaurin is a unique orally active PKC inhibitor which is highly specific for the PKC-β isoforms. Following oral administration to STZdiabetic rats, ruboxistaurin prevented diabetes-related increases in retinal and renal PKC activity in parallel with amelioration of glomerular hyperfiltration, microalbuminuria and increased retinal blood flow. • Ruboxistaurin shows an excellent tolerability profile in experimental diabetic animals, no doubt reflecting its specificity for inhibiting only two out of twelve PKC isoforms. Furthermore, in non-diabetic animals (in which there is no augmentation of PKC activity) ruboxistaurin has no significant effects on retinal or renal haemodynamics. Thus, the compound seems to be highly specific for PKC-β and only achieves therapeutic effects in experimental studies in which diabetes-related increases in PKC are present. • Large multicentre clinical trials with ruboxistaurin are on-going to assess its efficacy and safety in patients with diabetic retinopathy and peripheral neuropathy. In due course further studies will be established to define the role of this compound in other diabetes complications, including nephropathy and erectile dysfunction.

FURTHER READING Aiello LP, Bursell SE, Clermont A et al. Vascular endothelial growth factor-induced retinal permeability is mediated by protein kinase C in vivo and suppressed by an orally effective β-isoform-selective inhibitor. Diabetes 1997; 46: 1473–1480. Ishii H, Jirousek MR, Koya D et al. Amelioration of vascular dysfunction in diabetic rats by an oral PKC-β inhibitor. Science 1996; 272: 728–731. Kowluru RA, Jirousek MR, Stramm L et al. Abnormalities of retinal metabolism in diabetes or experimental galactosemia: V relationship between protein kinase C and ATPase. Diabetes 1998; 47: 464–469. Nakamura J, Kato K, Hamada Y et al. A protein kinase C-β-selective inhibitor ameliorates neural dysfunction in streptozotocin-induced diabetic rats. Diabetes 1999; 48: 2090–2095.

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

CHAPTER 25 Richard Donnelly MD, PhD, FRCP, FRACP CLINICAL TRIALS WITH RUBOXISTAURIN INTRODUCTION Ruboxistaurin is the first molecule in an exciting new class of PKC-β‚ specific inhibitors which ameliorate the structural and functional vascular abnormalities associated with hyperglycaemia in humans and experimental animals. A series of detailed molecular and experimental studies were conducted to document the effects of ruboxistaurin in retinal, neural and endothelial tissues. These were followed by a series of multicentre clinical trials to evaluate longer term efficiency and safety in patients with diabetes related complications. The design and execution of these trials has posed considerable challenges, and many of these trials are still ongoing. There are, however, a number of encouraging results already in the public domain from phase II studies.

RUBOXISTAURIN IMPROVES ENDOTHELIAL DYSFUNCTION Diabetes is associated with endothelial dysfunction, and hyperglycaemia impairs the endothelial-dependent vasodilator response to acetylcholine. In a placebo controlled, double blind crossover study in healthy volunteers, there was evidence that ruboxistaurin improved forearm blood flow in response to incremental arterial infusions of the endothelium-dependent vasodilator methacholine under hyperglycaemic conditions (Fig. 25.1). Thus, this novel experimental study has confirmed that inhibition of PKC-β in healthy volunteers prevents the reduction in endothelium-dependent (nitric oxide mediated) vasodilation induced by acute hyperglycaemia.

CLINICAL TRIALS OF RUBOXISTAURIN IN DIABETIC RETINOPATHY Experimental studies have shown that ruboxistaurin inhibits hyperglycaemia-induced PKC activation in the retina (Fig. 25.2). In addition ruboxistaurin prevents neovascularization in a porcine model of retinal vein occlusion (Fig. 25.3). These experimental data provide encouraging evidence that PKC-β‚ inhibition might have a favourable effect on macular oedema formation and new vessel formation (two sight threatening complications) in patients with diabetic retinopathy. The clinical development of ruboxistaurin began with phase I tolerability and pharmacokinetic studies in healthy volunteers, followed by phase II efficacy studies in patients with diabetes. In patients with type 1 or type 2 diabetes and minimal or no evidence of diabetic retinopathy, ruboxistaurin increased retinal blood flow in a dose-dependent manner, maximal after 32 mg daily for

221

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

Euglycaemia Hyperglycaemia p = 0.08

p = 0.001

Placebo

Ruboxistaurin

Forearm blood flow (ml / dl / min)

3

2

1

0

Fig. 25.1 Forearm blood flow in healthy volunteers during euglycaemia and hyperglycaemia after pretreatment with ruboxistaurin or placebo. The PKC-β‚ inhibitor improved the endothelial-dependent vasodilator response to methacholine under conditions of high glucose. Adapted from Beckman et al. Circulation Research 2002; 90: 107–111.

20

PKC activity (pmo/min/mg of protein)

222

Nondiabetic Diabetic 15

10

5

0 0

0.1

10.0

Ruboxistaurin (mg/kg/d)

Fig. 25.2 Ruboxistaurin attenuates the increase in retinal PKC activity in experimental rats with diabetes.

CHAPTER 25 • CLINICAL TRIALS WITH RUBOXISTAURIN

4

Neovascularization score

p = 0.03 3

2

1

0 Placebo

Ruboxistaurin 1 mg/kg/d, po

Fig. 25.3 Ruboxistaurin prevents neovascularization in a porcine retinal vein occlusion model of new vessel formation.

Extent of MCT abnormality at endpoint

1.2 1.0 0.8 0.6 0.4 0.2 0.0 Placebo

16 mg/d

32 mg/d

Fig. 25.4 Phase II study of ruboxistaurin in patients with type 1 or type 2 diabetes and retinopathy. In a double blind, placebo controlled study for four weeks, ruboxistaurin decreased mean retinal circulation time, i.e. improved retinal blood flow. Adapted from Aiello et al. Diabetes 1999; 48: A19.

223

224

SECTION IV • MECHANISMS OF HYPERGLYCAEMIA INDUCED VASCULAR DYSFUNCTION

one month (Fig. 25.4). Having confirmed the basic safety and tolerability of ruboxistaurin, and demonstrated that it has pharmacodynamic activity on retinal blood flow, large multi-centre clinical trials were initiated to evaluate the safety and efficacy of the treatment in larger patient groups during longer term administration (two to four years). The PKC-diabetic retinopathy study (DRS) and the PKCdiabetic macular oedema (DME) study were the first international randomized, placebo controlled trials to assess whether oral treatment with ruboxistaurin will delay progression in patients with moderate to severe non-proliferative diabetic retinopathy at base-line, including progression from non-clinically significant to clinically significant macular oedema (CSMO). The results for approximately 1,000 patients followed for an average of 36–46 months will be announced soon. The clinical trials of ruboxistaurin in diabetic retinopathy have two key objectives. First, to determine if oral treatment with ruboxistaurin over three years will reduce progression of diabetic retinopathy or the need for laser photocoagulation in patients with moderately severe to very severe non-proliferative diabetic retinopathy in at least one eye. Second, to determine if all treatment with ruboxistaurin in patients with mild to moderate non-proliferative diabetic retinopathy and non-visually threatening diabetic macular oedema will delay development of clinically-significant macular oedema or the need for laser photocoagulation. The clinical trials with ruboxistaurin use 7-field stereoscopic fundal photographs and measurements of visual acuity as markers of drug efficacy.

CLINICAL TRIALS OF RUBOXISTAURIN IN DIABETIC PERIPHERAL NEUROPATHY A number of large phase III clinical trials are in progress to evaluate the effects of ruboxistaurin on various unpleasant symptoms of diabetic neuropathy and longer term outcomes in relation to nerve function and neurophysiological endpoints. These trials use a combination of symptom scores and nerve function measurements to assess efficacy.

CLINICAL TRIALS OF RUBOXISTAURIN IN OTHER DIABETESRELATED COMPLICATIONS Further clinical trials are in progress to evaluate the effects of ruboxistaurin on renal function and proteinuria, and on endothelial function in relation to lower limb ischaemia and macrovascular end-points.

CHAPTER 25 • CLINICAL TRIALS WITH RUBOXISTAURIN

CURRENT ISSUES • The results of large multicentre clinical trials of ruboxistaurin in patients with diabetic retinopathy and neuropathy are eagerly awaited in 2005–7. • As a completely novel drug, ruboxistaurin has posed unique challenges to the design and execution of international clinical trials in diabetes complications. • The safety and tolerability profile of ruboxistaurin is very encouraging. It is suitable for once-daily oral administration and has no adverse effects on immune function. • There are still some unanswered questions about the optimal dose of ruboxistaurin for each potential indication.

FURTHER READING Aiello LP, Bursell S, Devries T. Protein kinase C beta selective inhibitor Ruboxistaurin ameliorates abnormal retinal haemodynamics in patients with diabetes. Diabetes 1999; 48: A19. Aiello LP, David MD, Sheetz MJ. The PKC Inhibitor Diabetic Retinopathy Study Group. Design, baseline patient characteristics and high prevalence of severe to very severe nonproliferative diabetic retinopathy (NPDR) in the Protein Kinase C Diabetic Retinopathy Study (PKC-DRS). Diabetes 2002; 51 (suppl2): A209. Demolle D, de Suray JM, Onkelinx C. Pharmacokinetics and safety of multiple oral doses of LY333531, a PKC beta inhibitor, in healthy subjects. Clin Pharm Ther 1999; 65: 189. Demolle D, de Suray JM, Vandenhende F, et al. Ruboxistaurin single escalating oral dose study in healthy volunteers. Diabetologia 1998; 41: (Suppl 1): A354. Donnelly R, Idris I, Forrester J. Protein Kinase C inhibition and diabetic retinopathy: a shot in the dark at translational research. Br J Ophthalmol 2004; 88: 145–151.

225

Vascular Complications of Diabetes: Current Issues in Pathogenesis and Treatment, Second Edition Edited by Richard Donnelly, Edward Horton Copyright © 2005 by Blackwell Publishing Ltd

226

INDEX aminoguanidine 86, 114, 124, 186–7 renal protection 218 amitriptyline 116, 125 A amputation 17, 35, 70 ACE inhibitors 28, 40, 49 diabetic foot 105, 106, 107, 125 myocardial infarction 63–4 lower limb 7, 8, 9, 18, 61, 67 neuropathy 115, 124 smoking 67 renal protection 218 anaemia 24, 33, 186 retinopathy 134 angiography 26, 39, 43, 61, 66 versus beta-blockers 63 fluorescein 102, 153, 154, 155, 156, 174 acetyl-L-carnitine 124 angioplasty 40, 42, 44, 62, 66 acetylcholine 207, 221 diabetic foot 107 acetylcysteine 26 angiotensin receptor antagonists 30, 70 actin 199, 200 angiotensin-converting enzyme inhibitors see acupuncture 118 ACE inhibitors acute sensory neuropathy 80 ankle reflex 79, 80, 81, 122, 125 adenosine 53, 100, 101 anti-arrhythmics 117 adenosine diphosphate 209 anti-oxidants 114 advanced glycated haemoglobin (Hb-AGE) 186 anti-platelet therapy 43 advanced glycation end-products (AGE) 86–7, anticoagulation 61 114, 179–87, 184 anticonvulsants 116–17, 117, 125 and age-adjusted death rate 180 antihypertensive agents 29, 32, 49, 73 cataract 172 erectile dysfunction 54 cross-linking 185 nephropathy 28–9 formation 183 antioxidants 30 harmful effects 185 apomorphine 55 inhibitors of 114 Appropriate Blood Pressure Control in Diabetes and oxidative stress 200 (ABCD) trial 87 receptors 185–6 arteriography 26 age factors arteriolohyalinosis (Kimmelstiel-Wilson kidney) cataracts 171 26 hypertension and proteinuria 28 AS-3201 124 retinopathy 129, 134 ascorbic acid (vitamin C) 179 age-adjusted death rate 180 ASPECT study 61 ALADIN-II study 87 aspirin 47, 48 albumin-creatinine ratio 25 myocardial infarction 43 albuminuria 22, 24 peripheral vascular disease 62, 68, 70 Albustix 24, 25 stroke 47, 48 alcohol dependency 51, 52 atenolol 63 aldose reductase 76, 85, 86, 124, 171, 187, 188, atherosclerosis 35 189 atorvastatin 67 aldose reductase inhibitors 85, 113–14, 113, autonomic neuropathy 80, 81 124 alkaline phosphatase 32, 110 B allodynia 80, 91, 116 balanitis 52 alpha-lipoic acid 114, 124 BDA Cohort Study 35, 36 alpha-oxoaldehydes 182, 183 beading 144 alprostadil (prostaglandin E1) 56 Becaplermin 108 ALT-462/486 186 Berlin Retinopathy Study 13, 14 alteplase 47, 69 beta-blockers 40, 63 biopsy Amadori rearrangement 179, 182, 183, 185 nerve 101–2 amblyopia 137 skin 102 Page numbers in italic refer to figures; those in bold refer to tables.

INDEX

biothesiometer 99 bisphosphonates 110 blindness 7, 8, 9, 75 neuropathy 129 retinopathy 129, 137 blood-retinal barrier 156 boat shaped haemorrhage 144 body mass index 5 bradykinin 207 brain-derived neurotrophic factor 88, 115 branch retinal vein occlusion 173, 174 British Diabetic Association 149 British Diabetic Association Cohort study 3 Bruch’s membrane 158, 168 Bypass Angioplasty Revascularization Investigation (BARI) trial 44, 66 C C-peptide 89, 124 C-reactive protein 66 calcium 190, 197, 200, 202, 204 caldesmon 200 calphostin 208, 209 capsaicin 118 captopril 29, 63, 73 carbamazepine 117, 125 carbonyl stress 182–3, 184 carboxymethyl-lysine 179 cardiomyopathy 37, 193, 206 carotid stenoses 46 carpal tunnel syndrome 80 cataracts 171–2, 188 treatment 172 central retinal vein occlusion 173 cerebral haemorrhage 43, 45, 48 cerebral infarcts 44 cerebral small-artery disease 46 cerebrovascular disease see stroke Charcot neuroarthropathy 108–10, 108, 109 management 110 X-rays 109 Charcot neuropathy 79 cholesterol and coronary heart disease 18, 19, 20, 36, 37, 38, 40, 43, 67 peripheral neuropathy 93 retinopathy 135 Cholesterol And Recurrent Events study 40 chorioretinal scarring 137 chronic inflammatory demyelinating polyneuropathy 83 chronic sensorimotor neuropathy 79–80 cilostazol 68, 69 circinate of exudates 142, 157 citalopram 116, 125

claudication 67–8, 68, 69 clinically significant macular oedema 142, 158 clomethiazole 49 clopidogrel 48, 62, 66, 71 Clopidogrel in Unstable angina to prevent Recurrent ischaemic Events (CURE) study 62 Clopidogrel vs. Aspirin in Patients at Risk of Ischaemic Events (CAPRIE) study 48 collagen 86, 179, 183, 185, 205 compound muscle action potential 100–1 connexin-43 199 CONSENSUS II trial 41, 63 corneal confocal microscopy 102, 103 coronary heart disease 18, 35–44 aetiology 36–7, 37 anti-platelet therapy 43 environmental factors 36 epidemiology 35–6, 36 and ethnicity 38–9 evidence-based practice 38–44 fetal nutrition 36 glycaemic control 41–3, 42 hazard ratios 16 incidence 35 lipid-lowering therapy 43–4 management 40–1 microalbuminuria 38 mortality 19, 35, 36, 37, 38, 43 obesity 35, 36 prevention 61–7 prognosis 37–8, 38 screening 39 statins 43 vascular risk assessment 38–9 see also myocardial infarction cost of diabetes 7–10, 9, 10 cotton wool spots 139, 141, 142, 143, 173, 174 cranial nerve palsy 174 cranial neuropathy 82 creatinine 26, 27, 31, 32, 73 cross-link formation 185 CT-angiography 26 cyclic GMP 53 cystoid macular oedema 157, 172 D DECODE study 182 delquamine 57 3-deoxyglucosone 179, 182 desipramine 116 Diabetes Control and Complications Trial (DCCT) 3, 14, 28, 93, 132 Diabetes in Early Pregnancy Study 136 diabetic amyotrophy 82, 83

227

228

INDEX

Diabetic Control and Complications Trial (DCCT) 113 diabetic foot 105–10 amputation 105, 106, 107, 125 angioplasty 107 callus formation 105 causes 105–6, 105 education 107 healing 107, 108 hyperbaric oxygen 107 management guidelines 125 PEDIS classification 106 prevention 107 recombinant platelet derived growth factor 108 revascularization 107 risk factors 80, 105 UTDWCS classification 106 Wagner classification 106 diabetic nephropathy see nephropathy diabetic neuropathy see neuropathy Diabetic Retinopathy Study 163 Diabetic Retinopathy Vitrectomy Study 168 diabetic truncal radiculoneuropathy 82–3 diacylglycerol (DAG) 87, 90, 189, 193, 205 DAG-PKC pathway 114, 187, 189 generation of 192 diagnosis 5 dialysis 23, 32 2,3-diaminophenazone 86 diet coronary heart disease 36, 40 nephropathy 30–1 DIGAMI therapy 48, 64 diplopia 82, 174 dipyridamole 48, 70, 71 disc new vessels 145, 164 dot and blot haemorrhages 141 E E-selectin 209 Early Treatment Diabetic Retinopathy Study (ETDRS) 43, 135, 142, 151 EDTA clearance 30 electrophysiology 99–101, 100 enalapril 29, 41, 63, 87 end-stage renal failure 22 endophthalmitis 167, 172, 175 endothelial permeability 197–204, 197, 198, 199, 201 and hyperglycaemia 200 intercellular gaps 198–9 and protein kinase C 200, 203 endothelin 53, 206

entrapment neuropathy 80, 82 environmental factors 4, 35, 136 epalrestat 114, 124 epidemiology 4–6 type 1 diabetes 4 type 2 diabetes 4–6 see also individual conditions Epidemiology of Diabetes Complications (EDC) Study 93 erectile dysfunction 51–7 autonomic neuropathy 52, 55 causes 55 clinical presentation 51–2 diagnostic features 54 drug-related 54 evidence-based practice 54–7, 55, 56 incidence 51 management 54–7 pathophysiology 52–3, 53 erythropoietin 33 essential fatty acids 124 ethnicity 21 coronary heart disease 38–9 epidemiology 5 nephropathy 23 EURODIAB Controlled Trial of Lisinopril in Insulin-Dependent Diabetes Mellitus (EUCLID) 75, 93, 134 EURODIAB IDDM Complications Study 93 European Stroke Prevention Study 48, 70 external ocular muscle palsies 174 extracellular signal-regulated kinases 86 exudates 141, 142 exudative maculopathy 151 F F-actin 199 F-waves 101 ferritin 32 fibrates 43 fibrinogen 37, 62, 209 fidarestat 124 flame-shaped haemorrhages 141 fluorescein angiography 102, 153, 154–5, 154, 155, 156, 173, 174 fluoxetine 116 folic acid 67 foot ulceration see diabetic foot fovea 141 foveola 141 foveolar avascular zone 151 Framingham study 19, 33, 35, 38, 45 free fatty acids 187, 192, 193 fructosamine 13, 182, 183

INDEX

G gabapentin 117, 125 gamma-linolenic acid 114, 124 gangrene 24, 35, 47 gastroparesis 33 gemfibrozil 43 gene expression 205 genetics 36 microvascular disease 135 nephropathy 19, 23 retinopathy 135–6 type 1 diabetes 4 type 2 diabetes 4 GF109203X 213 GISSI-3 41, 63 glaucoma 146, 172 ghost-cell 166 haemolytic 166 neovascular 146, 172 glitazones 43 glomerular filtration rate 22, 31 glomerulopathy 209–10 glomerulosclerosis 25, 26, 185 GLUT-1 transporters 187, 189 glycaemic control, and cardiovascular risk 41–2, 42 glycine antagonists 49 glycoprotein IIb/IIIa glycoprotein receptor inhibitors 62–3 glycosylated haemoglobin (HbA1C) 13, 19, 27, 185 and amputation 67 hazard ratios 70 and myocardial infarction 181 retinopathy 14, 15, 16, 136 glyoxal 182, 183 Gothenburg study 45 grid macular treatment 157 growth factors 88–9, 166 gustatory sweating 32, 81 GUSTO-1 study 63

intracranial 47, 69 intraretinal 141 preretinal 145, 147 retinal 140–1, 144, 168 subarachnoid 45 vitreous 139, 147, 166, 167, 169 hallux valgus 97 Heart Outcomes Prevention Evaluation (HOPE) study 29–30, 41, 42, 48, 64, 65, 67, 71 heparin 47 stroke 48, 69 surface modified intraocular lenses 172 high density lipoprotein (HDL) 18, 135 histamine 199, 200 homocysteinaemia 67 HOT study 62 hydronephrosis 26 hydroxymethylglutaryl CoA reductase inhibitors 88 hyperalgaesia 116 hyperbaric oxygen 107 hyperglycaemia 6, 13–19, 14–18 balanitis 52 cataracts 171 diabetic angiopathy 179, 181 endothelial hyperpermeability 200, 201, 202, 206 endothelial permeability 200 macrovascular disease 13, 16 microvascular disease 13, 18, 19, 132 polyol pathway 187 protein kinase C activation 189, 192, 193, 194, 207, 212 retinopathy 132, 133 stroke 46, 47, 48 treatment 64, 115 vascular injury 181 hyperglycaemic neuropathy 80, 85, 89, 93 hyperlipidaemia 5, 19–21, 20 coronary heart disease 36 erectile dysfunction 53 retinopathy 135 H hypertension 19–21, 20 haematuria 26 coronary heart disease 36, 39 haemodialysis 32 hazard ratios 72 haemoglobin peripheral vascular disease 24 advanced glycated (Hb-AGE) 186 and proteinuria 28 glycosylated see glycosylated haemoglobin retinopathy 25–6, 134 haemorrhage treatment 28–9, 29, 32, 49, 54, 73 boat shaped 144 hypertriglyceridaemia 20–1, 67 cerebral 43, 45, 48 hypoaesthetic neuropathy 92–3 dot and blot 141 hypotension, postural 32, 81 flame-shaped 141

229

230

INDEX

I and hyperglycaemia 13, 16 IgA nephropathy 26 and hypertension 30 imipramine 116, 125 see also coronary heart disease; stroke impaired glucose tolerance 85 macular oedema impotence see erectile dysfunction classification 147 insulin neuritis 80 clinically significant 142, 158 insulin-like growth factor 89, 115 cystoid 157, 172 intercellular adhesion molecule-1 (ICAM-1) 209 duration of diabetes 152, 153 International Working Group of the Diabetic and retinopathy 152 Foot 106 maculopathy 139, 141, 151–61 intracavernosal injection therapy 56 diagnosis 153–6 intraocular pressure 137, 169 diffuse oedema 157 intraretinal haemorrhage 141 epidemiology 151 intraretinal microvascular abnormalities 143, fluorescein angiography 153, 154–5, 155, 165, 165 156 iris neovascularization 146, 164, 165, 172 focal oedema 157 ischaemia, lower limb 4 grid macular treatment 157 leg ulcers 105, 106, 107 ischaemic 151, 153, 154 ischaemic maculopathy 151, 153, 154 laser therapy 157–60 isoquinolinesulphonamides 213 micropulsed diode laser 160 isosorbide dinitrate 115 optical coherence tomography 155–6 spray 118 pathogenesis 156 surgery 158 K treatment 157–8 kidneys see also macular oedema asymmetrical 32 magnesium 49 biopsy 26 magnetic field therapy 110 morphological changes 26 magnetic resonance imaging 102 see also renal Maillard reaction 179, 182 Kimmelstiel-Wilson kidney (arteriolohyalinosis) matrix metalloproteinase 86 26 membrane transport 209–10 metabolic syndrome 5, 19, 20, 53 L metformin 11, 32, 42 lactic acidosis 26, 32 methylglyoxal 182 lacunar infarcts 45, 46 metoprolol 63 lamotrigine 117 mexilitine 117 laser therapy 157–60, 167–8 micro-HOPE study 48 complications 168 microalbuminuria 13, 73 left ventricular hypertrophy coronary heart disease 38 coronary heart disease 38 nephropathy 19, 23–9, 24, 27, 32, 71, 135, nephropathy 24 217 lens proteins, glycation 171 retinopathy 14, 134–5 lidorestat 124 microaneurysms 140 LIPID study 49 micropulsed diode laser 160 lipid-lowering therapy 43–4 microvascular disease 4 lisinopril 29, 41, 63 and erectile dysfunction 52 low density lipoproteins (LDL) 18 genetic factors 135 lubeluzole 49 and hyperglycaemia 13, 18, 19, 132 LY333531 see ruboxistaurin and hypertension 20 smoking 21 M mitogen-activated protein (MAP) kinases 205 macrovascular disease MOLD 184 evidence-based interventions 61 monocytes 209

INDEX

mortality 3, 38, 39 coronary heart disease 19, 35, 36, 37, 38, 43 myocardial infarction 30, 37, 40 stroke 45 Multiple Risk Factor Intervention Trial study 36 myo-inositol 124 myocardial infarction 20 anticoagulation 61 DIGAMI therapy 48, 64 and glycosylated haemoglobin (HbA1C) 181 mortality 30, 37, 40 statins 66–7 thrombolysis 47, 61 myoinositol deficiency 86 myopia 137 myosin light chain kinase 199 N N-[carboxymethyl]-lysine 179, 183, 184 NADH 113, 114 NADPH 113 NADPH oxidase 206 necrobiosis lipoidica 21 nephropathy 21, 23–34 antihypertensive agents 28–9 clinical presentation 23–4, 24, 25, 26 diagnostic features 24–8, 27, 28 diet adjustment 30–1 and ethnicity 23 evidence-based practice 28–31, 29, 30, 71–3, 72, 73 genetics 19, 23 haematuria 26 IgA 26 left ventricular hypertrophy 24 microalbuminuria 19, 23–9, 24, 27, 32, 71, 135, 217 monitoring renal function 31–3, 33 and retinopathy 134–5 ruboxistaurin treatment 216–17, 218–19 nephrotic syndrome 22 nerve biopsy 101–2 nerve conduction velocity 100 nerve growth factor 124 neuropathy 79–84 acute sensory 80 autonomic 80, 81 blindness 129 Charcot’s 79 cholesterol 93 chronic inflammatory demyelinating polyneuropathy 83 chronic sensorimotor 79–80 classification 79, 123 clinical screening 97–8

cranial 82 diabetic amyotrophy 82, 83 diabetic truncal radiculoneuropathy 82–3 entrapment 80, 82 evidence-based interventions 75–6 hyperglycaemic 80, 85, 89, 93 hypoaesthetic 92–3 invasive assessment 101–2 management guidelines 121–5, 122–5 non-invasive assessment 102–3, 103 painful 91–2, 91 pathophysiology 85–90 advanced glycation end-products 86–7 growth factors 88–9 hydroxymethylglutaryl CoA reductase inhibitors 88 immune mechanisms 90 myoinositol deficiency 86 oxidative stress 87 polyol pathway 85–6, 113 protein kinase C-β 87–8 vascular factors 87 peripheral see peripheral neuropathy quantitative sensory testing 98–101, 99, 122 electrophysiology 99–101, 100 thermal thresholds 99 vibration perception threshold 88, 92, 98–9 staging 80, 81 symptoms 97 Neuropathy Disability Score (NDS) 97, 98 Neuropathy Impairment Score (NIS) 97 neuropeptide-Y 53 neurotrophins 88, 115 new vessels elsewhere 145, 146, 163, 164, 165 nisoldipine 87 nitric oxide synthase 207–9 nitroglycerine 57 non-ketotic hyperosmolar states 47 normoglycaemia 113 O obesity 5, 11 childhood diabetes 4, 11 coronary heart disease 35, 36 ocular perfusion pressure 137 oculoischaemic syndrome 173, 174 oculomotor nerve palsy 82 OPB-9195 187 ophthalmoscopy 148 opioids 125 optical coherence tomography 155–6 oral glucose tolerance test 6 oxidative stress 87, 200, 207 oxycodone 125

231

232

INDEX

P painful neuropathy 91–2, 91 pamidronate 110 pancreatic/islet cell transplantation 113 papaverine 56 paroxetine 116, 125 pars plana vitrectomy 168–9 PEDIS classification 106 pentosidine 183, 184 pericytes 140 peripheral neuropathy 91–5 combined assessments 93 negative sensory symptoms 92–3 positive sensory symptoms 91–2, 91 risk factors 94 ruboxistaurin 224 stages of 122 treatment 113–19 peripheral vascular disease diabetic foot 105–10 evidence-based interventions 67–8, 67 smoking 19–21, 20, 67 peritoneal dialysis 32 Peyronie’s disease 52 phentolamine 56, 57 phimosis 52 phosphodiesterase inhibitors 54 pimagedine 186 plasminogen activatory inhibitor-1 37 platelet derived growth factor, recombinant 108 platelet-mediated vasodilation 209 polycystic kidneys 26 polyol pathway 85–6, 113, 183, 187–8, 187 cataracts 171 ponalrestat 124 population screening 11 pravastatin 67 pregabalin 117, 125 pregnancy, retinopathy in 136 prevalence of diabetes 6 proliferative retinopathy 144–6, 163–70 definitions 163–4, 164, 165 diagnosis/natural history 164–6, 165 disc new vessels 145, 164 epidemiology 163 iris neovascularization 164, 165 laser therapy 167–8 new vessels elsewhere 145, 146, 163, 164, 165 pathogenesis 166 retinal detachment 145–6 treatment 166–7, 167 vascular permeability factor 166 vitrectomy 168–9 vitreous detachment 166

protein kinase C 189–95, 189 and endothelial permeability 200, 203 endothelial and vascular smooth muscle function 206–10, 207 and gene expression 205 isoforms 191 vascular permeability 197–204 protein kinase C inhibitors 213–20 protein kinase c-β 87–8 activation of 194 inhibitors of 114–15, 119 proteinuria 13, 17, 19, 29 public health impact 3–11 pyridoxamine 86, 187 pyrraline 87 R ramipril 29–30, 42 myocardial infarction 65 stroke 48, 65 reactive oxygen species 207 Reduction of Cholesterol in Ischaemia and Function of the Endothelium (RECIFE) trial 67 RENAAL study 30, 72 retinal detachment 145–6 retinal haemorrhage 144 retinal hypoxia 166 retinal ischaemia 142 retinal mean circulation time 216 retinal neovascularization 146 retinal photography 148 retinal pigment epithelium 156 retinal vascular abnormalities 173–4, 173 retinopathy 14–16, 14, 15, 139–50 ACE inhibitors 134 background 139–42 blindness 129, 137 circinate exudate rings 142 classification 146–7, 147 clinically significant macular oedema 142 cotton wool spots 139, 141, 142, 143, 173, 174 deep retinal haemorrhages 144 diagnosis 147 direct ophthalmoscopy 148 disc new vessels 145, 164 epidemiology 129–37 evidence-based interventions 74–5, 74, 75 exudates 141, 142 familial clustering 135 growth factors 166 haemorrhages 140–1, 140 incidence 132 intraretinal microvascular abnormalities 143, 165, 165

INDEX

iris neovascularization 146, 164, 165 and microalbuminuria 134–5 microaneurysms 140 neovascular glaucoma 146, 172 new vessels elsewhere 145, 146, 163, 164, 165 preproliferative 142–4, 142 prevalence 129–31, 130, 131 proliferative see proliferative retinopathy retinal detachment 145–6 risk factors 132–7, 133 age 129, 134 blood pressure 25–6, 134 cholesterol 135 duration of diabetes 132 genetics 135–6 glycosylated haemoglobin 15, 16, 136 hyperglycaemia 132, 133 hyperlipidaemia 135 nephropathy 134–5 ocular 137 pregnancy 136 ruboxistaurin 213–17, 215–17, 221–4, 222, 223 screening 148–9 slit-lamp biomicroscopy 148 stages of 139 type 1 diabetes 131 type 2 diabetes 131 UK national screening scheme 149 vascular permeability factor 166 venous abnormalities 144 see also macular oedema; maculopathy revascularization angina 66 diabetic foot 107 risk assessment 38–9 risk factors 13–22 rosiglitazone 43 ruboxistaurin 124, 213 clinical trials 221–4 endothelial dysfunction 221, 222 IC50 values 214 nephropathy 216–17, 218–19 peripheral neuropathy 224 retinopathy 213–17, 215–17, 221–4, 222, 223 structure 214 Rydel-Seiffer tuning fork 97 S San Luis Valley Diabetes Study 93 Scandinavian Simvastin Survival Study 67 selective serotonin-reuptake inhibitors 116, 125

Semmes-Weinstein monofilament 97 serine/threonine kinases 190 sildenafil 51, 54, 56 simvastatin 41, 68 slit-lamp biomicroscopy 148 smoking 19–21, 20, 67 sorbinil 124 Sorbinil Retinopathy Trial 135 sorbitol 187, 188 spinal cord stimulation 119 Starling forces 198 statins 43–4 coronary heart disease 43 myocardial infarction 40, 66–7 nephropathy 33 Steno-2 study 85 Stockholm Diabetes Intervention Study 113 string-of-sausages appearance 144 stroke 45–50 aetiology 46 clinical presentation 46–7 epidemiology 45 evidence-based practice 47–9, 68–71, 69, 70 hyperglycaemia 46, 47, 48 mortality 45 secondary prevention 48 thrombolysis 47, 61 treatment aspirin 47, 48 DIGAMI therapy 48 heparin 48, 69 ramipril 48, 65 substance P 118, 206 sugar cataracts 171 superoxide dismutases 87 sweating, gustatory 32, 81 SYDNEY study 87 T tadalafil 54 talin 199 tenilsetam 86 therapeutic intervention targets 179–95 thermal thresholds 99 thienopyridines 62 thrifty gene hypothesis 5 thrombolysis 47, 61 thrombotic thrombocytopenic purpura 48 ticlopidine 62 timolol 63 tissue plasminogen activator 61 tolrestat 124 tramadol 118, 125 trandalopril 87

233

234

INDEX

transforming growth factor-β 217 transient ischaemic attacks 47 triamcinolone 160 tricyclic antidepressants 116, 125 type 1 diabetes epidemiology 4 natural history 3–4 retinopathy 131 type 2 diabetes epidemiology 4–6 natural history 4 retinopathy 131 tyrosine kinases 190 U UK Prospective Diabetes Study (UKPDS) 4, 30, 42, 54, 59, 132, 181 University Group Diabetes Program 42 urinary albumin excretion rate 197 US Diabetes Control and Complications Trial (DCCT) 132 UTDWCS classification 106 uveitis 172, 174 V VA Cooperative Study on type 2 Diabetes Mellitus (VACSDM) 85 vardenafil 54 vascular adhesion molecule-1 209 vascular endothelial growth factor 89, 115 vascular permeability see endothelial permeability vascular permeability factor 166, 200, 202, 213

vascular risk assessment 38–9 vasodilators 124 Viagra (sildenafil) 51, 54, 56 vibration perception threshold 88, 92, 98–9 vimentin 200 vinculin 199 vitamin E 194–5 vitrectomy see pars plana vitrectomy vitreous detachment 166 Volk Quadraspheric lens 166 Volk Transequator 158 von Willebrand factor 37 W Wagner classification 106 warfarin 48, 61 WARIS study 61 WHO Multinational Study of Vascular Disease 298 Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR) 15–16, 67, 129, 151, 163 X xanthine oxidase 206 Y yohimbe 57 Z zenarestat 124 zopolrestat 124