Management of Diabetic Retinopathy

Management of Diabetic Retinopathy Clinical Practice Guidelines June 1997 NHMRC National Health and Medical Research Council © Commonwealth of Au...
Author: Rudolf Boyd
2 downloads 2 Views 353KB Size
Management of Diabetic Retinopathy

Clinical Practice Guidelines June 1997

NHMRC National Health and Medical Research Council

© Commonwealth of Australia 1997 ISBN 0 642 27260 3 This work is copyright. It may be reproduced in whole or in part for study or training purposes subject to the inclusion of an acknowledgment of the source and no commercial usage or sale. Reproduction for purposes other than those indicated above, requires the written permission of the Australian Government Publishing Service, GPO Box 84, Canberra ACT 2601. The strategic intent of the NHMRC is to work with others for the health of all Australians, by promoting informed debate on ethics and policy, providing knowledge based advice, fostering a high quality and internationally recognised research base, and applying research rigour to health issues. The production and distribution of these guidelines were funded under the National Diabetes Strategy.

National Health and Medical Research Council documents are prepared by panels of experts drawn from appropriate Australian academic, professional, community and government organisations. NHMRC is grateful to these people for the excellent work they do on its behalf. This work is usually performed on an honorary basis and in addition to their usual work commitments. Typesetting by Arawang, Canberra for Commonwealth Department of Health and Family Services

iii

About these guidelines

This document is a set of clinical practice guidelines developed by the National Health and Medical Research Council (NHMRC). These guidelines and modified versions for consumers, General Practitioners, Optometrists, Nurses and other health practitioners have all been developed by an NHMRC working party with representatives from general practice, optometry, ophthalmology, endocrinology, the nursing profession, consumer movement and Aboriginal and Torres Strait Islander people. This document is a general guide to appropriate practice, to be followed only subject to the clinician’s judgement in each individual case. These guidelines are designed to provide information to assist decision making and are based on the best information available at the date of publication. Should you wish to obtain further copies of Clinical Practice Guidelines for the Management of Diabetic Retinopathy they can be obtained by contacting: The Secretariat National Diabetes Strategy Diabetes Australia AVA House 5/7 Phipps Place DEAKIN ACT 2600 Phone: (02) 6285 3277 Fax: (02) 6285 2881 Copies of the consumer guide titled Diabetes and Your Eyes: A consumer guide for the management of diabetic retinopathy are available free of charge from the National and State/Territory Offices of Diabetes Australia.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

v

Foreword

In recognition of the need for procedures which ensure effective clinical practices, and thereby improve health outcomes, the National Health and Medical Research Council has embarked on a national program to develop clinical practice guidelines. The objective of the guidelines is not only to assist practitioners to make decisions about appropriate health care for specific clinical circumstances, but also to assist consumers by providing them with comprehensive information about choices available in their treatment. One of the basic premises of the guidelines development process is that they should be based on the best available evidence. The adoption of a multidisciplinary approach, involving all stakeholders is a further key principle. The management of diabetic retinopathy was chosen as one of the first areas for guideline development. This is an appropriate choice given the fact that diabetic retinopathy is one of the most easily preventable eye diseases with appropriate identification and treatment. I am confident that these guidelines will assist practitioners to identify, treat and manage the condition and reduce the incidence of diabetic retinopathy in Australia. The National Health and Medical Research Council is particularly indebted to the meticulous and enthusiastic way in which members of the Working Party and in particular Professor Paul Mitchell and Jie Jin Wang (technical writers/researchers) have drawn together the evidence and distilled it into these comprehensive and practical clinical practice guidelines.

Richard Larkins Chairman National Health and Medical Research Council

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

vii

Contents

About these guidelines ............................................................................. iii Foreword ..................................................................................................... v List of tables and figures .......................................................................... ix Summary of guidelines .............................................................................. x List of acronyms used .......................................................................... xviii Introduction ................................................................................................ 1 1.

Diabetic retinopathy ......................................................................... 3 1.1 1.2 1.3 1.4 1.5 1.6

2.

Screening for diabetic retinopathy ............................................... 23 2.1 2.2 2.3 2.4 2.5 2.6

3.

Definitions .......................................................................................................3 Pathogenesis .....................................................................................................4 Prevalence and incidence .................................................................................6 Risk factors .................................................................................................... 12 Grading .......................................................................................................... 18 Cataract .......................................................................................................... 22

Timing and frequency of eye examinations in people with diabetes .............23 Screening methods ......................................................................................... 26 Different examiners ....................................................................................... 28 Role of pupil dilatation .................................................................................. 29 Fluorescein angiography ................................................................................ 31 Co-ordinated care ...........................................................................................32

Treatment of diabetic retinopathy ................................................. 34 3.1 3.2 3.3 3.4 3.5

Laser treatment .............................................................................................. 34 Fluorescein angiography ................................................................................ 40 Vitrectomy surgery......................................................................................... 42 Other therapies ............................................................................................... 45 Cataract surgery ............................................................................................. 48

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

viii

4.

Management of diabetic retinopathy in Aboriginal and .............. 50 Torres Strait Islander people 4.1 4.2 4.3

5.

Prevalence of diabetes ................................................................................... 50 Age of onset of diabetes ................................................................................. 51 Access to eye care services and screening ..................................................... 51

Community impact of diabetic retinopathy .................................. 52 5.1

Cost of diabetic retinopathy ........................................................................... 52

References ................................................................................................ 58 Appendices ............................................................................................... 77 1. 2. 3. 4.

Working party terms of reference and membership ....................................... 77 The guideline development process ............................................................... 79 The task and scope of literature review ......................................................... 81 Glossary of common terms ............................................................................ 86

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

ix

Tables and Figures Tables Table 1 Table 2 Table 3 Table 4 Table 5 Table 6

Overall summary of key points in the management of diabetic retinopathy ............................................................................ xv Collective summary of guidelines for the management of diabetic retinopathy by level of evidence ..................................... xvi Quality of evidence ratings ...................................................................2 Incidence rates (% per year) in Australia ...........................................10 Classification of diabetic retinopathy into retinopathy stages (Wisconsin level) and predictive value of retinal lesions ...................20 Summary of management recommendations .....................................39

Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5

Figure 6 Figure 7 Figure 8 Figure 9

Prevalence of retinopathy for IDDM cases ..........................................8 Prevalence of retinopathy for NIDDM cases .......................................9 Meta-analysis of effects of intensive blood-glucose control on late complications of type I diabetes ............................................. 13 Cumulative incidence of a sustained change in retinopathy with patients with IDDM receiving intensive or conventional therapy ...... 14 Sensitivity of model (savings in 1986 dollars) to changes in the sensitivity of a single screening visit in detecting diabetic retinopathy, using dilated ophthalmoscopy annually for patients with no retinopathy and every six months for those with retinopathy ........................................................................25 Proportion of untreated eyes with proliferative diabetic retinopathy from Diabetic Retinopathy Study ....................................35 Annual savings (cost) with 80% compliance compared to biennial screening policy with 30% compliance ................................56 Yearly savings: increasing compliance from 30% to 100% ($34 exam) ..........................................................................................57 Yearly savings: increasing compliance from 30% to 100% ($68 exam) ..........................................................................................57

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

x

Summary of the guidelines

The following is a summary of guidelines contained in this document. The reader should consult the relevant section of the document for further details and a discussion of the evidence upon which each guideline is based. Some guidelines are accompanied by a quality of evidence rating (level I, II, III, IV). A level I rating indicates that the guideline is based on the highest quality evidence whereas a level III or IV rating indicates the statement or recommendation is based on lower quality evidence. Full details of this rating system are contained in Table 3 of the Introduction.

1.

Diabetic retinopathy

1.1

Definitions

Diabetic retinopathy may be defined as the presence of typical retinal microvascular lesions in an individual with diabetes. The earliest lesions visible with the ophthalmoscope are termed non-proliferative retinopathy (NPDR), including microaneurysms, haemorrhages, hard exudates, cotton wool spots, intraretinal microvascular abnormalities (IRMAs) and venous beading (VB). The proliferative stage (PDR) is characterised by growth of new vessels and fibrous tissue and pre-retinal or vitreous haemorrhage. New vessels on or within one disc diameter of the disc are termed new vessels on the disc (NVD) and in other locations, new vessels elsewhere (NVE). Leak from macular capillaries results in macular oedema, and when present close to the central macula, is termed clinically significant macular oedema (CSME). No individual retinopathy sign is specific for diabetes, but all may be observed in other disease processes. It is the pattern, symmetry and evolution of the lesions that characterises diabetic retinopathy. Two common types of diabetes are recognised, insulin-dependent diabetes mellitus (IDDM), and non insulin-dependent diabetes mellitus (NIDDM). Other types (secondary and gestational diabetes or diabetes occurring in genetic syndromes) are uncommon. In this report, IDDM refers to diabetes with onset before age 30 and treated with insulin, while NIDDM refers to diabetes diagnosed after age 30 and treated with either diet alone, oral therapy or insulin.

1.2

Pathogenesis

Many biochemical pathways link the altered glucose metabolism seen in diabetes directly to development and progression of diabetic retinopathy. A multifactorial biochemical pathogenesis is likely, involving products of the aldose reductase pathway, increased nonenzymatic glycation of proteins, activated protein kinase C with increased vasodilatory prostaglandins, and increased production of growth factors in the retina. Increased blood retinal barrier permeability and alterations in retinal blood flow may also be important in the pathogenesis.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

xi

1.3

Prevalence and incidence

In Australian community diabetic populations, the prevalence of retinopathy ranges from 21 to 36%, while vision-threatening retinopathy occurs in 6 to 13%. The annual incidence of retinopathy for people with no retinopathy at baseline is 6 to 14% per year, depending on the duration of diabetes and glycaemic control. IDDM accounts approximately for 10%, NIDDM for most of the remaining 90%, while other types of diabetes comprise around 2% of cases.

1.4

Risk factors (Guidelines 1–3, page 12)

The known duration of diabetes is one of the most important factors determining the presence of diabetic retinopathy (level I evidence). Elevated blood glucose is causally related to the development and progression of diabetic retinopathy (level I evidence). Other risk factors including hypertension, pregnancy, nephropathy and elevated blood lipids may also play a role.

1.5

Grading

The Wisconsin or modified Airlie House classification, is the basis for detailed grading of diabetic retinopathy and was used in all major risk factor studies and treatment trials, including the Diabetes Control and Complication Trial (DCCT), the Diabetic Retinopathy Study (DRS) and the Early Treatment Diabetic Retinopathy Study (ETDRS). The ETDRS quantified the risk of progression associated with severity of individual lesions, from photographic grading and developed a seven level classification: No retinopathy, Minimal, Mild, Moderate and Severe NPDR, PDR and High-Risk PDR. A simpler, qualitative European protocol was also developed to identify lesions requiring referral.

1.6

Cataract

A well-documented association (level II evidence) between early-onset cortical or posterior subcapsular cataract and diabetes has been found in population studies, both in the U.S. and Australia.

2.

Screening for diabetic retinopathy

2.1

Timing and frequency of eye examinations in people with diabetes (Guidelines 4–10, page 23)

Early detection of sight-threatening retinopathy by regular eye examinations conducted by trained personnel is the key to reducing visual loss and blindness from diabetic retinopathy. In computer simulation models, the most cost-effective strategy is to perform yearly or 2yearly examinations of all people with diabetes, from the time of diagnosis, followed by more frequent examinations if retinopathy is detected. For children with onset of diabetes before puberty, eye screening should commence at puberty or earlier in some circumstances. Non-ophthalmologist screeners should refer cases found with Mild NPDR routinely to an ophthalmologist. Cases with Moderate or Severe NPDR, patients with decreased corrected

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

xii

visual acuity, or people in whom the fundus can not be adequately examined, need a referral as soon as possible. If sight-threatening retinopathy (PDR, macular oedema or CSME) is detected, an urgent referral is needed. Referral may not be needed for Minimal NPDR if visual acuity is normal. Ophthalmologists should repeat eye exams every six to twelve months in cases with Mild NPDR, every three to six months if Moderate or Severe NPDR is found and more frequently if threatened maculopathy is present.

2.2/2.3

Screening methods (Guidelines 11–14, page 26) and different examiners

Sensitivities and specificities of different screening methods performed by different screeners for detecting diabetic retinopathy vary widely. The sensitivity of ophthalmoscopic screening is substantially enhanced if pupils are dilated, without significant discomfort or side effects. Measurement of visual acuity is an essential part of the examination. Where feasible, general practitioners, optometrists and physicians should more actively screen their patients for diabetic retinopathy. Their abilities to detect retinopathy need to be improved by regular education and more frequent practice to exceed a sensitivity target of 60%.

2.4

Role of pupil dilatation (Guidelines 15–16, page 29)

Short-term pupil dilatation is safe, with tropicamide 0.5% eye drops, and patient acceptance is very high. Most people can drive safely after pupils are dilated, with sunglasses. Acute angle closure glaucoma is an extremely rare side effect (level II evidence).

2.5

Fluorescein angiography (FA) (Guideline 17, page 31)

Fluorescein angiography is inappropriate to use in screening for diabetic retinopathy, as it is invasive and has a risk of complications.

2.6

Co-ordinated care

Development and evaluation of a national program of retinopathy detection and management is a priority and should be included in the ‘National Action Plan: Diabetes to the Year 2000 and Beyond’ from Diabetes Australia. Retinopathy screening needs to be part of the systematic care of all people with diabetes and can be performed by many different professionals. General practitioners have a responsibility to ensure that all their diabetic patients have regular eye examinations.

3.

Treatment of diabetic retinopathy

Medical management Every effort should be made to achieve strict glycaemic control for all patients with diabetes, while taking care to avoid severe hypoglycaemia (level II evidence). It is also important to adequately treat elevated blood pressure and serum lipids in people with diabetes (level III evidence).

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

xiii

3.1

Laser treatment (Guidelines 18–25, page 34)

The DRS and ETDRS trials have shown conclusively that laser treatment is effective therapy for patients with both PDR and CSME (level I evidence). Panretinal Photocoagulation (PRP) treatment with laser was associated with a 50% reduction in development of severe visual loss in PDR patients. Focal or grid macular laser reduced the risk of moderate visual loss by at least 50% in cases of CSME. For high-risk PDR, perform PRP as soon as possible. For early PDR, commence PRP after any signs of maculopathy are stabilised. Consider focal or grid laser treatment for eyes with CSME. Consider either laser or deferral, for macular oedema not meeting CSME criteria, depending upon progression of signs, the fellow eye, or ability to follow closely. For eyes with both PDR and CSME, but without high-risk PDR, delay PRP until focal or grid laser is completed. To achieve ETDRS results, maintain intensive follow-up after completing laser, and repeat laser treatment if risk signs fail to regress or re-develop. Carefully follow all cases of Mild or Moderate NPDR, to observe for signs of visionthreatening retinopathy.

3.2

Fluorescein angiography (Guidelines 26–31, page 40)

FA will identify sources of perimacular leakage and guide laser treatment of macular oedema (level II evidence). FA may not be needed in the treatment of PDR, but can be useful to assess signs of retinal ischaemia. In some cases, FA can identify new vessels not otherwise seen. Routine use of FA in managing diabetic retinopathy at present should be guided by clinical experience as little evidence is available to provide firm guidelines. The most frequent side effects of FA are nausea, with occasional vomiting or urticaria. Although more serious side effects are rarely reported, it is important that resuscitation equipment is available.

3.3

Vitrectomy surgery (Guidelines 32–37, page 42)

Pars plana vitrectomy within three months may benefit IDDM patients with severe persistent vitreous haemorrhage and very severe PDR (level II evidence). However, there is less benefit for vitrectomy in NIDDM patients. Early vitrectomy may also be considered in eyes with severe PDR, not responding to aggressive photocoagulation (level II evidence). For traction retinal detachment, vitrectomy can serve as a last resort to save vision (level IV evidence). Frequent complications include vitreous haemorrhage and cataract. Neovascular glaucoma, epiretinal membrane formation, macular scarring and rhegmatogenous retinal detachment are less frequent complications. PRP should always be attempted prior to considering vitrectomy to reduce activity of new vessels as much as possible.

3.4

Other therapies

Apart from intensive glycaemic control and laser treatment, no other therapies have been shown to influence development or progression of retinopathy, in multicentre randomised controlled trials (RCTs). Ticlopidine has shown benefit in a single-centre RCT. Some other agents have shown promise in small RCTs or case series, but benefits have not been confirmed in larger studies. Human trials are now planned for aminoguanidine and other advanced glycation endproduct inhibitors, which have shown promise in animal studies. The use of aspirin is safe in patients with diabetic retinopathy, when indicated for

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

xiv

concomitant vascular disease, but its use confers no other benefit for retinopathy (level II evidence).

3.5

Cataract surgery (Guidelines 38–42, page 48)

Consistent findings in case series and a meta-analysis, suggest that diabetes may significantly worsen the visual outcome after cataract surgery (level III evidence). Preoperative retinopathy, particularly macular oedema, influences the visual outcome after cataract surgery due to asymmetric retinopathy progression in the operated eye and an increased risk of neovascular glaucoma (level IV evidence). Carefully assess retinopathy in diabetic patients with significant cataract. Where possible, treat any macular oedema or threatened maculopathy with focal or grid laser, prior to surgery and consider delaying surgery until retinopathy and maculopathy is stabilised. This may take two months or more to stabilise after laser treatment.

4.

Management of diabetic retinopathy in Aboriginal and Torres Strait Islander people (Guidelines 43–46, page 50)

Aboriginal and Torres Strait Islander communities have a higher prevalence and earlier onset of NIDDM than in Caucasian populations (level II evidence). Poor access to and low utilisation of services may contribute to their high rate of diabetic complications. Involvement of Aboriginal health workers in community-based screening for diabetic retinopathy is desirable. Links should be established and fostered between Aboriginal and Torres Strait Islander communities and the nearest accessible ophthalmologists and optometrists. Further data on the prevalence, incidence and progression of diabetic retinopathy in Aboriginal and Torres Strait Islander populations are needed to plan eye care services

5.

Community impact of diabetic retinopathy

Australian data indicate that a significant proportion of people with diabetes are not receiving adequate screening or follow-up for diabetic retinopathy. Substantial cost savings have been estimated from screening programs for diabetic retinopathy in the US population. An estimate of the cost of diabetic retinopathy to the Australian community is around $46 million per annum. Annual savings from a one or two-yearly diabetic retinopathy screen for all people known to have diabetes are predicted to be substantial.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

xv

Table 1

Overall summary of key points in management of diabetic retinopathy

A multifactorial biochemical pathogenesis of diabetic retinopathy is likely, involving the aldose reductase pathway, increased non-enzymatic glycation of proteins and increased production of growth factors and prostaglandins. The retinopathy prevalence in Australian diabetic populations ranges from 21 to 36%, with visionthreatening retinopathy in 6% to 13%. Incidence rates for new retinopathy lesions vary from 6–14% per year and are related to diabetes duration and glycaemic control. Combined effects of diabetes duration and glycaemic control are the two principal factors determining both development and progression of diabetic retinopathy, in both IDDM and NIDDM. Because of its frequency, the overwhelming majority of retinopathy occurs in NIDDM cases. Well-documented associations have been reported between diabetes and early-onset cataract (cortical and posterior subcapsular). Pupil dilatation is both safe and acceptable, and is essential for effective eye examinations. The timing of regular eye examinations should be planned and based on the presence and severity of diabetic retinopathy found at the first examination. The DRS and ETDRS trials have shown conclusively that laser treatment is effective in reducing the risk of visual loss in patients with both PDR and CSME. The benefits of early vitrectomy for non-resolving severe vitreous haemorrhage are less for NIDDM patients. A meta-analysis of ten case series suggests that diabetes may significantly worsen the visual outcome after cataract surgery. Pre-operative retinopathy, particularly maculopathy, influences the visual outcome after cataract surgery due to retinopathy progression in the operated eye. Apart from intensive glycaemic control and laser treatment, no other therapies to date have been shown to influence the development or progression of retinopathy in large multicentre randomised controlled trials.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

xvi

Table 2

Collective summary of guidelines for the management of diabetic retinopathy by level of evidence

Collective guidelines

Level of evidence

Strict glycaemic control is the most important modifiable factor in preventing development of retinopathy. Efforts to achieve strict glycaemic control should be undertaken in all diabetic patients, while taking care to avoid severe recurrent hypoglycaemia.

I

All people diagnosed with diabetes need regular eye examinations by appropriately trained screeners, using a test with adequate sensitivity and specificity. This should exceed a sensitivity of at least 60%.

II

Commence eye examinations at the time of diabetes diagnosis. For children with onset of diabetes before puberty, commence screening at puberty. Earlier examinations may be appropriate in some circumstances.

II

If no retinopathy is present, repeat the eye examination at least every two years.

II

Minimal NPDR (isolated microaneurysms only) may not require referral to an ophthalmologist if visual acuity is normal, but needs review examination at least yearly.

IV

Refer routinely to an ophthalmologist if Mild NPDR is found.

IV

Refer as soon as possible if Moderate or Severe NPDR is found, if vision is reduced or has worsened or if the fundus can not be adequately examined.

IV

Refer urgently if PDR or macular oedema is found.

II

Once NPDR without macular oedema is found, repeat eye examinations every six to twelve months if Minimal or Mild NPDR and every three to six months if Moderate or Severe NPDR.

IV

Individual physicians, general practitioners and optometrists should actively screen for diabetic retinopathy and need regular education to maintain a sensitivity of at least 60% for detecting retinopathy using dilated ophthalmoscopy.

III-2

Involve aboriginal health workers in Aboriginal and Torres Strait Islander community-based screening for diabetic retinopathy. Establish links between Aboriginal and Torres Strait Islander communities and the nearest accessible ophthalmologists and optometrists.

IV

For high-risk PDR, perform PRP as soon as possible. For other PDR, commence PRP after completing macular laser.

II

Consider focal or grid macular laser for CSME or macular oedema not meeting CSME criteria.

II

Perform fluorescein angiography if macular oedema is present and use the angiogram to guide focal or grid laser.

II

Consider fluorescein angiography to evaluate visual loss or determine the source of vitreous haemorrhage if new vessels are suspected, but not seen.

IV

Consider early vitrectomy for severe vitreous haemorrhage, particularly in IDDM or if advanced new vessels do not respond to extensive PRP.

II

Always attempt panretinal laser prior to considering vitrectomy.

II

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

xvii

Collective guidelines

Level of evidence

Consider vitrectomy to relieve macular/ other traction in advanced PDR.

IV

Carefully assess retinopathy in diabetic patients with significant cataract. Where possible, treat any macular oedema or threatened maculopathy with focal or grid laser, prior to cataract surgery.

III-3

Delay cataract surgery until retinopathy and macular oedema are stabilised.

III-3

Aspirin can be used safely in patients with diabetic retinopathy, when indicated for vascular disease, but its use confers no benefit for retinopathy.

II

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

xviii

List of acronyms used AACG

acute angle closure glaucoma

AGEs

Advanced glycation endproducts

ARIs

Aldose reductase inhibitor drugs

BDES

Beaver Dame Eye Study

BMES

Blue Mountains Eye Study

CSME

Clinically Significant Macular Oedema

CWS

Cotton Wool Spot

DCCT

Diabetes Control & Complications Trial

DR

Diabetic Retinopathy

DRS

Diabetic Retinopathy Study

DRVS

Diabetic Retinopathy Vitrectomy Study

EDCS

Pittsburgh Epidemiology of Diabetes Complications Study

ETDRS

Early Treatment Diabetic Retinopathy Study

FA

fluorescein angiography

FGF

fibroblast growth factor

GH

growth hormone

HEx

hard exudates

HRC

High Risk Characteristics

IDDM

Insulin Dependent Diabetes Mellitus

IGF

insulin-like growth factor

IGT

Impaired Glucose Tolerance

IRMA

Intra Retinal Microvascular Abnormalities

MVIP

Melbourne Visual Impairment Project

NIDDM

Non Insulin Dependent Diabetes Mellitus

NPDR

Non-Proliferative Diabetic Retinopathy

NVD

New Vessels on the (optic) Disc

NVE

New Vessels Elsewhere

PDR

Proliferative Diabetic Retinopathy

PRP

Panretinal Photocoagulation

RCT

Randomised Controlled Trial

VB

Venous Beading

WESDR

Wisconsin Epidemiologic Study of Diabetic Retinopathy

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

1

Introduction The National Health and Medical Research Council through its Standing Committee on Quality of Care and Health Outcomes (QCHOC), established a working party in 1995 to develop clinical practice guidelines for the management of diabetic retinopathy. This is part of the national program to promote the development of evidence-based clinical practice guidelines which are aimed at improving the quality of health care and patient health outcomes. Diabetic retinopathy was chosen as a target disease for development of clinical practice guidelines because of its importance as a cause of blindness in our young and middle aged population. Data from randomised clinical trials indicate that visual loss and blindness from diabetic retinopathy is preventable with timely photocoagulation treatment. Yet many people with diabetes suffer preventable visual loss as they present too late in the course of the disease for effective treatment. There is now a need to disseminate information about diabetic retinopathy to both health professionals and to consumers, those people with diabetes. This may assist both groups to comply with appropriate management, thereby reducing the toll of avoidable visual loss from diabetic retinopathy. The development of these guidelines were based on available evidence, following an extensive review of the literature from 1966 until May 1996. During the literature review process, answers were generated to a series of twelve questions set by the committee. (Appendix 3: The task and scope of literature review). The guidelines resulted from deliberations of the working party, and used the literature review to provide levels of evidence underpinning the recommendations. Clinical Practice Guidelines for the Management of Diabetic Retinopathy aims to be useful to both health professionals and consumers. It should educate all involved in the care of people with diabetes about diabetic retinopathy. It is designed to assist professionals in making management decisions about diabetic retinopathy, thus improving the quality of care. This book provides management guidelines and does not replace careful examination of the relevant literature. It should, however, assist the reader in identifying key references that provide evidence for treatment recommendations. Readers can also judge the strength of evidence on which the guidelines are based, from the quoted levels of evidence. In applying these guidelines to clinical practice, it is necessary to take into account all aspects of the clinical picture. The guidelines aim to provide information for use in making management decisions, rather than dictate a particular course of management. A four-point rating scale is used to indicate the evidence level for key decision points. This scale is recommended by NH&MRC1, after adaptation from the US Preventive Services Taskforce.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

2

Table 3

Quality of evidence ratings Controlled trials1

Prevalence, risk factors & sensitivity studies

I

evidence obtained from a systematic review of all relevant randomised controlled trials

evidence obtained from a systematic review of all available population-based studies

II

evidence obtained from at least one properly-designed randomised controlled trial

evidence obtained from a well-designed population-based study or representative cohort study

Levels

III-1

evidence obtained from well-designed controlled trials without randomisation

III-2

evidence obtained from well-designed cohort or case-control analytic studies, preferably from more than one centre or research group

III-3

evidence obtained from multiple time series with or without the intervention

IV

evidence obtained from a well-designed casecontrol study, cohort study or less welldesigned population-based study

opinions of respected authorities, based evidence obtained from descriptive case on clinical experience, descriptive studies, series, clinical experiences, respected or reports of expert committees authorities, or reports of expert committees

Desirable study design features were also used to assess the level of evidence: 1.

Avoidance of selection bias, measurement error or misclassification in determining diagnosis of retinopathy in prevalence and incidence studies.

2.

Control for known and potential confounders while investigating effects of other variables on development and progression of retinopathy in risk factor studies.

3.

Inclusion of a wide spectrum of retinopathy, from mild to severe in screening studies2.

4.

For intervention trials, including randomised control trials, additional criteria3, such as randomised allocation of treatment, analysis by intention-to-treat, masking and homogeneity, were used.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

3

1. Diabetic retinopathy 1.1

Definitions

Diabetic retinopathy Diabetic retinopathy may be defined as the presence of typical retinal microvascular lesions in an individual with diabetes. Microaneurysms (Ma), haemorrhages, hard exudates (HEx), cotton wool spots (CWS), intraretinal microvascular abnormalities (IRMAs), venous beading (VB), new vessels and fibrous tissue comprise the clinical features of diabetic retinopathy. However, none of these individual lesions are specific for diabetes, as they may occur in other disease processes such as hypertension, hyperviscosity, inflammation or radiation. It is the pattern, symmetry and evolution of the lesions that characterises the appearance as diabetic retinopathy. Diabetic retinopathy is first evident ophthalmoscopically as non-proliferative (background) retinopathy (NPDR), which is characterised by Ma, dot, blot or flame haemorrhages, HEx, CWS (soft exudates), IRMAs and VB. The proliferative stage of diabetic retinopathy (PDR) is characterised by the growth of abnormal new vessels and fibrous tissue in response to retinal ischemia, and the development of pre-retinal or vitreous haemorrhage. If new vessels appear on or within one disc diameter of the disc margin, they are known as new vessels on the disc (NVD). In any other location, they are referred to as new vessels elsewhere (NVE). Leak from capillaries in the macular or perimacular region results in retinal thickening or macular oedema, defined as thickening located within two disc diameters of the centre of the macula. When this is present within or close to, the central macula, it is termed clinically significant macular oedema (CSME)4. The presence of CSME is best assessed by stereo slit lamp biomicroscopy (with or without a contact lens) or from stereo photographs of the macula.

Diabetes Two common types of diabetes are recognised, insulin-dependent diabetes mellitus (IDDM), also known as type I diabetes, and non insulin-dependent diabetes mellitus (NIDDM), also known as type II diabetes. Other types of diabetes are relatively uncommon and include diabetes secondary to pancreatic diseases, gestational diabetes and diabetes occurring as part of a genetic syndrome. Differentiation between IDDM and NIDDM may not be obvious and there is some overlap in the age at onset. For the purposes of this report, however, cases with diabetes onset prior to age 30 and treated with insulin (younger-onset) will be considered to have IDDM, while people with diabetes diagnosed from age 30 (older-onset), and treated with either diet alone, oral therapy or insulin, will be considered to have NIDDM.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

4

1.2

Pathogenesis

Key Points •

Many biochemical pathways link the altered glucose metabolism seen in diabetes directly to development and progression of diabetic retinopathy.



A multifactorial biochemical pathogenesis is likely, involving products of the aldose reductase pathway, increased non-enzymatic glycation of proteins, and increased production of growth factors and prostaglandins in the retina.



These biochemical changes are accompanied by increased blood retinal barrier permeability and alterations in retinal blood flow.

What are the current theories about pathogenesis of diabetic retinopathy? Following a series of animal studies of diabetic retinopathy5–7, clinical trial results from the Diabetes Control and Complications Study (DCCT)8–10 and the epidemiological data from the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR)11,12 have emphasised the strong relationship of glycaemic control to the development and progression of diabetic retinopathy. Chronic hyperglycaemia is now accepted as the common pathway leading to diabetic retinopathy13,14. A number of mechanisms appear to be involved15, including the sorbitol or aldose reductase pathway, non-enzymatic glycation or glycosylation of proteins, production of growth factors in the retina, as well as retinal capillary blood flow changes and increased capillary permeability. These biochemical links between hyperglycaemia and diabetic retinopathy were recently summarised by Larkins16, who described a number of plausible biochemical pathways linking glucose metabolism directly to the development of diabetic retinopathy: the aldose reductase pathway, increased protein kinase C activity with increased vasodilatory prostaglandins production, increased non-enzymatic glycation, and glucose induced auto-oxidative damage. Aldose reductase (sorbitol) pathway The pathway is active during hyperglycaemia16,17, with accumulation of sorbitol. In critical retinal vascular tissues, it causes damage to retinal pericytes and basement membrane thickening, with closure of retinal capillaries13. These changes have long been identified as the key early lesions of diabetic retinopathy18,19. Aldose reductase inhibitor drugs (ARIs) have the potential to influence the sorbitol pathway and have been subject to a number of trials. However, conflicting results have been found in trials using animal models, with consistent beneficial effects on the pericyte and basement membrane changes found by one group of investigators20–22, but not by a second23. Human trials have been conducted for a number of ARIs, including Sorbinil and Ponalrestat24,25, but no trial has shown any convincing benefit. Experimental work using this model is continuing26,27. Increased nonenzymatic glycation Nonenzymatic glycation (glycosylation) of many different proteins accompanies diabetes and may play a significant role in the pathogenesis of retinopathy28,29. High glucose

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

5

concentrations modify long-lived macromolecules, forming advanced glycation endproducts (AGEs). These cause changes in extracellular matrix components, including cell adhesion and other cell function changes, associated with retinopathy. The potential for inhibiting AGEs by aminoguanidine is now being explored in human trials29, as it has been shown to decrease diabetic retinopathy signs in animal models by decreasing endothelial cell proliferation and reducing pericyte dropout30. Activated protein kinase C with vasodilatory prostaglandins production In hyperglycaemic conditions, there is an increased flux of glucose through the pentose phosphate pathway with increased protein kinase C activation and increased vasodilatory prostaglandins production. This could be a mechanism for the increased retinal blood flow and vascular permeability which occurs in retinopathy and may also be linked to the aldose reductase pathway16. Production of growth factors in the retina Considerable research has investigated factors that might increase or inhibit retinal blood vessel proliferation31 with a number of ‘growth factors’ identified14,31. These include acidic and basic fibroblast growth factor (aFGF and bFGF), insulin-like growth factor (IGF), vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF). Release of these factors in response to ischaemia is a current hypothesis for the pathogenesis of diabetic retinopathy32. Tissue immunocytochemical studies have localised FGF in patients with diabetic neovascularisation14,33. Similarly, elevated serum IGF-1 levels has been measured in patients with proliferative retinopathy, both early-onset34 and late-onset diabetes35. VEGF is a multifunctional cytokine expressed and secreted by many cells, and is a potent permeability factor. It has recently been shown to influence blood-retinal barrier function in diabetic retinopathy36 and found in increased vitreous concentrations in eyes with PDR37. A role for PDGF ligands and receptors has been postulated in the pathogenesis of different proliferative diseases, such as proliferative vitreoretinopathy, as well as in PDR38. Alterations to retinal blood flow in diabetic retinopathy Laser Doppler studies have shown that retinal blood flow is increased in diabetic retinopathy by comparison with non-diabetic controls and diabetic subjects with no retinopathy39. However, in the presence of severe capillary non-perfusion and PDR, blood flow rates are reduced40. Retinal arteriolar dilatation41,42 appears to counteract increasing resistance in smaller vessels41, leading to vascular leakage42. Retinal capillary pericyte contractility is inhibited by high glucose concentrations, which is consistent with the hypothesis that increased retinal blood flow is involved in the early pathogenesis of diabetic retinopathy43. These blood flow changes have implications for control of both hypertension and hyperglycaemia in reducing retinopathy39. Biochemical, haemodynamic and hormonal mechanisms may interact together to produce the typical lesions of vascular occlusion, microaneurysms, haemorrhages, hard exudates and new vessels44. In such a model, increased retinal blood flow is initiated by high blood glucose, but made worse by high blood pressure45 and impaired autoregulation. High glucose levels have a direct damaging effect on both pericytes and endothelial cells, with further vessel wall damage, occlusion of vessels and ischaemia, resulting in PDR44. These studies support early and effective treatment of elevated blood pressure in diabetes46.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

6

Changes in the blood-retinal barrier Increased blood-retinal barrier permeability has been shown in diabetic retinopathy using vitreous fluorophotometry for many years47,48. Using this technique, the blood-retinal barrier is stable until puberty, with a progressive decline from this age49. This early sign may also predict an unfavourable course over the long-term50.

1.3

Prevalence and incidence

Key Points •

Reported retinopathy prevalence rates in Australian community diabetic populations range from 21 to 36%, while the prevalence of vision-threatening retinopathy has been reported from 6% to 13%.



The incidence of new retinopathy for people with no retinopathy at baseline has been reported at 6–14% per year, and is related to the duration and glycaemic control of the diabetes.



IDDM (type I), typically with onset before age 30, accounts for 10% of cases, NIDDM (type II) typically with onset after age 30, accounts for most of the remaining 90%, with gestational diabetes and diabetes in genetic syndromes comprising around 2% of cases.

Diabetes mellitus (IDDM and NIDDM) in Australia The prevalence of diabetes mellitus in adult Caucasian Australians varies from 2% to 7%, while in the Aboriginal population, it ranges from 5% to 19%, depending on the age range examined51. According to recent estimates, in 1996 around 780,000 Australians had diabetes, rising to an estimated 900,000 Australians with diabetes by the year 2000. Among the two major types of diabetes, insulin-dependent diabetes mellitus (IDDM or type I diabetes), usually with onset before age 30, accounts for around 10% of diabetes cases, while non-insulin dependent diabetes mellitus (NIDDM or type II diabetes), usually with onset from age 30, accounts for almost 90% of cases in Australia. Gestational diabetes, and diabetes occurring as part of a genetic syndrome, make up only 2% of cases. Diabetes mellitus in adults can be either IDDM or NIDDM, but the overwhelming majority of people who develop diabetes after age 30 will have NIDDM. Diagnosed NIDDM is present in 5% of our population aged 30–64, and in 6% of older ages, and is far more common than IDDM, which is present in 0.08% or less of the population aged 0–1951. There are many undiagnosed NIDDM cases in the Australian population, though few estimates of this number are available. According to US data from the 1976–80 National Health and Nutrition Examination Survey52, the proportion of undiagnosed NIDDM cases aged 20–74 (3.2%) was almost equal to the proportion of previously diagnosed NIDDM cases (3.4%). This suggests that the actual prevalence of NIDDM in Australia may also be underestimated by many cross-sectional studies. The Blue Mountains Eye Study found a 5.9% prevalence of previously-diagnosed diabetes, and only a 1.1% rate of undiagnosed diabetes, using fasting blood glucose measurements (unpublished data). It is likely that a higher proportion of undiagnosed diabetes would be detected from 2-hour post-glucose

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

7

load measurements. By comparison, in the Melbourne Visual Impairment Project, including rural areas, 5.1% of people aged 40 or older had known diabetes (unpublished data). The incidence of IDDM was estimated as 12–15 per 100,000 children aged 0–14 or 0–19 from five Australian studies during the period 1983 to 199251.

Prevalence of diabetic retinopathy in Australia Six large studies have provided data on diabetic retinopathy in Australia, including the Newcastle Diabetic Retinopathy Study53–56, a West Australian Study57,58, Sydney Children’s Hospital Study59, Adelaide Queen Elizabeth Hospital Study60, the Melbourne Visual Impairment Project (MVIP)61,62 and Blue Mountains Eye Study (BMES)63. Although the Newcastle Diabetic Retinopathy Study53–56 is likely to have included the majority of known cases of diabetes within the Newcastle area, it was not strictly population-based. Selection bias could also have influenced findings in the Sydney Children’s Hospital Study59 and the study from the Queen Elizabeth Hospital, Adelaide60. In the West Australian Study57,58, screening of voluntary cases could also be subject to selection bias. Both the MVIP61,62 and BMES63 were population-based, but are confined to older populations (aged 40 or 49 and older, respectively). In the MVIP, there was no ability to detect undiagnosed diabetes as blood glucose was not measured. To summarise findings from these studies, the prevalence of diabetic retinopathy among diabetic clinic patients has varied from 35% to 49%, and for diabetes cases identified in the community, from 22% to 36%. Overall, diabetic retinopathy was found in 1.1% to 2.2% of older Australian population samples61,63. Most studies have been of predominately Caucasian samples. While there is much information regarding the prevalence of diabetes in Aboriginal communities51,64–67, there are few data providing estimates of the prevalence of diabetic retinopathy in these communities. The West Australian Study68 included 134 subjects of Aboriginal descent, finding a retinopathy prevalence of 44% in early and 29% in late-onset diabetes cases. Reliable data are available for Pacific Island populations. For example, in Nauru, which has one of the highest prevalence rates ever recorded for diabetes, 27% of adults have diabetes and 24% of those have retinopathy69.

Clinical Practice Guidelines for the Management of Diabetic Retinopathy

8

Figure 1. Prevalence of any retinopathy, proliferative retinopathy and macular oedema by known duration of diabetes, for IDDM cases, defined as diabetes diagnosed before age 30, in the Newcastle Study 53–56.

100

any retinopathy proliferative

80

macular oedema

60 % 40

20

0