Visual Fields (Ophthalmology Monographs)

Citation preview

Visual Fields

Ophthalmology Monographs A series published by Oxford University Press in cooperation with the American Academy of Ophthalmology Series Editor: Richard K. Parrish, II, MD, Bascom Palmer Eye Institute American Academy of Ophthalmology Clinical Education Secretariat: Louis B. Cantor, MD, Indiana University School of Medicine Gregory L. Skuta, MD, Dean A. McGee Eye Institute

1. Retinal Detachment: Principles and Practices, third edition Daniel A. Brinton, MD, and Charles P. Wilkinson, MD 2. Electrophysiologic Testing in Disorders of the Retina, Optic Nerve, and Visual Pathway, second edition Gerald Allen Fishman, MD, David G. Birch, MD, Graham E. Holder, MD, and Mitchell G. Brigell, MD 3. Visual Fields: Examination and Interpretation, third edition Edited by Thomas J. Walsh, MD 4. Glaucoma Surgery: Principles and Techniques, second edition Edited by Robert N. Weinreb, MD and Richard P. Mills, MD 5. Fluorescein and Indocyanine Green Angiography: Technique and Interpretation, second edition Joseph W. Berkow, MD, Robert W. Flower, David H. Orth, MD, and James S. Kelley, MD 6. Neuroimaging in Ophthalmology, second edition Michael C. Johnson, M.D., F.R.C.S.C, Bruno A. Policeni, M.D., Andrew G. Lee, M.D., and Wendy R. K. Smoker, M.D., F.A.C.R. [First edition published as Magnetic Resonance Imaging and Computed Tomography: Clinical Neuro-Orbital Anatomy by Jonathan D. Wirtschafter, MD, Eric L. Berman, MD, and Carolyn S. McDonald, MD] 7. Cataract Surgery and Intraocular Lenses: A 21st-Century Perspective, second edition Edited by Jerry G. Ford, MD, and Carol L. Karp, MD 8. Surgery of the Eyelid, Orbit, and Lacrimal System Edited by William B. Stewart, MD 9. Surgical Anatomy of the Ocular Adnexa: A Clinical Approach David R. Jordan, MD, and Richard R. Anderson, MD 10. Optic Nerve Disorders, second edition Edited by Lanning B. Kline, MD, and Rod Foroozan, MD 11. Laser Photocoagulation of the Retina and Choroid [with slide set] James C. Folk, MD, and José S. Pulido, MD 12. Low Vision Rehabilitation: Caring for the Whole Person Edited by Donald C. Fletcher, MD 13. Glaucoma Medical Therapy: Principles and Management, second edition Edited by Peter A. Netland, MD, PhD 14. Diabetes and Ocular Disease: Past, Present, and Future Therapies, second edition Edited by Ingrid U. Scott, MD, MPH, Harry W. Flynn, Jr., MD, and William E. Smiddy, MD 15. HIV/AIDS and the Eye: A Global Perspective Emmett T. Cunningham, Jr., MD, and Rubens Belfort, Jr. 16. Corneal Dystrophies and Degenerations: A Molecular Genetics Approach [with CD] Edited by Ming X. Wang, MD, PhD 17. Strabismus Surgery: Basic and Advanced Techniques Edited by David A. Plager, MD; written by Edward G. Buckley, MD, David A. Plager, MD, Michael X. Repka, MD, and M. Edward Wilson, MD; contributions by Marshall M. Parks, MD and Gunter K. von Noorden, MD www.oup.com/us/aao/plager/strabismus 18. A Compendium of Inherited Disorders and the Eye Elias I. Traboulsi, MD www.oup.com/us/aao/traboulsi/genetic

VISUAL FIELDS Examination and Interpretation Third Edition Edited by

Thomas J. Walsh, MD Published by Oxford University Press in cooperation with the American Academy of Ophthalmology

1 2011

1 Oxford University Press, Inc., publishes works that further Oxford University’s objective of excellence in research, scholarship, and education. Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Tai pei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright © 2011 Oxford University Press Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 Oxford is a registered trademark of Oxford University Press 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, without the prior permission of Oxford University Press. Library of Congress Cataloging-in-Publication Data Visual fields : examination and interpretation / edited by Thomas J. Walsh. — 3rd ed. p. ; cm. — (Ophthalmology monographs ; 3) Includes bibliographical references and index. ISBN 978-0-19-538968-5 1. Perimetry. 2. Visual fields. I. Walsh, Thomas J. (Thomas Joseph), 1931- II. Series: Ophthalmology monographs ; 3. [DNLM: 1. Perimetry. 2. Eye Diseases—diagnosis. 3. Visual Fields. W1 OP372L v.3 2010 / WW 145 V8343 2010] RE79.P4V58 2010 617.7’15—dc22 2010000009

987654321 Printed in China.

Legal Notice

T

he American Academy of Ophthalmology provides the opportunity for material to be presented for educational purposes only. The material represents the approach, ideas, statements, or opinion of the authors, not necessarily the only or best method or procedure in every case, nor the position of the Academy. Unless specifically stated otherwise, the opinions expressed and statements made by various authors in this monograph reflect the authors’ observations and do not imply endorsement by the Academy. The material is not intended to replace a physician’s own judgment or to give specific advice for case management. The Academy does not endorse any of the products or companies, if any, mentioned in this monograph. Some material on recent developments may include information on drug or device applications that are not considered community standard, that reflect indications not included in approved FDA labeling, or that are approved for use only in restricted research settings. This information is provided as education only so physicians may be aware of alternative methods of the practice of medicine, and should not be considered endorsement, promotion, or in any way encouragement to use such applications. The FDA has stated that it is the responsibility of the physician to determine the FDA status of each drug or device he or she wishes to use in clinical practice, and to use these products with appropriate patient consent and in compliance with applicable law. The Academy and Oxford University Press (OUP) do not make any warranties as to the accuracy, adequacy, or completeness of any material presented here, which is provided on an “as is” basis. The Academy and OUP are not liable to anyone for any errors, inaccuracies, or omissions obtained here. The Academy specifically disclaims any and all liability for injury or other damages of any kind for any and all claims that may arise out of the use of any practice, technique, or drug described

v

vi

Legal Notice

in any material by any author, whether such claims are asserted by a physician or any other person.

DISCLOSURE STATEMENT Unless otherwise noted below, each author states that he or she has no significant financial interest or other relationship with the manufacturer of any commercial product discussed in the chapters that he or she contributed to this publication or with the manufacturer of any competing commercial product.

Preface

T

he ophthalmologist has many instruments available for examining different parts of the visual system. The slit lamp examines the anterior chamber of the eye. The direct and indirect ophthalmoscopes afford a detailed inspection of the central and peripheral retina. The art of examining the visual fields is probably the only ophthalmic technique that allows investigation of each and every part of the visual system. The defects that are found must then be interpreted as being in one particular area of the area of the visual system or as being the result of other conditions, such as opacities of the lens or of the media, which point to an organic abnormality in the visual system. It is my experience that although many examinations may be delegated to the ophthalmic technician, it is the ophthalmologist who is experienced in diagnosing ocular disease and therefore the person who should select the visual field technique and program that best serve the patient. The refinement of radiographic studies, such as computer tomography (CT) and magnetic resonance imaging (MRI), has permitted the identification of discrete lesions in the visual pathway. However, these technologies are not infallible, so that many times no lesion is visualized. It is the experienced physician who puts the clinical picture and field abnormality together in one anatomic location: The amount of information that can be seen in a current printout does not always give the answer, so we have to use other, new tests to provide the complete picture. It is the same physician, the ophthalmologist, who has mastered the different parametric techniques to obtain the most accurate information from any field examination. The principal impetus behind the current revision of this monograph has been the widespread popularity of computerized perimetry in recent years. The discussion of the major changes in perimetry from the first edition of this book as a monograph,

vii

viii

Preface

published in 1990, will continue in this edition. And we will not forget some of the other techniques that have been useful throughout the years and need to be learned by the well-founded perimetrist. In previous editions, we have put the tangent screen illustrations next to the computerized illustrations. In this edition, we have put the tangent screen in the first chapter to show that it still can be a valuable technique when computerized fields cannot be used. Chapter 1 “FIELD OF Vision”, reviews the type of defects that we find in doing visual fields. There is also a review of techniques that preceded automated perimetry and have uses even today in special situations. Chapter 2, “The Anatomic Basis and Differential Diagnosis of Field Defects”, reviews the neurologic significance of certain types of field defects and identifies the location of the defect along the long visual pathway. Chapter 3, “Essentials of Automated Perimetry,” discusses several programs such as SITA Standard and SITA Fast as well as other programs useful for glaucoma. This is followed by Chapter 4, “Automated Perimetry in Glaucoma,” which demonstrates the usefulness of these new programs in diagnosing and in monitoring progression of the disease. In Chapter 5, “Inherited or Congenital Optic Nerve Diseases,” and Chapter 6, “Acquired Optic Nerve Diseases,” the authors comment on the use of fluorescein, MRI, and OCT in making a diagnosis. Chapter 7, “Visual Field Defects in Chorioretinal Disorders”, shows examples of the use of perimetry as an important adjunct in the management of Retinal disease in this era of OCT and photography. Chapter 8, “Optic Chiasm Field Defects,” discusses the current thoughts regarding Wilbrand’s knee junction scotoma and the field defects associated with lesions of the optic chiasm. In “Optic Tract and Lateral Geniculate Body Field Defects,” Chapter 9, radiation was looked at again in terms of the history of congruity. Some new material is also provided about the size of the calcarine cortex and how much of the area is seen in a 24-2 program. There are also new chapters on chorioretinal disease and another on the optic tract. Chapter 10,”Retrogeniculate Visual Field Defects”, is more important in this era of MRI and other radiologic modalities that allow us to identify the type of lesion and its extension. In this era of more sophisticated surgery, radiation and chemotherapy it contributes significantly to therapy and not just as a marker.Chapter 11,”Functional Visual Loss,” is one of the most important features of doing fields. Many people produce a false field or do it so poorly as to obscure the real diagnosis. It is important to note that people with functional fields get sick also and we must be able to perform different techniques in order to get the real answer and provide the correct information to their physician to order the correct test which may be radiologic or from the laboratory. From the laboratory. In the centennial year of 1996, celebrating the founding of the American Academy of Ophthalmology, the publication of the second edition of this book on visual fields had special significance. It was 50 years earlier that the first edition of what was then called a manual on this topic was published. My contributors and I are indebted to

Preface

ix

C. Wilbur Rucker, MD, the author of that manual, published in 1946, on which this monograph is based. Dr. Rucker emphasized the importance of doing visual field examinations as well as demonstrating some of the techniques for performing them successfully. Many others have contributed to earlier editions, such as the late Thomas Hedges, Jr., who is now followed by his son in this edition. In recent editions of this monograph, we have been helped by perimetrists of this era like Joseph Caprioli and Mark Weitzman, who have given valuable help in the area of glaucoma, as well as Brian Younge: all have added their knowledge and perimetry skills so that their insights can be carried over to this edition. We have also been fortunate in our department at Yale University to have the experience of Bruce Shields and James Tsai, who have added the fruits of their own extensive experience to this complex subject. Jonathan Horton, in the last edition as well as in this one, has contributed advice and illustrative explanations concerning both Wilbrand’s knee in the chiasm and the visual cortex and how it is transcribed into the Humphrey 30-2 and 24-2. One of the sad features of compiling this third edition is the loss of Jonathan Wirtschafter, both as a skilled clinician and a dear friend. He was a giant in the field. His chapter in this volume brings together the field defects we see in the perimeter printout and shows us their clinical neuro-ophthalmologic application. His original chapter on blending neuroanatomy and perimetry brought increased clarity to the diagnosis of visual field disorders and remains a classic in perimetry. Those of us who have followed in his and Traquair’s tradition have endeavored to update and further contribute to the state of the art of perimetry. The updates and techniques outlined in this third edition of the monograph will continue to be valuable to the ophthalmology resident as well as to the practicing ophthalmologist. THOMAS WALSH, MD New Haven, Connecticut

This page intentionally left blank

Contents

Chapter 1.

Contributors A History of Perimetry

xvii xix

Overview of Perimetry

3

Thomas J. Walsh, MD 1-1 Overview of Perimetry 1-1-1 Peripheral Fields 1-1-2 Central Fields 1-1-3 Physiologic Blind Spot 1-1-4 Recording the Fields 1-2 Structure of the Visual Pathway 1-3 Interpretation of Defects in the Fields 1-4 Techniques of Field Testing 1-4-1 Confrontation Technique 1-4-2 Central Field Technique 1-4-3 Chamlin Step Technique 1-4-4 Peripheral Field Technique 1-4-5 Amsler Grid 1-4-6 Color Testing

3 5 6 7 7 12 16 21 21 24 31 34 36 36

xi

xii

Contents

Chapter 2.

Anatomic Basis and Differential Diagnosis of Field Defects

41

Jonathan D. Wirtschafter, MD, and Thomas J. Walsh, MD 2-1 Categories of Field Defects 2-2 Overview of the Visual Pathway 2-2-1 Occipital Lobe 2-3 Monocular Field Defects 2-3-1 Localized Defects 2-3-1-1 2-3-1-2 2-3-1-3 2-3-1-4 2-3-1-5 2-3-1-6 2-3-1-7

Wedge-Shaped Temporal Field Defect Arcuate and Paracentral Field Defects Central Scotoma or Depression Enlarged Physiologic Blind Spot Centrocecal Scotoma or Depression Equatorial Annular Scotoma or Depression Altitudinal Hemianopia

2-3-2 Generalized Defects 2-3-2-1 Generalized Depression or Peripheral Contraction

2-4 Binocular Field Defects 2-4-1 Homonymous Hemianopias 2-4-1-1 Complete: Macular Splitting 2-4-1-2 Incomplete Congruous: Horizontal Sectoranopia 2-4-1-3 Incomplete Congruous: Paramidline-Sparing Vertical Hemianopia 2-4-1-4 Incomplete: Macular Sparing 2-4-1-5 Incomplete: Two Scotomas 2-4-1-6 Incomplete Incongruous 2-4-1-7 Incomplete: Unilateral Sparing of Temporal Crescent 2-4-1-8 Incomplete: Unilateral Defect of Temporal Crescent

2-4-2 Bitemporal Hemianopias 2-4-2-1 Complete 2-4-2-2 With Central Depression, Scotomatous

2-4-3 Binasal Field Defects 2-4-3-1 Complete 2-4-3-2 Incomplete

2-4-4 Altitudinal Field Defects 2-4-4-1 Noncongruous Binocular and Monocular 2-4-4-2 Congruous

2-4-5 Quadrantanopias 2-4-5-1 Superior Homonymous, Incomplete 2-4-5-2 Inferior Homonymous, Complete

2-4-6 Bilateral Central Field Defects 2-4-6-1 Scotoma or Depression

2-4-7 Bilateral Peripheral Field Defects 2-4-7-1 Generalized Depression or Peripheral Contraction

2-4-8 Bilateral Checkerboard Scotomas 2-4-9 Bilateral Homonymous Hemianopias 2-5 Junctional Field Defects 2-5-1 Complete Monocular Plus Incomplete Contralateral Ocular 2-5-2 Homonymous Hemianopia Plus 2-5-3 Bitemporal Hemianopia Plus

41 43 46 48 48 48 52 54 56 57 58 59 59 59 60 60 60 62 63 64 64 65 66 67 67 67 69 70 70 71 72 72 73 75 75 76 77 77 78 78 79 80 80 80 81 83

Contents

Chapter 3.

Essentials of Automated Perimetry

xiii 85

George Shafranov, MD 3-1 Introduction 3-2 Historical Overview 3-3 Principles of Field Testing 3-3-1 Kinetic Perimetry 3-3-2 Static Perimetry 3-3-2-1 Suprathreshold Techniques 3-3-2-2 Threshold Techniques

3-3-3 Frequency-of-Seeing Curves and Fluctuations 3-4 Test Selection and Algorithms 3-4-1 Swedish Interactive Threshold Algorithm (SITA) 3-4-2 Foveal Threshold 3-4-3 Initial Values 3-4-4 Fixation Monitoring 3-4-5 Threshold Testing 3-5 Single Test Printout 3-5-1 Test Selection (and General Information) 3-5-2 Reliability Indices 3-5-3 Numeric Results and Grayscale Results (Raw Data) 3-5-4 Total Deviation 3-5-5 Pattern Deviation 3-5-6 Glaucoma Hemifield Test 3-5-7 Global Indices 3-6 Custom Tests 3-6-1 Grid Size 3-6-4 FASTPAC 3-6-5 Programs 30-1 and 24-1 3-6-6 Program 10-2 and Macula Test 3-6-7 Peripheral 60 and 60-4 Program 3-6-8 Nasal Step Program 3-6-9 Stimulus Size Option 3-7 Follow-up Printout 3-7-1 Overview Printout 3-7-2 Change Analysis Printout 3-7-3 Glaucoma Progression Analysis Printout 3-7-4 GPA-Guided Progression Analysis 3-8 Visual Function Specific Perimetric Technologies 3-8-1 Short-Wavelength Automated Perimetry (SWAP) 3-8-2 Frequency Doubling Perimetry 3-8-3 High-Pass Resolution Perimetry 3-8-4 Tendency-Oriented Perimetry 3-9 Learning Effect and Artifacts 3-9-1 Miosis and Mydriasis 3-9-2 Media Opacities 3-9-3 Eyelid and Nose Effects 3-9-4 Refractive Errors 3-9-5 Corrective Lens/Frame Artifacts 3-10 Role of the Visual Field Technician

85 85 86 88 90 90 92 92 94 94 95 95 96 99 99 99 99 101 101 101 102 105 105 108 108 109 109 110 110 111 111 111 111 114 114 114 117 117 117 118 118 118 119 119 119 120 121

xiv

Contents

Chapter 4.

Automated Perimetry in Glaucoma

127

Hylton R. Mayer, MD, Marc L. Weitzman, MD, and Joseph Caprioli, MD 4-1 4-2 4-3 4-4

Introduction Glaucomatous Field Loss Automated Perimetry Options Evaluation of a Single Test 4-4-1 Patient Reliability 4-4-1-1 4-4-1-2 4-4-1-3 4-4-1-4 4-4-1-5 4-4-1-6

Test Duration Fixation Losses False-Positive Responses False-Negative Responses Short-Term Fluctuation Stimuli Number

4-4-2 Criteria for Abnormality 4-4-3 Staging of Field Loss 4-4-4 Test Selection 4-4-5 Follow-up of Advanced Field Loss 4-5 Evaluation of a Series of Tests 4-5-1 Baseline Establishment 4-5-2 Nature of Progression 4-5-3 Artifacts 4-5-4 Long-Term Fluctuation 4-5-5 Criteria for Progression 4-5-6 Glaucoma Progression Analysis 4-5-7 Trend Analysis 4-5-8 Follow-up of Central Abnormalities 4-6 Future of Automated Perimetry 4-6-1 Altered Stimuli 4-6-1-1 4-6-1-2 4-6-1-3 4-6-1-4 4-6-1-5 4-6-1-6 4-6-1-7

Short-Wavelength Automated Perimetry Frequency-Doubling Technology Flicker Perimetry Ring Perimetry Motion Automated Perimetry Pattern-Discrimination Perimetry Color Perimetry

4-6-2 Altered Strategies 4-6-3 Interpretive Aids

Chapter 5.

Inherited or Congenital Optic Nerve Diseases

127 127 129 132 132 133 133 133 136 136 136 140 143 144 144 144 146 149 149 154 155 157 159 159 161 161 161 165 165 167 168 168 168 169 171 177

Peter A. Quiros, MD, Carlos Filipe Chicani, MD, PhD, and Alfredo A. Sadun, MD, PhD 5-1 Introduction 5-2 Congenital Optic Disc Anomalies 5-2-1 Aplasias and Dysplasias 5-2-2 Optic Nerve Colobomas and Pits 5-2-3 Anomalous Disc Elevations: Pseudopapilledema With or Without Hyaline Bodies (Drusen) 5-2-4 Tilted Disc and Crescents

177 177 177 179 181 181

Contents

Chapter 6.

xv

5-3 Heredodegenerative Optic Atrophies 5-3-1 Leber’s Hereditary Optic Neuropathy 5-3-2 Dominant Optic Atrophy 5-3-3 Recessive Optic Atrophy

182 183 185 189

Acquired Optic Nerve Diseases

193

Carlos Filipe Chicani, MD, PhD, Peter A. Quiros, MD, and Alfredo A. Sadun, MD, PhD

Chapter 7.

6-1 Introduction 6-2 Optic Neuritis 6-3 Ischemic Optic Neuropathy 6-3-1 Nonarteritic Ischemic Optic Neuropathy 6-3-2 Arteritic Ischemic Optic Neuropathy 6-4 Metabolic Optic Neuropathies 6-5 Compressive Optic Neuropathy 6-6 Papilledema

193 194 196 196 196 198 198 200

Visual Field Defects in Chorioretinal Disorders

207

Ron A. Adelman, MD, MPH, FACS, and Patricia Pahk, MD 7-1 7-2 7-3 7-4 7-5 7-6 7-7

Chapter 8.

Introduction Macular Diseases Vascular Diseases Congenital and Genetic Diseases Inflammatory/Infectious Diseases Toxicity Peripheral Retina

Optic Chiasm Field Defects

207 207 209 211 217 226 226 233

Christine E. Lin, and Jeffrey G. Odel, MD 8-1 History and Overview 8-2 Anatomy of the Chiasm 8-2-1 Gross Anatomy 8-2-2 Nerve Fiber Anatomy 8-3 Tests for Field Defects 8-4 Chiasmal Visual Field Defects 8-5 Chiasmal Region Lesions 8-5-1 Common Lesions Affecting the Optic Chiasm 8-5-2 Pituitary Tumors 8-5-3 Meningiomas 8-5-3-1 Suprasellar Meningiomas 8-5-3-2 Suprachiasmatic Meningiomas 8-5-3-3 Parasellar Meningiomas

8-5-4 Craniopharyngiomas 8-5-5 Aneurysms 8-5-6 Dilatation of Third Ventricle 8-5-7 Miscellaneous Lesions 8-6 Pseudo Temporal and Bitemporal Hemianopia

233 234 234 236 240 241 247 247 247 248 248 249 249 249 249 250 250 250

xvi

Contents

Chapter 9.

Optic Tract and Lateral Geniculate Body Field Defects

253

Thomas R. Hedges III, MD 9-1 Optic Tract Field Defects 9-2 Lateral Geniculate Body Field Defects

Chapter 10. Retrogeniculate Visual Field Defects

253 258 263

Thomas R. Hedges III, MD 10-1 Testing for Reticulogeniculate Visual Field Defects 10-2 Localization and Congruity of Optic Radiation and Calcarine Cortex Visual Field Defects 10-3 Temporal Lobe Field Defects 10-4 Parietal Lobe Field Defects 10-5 Occipital Lobe Field Defects 10-5-1 Types of Occipital Cortex Field Defects 10-5-2 Field Defects Unique to the Occipital Cortex 10-5-3 Color Field Defects in the Occipital Lobe

Chapter 11. Functional Visual Loss.

263 264 267 269 272 273 282 285 293

Thomas J. Walsh, MD 11-1 Types of Patients 11-2 Types of Field Loss 11-3 Tests for Functional Field Loss

293 294 295

Index

299

Contributors

Ron A. Adelman, MD, MPH, FACS

Peter A. Quiros, MD

Yale University

Doheny Eye Institute

Joseph Caprioli, MD

Alfredo A. Sadun, MD, PhD

Jules Stein UCLA School of Medicine

Doheny Eye Institute

Carlos Filipe Chicani, MD, PhD

George Shafranov, MD

Doheny Eye Institute

Yale University

Thomas R. Hedges III, MD

Thomas J. Walsh, MD

Tufts University Medical School

Yale University

Christine E. Lin

Marc L. Weitzman, MD Assistant Clinical Professor Yale University School of Medicine

Columbia University

Hylton R. Mayer, MD Yale University

Jonathan D. Wirtschafter, MD

Jeffrey G. Odel, MD

University of Minnesota School of Medicine

Columbia University

Patricia Pahk, MD Mount Sinai Medical Center

xvii

This page intentionally left blank

A History of Perimetry THOMAS J. WALSH, MD

T

he history of perimetry is the history of anatomy. In the early days of perimetry, in the late 1800s, the study of anatomy involved learning the muscles, vessels, and bony structures of the body. The brain was recognized as an organ, but the special functions of its different structures were generally unknown. Before the availability of current methods to study the brain—including those based on histology, electron microscopy, and electrophysiology and magnetic resonance imaging, magnetic resonance angiography, computed tomography, computed tomography angiography, optical coherence tomography, and positron emission tomography—our predecessors used their observations of visual functions to devise a technique to examine those functions: this technique is called perimetry. Just as the field of anatomy has been dynamic, so has the development of techniques in perimetry. With the advent of more sophisticated techniques, we have been able to diagnose diseases earlier and treat them more successfully. However, some of the older techniques can still be used when sophisticated techniques do not work, especially in neuro-ophthalmology. Perimetric techniques are as variable as are the patients and the diseases that are treated. Although most of this monograph is directed at computerized perimetry, techniques such as confrontation and the tangent screen are also discussed. Just as the range of defects varies, so, too, must the range of different techniques mark the ability of an accomplished perimetrist. As newer techniques and newer computerized programs come online, updating the sophistication of perimetry, so, too, should the perimetrist continually hone his or her skills. The contributors to this monograph are humbled by the giant steps in the development of perimetry made by the observations and ingenuity of those who have preceded us. In this chapter, we present the contributions and persons who are important to the history of our understanding of perimetry. The Imaging and Perimetry Society (formerly the International Perimetric Society)1 has added to our knowledge of those who have gone before us, including a history of perimetry by Thompson

xix

xx

A History of Perimetry

and Wall (see http://www.perimetry.org/PerimetryHistory/index.htm). We are also indebted to Drs. Simpson and Crompton for their exhaustive work on those who have advanced the field of perimetry.2 They searched in many a medical library and resurrected almost-forgotten dusty manuscripts. Among their sources is a library that I have relied on in the past and have been fortunate to have virtually at my doorstep—the Harvey Cushing/John Hay Whitney Medical Library of Yale University’s School of Medicine, which I used extensively in developing this review. This library has a long distinguished history and has played an important role in the history of perimetry. The name of the library honors Dr. Cushing,3 whose lifelong interest in surgery for pituitary disease was as keen as his observation of visual defects. In this, he was helped by Dr. Clifford Walker, who worked in his clinic and conducted perimetric examinations. Dr. Cushing and Dr. Eisenhardt, in their studies of the optic chiasm in 1926, took exception to Wilbrand’s theory of the location of the inferior nasal optic fibers as the fibers crossed through the chiasm. This issue was elaborated on by Jonathan Horton4 in 1997 and is discussed further in Chapter XX. Determining who was first to make observations in diagnoses based on perimetry is difficult. The technique of confrontation was probably used much earlier, but late in the 1800s, William Gowers popularized it as a field method.5 In many cases, the phenomenon of a particular defect was observed by one and its anatomic significance was observed by another. Hippocrates was probably the first to describe hemianopia, in the fifth century BC. It was Ptolemy, around 150 BC, who started to quantify the size of the peripheral field.6 Leonardo da Vinci in 1510 also studied this and reported that the temporal field was 90°. This was an important observation, as we will see, with future study changing field testing from a flat screen to an arc perimeter, which, of course, does not diminish the importance of Hippocrates’ observations. Issac Newton made the observation of a partial decussation of the chiasm in 1602,7,8 which Dr. Johann Spuryheim confirmed in 1820.9 In 1664, Thomas Willis published his work on the functional anatomy of the brain and reaffirmed the crossing of fibers in the chiasm.10 William Wollaston, in 1824, described a case of left homonymous hemianopia11: the patient was himself. The defect cleared, but it was followed by another homonymous hemianopia on the right, which was due to a tumor and which did not clear. He realized each time that the defect concerned only one half of each eye and that the fibers from the eye decussated. Albrecht von Graefe’s name is prominent in many areas of ophthalmology, including perimetry, ophthalmoscopy, and surgery.12 His friend Hermann von Helmholtz had perfected the first ophthalmoscope but was not impressed with its clinical use. It was von Graefe who introduced this instrument at the German Physical Society. He fell in love with the instrument and used it to make many observations throughout his life. He observed papilledema about 1850 and noted in his first report of that phenomenon that it was seen in cases of increased intracranial pressure from brain tumors and declared that a way of making such a diagnosis should be quantified. Some others, such as Ulmus of Padua in 1602, illustrated a visual field he had observed. In 1856, von Graefe then described a clinical technique to measure this phenomenon. He produced a board with a fixation object and

A History of Perimetry

xxi

with peripheral distances marked off (which he called isopters) so that he could reproduce these defects in the future. This was the beginning of the current tangent screen. Although the technique was a major step forward, there were those who could see the potential for perfecting this technique and immediately improved upon it. The flat screen was adequate for the central field but not for the peripheral field, which has been described as reaching 90° by da Vinci and again by Thomas Young in 1801 and Johannes Purkinje in 1825. In 1817, Joseph Bier in Vienna was studying the central field concept and identified defects such as central scotoma, paracentral scotoma, and central contraction. Richard Foster in 1860 constructed an arc-type perimeter to cover the peripheral field, but it was of no use for diagnosing a central scotoma. Schweiger then produced a patient hand-held arc-type perimeter, but it was not better and it was very cumbersome. In 1889, Jannick Bjerrum became very interested in the tangent type of technique and constructed a revised one on his office door. He standardized it so that a defect could be reproduced. Later, others added different-sized test objects to make the more subtle defects identifiable. Behrens thought that the flat test objects were inaccurate because the test object could move from the center of the visual field, so he devised a set of test objects in the form of balls. The test objects would then cast the same size image on the retina regardless of where they were projected in the field. However, these were always visible and could not be “turned off” from the patient’s perception. Perimetry was developing a following, and in 1874, a book on perimetry was written by William Schon.13 Published 1874Schon worked with Dr. Horner of Zurich, renowned for his study of the pupil. Hermann Munk in 1887 performed experiments on dogs, first ablating one half of the occipital lobe and demonstrating hemianopia in each eye. Then he performed a similar experiment in another dog, on the other side of the lobe, demonstrating reverse hemianopia. This location of different types of defectc brings us to interest in decussation of the chiasm. Dr. Edward Schafer performed the same experiment as Horner’s in monkeys and achieved the same result. However, he added another observation to our fund of knowledge. He saw that there was a difference between an occipital hemianopia and one in the tract. Optic tract A tract hemianopia would have a hemianopic pupillary abnormality, which had been observed previously by Wernicke. There would not be a pupillary response in the blind field of a tract hemianopia. Although this is an established fact, the defect is a difficult clinical sign to elicit. By the end of the 1800s, interest in perimetry was well established. Dr. Salomon Eberhard Henschen identified the location of the visual fibers on the medial surface of the occipital lobe and that occipital cortical location was the same for the retinal field location. Adolph Meyer reported in 1907 the course of the visual fibers from the lateral geniculate body. He reported that the lower visual fibers did not return directly through the temporal lobe.14 They went anteriorly into the temporal lobe and fanned out before they turned posteriorly along the course of the lateral ventricle. He also noted that they did not reach into the most anterior part of the temporal lobe. This information had neurosurgical as well as perimetric implications. In the 1950s, Falconer and Wendland were performing temporal lobectomy on patients

xxii

A History of Perimetry

with epilepsy.15,16 As part of that operation, meticulous measurements were made from the tip of the temporal lobe to the area resected, and a correlation was made as to how much of a field defect was created. Up to 5 cm from the temporal tip could be resected before any field defect was created. Because of their work, surgery for epilepsy is done differently today, resultng in fewer field defects and better therapeutic results in most patients. That their results were truly anatomic rather than only perimetric confirmed the statement made earlier by Traquair in 1933 that hemianopias in that area are congruous. Some authors disagree with that proposition, so recently Kedar and colleagues17 raised the question again. This is discussed in the chapter on The radiations radiation. That is what it is called and that author is Thomas Hedges It is title of tracts from the temporal, parietal and occitpital lobes. Many physicians have looked at field defects in the radiations and have noted other findings seen with different deficits. Purkinje noted one of these phenomena while watching a parade in Vienna.18 He noted that the people across from him all had horizontal nystagmus as the lines of soldiers moved past. Then, later in his clinic, he noted abnormalities in one direction and not in the other in persons with hemianopia. He thought that this might be a new way to diagnose hemianopia in nonverbal patients, but this was not found to be valid in all patients with hemianopia. Ohms corrected this theory and localized the hemianopia to the deep parietal lobe. As terrible as war is in terms of human sacrifice and suffering, it sometimes results in new methods of treatment by caring and bright physicians. During the early twentieth century, Dr. Tatsusi Inouye examined soldiers with head wounds and eye complaints during the Russo-Japanese war.19 He used an arc perimeter for his observations and recorded the tract of the wounds and the type of field defect that would suggest the cortical location of those wounds. During World War I, Gordon Holmes, together with Chief British Army Ophthalmologist Lister, used a hand-held perimeter to confirm previous work showing that the upper calcarine cortex corresponded to the upper part of the retina and that similar correspondences held for other parts of the calcarine cortex. He also supplied evidence that acuity is spared even if an entire occipital lobe is removed. However, he did not support the theory that the macula is duplicated in both occipital lobes. Then, in World War II, the neurologist John Spalding studied 180 cases of missile wounds with defects limited to the posterior radiations and occipital cortex and confirmed the work of Holmes.20 Holmes created a map of the defects occurring in the occipital cortex. He allowed about 25% of the striate surface to represent central vision. Jonathan Horton and William Hoyt enlarged the field map, showing a much larger area for the macula.21 They also proposed that a 30–2 field would cover 89% of the field and would be adequate except for cases of central and peripheral defects that did not involve the monocular temporal crescent. The arc perimeter had its supporters, and when I started performing perimetry, the Aimark perimeter was very popular. However, another perimeter was to take its place. In 1945, Dr. Hans Goldmann of Bern had devised a projection perimeter that took the shape of a bowl and that provided a way to change the stimulus. This again changed the way perimetry is performed. However, it was inadequate

A History of Perimetry

xxiii

for static perimetry. The value of static perimetry was pointed out in 1933 by Dr. Louise Sloan, but it did not gain ready acceptance. The next development was the Tübingen perimeter in 1959, and then in 1971 Stephen Drance and Mansour Armaly introduced suprathreshold testing.22 Dr. William Hart, Jr. developed the first perimeter to measure color contrast. Then Lars Frisén developed a technique of high-pass resolution, leading to the Octopus and Humphrey perimeters that are so familiar today and that are now equipped with software programs that make the test even more user-friendly. In 1951, Max Chamlin pointed out the significance of a step defect at the vertical meridian, suggesting an early sign of a field defect such as a bitemporal defect that might otherwise not have been seriously considered. This advance was achieved with just a tangent screen, but it can be performed with a Goldmann perimeter as well. Harrington then published his text on perimetry, which represented the state of the art at that time. Wilber Rucker in 194623 also began to publish his studies on perimetry under the auspices of the American Academy of Ophthalmology (AAO). We are continuing that tradition today with this book in the Ophthalmology Monographs series published by Oxford University Press in cooperation with the AAO. Rucker also provided an in-depth review of the history of the semidecussation of the optic nerves. Rucker,C W The concept of a semidecussation at the chiasm. Arch. Ophthalmol 1958:59: 159–171. Here we sought to give the reader an idea of how many people it takes to make an idea evolve. This is true in all branches of medicine. New testing techniques will continue to be developed in the future, and those physicians who succeed us will look back at us just as we are looking back at those practitioners who were using A confrontation technique. Although we look to computerized perimetry as the technique of choice, a good perimetrist must learn other techniques, such as confrontation and the tangent screen, because they are still very useful when computer-based techniques are not appropriate. If a patient cannot be examined adequately with the use of a computer program, a well-executed tangent screen test may be the answer. The reader can see from this short historical glimpse that this is a very dynamic area and that advances are still occurring. The perimetrists mentioned in this review put us on our current path as we await the next developments in the field.

REFERENCES 1. Imaging and Perimetry Society. 2010. webeye.ophth.uiowa.edu; http://www.perimetry. org/. 2. Simpson DA, Crompton JL. The visual field: An interdisciplinary history, 1. The evolution of knowledge. J Clin Neurosci. 2008;Feb:579–609. 3. Cushing H. Selected Papers in Neurosurgery. New Haven, Conn: Yale University Press; 1969. 4. Horton JC. Wilbrand’s knee of the primate optic chiasm is an artefact of monocular enucleation. Trans Am Ophthalmol Soc. 1997:579–609. 5. Gower WR. A Manual and Atlas of Medical Ophthalmoscopy. 2nd ed. London, England: Churchill; 1882:57–61. 6. Heitz R. The history of contact lenses. 2003;1:50.

xxiv

A History of Perimetry

7. Brester D. Memories of the Life, Writing and Discoveries of Sir Isaac Newton. Vol 2. Edinburgh, UK: 1855. 8. Newton I. Opticks: Or a Treatise of the Reflextions, Refraction and Colours of Light. New York: McGraw-Hill; 1931. 9. Spuryheim G. The Anatomy of the Brain, With a General View of the Nervous System. TR Willis, London, England: Highly, 1826. 10. Willis T. The anatomy of the brain and nerves. In: Feindel W, ed. Classics of Neurology and Neurosurgery. Trans. by Pordage S. Facsimile 1st English ed. 1681. Birmingham: 1983:63. 11. Wollaston WH. On semi-decussation of the optic nerves. Phil Trans R Soc Lond. 1824;114:222–231. 12. von Graefe A. Ueber die unterschung des Gesechtsfeldes bei amblyopischen affectionen. Graefes Arch Ophthalmol. 1856;2:255–298. 13. Ferrier D. Munk on localization of function of the brain. Rev Brain. 1878;1:229–231. 14. Meyer A. The connections of the occipital lobes and the present status of the cerebral visual affections. Trans Assoc Am Phys. 1907;22:7–15. 15. Falconer MA. Visual field changes following anterior temporal lobectomy. Brain. 1958;81:1–14. 16. Windland JP. Visual field studies after temporal lobectomy for epilepsy. Arch Ophthalmol. 1960;64:195–200. 17. Kedar S, Zhang X, et al. Congruity in homonymous hemianopia. Am J Ophthalmol. 2007;143:856–858. 18. Smith JL. Optokinetic Nystagmus. Springfield, Ill: Charles C Thomas; 1963:3–4. 19. Inouye T. Visual disturbances following gunshot wounds of the cortical visual area based on observations of the wounded in the recent Japanese war. Brain. 2000;spec suppl:123. 20. Spalding J. Wounds of the visual pathway, 1. The striate cortex. J Neurol Neurosurg Psychiatr. 1952;15:169–183. 21. Horton J, Hoyt W. The representation of the visual field in human striate cortex. Arch Ophthalmol. 1991;109:816–824. 22. Armaly M. Selective perimetry for glaucomatosis defects in ocular hypertension. Arch Ophthalmol. 1972;87:518–524. 23. Rucker W. The concept of a semidecussation of the optic nerves. Arch Ophthalmol. 1958;59:159–171.

Visual Fields

Visual Fields

1 Overview of Perimetry THOMAS J. WALSH, MD

1-1 OVERVIEW OF PERIMETRY Like a painter, the practitioner of perimetry must learn his or her profession from experience. Just as a painting does not spring from the paint and brushes alone, the perimetrist does get his or her diagnosis from just a printout of the field test. Rather, the perimetrist’s experience in interpreting field test results, his clinical skill in examining the validity of the patient’s performance, and his selection of the needed field technique chosen under the appropriate clinical circumstances combine to produce a suitable test and interpretation of results. In this age of computerization, we tend to accept the infallibility of perimetry. It is true that new developments have corrected some of the errors in technique that have been troubling in earlier methods such as the tangent screen and Goldmann perimeter. However, in our rush to embrace these new techniques, we might forget that there is still a place for these older techniques in selected cases. Among other things, these older techniques may allow for a human element to be introduced when the patient is overwhelmed by technology—that is, a well-performed tangent screen is more valuable on a given occasion than a poorly performed computerized field examination. Such circumstances occur almost always with neuro-ophthalmology patients, who are usually ill in other ways than just visually and need more help in performing the test. Most other patients, particularly those with glaucoma, are much more reliable in their responses in using the newer techniques. They frequently start testing at a younger age and do their testing frequently so they become skilled at performing the test. Many neuro-ophthalmologic patients do not have that experience.

3

4

Visual Fields

Interpreting the blind spot remains a standard part of any field examination. Interpreting the blind spot size requires experience. The blind spot may be enlarged because the patient is a slow responder or because a large myopic crescent is present. An important use of measuring the blind spot is to show the patient what a scotoma is and to test his validity of fixation by putting the target in the blind spot from time to time. A cataract causes contraction of the peripheral field. Miosis due to the use of miotics and a small cataract can cause different defects that suggest a worsening of the patient’s glaucoma, which may be fallacious. In kinetic testing, if the examiner moves the test object too quickly, the sensitivity is altered and a defect is missed. If the kinetic test object is moved too slowly, the patient becomes distracted and loses fixation. (These problems are somewhat obviated by the use of computerized testing machines.) There are many different programs from which to select depending on the defect being sought and the patient’s field testing history and defects. The selection of the most ideal program to produce a good review of both central and peripheral fields was reviewed by Horton and Hoyt1 on an anatomic basis. They believe that a 30-2 program meets the requirements for central and peripheral field evaluation, except for monocular temporal crescent syndrome. Another variable in testing may be which eye to test first. If this is a patient’s first time with field testing, then the eye with better vision eye should be tested first to give the patient experience. If the patient has performed the test previously, then perhaps the worse eye should be tested first while the patient is fresh. All of these considerations are part of the performance and evaluation of field testing. Interpretation of the fields of vision forms a key part of ophthalmic and neurologic examinations. If the visual field is not intact, something is wrong in the retina or in the visual pathway—a bundle of nerve fibers that extends from the eyes, across the entire lower portion of the brain, to the visual centers of the occipital lobes. Not only will a field defect reveal almost any disease process occurring in the region, it can also help locate the site of the lesion. When lesions interrupt various parts of the visual pathway, they cause specific types of field defects; the nature of the defect frequently helps pinpoint the location of the lesion. The function of perimetry is to indicate the site and the extent of involvement of the visual pathway. For purposes of clinical perimetry, the field of vision is usually regarded as the inner surface of part of a hemisphere. On this hemisphere, the level of vision is determined from the point of fixation of gaze by the use of targets of various sizes and colors. A line, or an isopter, is drawn on the chart of the visual field to represent the limit of the area within which a specific target can be recognized. The chart of the visual field usually contains several concentric isopters, one for each target. This is best demonstrated in the Goldman perimeter. Now with computerized printouts, grayscale and numeric readouts in decibels are used to demonstrate the location and density of the defect. The Snellen acuity charts evaluate the foveal part of the visual field. The remainder of the field cannot be evaluated as easily as the central vision is evaluated with the Snellen chart. However, a 20/20 Snellen report does not rule out a paracentral scotoma, which may be the cause of the patient’s complaint about their reading.

Overview of Perimetry

5

We do not have a Snellen system as such for the rest of the field, nor does every part of the field have the same sensitivity as other parts. The ability to just see a certain test object is called the threshold for that area of the field. Field testing is unlike Snellen testing, which is the same from person to person. However, the lightsensitivity threshold for given degrees away from fixation has been established in all four quadrants until the outside limits of the island of vision have been reached. More details on this subject are explored in Chapter 3. Moving a test object of the same size and brightness from a nonseeing to a seeing area is called kinetic perimetry. If instead a stationary test object that increases in brightness until the patient sees it is used, this method is called static perimetry. Static suprathreshold testing uses a single test object that is above the expected threshold for that part of the field. The threshold for a kinetic test object is lower than that of a static test object. This finding is consistent with the work of Riddoch,2 who found that moving objects are seen sooner than are stationary areas in a patient recovering from an occipital lobe lesion. There are many ways to evaluate the visual system. The tangent screen is the traditional way of examining the central fields. The recording of distant and near acuities, color vision, and fusion testing, as well as the dazzle test and pupillary responses, all primarily evaluate the macular area. It is equally important to evaluate the peripheral field, which can be regarded as an extension of macular vision. Too often, physicians make the mistake of assuming that a good central acuity obviates the need for doing a central field. Patients with glaucoma defects, however, expose that error. Good perimetrists learn many techniques, both gross and sophisticated, for examining each area of field. They choose the type of test for the specific areas where a defect is most likely to occur, based on the patient’s history, state of alertness, and ability to communicate. The mark of good perimetrists is not that they can conduct one type of field test superbly but rather that they can evaluate the individual patient well and select the appropriate field technique for that patient. Most of this monograph will deal with computerized perimetry, which is used in the overwhelming number of patients. I believe it is important in this introduction to point out the usefulness of these techniques when the occasion presents itself. Examples of these techniques are shown for comparison in different sections of this monograph.

1-1-1 Peripheral Fields. The peripheral portions of the field are explored on a perimeter with the aid of an arc that can be turned in any desired meridian and that is marked from 0° to 90°. One technique requires that a white test object 3 mm in diameter outside an area of seeing be carried from the point of fixation along the arc until it disappears from view and then is returned to the seeing field. The point at which the test object reappears is recorded as the edge of the field for a test object of that size. Larger targets are used if the patient is not able to see well enough to permit use of the 3-mm test object or if there is a dense defect in the field. At least 12 radii of each field should be investigated. It is not sufficient to determine merely the peripheral limits of the field; the interior part of the field should also be

6

Visual Fields

evaluated, using a tangent screen or a sophisticated field-testing instrument such as a Goldmann or computerized perimeter. The central field itself must be searched for defective or depressed areas, known as scotomas. The temporal field constricts with age after the sixth decade of life. This constriction may be due partly to age-related miosis but is more probably due to decreased oxygenation of the peripheral retina with age. Oxygen concentration has been reduced experimentally during field measurements. Peripheral loss was reported by dive-bomber pilots during World War II; they noticed peripheral loss as they went into power dives and increased the G forces, reducing the oxygenation of the head.

1-1-2 Central Fields. Different sizes and intensities in test objects are necessary to evaluate the actual density in limits of any field defect. The use of the smallest or most sensitive test objects is necessary to outline the outside limits of the defect. The largest and brightest test object outlines a field defect that is still demonstrable. If a larger or brighter test object is used, no defect is found. This later isopter is necessary because it is least sensitive and is the first to vary as a field defect improves. Thus, this isopter is useful for monitoring improvement. Vascular lesions generally have steep margins to the defect. In those cases, the outside limits of the defect are almost the same as the most and the least sensitive test objects. Tumors usually have relative defects. The central portion, which is denser at the site of the disease process, is surrounded by an edematous area that represents a variable defect in the field. These statements about vascular lesions and tumors are good general rules of thumb, but they are not absolute. Some vascular lesions can be total infarctions with absolute defects or have areas of surrounding edema and cause relative defects in the outside isopters. Some tumors can present with dense defects when the tumor encroaches on a vascular supply. It can be very difficult to outline a central scotoma when the acuity is minimally decreased. Retinal sensitivity can decrease from 20/20 to 20/40 by moving the test object only 2.5° from the fovea. To explore the central area of the visual field in detail, the perimeter offers such a short working distance that the usual test objects are too small for practical handling. The accepted solution is an instrument such as a Goldmann or computerized perimeter, which permits exploration of the field out to 30° and allows one to perform central field testing at 0.33 m rather than at 1 or 2 m, which is required for a tangent screen. The tangent screen, on the other hand, allows personal contact with the patient, which aids in judging the quality of responses. Black felt is generally used to cover the screen. As exploration of the field progresses, the various isopters can be drawn directly on the felt with a dark-colored pencil and readily brushed off. This manner of recording has supplanted the use of black-headed pins and will be described later in this chapter. According to Chamlin and Davidoff,3,4 a 1-mm white test object should be visible over most of the screen at 1 m (except at the physiologic blind spot) if there is no visual field defect present. This, however, is not always the case, particularly if it is a patient’s first field test or if a patient’s level of consciousness is depressed (with such patients, I frequently use a 2- or 3-mm white test object). In patients with glaucoma who are well otherwise and have their fields tested repeatedly, however,

Overview of Perimetry

7

the 1-mm white test object is more valuable for discovering the most subtle defects. After a defect is discovered and charted with a small target, it is explored with larger targets, which may be white test objects fastened to the ends of dull-black wands. The largest target made to disappear indicates the density of the defect. It should be noted that when central vision is greatly reduced, fixation is frequently unsatisfactory. This difficulty can be alleviated by using a pair of crossing lines or a circle of adequate size as a fixation target and asking the patient to look at its center, even if the patient cannot see anything in the center. Exploration of the visual field with white targets of various sizes is known as quantitative perimetry. The use of colored targets, or qualitative perimetry, is of value in distinguishing disturbance of the percipient elements of the retina from interruption of conduction fibers. Colored test objects are also of value in eliciting and mapping scotomas in the central portion of the field. Charting the fields in a dim light exaggerates all relative defects, especially those resulting from implication of the rods and cones.5,6 (Various uses of colored test objects and their specific applications are amplified throughout this monograph, especially in Chapters 7 & 10.)7 this is for a reference.

1-1-3 Physiologic Blind Spot. Plotting the physiologic blind spot early in the course of the examination will make patients aware of the possibility of other blind areas, as well as help them understand what is expected of them. Demonstrating a physiologic scotoma to patients early in the field examination will help their fixation when a pathologic scotoma is found. The cause of enlargement of the blind spot as a sign of papilledema has been debated since the time of Knapp’s description.8 Retinal displacement and retinal elevation of percipient elements 9,10 and the Stiles-Crawford effect11 are three of the more traditional explanations. More recently, reports suggesting a relative scotoma on the basis of induced hyperopia were suggested by Dailey et al.12 and Corbett et al.13 The theory is that edematous elevations of the percipient elements away from the blood supply of the choriocapillaris result in a decrease in sensitivity.14 As pointed out by Corbett et al.,13 a purely mechanical displacement of the retinal percipient elements would produce an absolute, not a relative, scotoma. A relative scotoma is the defect produced by papilledema.15 This may not be true if the disc edema is due to another cause such as ischemic optic neuritis, in which the retinal or optic nerve tissues have some additional insult that alters their function. Corbett et al.13 found that in most cases the scotoma could be reduced by using plus lenses. There is previous experience to support their refractive theory as a cause of field defects, as reported by Young et al.16 in their cases of patients with tilted discs. In the tilted disc, there is a bitemporal hemianopia, which can be reduced by using minus lenses. Regardless of the explanation, these authors pointed out an important basic principle in performing fields: Unless the perimetrist pays scrupulous attention to the patient’s refractive error, false-negative or false-positive responses can be produced. 1-1-4 Recording the Fields. Artificial illumination is desirable if fields plotted on one day are to be compared with those plotted on another day; daylight is too variable

8

Visual Fields

to be satisfactory. A generally accepted standard of artificial illumination is 7 footcandles. Even if the light is not exactly 7 foot-candles, an ordinary light meter will allow one to standardize the illumination. Lighting in the room should be replaced from time to time. This is not a problem for the computerized instruments. The fields are recorded on charts that represent the field as the patient sees it. The field for the right eye is placed to the right on the chart: the upper portion, above; the temporal portion, to the right; and the nasal portion, to the left. The test object is recorded as a fraction, with the numerator indicating the size of the test object and the denominator indicating the distance at which the object was used. For example, 3/330 indicates that a 3-mm bead was used on the arm of a perimeter that has a distance of 330 mm (Figure 1-1). This notation system gives more information than merely stating the size of the target in degrees. To transpose the notation into degrees, the following formula can be used: object 180 × = number if degrees distance p For the 3/330 example, the formula would read: 3 180 × = 0.052° 330 301416 It is important to record not only the details of the test object size, color, and distance but also the patient’s state of cooperation and alertness. If the patient is obtunded or sick, the defect may seem worse than on a subsequent visit when the patient is well or has had more experience with field tests. The patient may appear to be improving on the second or third field test when in reality the change is due to increased cooperation or alertness. The degree of alertness is not a problem in the computerized perimeters because there are printouts of false-positive, falsenegative, and fixation losses. The boundary of the field for a given target is usually drawn on the chart as a solid line. Heavy lines may be used for large targets, and thin or broken lines are

Figure 1-1. Visual fields charts. Additional information given at the bottom of the charts aids in the interpretation of the fields later in the course of a patient’s disease.

Overview of Perimetry

9

used for smaller targets. Color may also be used as a coding device. The system that is adopted should be indicated at the bottom of the recorded field so that anyone reading the chart can easily interpret it. The choice of sizes of test targets depends in part on the alertness of the patient and on the nature of the defect in the field. It is generally advisable to start with the smallest target the patient can see (often 2 mm), proceed to one or two targets of medium size, and finally determine the largest target that will disappear in any portion of the defect. With experience, one learns to judge the ability of the patient and to guess in advance the probable or possible extent of visual loss. This approach varies with different diseases. For instance, small test objects are necessary to use to discover subtle and early changes in glaucoma. In neuro-ophthalmology, small test objects are used to find the field defect but are followed up with the largest test object that still demonstrates some of the defect, because the largest test object will show improvement first. The points at which the target is seen on the perimeter are recorded, in degrees, from the point of fixation. The arm of the perimeter is marked in degrees, making notations easy. Because it is preferable on the tangent screen to start with an unmarked surface, a method of transferring the results from the screen to the chart is needed. This can be done by superimposing on the screen a projected slide on which the degrees are ruled or by pulling down in front of the screen a similarly marked transparent cellophane sheet. Most good perimeters, such as the Goldmann, are self-recording and speed the taking of a field. A pantograph has been designed for transferring the results on the tangent screen to a record card fastened to the wall at the side of the screen. This device is generally not available; however, when used, it produces quite satisfactory results when performing tangent fields. The perimetrist should be aware that a vast number of methods and instruments have been designed over the years for the purpose of seeking evidence of impaired function of the visual pathway. The confusing array of available techniques, many of them offering distinct, if limited, advantages over others, stems from an uncertainty of exactly what should be measured in perimetry. The subject is still wide open for further fundamental research. The essence of doing a good field lies not in the perimeter used but in the acumen of the perimetrist; a tangent screen used skillfully by an experienced perimetrist is every bit as effective as the Goldmann or Humphrey perimeter. The introduction of computerized perimetry has meant a significant step forward in measuring defects in the visual field.17 There is no doubt that subtle defects not reported by the patient or not found on routine perimetry are being reliably found with computerized perimetry (Figures 1-2 and 1-3). However, in my experience, computerized perimetry occasionally has one drawback in the routine initial examination of neuro-ophthalmic patients: It is a prolonged procedure and may unduly fatigue a patient who is already ill with a neuro-ophthalmic disease that may impair concentration and the ability to respond reliably. The use of computerized perimetry as a way to follow a field defect or in patients who have experience in field examinations or who are physically well, such as patients with glaucoma, is an invaluable diagnostic instrument. For some of my neuro-ophthalmology patients, I use computerized perimetry as an adjunct, because initially it is more important

A Figure 1-2. A comparison of techniques using computerized perimetry, static perimetry, and tangent screen. (A) A Humphrey field can be read out in grayscale or a numeric scale. The latter is shown in comparison to the normal for that area of the field. (B) A normal Goldmann and tangent screen examination with a subnormal static profile across the 0° to 180° meridian of the right eye. (C) A computerized representation in grayscale with a normal field in the left eye but a depressed central area in the right eye. (Source: A, Courtesy California Pacific Medical Center. B, C, Reprinted by permission from Younge BR. Computer-assisted perimetry in visual pathway disease: neuro-ophthalmic applications. Trans Am Ophthalmol Soc. 1984;82:899–942.)

10

Overview of Perimetry

11

B

C Figure 1-2. (Continued)

to determine the anatomic focus of a defect and to establish a likely differential diagnosis. During the past few years, we have had extensive experience with the Octopus perimeter and more recently with the Humphrey. In this monograph, for the sake of brevity, the isopters of only one or two test targets are shown in cases using the Goldmann perimeter, although in actual practice several targets are usually used. The reader is also warned against assuming that the technique selected to demonstrate a particular field defect is the only technique that can or should be used. In any examination, both peripheral and central field examinations should be performed to establish the true limits of the defect.

12

Visual Fields

Figure 1-3. A subtle central scotoma shown on a computerized representation. (Source: Courtesy Humphrey Instruments.)

1-2 STRUCTURE OF THE VISUAL PATHWAY The visual pathway consists of bundles of nerve fibers connecting the retina of each eye with the visual cortex of the occipital lobes. In the retina, there are three layers of nerve cells: the rod and cone cells with their receptors pointing outward, a middle layer of bipolar cells, and an inner layer of large ganglion cells. Each layer sends a single axon back through the optic nerve, the optic chiasm, and one of the

Overview of Perimetry

13

optic tracts. The fibers of the optic tracts end in the lateral geniculate bodies. Other fibers leave the tract before its termination and turn medially to end in the pretectal region. These latter fibers are concerned with pupillary reflexes and are not shown in Figure 1-4. The cells of each lateral geniculate body send long axons backward, forming a thick band called the optic radiation. This structure ends in the visual cortex on the medial surface of the occipital lobe, in the region of the calcarine fissure. The visual pathway lies near the base of the brain. The anterior half of each optic nerve lies within the orbit; the posterior half lies within the optic canal of the sphenoid bone and within the cranial cavity. The chiasm is suspended in the basal cistern 5 to 10 mm above the hypophysis and forms part of the floor of the third ventricle. The optic tracts encircle the cerebral peduncles and are laterally covered by the foreparts of the temporal lobes. These fibers terminate in the lateral geniculate bodies, gray masses of cell bodies situated at the posterior lateral margin of the peduncles. The geniculocalcarine fibers, comprising the optic radiation, lie in the external sagittal stratum close to the outer walls of the lateral ventricles. The geniculocalcarine fibers first extend laterally. The upper fibers soon turn backward, but the lower ones loop forward a variable distance around the inferior horn of the lateral ventricle (forming Meyer’s loop) before they pass backward and join the upper fibers on their way to the visual cortex. They all lie deep within the temporal and occipital lobes. (The importance of the relationship between adjacent anatomy and a specific field defect is graphically outlined in Chapter 2.) Figure 1-4 is a diagram of the visual pathway viewed from above. The rectangle at the top represents the field of vision; the yellow dot at its center is the point of fixation. A vertical line and a horizontal line divide the field into quarters. The color of each quadrant corresponds to that of the portion of the visual pathway along which that quadrant is projected. The left half of the visual field (green) is projected onto the right half of each retina. From there, axons pass back through corresponding portions of the optic nerves to the chiasm, where fibers from the nasal half of the left retina cross to the opposite side, while fibers arising from the temporal half of the right retina remain uncrossed. Behind the chiasm, all the nerve fibers concerned with the left half of the visual field lie within the right half of the visual pathway. The macular fibers (yellow) occupy temporal sector of the optic nerve at first; as they pass backward, they dip into the nerve, lying centrally throughout its posterior portion. Axons arising in the nasal half of the macula cross the chiasm, mostly in its posterior portion; axons arising in the temporal half of the macula remain uncrossed. In the tract, the macular bundle is at first central; as it passes backward, it rises and ends in the upper and posterior portions of the lateral geniculate body. In the optic radiation, the macular bundle occupies the central third and spreads over a large portion of the visual cortex at the occipital pole. The macular bundle can be divided into quadrants similar to those of the more peripheral fibers, but for the sake of simplicity this is not shown in the figure. The left half of the visual field (green) is projected onto the right half of each retina. From there, axons pass back through corresponding portions of the optic nerves to the chiasm, where fibers from the nasal half of the left retina cross to the

Figure 1-4. The visual pathway, viewed from above. The different colors represent

the locations of the visual fibers from the retina to the optic nerve, the optic chiasm, the optic radiations, and the calcarine cortex. Anatomic cutouts on the lateral edges of the diagram show the variations for temporal and nasal fibers and particularly the macular.

14

Overview of Perimetry

15

opposite side, while fibers arising from the temporal half of the right retina remain uncrossed. Behind the chiasm, all the nerve fibers concerned with the left half of the visual field lie within the right half of the visual pathway. The macular fibers (yellow) occupy the temporal sector of the optic nerve at first; as they pass backward, they dip into the nerve, lying centrally throughout its posterior portion. Axons arising in the nasal half of the macula cross the chiasm, mostly in its posterior portion; axons arising in the temporal half of the macula remain uncrossed. In the tract, the macular bundle is at first central; as it passes backward, it rises and ends in the upper and posterior portions of the lateral geniculate body. In the optic radiation, the macular bundle occupies the central third and spreads over a large portion of the visual cortex at the occipital pole. The macular bundle can be divided into quadrants similar to those of the more peripheral fibers, but for the sake of simplicity this is not shown in the figure. The arrangement of the nerves within the pathway is best demonstrated by tracing axons arising in the various quadrants of the retina. For example, the upper left quarters of both retinas (light blue) are concerned with the lower right quadrant of the visual field. From their ganglion cells, axons pass back through corresponding sectors of the optic nerves. Those axons arising in the nasal portion of the right retina cross in the upper portion of the chiasm; those arising in the temporal portion of the left retina remain uncrossed. The lower nasal fibers of each optic nerve cross the chiasm and turn back into the chiasm and optic tract. In the optic tract, as these fibers pass backward, they turn mesially, ending in the medial portion of the lateral geniculate body. From there, nerve cells send axons back through the upper third of the optic radiation to end above the calcarine fissure on the cuneus. The lower left quarters of both retinas (blue) are concerned with the upper right quadrant of the visual field. They send axons back through corresponding sectors of the optic nerves. Those axons arising in the nasal portion of the right retina cross through the lower part of the chiasm, while those arising in the temporal portion of the left retina remain uncrossed. In the optic tract, as the axons pass backward, they turn outward and end in the lateral portion of the lateral geniculate body. From there, nerve cells send axons back through the lower third of the optic radiation, ending below the calcarine fissure on the lingual gyrus. The complexity of the internal arrangements of fibers in the optic nerves and tracts may be attributed in part to the intrusion of the macular bundle and to the rotation of the nerves and tracts on their longitudinal axis. The optic nerves rotate mesially as they approach the chiasm, their upper sectors becoming nasal. In the anterior portions of the tracts, as in the posterior portions of the nerves, fibers that arose in the upper quadrants of the retina lie dorsomedially; those that arose in the lower quadrants of the retina lie ventrolaterally. The rotation continues within the tracts to their termination, where fibers that were originally superior are now medial and those that were inferior are now lateral. The horizontal raphe of the retina is vertically projected on the lateral geniculate body. At the beginning of the optic radiation, a counterrotation returns the fibers to relationships that existed at the forward portions of the optic nerves.

16

Visual Fields

1-3 INTERPRETATION OF DEFECTS IN THE FIELDS Each of the many types of defects has a specific anatomic location. Some are peripheral, such as general contraction, quadrantanopsia, hemianopsia, altitudinal defects, and others. Those inside the peripheral limits of a normal field are central, paracentral, cecocentral, and arcuate scotomas. Some defects are best seen at the vertical meridian such as temporal lobe and pituitary lesions. Arcuate lesions may be seen along the horizontal meridian and are called nasal step defects. Not all defects of the central field cause loss of acuity. Therefore, good acuity does not rule out defects near the center that do not break out to the periphery. Hence, it is important to examine the central as well as the peripheral field. Lesions that interrupt the visual pathways behind the chiasm produce a homonymous hemianopia; that is, they impair the function of both eyes, causing defects in either the right or the left half of both visual fields without affecting the other half-fields. The term hemianopia is not restricted to its literal sense. Rather, it implies a loss in one of the half-fields, a loss that is not necessarily complete, but that may be quadrantic, partial, or even relative. Defects of the fields that are similar in the two eyes are called congruous (Figure 1-5). Exact congruity suggests that the location of the lesion causing the defects is in the posterior portion of the optic radiation. Lesions that interrupt the anterior portion of the radiation may produce defects that are slightly incongruous (in my experience, however, they are usually congruous). Lesions that interrupt the tracts cause incongruous homonymous defects that are grossly dissimilar (Figure 1-6). The reason is that while fibers from the corresponding points on the two retinas are projected onto identical areas of the visual cortex, they do not pursue exactly the same course in arriving there. Near the cortex, corresponding fibers lie close together. Farther forward in the optic tract, they are only roughly sorted out. Consequently, a lesion situated in the tract may interrupt fibers from a segment of the retina of one eye and from a larger or smaller segment of the retina of the other eye. Lesions situated at the chiasm, by interrupting the crossing nasal fibers, bring about a loss in the temporal portion of the field of each eye. This loss is called a bitemporal hemianopia (Figure 1-7). Such lesions are commonly tumors, which may grow toward one side more than the other and appear as incongruous defects. By interrupting an optic nerve, a central scotoma may be added to the bitemporal hemianopia, or one eye may be completely blind, leaving only a nasal half-field in the opposite eye (Figure 1-8). If the tumor grows posteriorly or there is an anteriorly placed chiasm, the tumor encroaches on the tract and causes an incongruous homonymous hemianopia (Figure 1-9). A scotoma is a field defect surrounded by a normal field (Figure 1-10). This differentiates it from a hemianopia or a quadrantanopia, both of which break out to the periphery and have no field remaining beyond their peripheral limits. Lesions that affect the retina or the optic nerve induce a defect in the field of the corresponding eye and spare the field of the other eye. The characteristic loss is in the form of a central scotoma, but contractions also occur and complete blindness may ensue. Certain poisons and inflammations may affect both optic nerves and

A

B Figure 1-5. These two central fields have the same degree of field defect to the same testing circumstances. (A) Tangent screen representation. (B) Computerized representation of left eye. The right eye is identical, as in the tangent screen representation (A).

17

A

B Figure 1-6. A grossly incongruous field defect. The 3-mm test object demonstrates

a different-size field defect in each eye. (A) Tangent screen representation. (B) Computerized representation of left eye. (C) Computerized representation of right eye.

18

Overview of Perimetry

19

C Figure 1-6. (Continued)

produce defects in the fields of both eyes. In these instances, the defects may be present in both the nasal and the temporal half of the fields and are not limited by the vertical midline, as are defects caused by postchiasmal lesions. Defects vary in their densities. In some cases, large targets disappear; in others, only small targets fade out. A defect so dense that not even light is perceived within it is said to be absolute; all other defects are relative. The density of a defect indicates the degree of interruption of the nerve fibers involved. The term sloping refers to the variation of the size of the field defect according to different sizes of test objects. The smaller the test object, the larger is the defect, so the edge of the defect looks like a children’s slide rather than a cliff. This type of defect is more commonly seen with a tumor that has a central absolute defect and surrounding tissue that is variably affected (Figure 1-11). This damaged tissue produces a defect resulting from that involvement to varying sizes of test objects or different shades of gray in the Humphrey perimeter that depicts the same variations.

20

Visual Fields

Figure 1-7. A bitemporal hemianopia is incongruous, typical of chiasmal tumors. (A) Tangent screen representation. (B) Computerized representation of left eye. (C) Computerized representation of right eye.

The term sharp margin indicates that the edge of the defect can be demonstrated by large or small test objects; the site of the defect remains the same regardless of the size of the test object (Figure 1-12).This is shown as no variation of the grayscale and numeric readout. Such a demonstration indicates a sharp margin of tissue involvement, which is usually seen with vascular infarctions. For this reason, test objects of several different sizes should be used to plot any field defect. Merely finding the defect is not sufficient. This is done in computerized perimeters by using a brackening technique. The defects affecting the optic nerve and the retina differ from those affecting the chiasm, the tract, and the optic radiation. In addition to the central scotoma (see Figure 1-10), there are defects of the following types: paracentral (Figure 1-13), cecocentral (Figure 1-14), and arcuate (Figure 1-15). Quadrantic defects also arise in the retina as a result of vascular disease, but they differ in one important aspect: Unlike quadrantic defects behind the globe, a quadrantic defect caused by

Overview of Perimetry

21

Figure 1-8. A chiasmal tumor not only involves the crossing fibers from both eyes

but also encroaches on the left optic nerve. (A) Tangent screen representation. (B) Computerized representation of left eye. (C) Computerized representation of right eye.

an infarction of the retina has the central apex of the lesion pointing toward the blind spot rather than toward the fixation point. A large test object that reveals a partial quadrantic defect may not demonstrate this, so smaller and smaller test objects should be used to enlarge the defect until it extends to the blind spot or to the fixation point. Neurologically, the field of vision divides at the fixation point (Figure 1-16A), whereas the vascular supply to the retina divides into quadrants at the optic nerve or at the blind spot (Figure 1-16B).

1-4 TECHNIQUES OF FIELD TESTING 1-4-1 Confrontation Technique. The easiest and most rudimentary method of field testing is the confrontation technique.18 The procedure may involve the examiner holding up fingers, which are the test object, and asking the patient whether they

22

Visual Fields

Figure 1-9. A tangent screen examination demonstrating an incongruous defect. Different isopters would still show the incongruous defect to a greater or lesser degree. (A) Tangent screen representation. (B) Computerized representation of left eye. (C) Computerized representation of right eye.

are moving. Because the patient has only two choices; however, this is not the best way to perform the test. Usually, the examiner stands several feet in front of the patient.18,19 The patient’s right eye is covered and the examiner’s left eye is closed, enabling the examiner to use the right eye as a check against the patient’s left eye for field size. The patient is to fixate on the examiner’s nose. The examiner shows fingers in all four quadrants of the patient’s left eye out to the limits the examiner can see them with the right eye. If the patient counts them correctly, there is presumably no gross defect. Most people can distinguish one, two, or five fingers accurately, but many cannot accurately tell three and then four fingers. If three and four fingers are used consecutively, incorrect answers should not be counted. How the examiner holds his or her hand is important. What appears to the examiner as two fingers

Overview of Perimetry

23

Figure 1-10. This defect is found only in the left eye and is presented to several

sizes of test objects. The defect does not respect the vertical midline and has a normal field peripheral to it in all quadrants. (A) Tangent screen representation. (B) Computerized representation of right eye.

(Figure 1-17A) may be perceived as only one finger if one finger is behind the other in the patient’s field of vision (Figure 1-17B). For the patient who has trouble fixating, finger counting offers an advantage over other methods of field testing. In this technique, the number of fingers can easily be shown only briefly before the patient is forced to change fixation. If the finger-movement technique is used, however, the fingers will theoretically be moving almost constantly as the examiner briefly projects a few fingers into one of the patient’s fields. The second step in the confrontation technique is to project a series of fingers into the nasal and temporal fields of one eye simultaneously. The combinations are one and one, two and two, one and two, and one and five fingers. Other combinations, such as one and three or three and four fingers, should not be used, because even a patient with normal fields will frequently miss an accurate numeric identification. The patient who fails this test of simultaneous stimulation may have a more subtle field defect than could be demonstrated by testing each quadrant separately. Failure of the test could also represent the extinction phenomenon (discussed in Chapter 10) as the radiation passes through the parietal lobe, or it could reflect the patient’s inability to add one finger and two fingers (a condition termed dyscalculia). The last step of the confrontation technique is to compare the temporal and nasal fields for subtle differences. This requires judgment on the part of the patient;

24

Visual Fields

Figure 1-11. As smaller or subtler test objects are used, the defect becomes larger and is described as having a sloping margin with a relative defect. (A) Tangent screen representation. (B) Computerized representation of the right eye, which demonstrates a variable defect, as does the left eye. The left eye has a congruous defect to the right eye.

therefore, the interpretation is not as clear-cut as are the results of the first two steps. The examiner shows one hand to each half-field of the patient’s left eye as the patient continues to fixate on the examiner’s nose. The patient then compares the clarity of the two hands. If the patient says the hand in the temporal field is blurred compared with the hand in the nasal field, a field defect may exist. When the right eye is similarly examined, the patient may give a similar response for the temporal field, suggesting chiasmal involvement. Blurring of the nasal field would suggest a homonymous defect, indicating a location in the optic radiation. If the patient gives no such response for the right eye, the possibility exists that the defect is not a true defect in the left eye—that it is instead in the nasal retinal portion of the left eye or in the nasal portion of the left nerve—or that there is a defect in the right eye that is much more subtle and was not picked up. The simultaneous comparison of nasal and temporal fields can further be refined by using colored test objects, such as the tops of mydriatic bottles. If you use two bottle tops at a time as comparison, it is important to repeat the test and switch bottle tops, because the difference noted by the patient may be a difference in manufacture.

1-4-2 Central Field Technique. The area encompassed by the central field is the central 30°, and in the central field technique this is examined at a distance of 1 m with a

Overview of Perimetry

25

Figure 1-12. The defect as shown is the same for the three test objects. If the defect

is found with all test objects, it is called absolute. (A) Tangent screen representation. (B) Computerized representation of the right eye. The left eye, which is not shown, is a mirror image of the right eye, as demonstrated (A).

1-mm white test object. Chamlin4 found that 30° was the average limit for seeing a 1-mm white test object. There are many sizes, colors, and shapes of test objects for performing central field examinations on the tangent screen. The advantage of a spherical test object is that it projects an image of the same size on the retina regardless of how it is held or in what part of the field it is held. The disadvantage is that the sphere cannot be turned off or turned over so that the patient cannot see it, because it is all of one color. The ability to hide a test object or to turn it on and off during the field test is important in testing the validity of the patient’s response. The flat test objects of the Bausch & Lomb type have different colors on each side. These test objects can be modified by blackening one side and turning them over so that the color is hidden and the dark side now blends into the black tangent screen. Unless the test object is held just right, however, it is seen at different angles and, therefore, presents a different size at different times. Even if the test object is held flush with the tangent screen, it projects a slightly different size on the retina as it is moved farther from fixation. This small problem, however, is overshadowed by the ability of the examiner to turn the test object on and off to test the validity of the patient’s response. It is important that the examiner learns to test the validity of

Figure 1-13. Paracentral defects are different in each eye. As shown, the defects have different densities. (A) Tangent screen representation. (B) Computerized representation of right eye.

Figure 1-14. A cecocentral defect differs from a central scotoma in that the

defect extends asymmetrically from the fixation point toward the blind spot and usually connects with it. (A) Tangent screen representation. (B) Computerized representation of right eye.

26

Overview of Perimetry

27

Figure 1-15. An arcuate defect follows the direction of arcuate fibers coming out

of the blind spot and stopping at the horizontal raphe, as shown in the left eye. The defect is not always continuous but may be segmental along the same meridian, as shown in the field of the right eye. (A) Tangent screen representation. (B) Computerized representation.

the patient’s response. It is also important that the examiner learns to twist the handle of the test object between the fingers so the examiner knows without looking when the test object is white or black. Turning one’s hand or head may give an unwanted clue to the patient. The projection perimeter machines, such as the Goldmann or a computerized perimeter (Figure 1-18), can switch the test object on and off and can project the same-size spot of light in all quadrants. An additional advantage of projection perimeters is that the examiner can move the test object randomly from one part of the field to another without the examiner moving or the patient being aware of the change. This feature eliminates the otherwise obvious clue to the patient that the test object will always appear from the side on which the examiner is standing. There is no absolute, standard way to conduct a central field examination, but there are general guidelines that should always be observed: 1. Before the examination begins, the patient is told what a field examination is, what is going to be done with the test object, and what is expected of him or her. This is especially important for the patient who is having a field examination for the first time: The patient will perform more accurately if he knows what is being asked of him. There is no question that the more often a patient performs a field, the more precise he will become. However, an initial examination can be improved immeasurably if the patient thoroughly understands the process.

28

A

Visual Fields

B

Figure 1-16. (A) The apex of this defect goes to the fixation point and represents

a lesion behind the lamina cribrosa in the nerve. (B) The apex of this defect goes to the blind spot and is due to a vascular lesion in the retina that divides the field at the blind spot, not the fixation point.

2. On a tangent screen examination, the patient is shown the test object and is told to give a verbal response when the white side is exposed and to give no response for the dark side. The patient is then asked to fixate on a target on the screen. At this point, before mapping the patient’s blind spot, the examiner explains that there are certain areas within which the patient will not be able to see the test object and that this is normal. The examiner then proceeds to map the blind spot as an example. The advance warning relieves the patient’s anxiety or curiosity about not seeing something essentially in front of him and allows him to fixate better. So when another scotoma is found, the patient should be able to fixate well during exploration of its limits. 3. After the blind spot is mapped, the peripheral limits of the field are examined. A 1-mm white test object is usually used, but a 2-mm white test object will be required if there is too much peripheral contraction. Recordings should be made about every 15°. In the usual testing format, the test object is brought from beyond the visible limits of the field to a point at which the patient just sees it. At this point, a recording mark that will later be transferred to a permanent record is made on the screen; it is here where a cardinal mistake is often made. The perimetrist must watch the patient all the time to make sure small lapses in fixation do not mislead him. Special care should be taken when using the Goldmann perimeter because the patient’s fixation must be watched through a small tube. With any perimeter, a lapse in fixation of which the patient is unaware may cause a more significant error than missing putting a mark on the tangent screen by 1°. Computerized perimeters have built-in fixation sensors.

Overview of Perimetry

29

4. After the outer limits of the central field are examined, the inner aspects are explored. Once a defect is found, the examiner tests for the defect in all directions to map its limits. This procedure again uses the principle of going from a nonseeing, or scotomatous, area to a seeing area. The density of the scotoma is also examined by using different sizes of test objects until the defect can no longer be found. During the testing of any area, the dark side is shown periodically so that the validity of the patient’s response can be evaluated. The usual tangent screen has subtle lines sewn in, marking out the degrees from fixation and the blind spot. The use of pins or chalk marks on the screen is clumsy as well as difficult. In addition, if the marks are too obvious, they act as a clue for the patient as to the limits of the field. A gray, No. 966 Eagle Prismacolor pencil is

A

B

Figure 1-17. (A) The correct way is to present two fingers to a patient so they

are side by side in a frontal plane. (B) The incorrect way is to show two fingers, with one behind the other so they appear as one finger to the patient. (Source: Photographs courtesy of Pamela Ossono.)

A

B

C Figure 1-18. (A) Anterior view of Goldmann perimeter. (B) Posterior view of

Goldmann perimeter. (C) Humphrey perimeter. (Source: Photograph courtesy of Pamela Ossono.)

30

Overview of Perimetry

31

an excellent marker. If the pencil is used lightly, the patient cannot see the marks. Different symbols can be used to outline the different sizes of test objects. When the examination is finished, all the room lights are turned on and the marks can be seen and transferred to a permanent record. The marks can easily be erased from the felt screen with a felt blackboard eraser, and the screen is ready for another examination.

1-4-3 Chamlin Step Technique. When the examiner finds a minimal peripheral field defect, he or she needs to determine whether it is truly an early defect or merely an artifact. Chamlin3,4 developed a theory and a method for evaluating this problem. He postulated not only that the peripheral defect affects the peripheral fibers but also that there is hemiretinal suppression. Therefore, all the examiner has to do is find a method to compare one half of the field with the other. This is done at the vertical meridian, which separates the two homonymous or bitemporal fields. In the experience of many perimetrists, a significant defect is one that differs at least 10° from one side to the other. If the difference is less, the test is not necessarily negative, but it is questionable. If the points of reference are at 15° on either side of 90° and at 90°, the line of the field can look like that in Figure 1-19. If, however, one measures 2° on each side of the vertical meridian, the difference between the two fields is obvious (Figure 1-20). The examiner, therefore, should always measure on each side adjacent to the vertical meridian and not exactly at the vertical or at 15° or 30° nasal and temporal to the 90° meridian. There are several ways to evaluate the problem. The examiner can simultaneously bring down on either side of the vertical meridian two small white test objects of the same size. The patient tells when and in which field he sees one test object and when he sees both. The patient’s fixation must be scrupulously watched, since recognition of the first object will encourage him to shift fixation. If the patient truly sees the two objects at the same time, there may be no difference in the two fields. The test can also be performed by comparing one side at a time using a single test object. Any difference of 10° or more is considered significant, and the defect, although subtle, is real. If the defect is consistently present but is equivocal such as 5° or 6°, a further refinement can be added—the use of color recognition. The perimetrist brings a colored test object down from the periphery and along each side of the vertical meridian until the color of the test object is recognized. The patient’s fixation must again be constantly and scrupulously watched; when he sees the test object as colorless, he may lose fixation and invalidate the test. The examiner should make sure the patient maintains fixation until he recognizes a color. For example, the usual Bausch & Lomb colored test objects are red and blue, back to back. By repeating the test and varying the color from red to blue, the perimetrist can keep the patient from identifying red automatically regardless of whether red is present. Perhaps the previous defect to a small white test object was 7°; now when the patient is asked to recognize red, he may miss an entire quadrant (Figure 1-21). The examiner shows the patient blue as well as red but only records responses to

32

Visual Fields

A

B

C Figure 1-19. The “X” marks are the points recorded and suggest only a right homonymous defect. The minimal slope leaves the perimetrist wondering if the defect is real. (A) Tangent screen representation. (B) Computerized representation of the left eye. (C) Computerized representation of the right eye.

the red test object. The field to a blue test object is greater than that to the same size of red test object. If the patient has a congenital color defect or even a brunescent cataract that changes his color perception, the test will not be seriously affected because the entire field has the color defect. If there is a hemianopic defect, there will be an even more pronounced color defect in the hemifield. A mistake that many perimetrists make is to select a colored test object that is too small. They assume that if the field defect is equivocal when a 1-mm white object is used, the smallest red test object available (usually 3 mm) must be used. This is fallacious and will produce disappointing results, because it is color, and not size, that is being evaluated. The perimetrist is measuring the decreased sensitivity of cells in the central field that are predominantly cones and that will show defects

Overview of Perimetry

33

A

B

C Figure 1-20. Reference points are on either side of the vertical meridian. Note that

the test object in the left homonymous field is seen at 30°, while just 4° away it is seen at 20°. (A) Tangent screen representation. (B) Computerized representation of the left eye. (C) Computerized representation of the right eye.

to colored test objects before defects to white test objects. All too often, patients with 20/20 or 20/25 vision miss all the colored testing plates because of optic nerve disease. This is a frequent and well-known phenomenon. A 9- or 12-mm red test object is much more effective in achieving valid results. Mindel et al.20 asserted that it is not the color but the intensity of the object that identifies the defect. They believed that equal intensity of red and white test objects should produce equal results. This view is supported by Safran and Glaser.21 However, it is much easier clinically to use the red test object than the white test object of equal intensity that I use the red test object in a patient with optic neuritis and 20/25 vision. The perimetrist avoids the difficulty of identifying the scotoma with a small white test object and demonstrates a central scotoma more easily with a red color test object.

34

Visual Fields

Figure 1-21. A Chamlin step to a 1-mm white test object is only 7°. A defect to a

larger, red test object in color recognition, however, is 90°.

A simple method of comparing the two homonymous fields is to use the red caps from two bottles of mydriatic eye drops or red test objects cut from red paper. The examiner holds one test object in the nasal field as the patient fixates centrally on the examiner’s nose. The patient is asked to comment on whether there is a difference in the color of the two test objects. If there is no defect, the patient will describe them as the same (Figure 1-22A). If there is a defect, the patient will describe one as red and the other as light red, barely pink, or colorless (Figure 1-22B). This suggests a defect on the side described as pink or colorless. If the defect is more subtle, the patient may just comment that there is a qualitative difference between the two test objects. Although subtle, this distinction is valid if it is reproducible. If the patient is then asked to compare one test object centrally and one or two peripherally at the same time, the central one may appear faded or colorless, suggesting a central scotoma (Figure 1-22C and D). If a patient’s test results suggest a particular homonymous defect but are mildly inconsistent, another technique is available. The premise on which the testing of the Chamlin step defect is based is the difference between one homonymous field and the other. Therefore, instead of smaller and more subtle test objects being brought down on either side of the vertical meridian, the test objects should be brought across the vertical meridian. As the test object comes from the supposedly defective side to the normal side, there will be a sudden change in color or in intensity. Repetition of this test will demonstrate that for at least part of the field there is not only a difference but also a vertical line dividing it. In exploring the characteristics of the peripheral field, Damgaard-Jensen22 found that about 50% of normal persons have a slight step at the vertical meridian but not 10°. He also found that the field was always larger on the temporal side of the vertical meridian. Therefore, any suggestion of a Chamlin step with the smaller field occurring on the temporal side should be considered pathologic.

1-4-4 Peripheral Field Technique. The examination of the peripheral field is performed in somewhat the same way as that of the central field. The peripheral limits of the field are examined every 15° and recorded. The area inside that peripheral limit is also examined for defects. The selection of a field machine usually depends on one’s training. The Goldmann perimeter is self-recording. It has the additional advantage

Overview of Perimetry

35

of being able to perform both central and peripheral field examination. The instrument has a variety of sizes, colors, and intensities of test objects available. The Riddoch phenomenon, in which the patient perceives movement before light in the peripheral field, was originally described as a sign of a recovering field defect in the occipital lobe. This can be demonstrated when present by the projection perimeters. In one case, the patient perceives the movement of the projected test light but does not see that same object when it is not in motion.21 They regarded its use as not as anatomically specific—that it could be demonstrated in most field defects using kinetic red stimuli.

A

B

C

D

Figure 1-22. (A) Two colored test objects presented simultaneously to the nasal and

temporal fields appear equal when the fields are normal. (B) The same test with test objects of equal intensity do not appear the same to this patient. In this instance, the temporal field in the right eye is defective. The temporal field test object is colorless or less color-saturated. (C) In this instance, three identical test objects can be used to compare the nasal, temporal, and central fields simultaneously. (D) The central target in this instance is much lighter to the patient than the peripheral field, indicating a central field deficit. (Source: Photographs courtesy of Pamela Ossono.)

36

Visual Fields

1-4-5 Amsler Grid. Although the Amsler grid is sometimes difficult for a patient to learn to use properly, it is particularly useful in subtle foveal problems or in central serous retinopathy.23 I consider using the Amsler grid for a patient who complains of a blur and whose vision is perhaps 20/30. A central scotoma is difficult to demonstrate by the usual field-testing methods. If the patient can learn to fixate steadily on the center of the grid (Figure 1-23), the center lines will appear either distorted or elevated in central serous retinopathy. The image will look as if someone were leaning on the center of a chain-link fence. Two problems the perimetrist may encounter with the test are (1) getting the patient to fixate steadily and to just be aware of the surrounding area rather than looking all around and (2) instructing the patient as to what he may see without suggesting any defects to him. If both of these problems can be resolved, the Amsler grid is a valuable test device. 1-4-6 Color Testing. The use of color has always been controversial. Even those who subscribe to the use of color do not agree about which technique to use. Some use color as an object of decreased sensitivity, while others use it as color recognition. I prefer to use color recognition. This experience agrees with Kollner’s law even if the law is not perfect. The decreased-sensitivity technique gives varied results even in the same patient from examination to examination. There is even some argument about which color is preferable for detecting central field defects. Red has been the most used and seems the most reliable. If color recognition is the technique, then using the smallest red object defeats the purpose. Certainly, smaller and smaller white test objects can reveal the same scotoma, but the progressive

Figure 1-23. The Amsler grid. This type of field covers about 10° around the fixation

point and is used primarily for foveal problems such as central serous retinopathy or subtle changes in macular degeneration.

Overview of Perimetry

37

decrease in the size of white test objects is governed more by the minimal visual threshold than by any other feature. If rods are affected, blue test objects seem to be more sensitive. If the optic nerve is the location of the defect, then red has traditionally been more sensitive. It is difficult to determine whether this increased sensitivity is due to lower intensity than the same-size white test object or to a different wavelength. As described previously, color is a useful method for demonstrating a Chamlin step; it is also particularly useful in demonstrating a central scotoma when small white test objects are either inadequate or inconsistent. When it is difficult to establish whether a decrease in acuity is the result of retinal involvement or of optic nerve involvement, colored test plates such as the HRR plates can pinpoint the location. A patient with optic nerve involvement and 20/30 acuity often misses all the HRR plates. Similarly, a patient with retinal disease and an acuity of 20/200 may see almost all the HRR plates. Even if the patient has some degree of color blindness, there will be a significant difference between the two eyes when one has optic nerve involvement. In central scotomas, cones are particularly involved and red test objects are most useful. Because color is the basis of the test, the perimetrist should not use the smallest colored test object in the set because it is too small to see. One of the more difficult field defects to map is the cecocentral scotoma. Early in the course of the disease, the acuity may still be only moderately depressed. Since a patient with a cecocentral scotoma tends to fixate unsteadily, it is difficult not only to map the defect but also to identify it as a cecocentral rather than a purely central scotoma. The distinction is important because if the examiner can establish it as a cecocentral defect, the etiologic possibilities will be limited. Cecocentral scotomas are usually due to nutritional amblyopia and rarely due to Leber’s optic neuritis or pernicious anemia. Central scotomas have a long list of possible causes. To overcome the problems of mapping the cecocentral scotoma, the “big pumpkin test,” a method involving a variation on the use of color, has been used at the Yale Medical Center.24 A large piece of orange poster board, about 2 feet square, is placed over the tangent screen so that it covers at least 7° on the nasal side of the fixation point and 20° on the temporal side in order to cover the blind spot (Figure 1-24). The patient is asked to look at a fixation spot and say if any part of the poster board lacks color. Even slight eye movements will generally keep the central and cecocentral areas on the orange board. In essence, the entire central field serves as a colored test object, obviating the need to explore that same area with a colored test object to observe where it is seen and where it is not. To help in this identification, the patient is given a 3-foot dowel rod and is asked to map out the defective area. If he continues to shift fixation, he will be somewhat frustrated in mapping out the scotoma. This, in turn, will encourage the patient to fixate better. A color defect, or dyschromatopsia, does occur although rarely on a cerebral basis (discussed in Chapter 10).24,25,26 Such a defect requires very careful examination to ascertain that it is not congenital or some expression of aphasia or visual agnosia.

A

B Figure 1-24. (A) Orange poster board cut out to cover beyond the cecocentral area.

The fixation object is large enough to fixate on. It is also displaced away from the center, to leave more room temporally for the cecocentral area. A right eye test circumstance is illustrated here. (B) The patient is pointing at the area on the poster board that is color-desaturated, corresponding to the cecocentral defect.

38

Overview of Perimetry

39

REFERENCES 1. Horton J, Hoyt W. The representation of the visual field in human striate cortex. Arch Ophthalmol. 1991;109:816–824. 2. Riddoch G. Dissociation of visual perceptions due to occipital injuries with especial reference to appreciation of movement. Brain. 1917;40:15–29. 3. Chamlin M. Minimal defects in visual field studies. Arch Ophthalmol. 1949;42: 126–139. 4. Chamlin M, Davidoff LM. Choice of test objects in visual field studies. Am J Ophthalmol. 1952;35:381–391. 5. Enoksson P. A study of the visual fields with white and colored objects in cases of pituitary tumors with special reference to early diagnosis. Acta Ophthalmol. 1953;31:505–511. 6. Alexander GL. Diagnostic value of colored fields in neurosurgery. Trans Ophthalmol Soc UK. 1956;76:235–240. 7. Feldman M, Todman L, Bender MB. Flight of color in lesions of the visual system. J Neurol Neurosurg Psychiatry. 1974;37:1265–1271. 8. Knapp H. The channel by which, in cases of neuroretinitis, the exudation proceeds from the brain into the eye. Trans Am Ophthalmol Soc. 1870;1:118. 9. Wilbrand H, Saenger A, eds. Die pathologie der Netzhaut. In: Neurologie des Augen. Wiesbaden: JF Bergmann; 1909;4(pt 1):568. 10. Paton L, Holmes G. The pathology of papilledema: a histological study of sixty eyes. Brain. 1911;33:389. 11. Stiles WS, Crawford BH. The luminous efficiency of rays entering the eye pupil at different points. Proc R Soc Lond. 1933;12:428. 12. Dailey RA, Mills RP, Stimac GK, et al. The natural history and CT appearance of acquired hyperopia with choroidal folds. Ophthalmology. 1986;93:1336. 13. Corbett JJ, Jacobson DM, Mauer RC, et al. Enlargement of the blind spot caused by papilledema. Am J Ophthalmol. 1988;105:261–265. 14. Reese A. Peripapillary detachment of the retina accompanying papilledema. Trans Am Ophthalmol Assoc. 1930;28:341. 15. Duke-Elder S, Scott GI. Neuro-ophthalmology. In: Duke-Elder S, ed. System of Ophthalmology. St. Louis: CV Mosby; 1971:12:59. 16. Young SE, Walsh FB, Knox DL. The tilted disc syndrome. Am J Ophthalmol. 1976;82:16. 17. Younge BR. Computer-assisted perimetry in visual pathway disease: neuro-ophthalmic applications. Trans Am Ophthalmol Soc. 1984;82:899–942. 18. Welsh RC. Finger counting in the four quadrants as a method of visual field gross screening. Arch Ophthalmol. 1961;66:678–679. 19. Frisén L. A versatile color confrontation test for the central visual field: a comparison with quantitative perimetry. Arch Ophthalmol. 1973;89:3–9. 20. Mindel JS, Safir A, Schare PW. Visual field testing with red targets. Arch Ophthalmol. 1983;101:927–929. 21. Safran BA, Glaser JS. Statokinetic dissociation in lesions of the anterior visual pathways: a reappraisal of the Riddoch phenomenon. Arch Ophthalmol. 1985;98:291–295. 22. Damgaard-Jensen L. Demonstration of peripheral hemiopic border steps by static perimetry. Acta Ophthalmol. 1977;55:815–819.

40

Visual Fields

23. Amsler M. Earliest symptoms of diseases of the macula. Br J Ophthalmol. 1953;37: 521–525. 24. Walsh T. Paracentral scotoma testing. Ophthalmic Surg. 1973;4:72–79. 25. Green GJ, Lessell S. Acquired cerebral dyschromatopsia. Arch Ophthalmol. 1977;95:121–128. 26. Zihl J, Van Cramon D. Colour anomia restricted to the left visual hemifield after splenial disconnection. J Neurol Neurosurg Psychiatry. 1980;43:719–724.

2 Anatomic Basis and Differential Diagnosis of Field Defects JONATHAN D. WIRTSCHAFTER, MD, AND THOMAS J. WALSH, MD

2-1 CATEGORIES OF FIELD DEFECTS The purpose of any medical test is to confirm or rule out a diagnosis based on the clinical facts. In performing perimetry, the printout of the defect is not the end of the test. For even the most experienced reader, the test results at best tell the location of the defect. The next step is to consider the causes of such a defect in that part of the vision system. The experienced perimetrist will look at the results and suggest a differential list of causes. The primary diagnostic list is frequently aided by adding to the perimetry the medical history and other physical signs. The results of both then lead to the next step: ordering tests to confirm the cause of the field defect. It may require the ordering of a magnetic resonance (MR) image, but that may not be the proper test if the original differential diagnosis is faulty. Sedimentation rate and C-reactive protein may be more appropriate tests if the clinical facts suggest cranial arteritis. If carotid disease is suspected, a computed tomography (CT) angiogram may be more appropriate. In the following discussion of these defects, there has been a melding of a discussion explaining anatomically why these defects occur in certain areas. Because the course and relations of the primary visual sensory pathway have been frequently and well described (including in other chapters of this monograph), this chapter concentrates on the multiple anatomic substrates that may explain each particular pattern of visual field abnormality (Table 2-1). Visual field abnormalities are represented by three categories: monocular, binocular, and junctional. Monocular field defects include those that can be caused by lesions of one eye or optic nerve. Binocular field defects include those that

41

42

Visual Fields

TABLE 2-1. Visual Field Defects Monocular Field Defects Localized Defects Wedge-shaped temporal field defect Arcuate and paracentral field defects Central scotoma or depression Enlarged physiologic blind spot Centrocecal scotoma or depression Equatorial annular scotoma or depression Altitudinal hemianopia Generalized Defects Generalized depression or peripheral contraction Binocular Field Defects Homonymous Hemianopias Complete: macular splitting Incomplete congruous: horizontal sectoranopia Incomplete congruous: paramidline-sparing vertical hemianopia Incomplete: macular sparing Incomplete: two scotomas Incomplete incongruous Incomplete: unilateral sparing of temporal crescent Incomplete: unilateral defect of temporal crescent Bitemporal Hemianopias Complete With central depression, scotomatous Binasal Field Defects Complete Incomplete Altitudinal Field Defects Noncongrous and monocular Congruous Quadrantanopias Superior homonymous, incomplete Inferior homonymous, complete Bilateral central Field Defects Scotoma or depression Bilateral peripheral Field Defects Generalized depression or peripheral contraction Bilateral Checkerboard Scotomas Bilateral Homonymous Hemianopias Junctional Field Defects Complete Monocular Plus Bitemporal Hemianopia plus

Anatomic Basis and Differential Diagnosis of Field Defects

43

may result from single or multiple lesions at one or more points along the visual pathway. Junctional field defects include three types of visual field defects resulting from a lesion at the junction of the optic nerve and optic chiasm or of the optic tract and optic chiasm.

2-2 OVERVIEW OF THE VISUAL PATHWAY The layers of the retina are illustrated in Figure 2-1. A general scheme of visual field defects related to the anatomy of the primary visual sensory pathway is depicted in Figure 2-2, and the blood supply of the visual pathway is shown in Figure 2-3. Other relevant anatomic points will be presented with the specific categories of visual field defects. The visual sensory pathway can be divided into “territories” according to various schemes (Table 2-2). The preretinal territory consists of those anatomic barriers

Figure 2-1. Functional microanatomy of the retina. (A) Cell types: A, amacrine; C, cone; DB, diffuse bipolar; DG, diffuse ganglion; H, horizontal; M, Müller cell; MB, midget bipolar; MG, midget ganglion; R, rod; RB, rod bipolar. (B) Layers and membranes of the retina.

Figure 2-2. Visual field defects related to the anatomy of the visual pathway. The arrow is perpendicular to the horizontal line of the retina throughout the visual pathway. Macular projections are shown as small circles; the monocular temporal crescent field is shown with stipple. Projections from the superior retina (inferior visual field) are shown in red; projections from the inferior retina {superior visual field) are shown in blue. Visual field defects and lesions: (1) Left optic nerve—blind left eye. (2) Vascular lesion of upper aspect of right optic nerve—right inferior altitudinal hemianopia. (3) Left optic nerve involving crossed inferior retinal fibers—junctional scotoma with contralateral superior quadrant field defect. (4) Optic chiasm, crossing fibers—bitemporal hemianopia. (5) Optic chiasm, uncrossed fibers compressed by internal carotid arteries and/or perichiasmal tumor—binasal hemianopia. (6) Optic tract—contralateral relative afferent pupillary defect and contralateral homonymous hemianopia, often incongruous. (7) Lateral geniculate body—contralateral homonymous hemianopia. (8) Optic radiation— contralateral homonymous hemianopia, less congruous anteriorly; optic radiation is external to lateral ventricle. (9) Parietal lobe (and nonstriate occipital visual cortex)— contralateral inferior quadrantanopia. (10) Temporal lobe (and nonstriate occipital visual cortex)—contralateral superior quadrantanopia, often incongruous. (11) Occipital lobe—contralateral homonymous hemianopia. (12) Occipital pole through midportion of calcarine fissure—contralateral hemianopic central scotomas. (13) Anterior portion of calcarine fissure and fibers to it—contralateral hemianopia with macular sparing. (14) Extreme anterior lip of calcarine fissure—contralateral temporal crescent field defect.

44

Anatomic Basis and Differential Diagnosis of Field Defects

45

Figure 2-3. Arterial supply of the visual pathway. Left optic nerve and carotid arteries viewed from left; right optic tract, optic radiation, middle and posterior cerebral arteries viewed from their medial surfaces; other cerebral structures removed. Note the dual blood supplies of the optic tract and lateral geniculate body from the anterior choroidal and posterior cerebral artery branches. The temporal isthmus is supplied by the anterior choroidal artery. Inset, Blood supply of the optic nerve head from anastomotic branches of the pial arterioles and posterior ciliary artery branches. The anastomotic vascular circle of Zinn-Haller is thought to be incomplete, explaining the watershed infarctions at the vertical poles of the optic nerve head that are characteristic of anterior ischemic optic neuropathy. Note that the intraneural course of the ophthalmic artery in the optic nerve is longer than that of the central retinal artery.

to the normal focusing of light onto the retina. There are four neural elements in the primary visual sensory pathway. Three are in the retina: the photoreceptor, the bipolar cell, and the retinal ganglion cell (see Figure 2-1). Afferent sensory systems are described as having three orders of neurons excluding the neurosensory transduction cell, which in the retina is the photoreceptor. The first-order neurons are the bipolar cells. The retinal ganglion cells and the axons in the nerve fiber layer of the retina and the optic nerve comprise the second-order neurons. These neurons are also called the anterior visual pathway and include the retinal ganglion cell layer, optic nerves, optic chiasm, and optic tracts. The third-order neurons extend from the lateral geniculate body to the calcarine cortex along the optic radiation. These neurons are often called the posterior visual pathway.

46

Visual Fields

TABLE 2-2. Territories of the visual pathway Structure

Neuron order

Visual Pathway

Territory

Optical media

—

—

Preretinal

Photoreceptor

—

—

Retinal

Bipolar cell

First

—

Retinal

Retinal ganglion cell body

Second

Anterior

Neural I

Retinal ganglion cell axon

Second

Anterior

Neural II

Optic nerve axon

Second

Anterior

Neural II

Optic chiasm axon

Second

Anterior

Neural III

Optic tract axon

Second

Anterior

Neural IV

Optic radiation

Third

Posterior

Neural IV

Abnormalities of the preretinal territory may produce a generalized depression of the visual field, a flattening of the hill of vision, or sometimes a localized depression but rarely a scotoma. Visual field defects relating to a retinal region such as the macula can involve the retinal territory (with a lesion of the outer retinal layers or choroid) or neural territory I (with a lesion of the inner retinal layers). A partialthickness macular hole involving only the internal layers of the retina would be designated as occurring in neural territory I as described by Trobe and Glaser.1 Neural territory I lesions never respect the vertical meridian but reflect the anatomy of the abnormal retinal sensory structures. Lesions of the retinal nerve fiber layer and optic nerve are designated as occurring in neural territory II, perichiasmal lesions as occurring in neural territory III, and retrochiasmal lesions as occurring in neural territory IV.

2-2-1 Occipital Lobe. Occipital lobe lesions may cause visual field defects due to involvement of any or all of these three anatomic sites: the optic radiation, the primary or striate visual cortex, the secondary visual cortex. Studies performed in the 1990s using MR imaging (MRI), functional MRI, and computer graphics have significantly altered our understanding of the retinotopic projection within the visual cortex. These studies have: 1. Demonstrated remarkable variability in the orientation of the calcarine fissure, from almost horizontal to almost vertical in different human brains. 2. Expanded the proportional area of the macular projection so that it takes up almost all but the anterior part of the striate cortex within and adjacent to the calcarine fissure of the occipital lobe2; the striate cortex is designated as VI. 3. Identified the borders of the secondary visual areas, including V2, V3, and others.3,4 The V designations for the visual areas have been carried over from studies in nonhuman primates.

Anatomic Basis and Differential Diagnosis of Field Defects

47

The medial surface of the occipital lobe is bordered anteriorly by the parietooccipital sulcus, which intersects with the anterior portion of the calcarine fissure that divides the occipital lobe into a superior and an inferior portion. The orientation of the calcarine fissure is now routinely observable on the sagittal MRI section. The primary projection of the fovea to the striate cortex is to the most posterior portion of the occipital lobe and usually extends 1 cm onto the lateral convexity of the occipital lobe (Figure 2-4A). The eccentricity of the projection for the fovea can be measured along the calcarine fissure to the junction with the parietooccipital fissure, where the most peripheral visual field is located; this distance is about 80 mm. The linear cortical magnification factor is one way of expressing the weighting given to the retinal projection within the striate cortex. Horton and Hoyt2 have estimated that the linear cortical magnification factor is 9.9 mm/degree at 1° eccentricity, 2.0 mm/degree at 5°, and 1.6 mm/degree at 10°.

A

B

Figure 2-4. Representation of the primary visual field on the human striate cortex. Blue, binocular representation; green, monocular representation. (A) The cortical banks surrounding the calcarine fissure have been slightly opened to better expose the borders of this region. The lower vertical visual field meridian is located on or close to the medial cortical surface of the upper bank, while the upper vertical visual field meridian is located on or close to the medial cortical surface of the lower bank. Most of the primary visual cortex is hidden within the fissure. (B) The calcarine fissure has been widely opened so that the horizontal meridian is indicated at the depth of the fissure. The fovea is represented at the most posterior portion of the fissure (the fundus), and the relative location of the physiologic blind spot of the contralateral eye is indicated by the dark ellipse. The most anterior 8-10% of the striate cortex (green area) represents the monocular temporal field of the contralateral eye. The scale bar is calibrated in centimeters. (Source: Redrawn with permission from Horton JC, Hoyt WF. The representation of the visual field in human striate cortex: a revision of the classic Holmes map. Arch Ophthalmol. 1991;109:816–824. Copyright 1991, American Medical Association.)

48

Visual Fields

The clinical implications of this magnified projection of the macula in the primary visual cortex are many. With regard to the selection of the automated visual techniques, this information diminishes the concern that a 24° radius visual field might miss some cortical lesion that could have been detected by a 30° radius examination. Actually, a 30° visual field tests 83% of the cortical area, while a 24° visual field tests 80%. Thus, only 3% of the cortical visual field is lost from the examination by the reduction of the tested area and the test time. Conversely, only the most anterior 10% of the striate cortex is involved in the uniocular temporal crescent, a cortical area equal to that devoted to the central 1° of the visual field. The large proportion of the striate cortex devoted to the macula also explains why the central 8° of stimulation during pattern-shift visually evoked cortical potential testing generates 60% of the voltage of the P100 wave form. The most important implication of this expanded representation of the macula is that there is a large anatomic substrate for the phenomenon of macular sparing. Survival of even a small portion of the striate cortex provides some possibility for retained macular function. Most of the striate cortex is contained within the superior and inferior banks of the calcarine fissure. The unfolded size of the striate cortex is 40 × 80 mm. The horizontal meridian for V1 is thus buried along the fold line separating the superior and inferior banks, while the lower and upper visual field vertical meridians are represented on the superior and inferior gyri of the calcarine fissure (Figure 2-4B). The horizontal and vertical visual field meridians meet at the posterior pole of the occipital lobe, where the fovea is represented, and again anteriorly at the anterior pole of the calcarine fissure, where the temporal crescent is projected. The reader can conceptualize this representation by imagining the right visual hemifield as an elastic sheet in visual space attached on an opened semicircular fan with a pivot at the fovea (Figure 2-5). The extrastriate visual cortex is situated above, below, and surrounding the striate cortex on the medial and lateral surfaces of the occipital lobe. This relationship is best seen with computer graphic techniques that unfold, and then cut and flatten the occipital lobe (Figure 2-6).3 These maps can now be made in normal subjects with physiologic, functional MRI and confirm maps made from pathologic cases. While such studies should have many applications with regard to our comprehension of human visual performance, the most immediate application is understanding that an extrastriate lesion that crosses both a horizontal and a vertical meridian of the secondary visual cortex (V2 and/or V3) can cause an inferior homonymous quadrantanopia in the absence of lesions in the parietal lobe or calcarine fissure (Figure 2-7).

2-3 MONOCULAR FIELD DEFECTS 2-3-1 Localized Defects 2-3-1-1 Wedge-Shaped Temporal Field Defect. This pattern of field defect has its apex pointing toward the physiologic blind spot and involves ganglion cell axons from the nasal retina (Figure 2-8).

Anatomic Basis and Differential Diagnosis of Field Defects

A

B

49

C

Figure 2-5. Topologic representation of the right visual hemifield onto the left visual cortex. (A) The reader can conceptualize by imagining the right hemifield as an elastic sheet in visual space attached on an open semicircular fan with a pivot at the fovea. (B) As the fan collapses around the foveal pivot, the vertical meridians become parallel (and inverted due to the inverted image in the eye) and the elastic sheet collapses along the temporal border. The foveal region is stretched toward the periphery, thus magnifying the proportional area of the representation of the central visual field. (C) The flattened representation of the hemifield in the left visual cortex. The upper and lower vertical meridians are parallel, and the temporal borders collapse at the most anterior part of the calcarine cortex. The green areas correspond to the monocular temporal crescent. HM, horizontal meridian; LTB, lower temporal border; LVM, lower vertical meridian; UTB, upper temporal border; UVM, upper vertical meridian. (Source: Redrawn with permission from Horton JC, Hoyt WF. The representation of the visual field in human striate cortex: a revision of the classic Holmes map. Arch Ophthalmol. 1991;109:816–824. Copyright 1991, American Medical Association.)

Retina (1). In the nasal retina, the ganglion cell layer contains the nuclei of the axons that will travel through the retinal nerve fiber layer to approach the optic nerve head in fan-shaped bundles: the superior (A) and inferior (B) radiating bundles. Any lesion causing complete destruction of a region of the inner retinal layers—retinal nerve fiber, ganglion cell, inner plexiform, or inner nuclear (bipolar cell) layer—is likely to cause a dense field defect. Occlusion of one of the nasal branches of the central retinal artery or nasal tributaries of the central retinal vein can cause a wedge-shaped field loss with its apex at the physiologic blind spot. Although the usual pattern of distribution of the branch retinal arteries is quadrantic, it should be noted that the central point of the quadrants is the optic nerve head, not the fovea. When analyzing a visual field defect that involves one or more quadrants of the visual field, one should determine if the quadrants are retinovascular and related to the physiologic blind spot or neurologic and related to the point of fixation. Embolic disease involves the superotemporal branch retinal arterioles more frequently than the nasal branch retinal arteries. The initial finding may consist of a dense scotoma reflecting a complete loss of function of the inner retinal layers, including the nerve fiber layer. Unless recovery occurs within a few days, the field loss may be permanent. The blind spot–based quadrantic field loss may not have its apex extend to the blind spot if the occlusion occurs at a peripheral bifurcation of a retinal arteriole. The inner retinal swelling may first manifest a white appearance (e.g., the cherry-red

50

Visual Fields

A

B

C

D

Figure 2-6. Retinotopic projection of human visual areas as derived from functional MRI studies. (A) MRI snowing original cortical surface of right cerebral hemisphere with retinal isoeccentricity indicated by a color code on the cortical surface. Red indicates the fovea, then blue, then green (for the parafovea), then yellow, then brown for the far periphery. (B) The cortical surface is unfolded by computer manipulation, thus expanding the apparent area of the brain. The local areas and local angles are preserved. The sulcal cortex is dark gray and the gyral cortex is light gray. (C) The cortical surface of the occipital lobe is flattened after the occipital lobe has been cut off, and an additional cut (dashed line in B) has been made in the fundus of the calcarine sulcus. (D) The schematic functional representation of the primate (striate, V1) and the secondary visual cortex (extrastriate visual areas, V2, and to some extent V3 are shown as repetitive mirror images surrounding the primary visual cortex). V4 is present only in the lingual gyrus of the inferior occipital lobe and is associated with color recognition. The arrow shows the location of the fovea. VP, ventroposterior area. (Source: Redrawn with permission from Sereno MI, Dale AM, Reppas JB, et al. Borders of multiple visual areas in humans revealed by functional magnetic resonance imaging. Science. 1995;268:889–893.)

spot of the macula that is seen if both the inferior and the superior temporal branch retinal arterioles are occluded). Occlusion of the superior branch retinal arteriole produces an inferior altitudinal field defect. Acute systemic hypotension can have effects similar to those of arterial occlusion. Thrombosis of the tributary retinal veins usually occurs at and distal to the arteriovenous crossings. Systemic arterial hypotension and ocular hypertension (glaucoma) are predisposing factors. The field defects following venous occlusion are often more peripheral and less dense, giving more hope for late recovery than is

Anatomic Basis and Differential Diagnosis of Field Defects

51

B

A Figure 2-7. (A) A schematic representation showing the retinotopic organization of the left posteromedial occipital lobe and the coordinates of the right visual field projected onto the striate (V1) and extrastriate (V2, V3) visual cortex. The hemivisual field projection is arranged from posterior to anterior so that the horizontal meridian (HM) and the vertical meridian (VM) alternate at the extrastriate area boundaries. A lesion (gray area) that crosses these boundaries will produce a congruous inferior quadrantic visual field defect. (B) The lesion shown is rather anterior so that the fovea 10° is spared. Occipital lobe lesions that do not involve the calcarine fissure can produce a quadrantic visual field defect mimicking a parietal lobe lesion. (Source: Redrawn with permission from Horton JC, Hoyt WF. The representation of the visual field in human striate cortex: a revision of the classic Holmes map. Arch Ophthalmol. 1991;109:816– 824. Copyright 1991, American Medical Association.)

Figure 2-8. Wedge-shaped temporal field defect.

51

52

Visual Fields

possible for the field defects of arterial occlusion. Secondary field changes may occur as a result of chronic macular edema or preretinal and vitreous hemorrhages. Lesions nasal to the optic nerve head will cause wedge-shaped scotomas or peripheral contraction in the field temporal to the blind spot. Lesions of the choroid or outer retinal layers will usually cause less dense defects, because the nerve fibers passing in the overlying nerve fiber layer and some of the input to the bipolar cells in the affected region may be spared. Complete loss of field occurs in regions where retinoschisis has split the retina, thus disconnecting the retinal ganglion cells from the photoreceptors. In contrast, retinal detachments produce milder depressions with less steep margins, at least in their early stages. Optic Nerve Head (2). The retinal course of the axons is such that the most posterior enter the peripheral portions of the optic nerve head while the most anterior retinal axons enter the central portion. Drusen of the optic nerve head produce local injury to axons at the optic nerve head and may cause progressive (rarely sudden) contraction of the visual field or localized depressions or scotomas. The arterial anatomy of the optic nerve head is invoked to explain various patterns of field loss. The ophthalmic artery is the ultimate source of supply to a continuous capillary bed that extends throughout the optic nerve head. The arterial blood is delivered by three systems: (1) branches of the short posterior ciliary arteries that enter the medial and lateral peripapillary sclera within 1 mm of the peripheral margin of the optic nerve head, (2) vessels of the pial plexus and retrobulbar optic nerve, and (3) branches of the central retinal artery. The posterior ciliary artery branches were formerly described as completing the arterial circle of Zinn-Haller, but an anatomically demonstrable arterial circle is generally not complete (see Figure 2-3). The circle of Zinn-Haller may also be supplied by pial arterioles supplying the anterior aspect of the optic nerve. The lack of functional anastomotic continuity is invoked to explain the altitudinal field defects that present in anterior ischemic optic neuropathy. Focal infarctions, usually at the vertical poles of the optic nerve head, may result from infarction in watershed zones between territories served by centripetal optic nerve head branches of the short posterior ciliary arteries (C). The posterior ciliary arteries are generally situated near the horizontal poles of the optic nerve head. They also give rise to centrifugal branches that supply the peripapillary choroid. Retrobulbar Optic Nerve (3). Monocular wedge-shaped nerve fiber bundle defects can occur with infarction, inflammation, tumors, or trauma anywhere between the optic nerve head and the optic chiasm. However, the likelihood of involving only the nasal fibers decreases at 1 cm behind the globe as fibers from the papillomacular bundle disperse throughout the optic nerve.

2-3-1-2 Arcuate and Paracentral Field Defects. Pericentral (within 5° of the fovea) or paracentral (within 20° of the fovea) scotomas or depressions in the Bjerrum region result from injury to retinal ganglion cell arcuate axons arising from the temporal retina (Figure 2-9). These arcuate nasal defects are often referred to as “nerve fiber bundle defects,” although they are just one particular pattern of nerve fiber bundle defect. Injury to either the superior or the inferior nerve fiber bundle will produce an arcuate scotoma terminating at the horizontal meridian. If the superior and inferior fibers are injured asymmetrically, a nasal step field defect will result.

53

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-9. Arcuate and paracentral field defects.

Retina (1A and 1B). During the embryogenesis of the eye, the fovea develops at the temporal periphery. At that time, all the nerve fibers have a fan-shaped radiation from the optic nerve head. Later in development, the fovea migrates toward the optic nerve head. As a result, the temporal nerve fiber layer arches around the fovea, causing the nerve fiber layer to assume an arcuate shape—the superior (1A) and inferior (1B) arcuate bundles. Peripheral axons in the arcuate bundles originate in ganglion cells on either side of the horizontal retinal raphe. The initial course of these axons is parallel to the horizontal raphe (C) where the axons are in the superior or inferior temporal retinal raphe. The arcuate portion of the retinal nerve fiber layer also contains the blood vessels coursing around the macular region. The arcuate portion of the retinal nerve fiber layer is the thickest (0.5 mm), and its bundles reflect bright white striations especially when viewed with red-free (green-filtered) light. The thickness of the nerve fiber layer decreases 1 or 2 disc diameters from the optic disc margin. Loss of the retinal nerve fiber layer may occur in various patterns ranging from diffuse atrophy to highly localized atrophic wedge defects. The minimum ophthalmoscopically detectable wedge defect is the loss of a 50-μ thickness of nerve fibers adjacent to 50 μ of intact nerve fibers. This would represent a loss of 15,000, or about 1% of retinal axons.5 Optic Nerve Head (2). The retinal ganglion cell axons pass as nerve fiber bundles through the pores of the collagenous scleral lamina within the scleral canal.

,

54

Visual Fields

The upper and lower quadrants of the optic nerve head contain the least dense glial and connective tissue and the largest laminar pores. It has been suggested that the vertical quadrants through which the arcuate area ganglion cell axons pass are the most susceptible regions of the optic nerve head to the effects of elevated intraocular pressure.6 This sensitivity may explain the isolated paracentral scotomas 5° to 25° from fixation that are usually the earliest field defects of glaucoma. Conversely, the central 5° and the temporal fields are preserved until the later stages of glaucoma. Glaucoma causes nerve fiber bundle defects that occur first in fibers of the superior (A) or inferior (B) arcuate bundles. Glaucoma is the most common cause of arcuate field defects. The horizontal raphe is the basis for nasal step field defects. The changes are thought to result from localized ischemia of the fibers within the optic nerve head, where they are supplied by branches of the arterioles of the peripapillary region. Optic nerve head drusen, microvascular disease, low-tension glaucoma, tumors, myopia, and congenital defects of the optic nerve head can also cause arcuate field defects. Retrobulbar Optic Nerve (3). Discrete lesions of the retrobulbar optic nerve that are within 1 cm of the globe can spare fibers of the papillomacular bundle. Occasionally, monocular arcuate field defects may appear, with lesions as far posterior as the optic chiasm. It has been observed that any anterior visual pathway lesion can mimic any other. Thus, caution should be applied when localizing a field defect with excessive certainty.

2-3-1-3 Central Scotoma or Depression. This pattern of field defect usually results from a lesion in the papillomacular bundle of retinal ganglion cell axons (Figure 2-10). Retina (1). The papillomacular bundle consists of many small bundles of retinal axons entering the temporal border of the optic nerve head. Axons on the horizontal meridian (A) serve the nasal aspect of the macula and project temporally to the vertical meridian of the visual field. The superior and inferior portions of the papillomacular bundle serve the temporal aspect of the macula. These fibers arch around the fovea to meet at the horizontal meridian (B). The temporal fibers of the papillomacular bundle arise at right angles to the horizontal meridian, while the arcuate fibers arise parallel to the horizontal raphe (C). This may explain why nasal step defects are more marked outside the macular region of the visual field. Some fibers within 3° of the macula segregate to the “wrong” side of the optic chiasm and thus may provide some explanation for some cases demonstrating up to 3° of “macular sparing” in retrochiasmal homonymous hemianopias. However, research using the scanning laser ophthalmoscope has shown that most cases of macular sparing are accompanied by nystagnoid searching movements toward the seeing hemiretina, and thus result from a perimetric artifact rather than an anatomic anomaly.7 Occlusion of the macular branches of one or more branch retinal arterioles or of a cilioretinal artery may also explain a central scotoma or depression. Photocoagulation of outer retinal layer and choroidal lesions can often be accomplished without causing a central scotoma. Initial photocoagulation may scar the nerve fiber layer to the underlying (outer) layers, so that repeated

55

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-10. Central scotoma or depression field defect.

photocoagulation is likely to cause a field defect. The papillomacular bundle contains more than half the optic nerve axons. Significant depressions of visual acuity, central visual fields, spatial contrast sensitivity, and color perception may follow lesions in the macular region. The differential diagnosis of anatomic lesions in the macular region is beyond the scope of this chapter. The projection of human retinal ganglion cell axons of various sizes and physiologic characteristics must be deduced from studies in monkeys and cats, and these studies have not given wholly consistent results. It is generally agreed that the largest, fastest-conducting axons, including 25% of the axons originating in the parafoveal region, terminate in the magnocellular layers of the lateral geniculate nucleus. These are designated as magnocellular (M) cells and in many respects are similar to the Y cells of cats. The larger-diameter axons are thought to be important in such functions as achromatic vision, blue-yellow vision, and high temporal frequency vision (loss of the larger-diameter axons results in lowering of the critical flicker fusion frequency). Glaucoma seems to be a disorder whose initial effects are on the more peripheral M (Y-like) cells. Contrast-sensitivity testing seems to be particularly sensitive to M-cell dysfunction because M cells have large receptive fields, high contrast sensitivity, and a broad band of spectral response.

,

56

Visual Fields

Smaller-diameter, slower-conducting axons arise throughout the entire retina and terminate in the parvocellular regions of the lateral geniculate nucleus. These are designated as parvocellular (P) cells and are similar to the X-like cells of cats. The parvocellular cells include the midget ganglion cells with small receptive fields that account for central vision acuity and color opponent features. They have low contrast sensitivity. Optic Nerve Head (2). Lesions of the optic nerve head may involve sufficient numbers of papillomacular axons to cause a central scotoma. Retrobulbar Optic Nerve (3). Axons arising from the midget ganglion cells of the macula are generally smaller than other axons in the optic nerve. Although they enter on the temporal side of the nerve initially, they mingle with all the bundles from the retinal quadrants within a short distance. The fibers that will cross to the contralateral optic tract are probably segregated a few millimeters prior to the junction of the optic nerve with the optic chiasm. Within the optic chiasm, the macular fibers are generally superior to those from the peripheral retina.

2-3-1-4 Enlarged Physiologic Blind Spot. This pattern of visual field defect results at or in the immediate vicinity of the optic nerve head. It is sometimes designated as a pericecal field defect (Figure 2-11). Peripapillary Retina (1). Papilledema causes swelling of the optic nerve head axons (A) in the retina. These axons are not free to expand at the lamina scleralis (B) but

Figure 2-11. Enlarged physiologic blind spot (a pericecal field defect).

,

Anatomic Basis and Differential Diagnosis of Field Defects

57

are free to expand at the lamina choroidalis (C) and even more free to expand at the lamina retinalis (D), resulting in displacement of the adjacent retina and choroid. Because the peripapillary receptors are displaced laterally in all directions, this could partially explain the increase in the size of the absolute scotoma that comprises the physiologic blind spot. However, the major cause may be swelling of the axons and intercellular retinal edema (E). This masks some of the light falling on the peripapillary retinal receptors and, in turn, causes a relative scotoma or depression outside the absolute scotoma. Moreover, some of the enlargement of the blind spot is refractive, due to peripapillary hyperopia.8 Increased hyperopic refractive correction usually eliminates or reduces the size of an enlarged blind spot resulting from the optic nerve head swelling of increased intracranial pressure (papilledema). Other peripapillary disorders may cause absolute and relative increases in the size of the physiologic blind spot, but these are not usually as symmetric. Often they result in a pericecal depression or scotoma. Advanced papilledema may be associated with swelling of the outer plexiform layer of the retina (F, Henle’s layer). The anatomic relationships of the nerve head to the macula are reversed in the visual field, so the nerve head is anatomically superior to the macula on the globe but the blind spot is below the fixation point in the visual field. It is useful to know certain quantitative anatomic relationships of the optic nerve head (Table 2-3), based on the assumption that 1 mm of circumference at the retina corresponds to about 5.0° of visual field. Optic Nerve (2). Some forms of optic neuropathy, particularly ethanol-induced toxic optic neuropathy, produce centrocecal scotomas.

2-3-1-5 Centrocecal Scotoma or Depression. This phrase describes a visual field abnormality encompassing both the central region of fixation and the physiologic blind spot (Figure 2-12). Retina (1). The papillomacular bundle can be involved by regional choroidal, outer retinal, and inner retinal layer lesions. Causes can include optic pit with serous retinal detachment, macular degeneration or inflammation, or cilioretinal artery (1) occlusion. Superior or inferior cilioretinal arteries are present as congenital developmental variations in about 20% of normal eyes. They are retinal branches of the ophthalmic artery via a posterior ciliary artery rather than via a central retinal artery. A cilioretinal artery may be occluded alone, sparing the retinal regions supplied by the central retinal artery, or it may be uninvolved with occlusion of the central retinal artery leaving a small island of vision with good acuity. Optic Nerve (2). Many types of optic nerve lesions including toxic and nutritional amblyopias may produce centrocecal scotomas. The monofixation syndrome results

TABLE 2-3. Quantitative Anatomy of the Optic Nerve Head Diameter of Optic Nerve Head

Relation of Nerve Head to Macula

Horizontal: 1.1 mm, 5.5°

Nasal: 3.0 mm, 15.0°

Vertical: 1.5 mm, 7.5°

Inferior: 0.3 mm, 1.5°

Visual Fields

,

58

Figure 2-12. Centrocecal scotoma or depression.

in a central scotoma of less than 3° radius in one eye. It can be detected only with binocular perimetry (e.g., red and green targets viewed with a red-filtered right eye and a green-filtered left eye).

2-3-1-6 Equatorial Annular Scotoma or Depression. These visual field defects usually begin as depressions of the field at a 25° to 45° radius from the point of fixation (Figure 2-13). Preretinal Media (1). An annular, or ring, scotoma can be produced by aphakic spectacle lenses and relieved by removing them. The scotoma is produced by light rays striking the highly convex surface from an oblique direction and their being refracted outside the pupil. Equatorial Retina (2). The term retinitis pigmentosa describes a heterogeneous group of genetically determined disorders. Although there are notable exceptions (e.g., unilateral pigmentary retinopathy and sector pigmentary retinopathy), the usual pathologic and visual field changes occur in a circular region measuring 25-50° from fixation. The earliest changes of retinitis pigmentosa may be confused with the arcuate and paracentral nerve fiber bundle defects of glaucoma, where the ring scotoma may originate with vertical enlargement and baring of the physiologic blind spot, giving two arcuate scotomas that complete a ring (see Figure 2-9).

59

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-13. Equatorial annular scotoma or depression.

2-3-1-7 Altitudinal Hemianopia. For a discussion of monocular altitudinal hemianopia, see Section 2-4-4-1. 2-3-2 Generalized Defects 2-3-2-1 Generalized Depression or Peripheral Contraction. Generalized depression or peripheral contraction of the visual field results when a focal light stimulus greater than the background light produces less sensory appreciation than normal at any and all locations of the visual field. The depression may be equal in all meridians or may be unequal. An unequal diffuse field abnormality may be the sum of the abnormalities induced by several factors. An extreme example would be a superior altitudinal defect from a ptotic eyelid plus diffuse depression of the media from a cataract plus a central scotoma from a welding burn of the fovea plus a nasal scotoma from glaucoma. Preretinal Media. The artifact caused by a lens used for central perimetry may often affect the outer points of a central 30° visual field examination and be particularly confusing with programs that merge central and peripheral examinations on a single chart. Body parts such as the nose and upper eyelid frequently cause localized abnormalities. Media opacities include contact lenses, corneal edema, scars, and the astigmatic aspects of keratoconus. Cataracts are the most frequent media opacity and may cause either generalized, localized, or combined depression, contraction,

,

60

Visual Fields

or scotomas. Cataracts cause field defects more by scatter and defocusing of the stimulus light with regard to the background light (decreased contrast) than by absorption or backscatter of the stimulus light. Vitreous hemorrhage can also depress the visual field. Retina. Diffuse disorders of all retinal elements (photoreceptors to ganglion cells) can cause generalized depression of the visual fields. A description of all the disorders that can cause diffuse monocular pathology is beyond the scope of this chapter. Optic Nerve. Most monocular optic nerve disease is compressive, demyelinating, traumatic, or vascular. A detailed description of these disorders is beyond the scope of this chapter. About half the patients with optic neuritis present with diffuse visual field loss; 20% have altitudinal or other nerve fiber bundle-type defects, while only 8% have central or centrocecal scotomas.

2-4 BINOCULAR FIELD DEFECTS 2-4-1 Homonymous Hemianopias. Any retrochiasmal lesion may present with a congruous homonymous hemianopia. The occipital lobes and the parietal lobes are more frequent than the temporal lobes as the locations of lesions causing the visual field defects. 2-4-1-1 Complete: Macular Splitting. A complete homonymous hemianopia with macular splitting refers to the loss of either the entire right or the entire left visual field of both eyes. The central 5° of the visual field is designated as the macular portion of the visual field. It may be included in the complete hemianopic field defect; thus, the macular region is “split.” Conversely, it may not be included, in which case it is “spared” (Figure 2-14). Optic Tract (1). Any lesion ascending the visual pathway beyond the first few millimeters of the optic tract may cause a complete homonymous hemianopia. A lesion at the junction with the optic chiasm may cause a junctional scotoma. Optic tract lesions are associated with wedge-shaped retinal axonal degeneration and death of the retinal ganglion cell layer nasal to the fovea of the contralateral eye. Ophthalmoscopy with red-free light reveals homonymous hemiretinal atrophy at the nasal aspect of the contralateral optic nerve head and in a narrow band along the horizontal meridian at the temporal aspect of the nerve head. It should be noted that the nasal aspect of the fovea is represented by fibers in the middle of the papillomacular bundle. The ipsilateral eye will reveal bow-tie atrophy of the superior and inferior arcuate bundles of the retinal nerve fiber layer. The optic tract syndrome consists of a homonymous hemianopia, atrophy of the nasal fibers of the contralateral optic nerve, atrophy of the temporal fibers of the ipsilateral optic nerve, and a contralateral monocular relative afferent pupillary defect. The presumed basis is that about 60% of afferent pupil fibers decussate at the chiasm, giving rise to unequal representation of the two eyes in each optic tract. Anatomic studies in monkeys have shown that the retinal vertical midline is almost absolute. Above and below the macula, there is only a 1° overlap across the vertically oriented midline strip when the axonally transported tracer horseradish

61

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-14. Complete homonymous hemianopia with macular splitting.

peroxidase is injected into the lateral geniculate body and observed in the retinal ganglion cells. At the macula the overlap widens to 3°. Thus, there is no anatomic basis for macular sparing greater than 3° in patients with a complete retrochiasmal hemianopic lesion. Any macular sparing exceeding this amount must be ascribed to errors in fixation or response. (For further discussion, see Section 2-3-1-3.) There is no evidence of any representation of ipsilateral visual field in the calcarine cortex. Lateral Geniculate Body (2). Both the lateral geniculate body and the optic tract have dual arterial supplies. The lateral and anterior aspect of the lateral geniculate body is supplied by the anterior choroidal artery, a branch of the internal carotid artery (A), while the dorsal and medial aspect is supplied by the lateral choroidal artery (B), a branch of the posterior cerebral artery (C). If both arteries are occluded, a congruous homonymous hemianopia will result. Optic Peduncle (3). The optic peduncle is the first portion of the optic radiation. It may be considered a part of the internal capsule. It is posterior to the lentiform nucleus (D), the posterior limb of the internal capsule (E, sensory), and the anterior limb (F, motor). Occlusion of the anterior choroidal artery (A) or of the posterior choroidal artery (G) can produce contralateral hemianopia, hemianesthesia, and, sometimes, hemiplegia. Optic Radiation (4). The optic radiation passes lateral to the inferior horn (H) and the posterior horn (I) of the lateral ventricle. Field defects are often incongruous,

,

62

Visual Fields

with anterior lesions becoming more congruous as the fibers in the optic radiation approach the calcarine cortex, except for the portion supplied by the deep optic branch (J) of the middle cerebral artery (K). The optic radiation is supplied by branches of the posterior cerebral artery (C), especially the calcarine branch (L). Calcarine Cortex (5). Field defects are congruous, except when the lesion involves only the extreme anterior portion of the calcarine cortex (6), where the contralateral temporal crescent is represented. The blood supply to the calcarine cortex is from the posterior cerebral artery (C). Occasionally, the calcarine cortex receives an anastomotic branch (M) from the middle cerebral artery. In many cases, the macular representation on the lateral convexity of the occipital lobe may be supplied by branches of the middle cerebral artery.

2-4-1-2 Incomplete Congruous: Horizontal Sectoranopia. This phrase describes a homonymous, congruous wedge-shaped field defect extending from the point of fixation to the periphery (Figure 2-15). Lateral Geniculate Body (1). This field defect may result from a vascular lesion (e.g., a lateral choroidal artery infarction) damaging the dorsal portion of the lateral geniculate body that serves the macular function yet causing little damage to the medial portion with the terminations of axons from the inferior retina or to the lateral portion with the terminations of fibers from the superior retina. Frisén9 has presented a case of a congruous homonymous sectoranopia sparing the

Figure 2-15. Horizontal sectoranopia.

Anatomic Basis and Differential Diagnosis of Field Defects

63

,

horizontal sector. Computed tomography (CT) scanning demonstrated changes in the lateral geniculate body that followed ligation of the distal portion of the anterior choroidal artery. Incongruity with lesions of the lateral geniculate body is rare and, if present, may result from simultaneous involvement of the optic tract. Wedge-shaped defects that do not fully extend from fixation to the periphery have been reported with occipital infarction (2).

2-4-1-3 Incomplete Congruous: Paramidline-Sparing Vertical Hemianopia. This phrase describes an incomplete congruous hemianopia in which the paramidline portion of the visual field is spared. The spared portion is bordered by a vertical line tangent to a horizontal radius (Figure 2-16). Posterior and Middle Cerebral Arteries. After recovery from coronary artery bypass surgery, cardiac arrest, or similar hypoxic events, dysfunction may be present in the watershed zones (1) at the borders of perfusion of the middle (A) and posterior (B) cerebral arteries. Thus, occlusion can give rise to a field defect stretching from the top to the bottom of the visual field about 5-10° from the vertical midline and may coincide with the lateral border of cortical area 17. Prenatal and infantile hypoxia in these regions may lead to a dense incomplete hemianopic field defect associated with a porencephalic cyst (2). The injured fibers of the optic radiation are the most medial fibers, especially those at its upper and lower margins, while the more lateral and intermediate regions of the optic radiation are spared.

Figure 2-16. Paramidline-sparing vertical hemianopia.

Visual Fields

,

64

Figure 2-17. Homonymous hemianopia with macular sparing.

2-4-1-4 Incomplete: Macular Sparing. If a congruous homonymous hemianopia is complete except for the central radius of less than approximately 5°, it is said to have sparing of the macular representation of the visual field (Figure 2-17). Immediately Postchiasmal Optic Tract (1). Lesions such as craniopharyngiomas may cause a homonymous hemianopia with macular sparing, a rare form of congruous field defect. When it occurs, the region of sparing is likely to be less than 2°; perhaps only a few macular fibers remain. Calcarine Cortex (2). The more anterior the lesion of the calcarine cortex, the more the central field is spared. Lesions with macular sparing are usually the result of calcarine artery occlusion or tumors. The anastomotic branch of the middle cerebral artery (A) may supply the posterior lateral portion of the calcarine cortex. 2-4-1-5 Incomplete: Two Scotomas. These field defects result from two separate lesions in one retrochiasmal visual pathway (Figure 2-18). Postchiasmal Visual Pathway (1). While any location may be involved, the most typical location would be due to occlusions of two branches of one posterior cerebral artery or to occlusion of the ipsilateral posterior and middle cerebral arteries. One or both of the hemianopic homonymous scotomas may involve or spare the macula, as illustrated.

65

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-18. Two ipsilateral homonymous hemianopic scotomas.

2-4-1-6 Incomplete Incongruous. Incongruity results when the retrochiasmal fibers of the projection of one eye are differentially positioned or injured than the fibers of the opposite eye (Figure 2-19). Anterior Optic Tract (1). Extreme incongruity, especially near the vertical meridian, is usually related to the difference in the projections of fibers from the two eyes in the anterior portion of the optic tracts. The lesions are usually caused by tumors of adjacent structures. The field defects may have steeply sloping isopter borders as well as incongruity. Optic Radiation (2). After the fibers of the optic radiation leave the lateral geniculate body through the optic peduncle of the internal capsule (A), they enter the external sagittal striatum to form the midportion of the optic radiation, which lies in the temporal and temporoparietal lobes (B). Incongruous defects may occur in the temporal lobe, because fibers serving homologous points in the visual fields of the two eyes are not adjacent. The blood supply to the temporal lobe comes mostly from branches of the middle cerebral artery. The fibers of the optic radiation become more compact and the arrangement more homologous as the fibers pass into the posterior portion of the optic radiation in the parietal and parieto-occipital lobes (C). Incongruity and sloping margins are less frequent here. Field defects may result from tumors or middle cerebral artery disease. Because the more medial aspects of the parietal lobe are supplied

Visual Fields

,

66

Figure 2-19. Incongruous homonymous hemianopia.

by branches of the anterior cerebral artery, optokinetic nystagmus (OKN), which is generated in the parietal lobe, can be normal and symmetric after occlusions of the posterior or middle cerebral artery. This gives rise to the observation (“Cogan’s rule”) that a hemianopia with asymmetric OKN suggests a tumor; a hemianopia with symmetric OKN suggests a vascular lesion. Occipital Lobe. Incongruous field defects rarely occur in the calcarine cortex, except for visual field defects involving or sparing the uniocular temporal crescent of the visual field. Incongruity suggests a lesion elsewhere.

2-4-1-7 Incomplete: Unilateral Sparing of Temporal Crescent. There is a complete homonymous hemianopia sparing the most peripheral visual field of one eye (Figure 2-20). Posterior Calcarine Cortex. The peripheral temporal visual field of the contralateral eye is represented in the cortex above and below the most anterior portion of the calcarine sulcus anterior to the junction of the parieto-occipital fissure (A) with the calcarine fissure (B). The lesion responsible for a homonymous hemianopia with temporal crescent sparing would anatomically spare the anterior portions of the lingual gyrus (1C) and the precuneus (ID) while injuring the cuneus (1E) and the posterior portion of the lingual gyrus (1C).

67

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-20. Homonymous hemianopia with temporal crescent sparing.

2-4-1-8 Incomplete: Unilateral Defect of Temporal Crescent. In its purest sense, this is a monocular field defect involving the temporal visual field of one eye and could even mimic retinoschisis of the nasal retina. However, many patients with such defects have had or will have other homonymous hemianopic field defects that are binocular (Figure 2-21). Anterior Calcarine Cortex (1). A lesion of the anterior calcarine cortex may produce a peripheral temporal field defect corresponding to the nasal retina of the contralateral eye. The ipsilateral eye visual field may later become involved as the lesion extends posteriorly. At this time, the true homonymous hemianopic nature of the problem will appear. In the optic radiation, the fibers that will terminate in the anterior calcarine cortex are situated at the upper and lower edges of the superior (A) and inferior (B) bundles, respectively. They terminate in the precuneus (C) superiorly and in the lingual gyrus (D) inferiorly. The arterial supply comes from the posterior cerebral arteries. 2-4-2 Bitemporal Hemianopias 2-4-2-1 Complete. A bitemporal hemianopia is a temporal field defect of each eye (Figure 2-22). Decussating Fibers of Optic Chiasm (1). The optic chiasm is within the subarachnoid basal cistern. It is surrounded by cerebrospinal fluid. At its posterior margin, the

Visual Fields

,

68

Figure 2-21. Homonymous hemianopia minus (a unilateral temporal crescent defect).

optic chiasm indents the third ventricle (A). Anteriorly, it is continuous with the optic nerves that dive toward the optic canals at an angle of approximately 45° from the plane of the diaphragm of the sella. If the dive angle of the optic nerve is more vertical, the optic chiasm is said to be prefixed with regard to the sella; if the angle is more horizontal, the optic chiasm is described as postfixed. One may think of the brain as rotated anteriorly in the former case, while it is rotated posteriorly toward the tentorium cerebelli in the latter case. Because the optic chiasm is usually suspended 1 cm above the posterior two thirds of the diaphragm of the sella turcica, most pituitary tumors (B) must grow about 2 cm above the sella turcica before causing a field defect. Prefixed optic chiasms are associated with optic tract field defects when there is extrasellar extension of pituitary tumors, while postfixed optic chiasms are associated with optic nerve field defects. Miller10 provided a critique of nerve fiber schemata of the optic chiasm. Because the inferior fibers serve the superior fields, a bitemporal superior quadrantanopia precedes a complete bitemporal hemianopia. The progression is clockwise in the right eye and counterclockwise in the left eye. Eventually, the defect may involve uncrossed fibers at the lateral aspect of the optic chiasm. A complete bitemporal hemianopia gives no clue as to which surface of the optic chiasm was involved by extrinsic processes first. A bitemporal hemianopia progressing from the inferior to the superior temporal quadrants suggests initial

69

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-22. Complete bitemporal hemianopia. (A) Third ventricle. The arrows represent the progression of the visual field defect associated with the suprasellar extension of a pituitary adenoma (B) from a superior bitemporal quadrantanopia to a complete bitemporal hemianopia.

compression of the superior retinal fibers, as may occur with a craniopharyngioma. Chiasmal visual field defects are caused by tumors, demyelinating disorders, trauma, and other conditions.

2-4-2-2 With Central Depression, Scotomatous. In this pattern, the field defect is similar to a bilateral centrocecal scotoma but does not involve any aspect of the visual field nasal to fixation. The visual acuity may be normal (Figure 2-23A). Nasal Portions of Both Maculas (1). Binocular macular lesions are unlikely to have a sharp vertical border through the point of fixation. Optic Chiasm Fibers from the nasal portions of both eyes cross throughout the optic chiasm (2). Incomplete, diffuse, or posterior lesions may most affect the smallerdiameter macular fibers, causing incomplete central hemianopic depressions, which are best detected with minimal-stimulus contrast targets. The sharp vertical line through fixation is diagnostic. Since the macular fibers are especially dense in the posterior region of the optic chiasm, this field defect is even more apt to happen when pituitary tumors occur with a prefixed optic chiasm or when the third ventricle is enlarged, as in hydrocephalus. When the optic chiasm is prefixed, the intracranial portions of the

70

Visual Fields B

,

A

Figure 2-23. (A) Bitemporal hemianopia with central depression, scotomatous. (B) Relationship of the optic chiasm to the turcica, lateral view.

optic nerves are relatively short and the optic chiasm lies closer to the diaphragm of the sella turcica: The optic chiasm is said to be prefixed if the anterior portion of the optic chiasm covers the tuberculum sellae (A). This occurs in 5% of patients. The optic chiasm is above the center of the sella turcica (B) in 12% of patients. In 79% of patients, it is above the posterior half of the sella turcica. In 4% of patients, the intracranial portions of the optic nerves are relatively long and the optic chiasm is above the dorsum sella (C). In such patients, the optic chiasm is said to be postfixed, and pituitary adenomas must grow a considerable distance beyond the diaphragm of the sella turcica to involve the optic chiasm (Figure 2-23B).

2-4-3 Binasal Field Defects 2-4-3-1 Complete. In this pattern, the field defect involves the entire area nasal to fixation in each eye (Figure 2-24). Nondecussating Fibers of Optic Chiasm (1). The optic nerve fibers serving the temporal retina and the nasal visual field of each eye do not cross and are in the lateral portion of the optic chiasm. The sharp vertical division of the field at the fovea implicates the nondecussating fibers of the optic chiasm. Binasal hemianopiç visual field defects are rare and may be the result of fusiform, dolichoectatic, sclerotic, or aneurysmal enlargement of one or both internal carotid arteries (A), which lie lateral to the optic chiasm. The optic chiasm becomes trapped between the dilated

71

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-24. Complete binasal hemianopia.

internal carotid and anterior cerebral arteries lying in the anterior optic chiasmal notch. These are rarely pure nasal hemianopias. Usually, the defects blend with those of one optic nerve (B) or of one optic tract (C) (junctional scotoma).

2-4-3-2 Incomplete. These field defects involve a portion of the nasal visual field in each eye. It is rare for them to be completely symmetric (Figure 2-25). Bitemporal Retinas (1). The nasal fields may be depressed, often asymmetrically, in disorders of the peripheral retina. Optic Nerve Head (2). Glaucomatous visual fields involve the nasal portions of the visual fields first. Optic Nerve (3). Compression of the superolateral surfaces of the optic nerves against the anterior communicating artery and the A-1 segments of the anterior cerebral arteries (A) may occur when there are aneurysms or rarely as a result of expanding sellar tumors. More often than not, these vessels are stretched over the optic chiasm, rather than over the optic nerves. Optic Chiasm (4). Binasal field defects resulting from optic chiasmal lesions are much rarer than binasal defects resulting from retinal or optic nerve head disorders (see Section 2-4-3-1). Optochiasmic arachnoiditis may also cause a binasal hemianopia. Dilation of the third ventricle (B) may be accompanied by stretching of the lateral aspect of the optic chiasm. This may explain the binasal hemianopia seen occasionally in patients with hydrocephalus.

Visual Fields

,

72

Figure 2-25. Incomplete binasal hemianopia.

2-4-4 Altitudinal Field Defects 2-4-4-1 Noncongruous Binocular and Monocular. Because noncongruous binocular altitudinal visual field defects can arise as a result of separate processes in each eye and in the anterior visual pathway, monocular altitudinal visual field defects are also considered in this section. Monocular disease may lead to a symmetric or an asymmetric loss of all or part of the superior and/or inferior visual field of each eye. Retina (1). While the most significant retinal cause of altitudinal field loss is an embolism to the superior or inferior branch retinal artery of one eye, there is little likelihood of almost symmetric bilateral lesions without other evidence of stroke. The most common retinal causes of noncongruous altitudinal defects are disorders of the retina such as bilateral rhegmatogenous detachments or bilateral exudative detachments as seen in Harada disease. Disorders that do not directly involve the inner retinal layers may have field defects with sloped margins. Altitudinal visual field defects with step margins may result from the absence of retina, as occurs in inferior colobomas (Figure 2-26). Optic Nerve Head (2). The short posterior ciliary arteries supply most of the optic nerve head. Occlusion of the short posterior ciliary arteries can cause altitudinal field defects in each eye and may be the presenting symptom of anterior ischemic

73

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-26. Noncongruous altitudinal hemianopia.

optic neuropathy. Bilateral optic nerve head colobomas cause superior altitudinal field defects. Far-advanced glaucoma could mimic bilateral altitudinal field defects if each eye had a large nasal step and breakthrough from the blind spot to the temporal periphery. An apparent altitudinal defect found on a central visual field examination should be further examined with peripheral testing to be certain that it is not a nerve fiber bundle defect breaking through to the periphery. Automated perimetry incorporates altitudinal hemifield comparisons to test for threshold asymmetries above and below the horizontal meridian as well as losses that are symmetric around the horizontal meridian. Bilateral optic nerve head drusen may also cause altitudinal visual field defects. Optic Nerve (3). While most adult-acquired altitudinal visual field defects were thought to result from anterior ischemic optic neuropathy and other vascular causes, it is now known that approximately 20% of persons presenting with optic neuritis have altitudinal or other nerve fiber bundle defects.11 Contusions of both optic nerves of the optic chiasm produce altitudinal visual field defects. Intracranial tumors and aneurysms of the anterior cerebral and communicating arteries may compress the upper surfaces of both optic nerves.

2-4-4-2 Congruous. Altitudinal hemianopias may symmetrically involve either the superior or the inferior visual fields, with or without macular sparing

Visual Fields

,

74

Figure 2-27. Congruous altitudinal hemianopia.

(Figure 2-27). Clinical observations in humans and neurophysiologic studies in monkeys have revealed distinct symptom patterns depending on whether the lesion involves the inferior portions of the occipital lobes, the adjacent temporal lobes or the superior portions of the occipital lobes, and the adjacent parietal lobes. In addition to causing bilateral superior altitudinal hemianopia, bilateral infarcts of the inferior occipital and medial temporal lobes (including the fusiform and lingual gyri) can be associated with dyschromatopsia or achromatopsia of the remaining inferior hemifields so that the patient may demonstrate impaired color vision. The color vision function corresponds to monkey cortical area V4 (see Figure 2-6). Lesions in the inferomedial temporal lobes may also produce impairment of contrast sensitivity and form vision.12,13 A unilateral lesion of one inferior occipital lobe and adjacent temporal lobe may produce a contralateral superior quadrantanopia and a contralateral inferior quadrant dyschromatopsia. Conversely, patients with inferior altitudinal visual field defects resulting from bilateral superior occipital and adjacent parietal lobe lesions may also have simultanagnosia, manifested by problems in visual/spatial processing and attention.14 Bilateral Superior or Inferior Calcarine Cortex (1). Occlusion of the calcarine branches (A) of both posterior cerebral arteries can produce a congruous altitudinal

,

Anatomic Basis and Differential Diagnosis of Field Defects

75

hemianopia if the regions below or above both calcarine sulci (B) are infarcted. Disease of both middle cerebral arteries (C) may involve both superior lips of the calcarine cortex. Congruous inferior altitudinal field defects may follow trauma superior to the calcarine sulcus. Posttraumatic superior altitudinal field defects following injury to the inferior calcarine cortex may be less likely to occur because patients may not survive if the injury includes laceration of the dural sinuses, including the straight sinus. The straight sinus (D), which is within the tentorium cerebelli, joins the transverse sinuses (E) (connecting the sigmoid sinuses and the internal jugular veins) and the superior sagittal sinus (which runs with the falx cerebri) at the confluence of the sinuses (F) at the internal protuberance of the occipital bone.

2-4-5 Quadrantanopias 2-4-5-1 Superior Homonymous, Incomplete. In this pattern, there is a loss of the congruous superior right or left visual field of each eye. The example shown is of an incomplete field defect, sparing some of the affected quadrant (Figure 2-28). Temporal Lobe (1). The optic radiation arises from the lateral geniculate body (A) as the optic peduncle and then passes anteriorly into the temporal lobe. Fibers serving the inferior peripheral hemiretina (B) go most anteriorly to the tip of the inferior horn of the lateral ventricle (C). The fibers are known as Meyer’s loop (1).

Figure 2-28. Incomplete superior homonymous quadrantanopia.

,

76

Visual Fields

The fields shown will result if the temporal lobe is amputated more than 4 cm from the tip. The defects are usually congruous and have a sharply defined vertical meridian. The horizontal meridian may be included, with larger lesions yielding a congruous superior quadrantanopia with or without macular sparing. If more than 8 cm of the temporal lobe is amputated, a homonymous hemianopia will result. Tumors usually produce greater incongruity and sloping of isopters than do vascular lesions of the anterior portion of the retrochiasmal visual pathway. Lingual Gyrus of Occipital Lobe (2). Occipital lesions that produce a pure quadrantanopia do not usually result from pure involvement of the primary visual cortex (area 17), as too long a strip of tissue along the lip of the calcarine cortex would have to be injured. It has recently been demonstrated that the visual association cortex (areas 18 and 19) maps the peripheral visual field along the horizontal meridian in a compact region near the posterior pole of the occipital lobe. Thus, a quadrantanopia can be thought of as resulting from the loss of fibers going to or from the primary visual cortex.

2-4-5-2 Inferior Homonymous, Complete. In this pattern, there is loss of the overlapping inferior right or left visual field of each eye (Figure 2-29). Parietal Lobe (1). The optic radiation passes posterolaterally from the lateral geniculate body (A) into the parietal lobe. After initial dispersion, the fibers that

Figure 2-29. Complete inferior homonymous quadrantanopia.

Anatomic Basis and Differential Diagnosis of Field Defects

77

,

project from the superior retinal quadrants (B) collect into a compact bundle sooner than the temporal lobe fibers serving the inferior retinal quadrants. It is for this reason that complete quadrantic loss occurs more frequently in parietal lobe lesions than in temporal lobe lesions. Cuneate Gyrus of Occipital Lobe (2). An inferior quadrantanopia of occipital origin involves the visual association area, not just the primary cortex (see earlier section “Lingual Gyrus of Occipital Lobe”).

2-4-6 Bilateral Central Field Defects 2-4-6-1 Scotoma or Depression. The central visual fields of both eyes are abnormal in this pattern of field defect and are usually accompanied by decreased visual acuity (Figure 2-30). Media Opacities (1). See Section 2-3-1-3. Macula Denser scotomas occur with lesions that involve the inner retinal layers (2), in contrast to lesser depressions with outer retinal and choroidal disorders in the macular regions. Nutritional and toxic amblyopias may involve ganglion cells originating in the macula.

Figure 2-30. Bilateral central scotomas or depressions.

,

78

Visual Fields

Optic Nerve (3). The maculopapillary bundle is in the temporal quadrant of the optic nerve, immediately posterior to the optic nerve head. Axial neuritis results in dense central or centrocecal scotomas. Optic Nerve, Diffuse (4). After passing 1 cm posterior to the optic nerve head, many of the fibers of the maculopapillary bundle reach the center of the optic nerve while other macular fibers distribute to the peripheral quadrants of the optic nerve. Calcarine Cortex (5). The macula is represented on the posterior portion of the calcarine sulcus (6) and may extend onto the posterolateral portion of the occipital lobes. This is important in the analysis of trauma to the occipital lobes. Contusion of the superior aspect of both occipital lobes may result in an inferior altitudinal congruous field defect (see Figure 2-27). 2-4-7 Bilateral Peripheral Field Defects 2-4-7-1 Generalized Depression or Peripheral Contraction. In this pattern, the outer isopters of the visual field collapse or contract (Figure 2-31). Because perimetry is a subjective examination, contraction can result from patient or examiner factors. In addition to the lesions noted below, deliberate malingering, too rapid target movement, slow response times, inappropriate contrast of stimulus-target/ background ratio, and an inadequately dilated pupil can all contract the outer isopters.

Figure 2-31. Peripheral contraction.

,

Anatomic Basis and Differential Diagnosis of Field Defects

79

Media Opacities (1). Corneal scars, cataracts, Large Vitreous Hemorrhage. Peripheral Retina (2). Quinine, salicylates, thioridazine hydrochloride, and carbon monoxide poisoning are among the causes of contraction of the visual fields. Optic Nerve Head (3). Irregular, progressive constriction may occur. Sector sparing with sector and optic nerve defects may help determine the diagnosis. Glaucoma and optic nerve head swelling may lead to peripheral depressions of the visual fields but usually in combination with other more characteristic field defects. Optic Nerve (4). Peripheral optic neuritis may occur with tertiary syphilis. When peripheral constriction is combined with central depression, retrobulbar neuritis and optic atrophy from any cause should be considered. Anterior Calcarine Sulcus (5). To cause a binocular peripheral constriction, paired visual cortical lesions would need to injure the striate cortex or pathways related with the most anterior portions of both calcarine sulci. This portion of the striate cortex serves the visual field of the unpaired temporal crescent visual field of the contralateral eye. Limited extension of the lesion posterior to the junction of the calcarine with parieto-occipital sulcus would involve the nasal field of each ipsilateral eye. 2-4-8 Bilateral Checkerboard Scotomas. This pattern of visual loss consists of two congruous hemianopic scotomas, such as a right superior and left inferior field defect (Figure 2-32).

Figure 2-32. Bilateral checkerboard scotomas.

80

Visual Fields

,

Posterior Cerebral Arteries (1). Occlusions of the posterior cerebral arteries may occur simultaneously in severe systemic hypotension or in basilar-vertebral artery disease (“top-of-the-basilar-artery” syndrome); each posterior cerebral artery may be occluded at a different time. Identification of the old or new lesion depends on the history or on other neurologic findings. These are called “checkerboard scotomas” when the hemianopic visual fields are not symmetrically affected. 2-4-9 Bilateral Homonymous Hemianopias. This pattern of field defect consists of complete or incomplete homonymous hemianopias involving both the right and the left visual fields (Figure 2-33). Posterior Cerebral Arteries (1). The area of damage from occlusion of these arteries is usually more extensive than that described for bilateral checkerboard scotomas. 2-5 JUNCTIONAL FIELD DEFECTS 2-5-1 Complete Monocular Plus Incomplete Contralateral Ocular. This pattern of field loss involves the entire visual field of one eye and a portion (usually superior temporal) of the visual field of the opposite eye (Figure 2-34).

Figure 2-33. Bilateral homonymous hemianopias.

81

,

Anatomic Basis and Differential Diagnosis of Field Defects

Figure 2-34. Complete monocular plus an incomplete contralateral ocular (a junctional defect).

Both Eyes or Optic Nerves (1). One reason to perform a complete ocular examination on each patient is to avoid missing a second lesion. However, a complete monocular loss of vision plus an incomplete superior temporal quadrantanopia in the opposite eye with a short vertical meridian passing through fixation is diagnostic of the single lesion described next. One Optic Nerve at Junction of Optic Chiasm (2). Inferior nasal peripheral fibers from the opposite optic nerve have crossed the optic chiasm and are involved while passing by the junction with the contralateral optic nerve into the contralateral optic tract. (This arrangement was formerly known as the anterior knee of Wilbrand, but it has now been shown that the fibers do not actually enter the contralateral optic nerve.) 2-5-2 Homonymous Hemianopia Plus. This field defect includes a (usually) complete and congruous hemianopia with an additional field defect in one eye (Figure 2-35). Junction of Optic Tract and Optic Chiasm (1). A complete right homonymous hemianopia of a left postchiasmal lesion may also involve some of the left eye axons that decussate in the optic chiasm. This adds a left eye temporal field defect. The left superior temporal visual field is spared because the inferonasal fibers serving this field decussate anteriorly within the optic chiasm. The fibers from the superior

Figure 2-35. Homonymous hemianopia plus (a junctional scotoma).

Figure 2-36. Bitemporal hemianopia plus (a junctional scotoma).

82

Anatomic Basis and Differential Diagnosis of Field Defects

83

,

nasal retina of the left eye are injured because they decussate posteriorly, giving rise to the left inferior temporal visual field constriction. (Compare this example with Figure 2-34.)

2-5-3 Bitemporal Hemianopia Plus. In this pattern of field defect, there is a nasal field defect in one eye or both eyes in addition to temporal field defects in both eyes (Figure 2-36). Junction of Optic Chiasm and Optic Nerve (1). A bitemporal hemianopia can still be recognized by the fragment of sharp vertical meridian remaining in the right visual field, but there is also involvement of uncrossed fibers in the optic nerve. This may occur in pituitary tumors if the optic nerves are very long and if the optic chiasm is postfixed with regard to the sella turcica. Sometimes, pituitary tumors progress asymmetrically. Aneurysms of the anterior communicating artery (A) are well situated to produce these fields.

REFERENCES 1. Trobe JD, Glaser JS. The visual fields manual. Gainesville, Fla: Triad Publishing Co; 1983. 2. Horton JC, Hoyt WF. The representation of the visual field in human striate cortex: a revision of the classic Holmes map. Arch Ophthalmol. 1991;109:816–824. 3. Sereno MI, Dale AM, Reppas JB, et al. Borders of multiple visual areas in humans revealed by functional magnetic resonance imaging. Science. 1995;268:889–893. 4. Horton JC, Hoyt WF. Quadrantic visual field defects: a hallmark of lesions in extrastriate (V2/V3) cortex. Brain. 1991;114:1703–1718. 5. Quigley HA, Sommer A. How to use nerve fiber layer examination in the management of glaucoma. Trans Am Ophthalmol Soc. 1987;85:254–268. 6. Quigley HA, Addicks EM. Regional differences in the structure of the lamina cribrosa and their relation to glaucomatous optic nerve damage. Arch Ophthalmol. 1981;99: 137–143. 7. Bischoff P, Lang J, Huber A: Macular sparing as a perimetric artifact. Am J Ophthalmol. 1995;119:72–80. 8. Corbett JJ, Jacobson DM, Mauer RC, Thompson HS: Enlargement of the blind spot caused by papilledema. Am J Ophthalmol. 1988;105:261–265. 9. Frisén L: Quadruple sectoranopia and sectoral optic atrophy: a syndrome of the distal anterior choroidal artery. J Neurol Neurosurg Psychiatry. 1979;42:590–594. 10. Miller NR: Walsh and Hoyt’s Clinical neuro-ophthalmology. 4th ed. Baltimore: Williams & Wilkins; 1982:3–173. 11. Keltner JL, Johnson CA, Spurr JO, Beck RW: Baseline visual field profile of optic neuritis: the experience of the Optic Neuritis Treatment Trial. Arch Ophthalmol. 1993;111:231–234. 12. Newman RP, Kinkel WR, Jacob L: Altitudinal hemianopia caused by occipital infarcts: clinical and computerized tomographic corrections. Arch Neurol. 1984;41:413–418. 13. Schiller P: The effects of V4 and middle temporal (MT) area lesions on visual performance in the rhesus monkey. Visual Neurosci 1993;10:717–746. 14. Rizzo M, Robin DA: Simultanagnosia: a defect of sustained attention yields insights on visual information processing. Neurology. 1990;40:447–455.

,

This page intentionally left blank

3 Essentials of Automated Perimetry GEORGE SHAFRANOV, MD

3-1 INTRODUCTION Standard automated perimetry is a standard method of measuring peripheral visual function.1 Automated static perimetry gained wide acceptance among clinicians due to the test’s high reproducibility and standardization and ability to store, exchange, and statistically analyze digital data. Advances in the computerized visual field assessment have contributed to our understanding of the role that field of vision plays in clinical evaluation and management of patients. The Humphrey Visual Field Analyzer/HFA II-i is the most commonly used automated perimeter in the United States, and the examples in this chapter have been obtained with this instrument.

3-2 HISTORICAL OVERVIEW Aubert and Förster2 in the 1860s developed the arc perimeter, which led to the mapping of peripheral neurologic visual field abnormalities and advanced glaucomatous field defects. Analysis of the central visual field was not seen as clinically important by most clinicians until 1889, when Bjerrum2 described a detected arcuate paracentral scotoma. Later, Traquair3 further contributed to kinetic perimetry on the tangent screen. In 1893, Groenouw proposed the term “isopter” for lines with the same sensitivity on a perimetry chart. Rønne further developed kinetic isopter perimetry in 1909 and described the nasal step in glaucoma.

85

,

86

Visual Fields

Although the first bowl perimeter was introduced in 1872 by Scherk, due to problems with achieving even illumination on the screen, it did not become popular. The version of the bowl perimeter introduced by Goldmann in 1945 became widely accepted and is a significant contribution to clinical perimetry. The Goldmann perimeter incorporated a projected stimulus on an illuminated bowl, with standardization of background illumination as well as size and intensity of the stimulus, and allowed effective use of both static and kinetic techniques. For these reasons, the Goldmann instrument has remained the clinical standard throughout the world until widespread acceptance of automated perimetry. Harms and Aulhorn4 later designed the Tübingen perimeter with a bowl-type screen exclusively for the measurement of static threshold fields, using stationary test objects with variable light intensity.5 While excellent threshold measurements were possible with this instrument, the time and effort involved in such measurements prevented this perimeter from becoming widely used. Automated perimetry has progressed rapidly over the past few decades, largely to the credit of Fankhauser,6,7 Heijl and Krakau,8 Flammer et al.,9 and others. Automated perimeters have substantially enhanced the clinical practice due to accurate, standardized measurements of the visual function. Most of these instruments use computerized static threshold algorithms, which have proved invaluable for clinical research and patient care, and are very familiar to all ophthalmologists. Despite the complex statistical algorithms used by visual field analyzers, standardized quantitative results allow easy and practical application of statistical methods to measure early functional loss.8-11 The high sensitivity of the methods has required attention to the multiple psychological and physiologic variables that may affect measured thresholds. Automated perimetry has improved standardization by reducing variability in the examination technique. Of course, the test still depends on the reliability of the patient’s responses and may be affected by optic, neural, and psychological factors. Static automated threshold perimetry is one of the most important tests in the care of glaucoma patients. Detailed measurements of the visual field can be obtained, and some uncertainties regarding diagnostic and therapeutic decision making in glaucoma can be diminished. On the other hand, the large volume of data introduces uncertainties. Differentiating long-term fluctuation (LTF) from progressive loss remains one of the greatest clinical challenges in visual field interpretation.11 This chapter is designed to help in developing skills in the selection of automated perimetric tests and to afford familiarity with the various printouts used by the Humphrey perimeter. Chapter 4 [cross-check] is dedicated to the use of automated perimetry in the care of patients with glaucoma.

3-3 PRINCIPLES OF FIELD TESTING The human visual system has a relatively poor ability to estimate absolute magnitudes of light, mostly to the exceptional adaptation of that system. The human visual system has, however, a remarkable ability to perceive contrast (relative magnitudes

,

Essentials of Automated Perimetry

87

of light). Thus, it is the differential light sensitivity of a stimulus against a constantilluminated background that is measured with static perimetry. Light can be measured using various units. A point source with an output of 1 candela emits a total of 4 Π lumens. The illuminance on a surface is the number of lumens per square meter incident to that surface. Some of this energy is absorbed and transformed to another form of energy (heat), and another part is reflected and emitted as light, known as luminance. The apostilb (asb) is the unit of measurement for the luminance of a perfectly diffusing surface that is emitting or reflecting 1 lumen per square meter (m2). Most commercially available perimeters generate light of varying intensities by interposing neutral density filters, graded in decibels (dB), over a maximally emitting bulb. Retinal locations of reduced sensitivity require brighter stimuli to reach threshold, represented by lower decibel values. Similarly, higher decibel threshold values represent more sensitive retinal locations (Figure 3-1). Each decibel equals 1/10th of 1 log unit. Thus, 10 dB equals 1 log unit or a 10-fold change in intensity, and 30 dB equals 3 log units or a 1000-fold change in intensity. The maximum bulb intensities vary; Goldmann and Octopus perimeters generate a maximum stimulus luminance (0 dB) of 1000 asb, while the Humphrey perimeter uses a 10,000-asb bulb (0 dB). Background luminance also varies; Humphrey Visual Field Analyzer uses 31.5 asb. The hill of vision may be mapped by using moving kinetic or stationary static stimuli. The visual threshold at a specific location in the retina is defined as the luminance at which 50% of the stimulus presentations are identified by the patient as seen. A patient undergoing a threshold examination may see only half of the presented stimuli. This can be a source of frustration to patients, who may feel they have performed poorly. Understanding the phenomenon requires knowing the bracketing strategy used to make threshold measurements at each location of

Figure 3-1. Graytone symbols used in the older version of Humphrey Visual

Field Analyzer. As retinal sensitivity decreases (measured in decibels [dB] and depicted by progressively darker graytone symbols), the luminance (in apostilbs [asb]) increases to a maximum value of 10,000. Neutral density filters (graded in dB) are used to reduce the luminance of the 10,000-asb bulb.

,

88

Visual Fields

the visual field. Full threshold strategy algorithms use a double crossing of the threshold. For instance, if the initial stimulus is subthreshold (not seen), intensity is increased in steps of 4 dB until the patient responds with a “yes” (seen). The stimulus intensity is then decreased in steps of 2 dB until the patient does not respond (not seen). The visual threshold is thereby crossed twice. Suprathreshold stimuli are brighter than the expected thresholds at given locations in the retina. The Humphrey Visual Field Analyzer reports threshold values as the last seen stimulus using the 4-2 strategy (Figure 3-2). If the initial stimulus is suprathreshold, stimulus intensity is decreased by 4-dB steps until the threshold is crossed, then increased in 2-dB steps (the threshold again is doubly crossed). The choice of the initial stimulus luminance is explained below. With this strategy, accurate threshold estimates are achieved by presenting on average approximately five stimuli per test location. Stimulus presentations are not performed sequentially at a single location but are moved randomly throughout the entire visual field. This discourages cheating, because the patient does not know where to expect the next stimulus presentation.

3-3-1 Kinetic Perimetry. Kinetic perimetry uses a stimulus of constant size and intensity that moves from nonseeing to seeing areas of the visual field (Figure 3-3), that is, from the periphery toward fixation when outlining an isopter and from the center of the blind spot or a scotoma. Kinetic techniques are not very accurate for the examination of relatively flat areas of the visual field, requiring a sloping hill of vision in the tested area. Enlargement of the blind spot and Seidel’s scotoma (an arcuate-shaped elongation of the blind spot) have been described as early defects

Figure 3-2. The bracketing strategy used by the Humphrey Visual Field Analyzer to measure the visual threshold. The threshold is initially approached by stimuli at 4-dB steps. When the stimulus is seen, the direction is reversed and the threshold is crossed in 2-dB steps. Thus, each threshold is crossed twice with a measurement resolution of ±1 dB.

89

,

Essentials of Automated Perimetry

Figure 3-3. Kinetic perimetry uses a stimulus of specific size and intensity that is

moved from nonseeing to seeing areas of the visual field. A distinct end point requires a sloping hill of vision.

in glaucoma; they may in part be artifactual because slightly suprathreshold kinetic stimuli are used to explore the relatively flat area of the field just superior to the blind spot, and the technique of kinetic perimetry is not particularly sensitive to early localized depressions (Figure 3-4). Although computerized perimetry is presently somewhat limited to static algorithms, a number of computerized perimeters can use kinetic techniques

Figure 3-4. Localized, shallow depressions of the visual field can be missed with

kinetic techniques, particularly when they occur in a relatively flat area of the visual field.

,

90

Visual Fields

as well. Computerized kinetic perimetry was historically thought to be less reproducible than automated static perimetry because of the complex algorithms required. Nevertheless, recent studies suggest the opposite,12 and the reliability and reproducibility of the kinetic measurements have yet to be established. The optimal rate of movement of the target is about 4° per second,13 but a slower velocity of 2° per second may provide more reproducible results in some patients.14 A combination of central static threshold perimetry and peripheral kinetic perimetry has been used in glaucoma and neuro-ophthalmology patients.15 The optimal technique for examining the peripheral visual field that provides the most information in the shortest period of time has not yet been determined.

3-3-2 Static Perimetry. Standard automated static perimeter typically presents a projected stationary stimulus of known size, intensity, time, and location against a standard white background, also with known brightness level (Figure 3-5). Automated techniques of stimulus presentation are particularly suited to static measurements because the computer algorithms are relatively straightforward. Suprathreshold screening techniques may be used to make qualitative estimates of the visual field. Threshold measurements are required to obtain the quantitative data needed for the early diagnosis and careful follow-up of glaucoma patients. 3-3-2-1 Suprathreshold Techniques. Suprathreshold static perimetry is a technique in which a bright stimulus with an intensity that is above the anticipated threshold for the retinal area being tested and is expected to be seen in all parts of normal visual field. The locations at which the patient fails to recognize the target are noted as visual field defects. It is a quick way of detecting areas of blindness, usually within the central visual field. Stimulus intensity for suprathreshold perimetry may be constant over the entire field (Figure 3-6) or may correspond to the slope

Figure 3-5. Static techniques use a stationary stimulus of variable intensity to measure the visual threshold at a specific location in the visual field.

91

,

Essentials of Automated Perimetry

Figure 3-6. Constant stimulus intensity may be used with suprathreshold

techniques to obtain qualitative information.

of the visual hill (threshold-related, Figure 3-7). Suprathreshold static targets are presented for a longer time, usually 0.5 to 1 second. Some automated perimeters predict the slope of the visual hill based on several initial threshold measurements in the peripheral visual field, a solution that is not always satisfactory. While this type of screening is rapid, it does not always detect early abnormalities with localized partial depressions or increased variability of responses in localized areas.

Figure 3-7. Suprathreshold test may use a stimulus that conforms to the slope

of the visual hill. This is sometimes called a threshold-related test and can be used to obtain qualitative information about the visual field.

92

Visual Fields

,

The static suprathreshold strategy is used as a screening test, to determine if visual function falls within or outside the normal range.8

3-3-2-2 Threshold Techniques. Computerized perimetry allows the performance of static threshold measurements in a short period of time under standardized conditions. Static threshold measurements are relatively sensitive to shallow depressions of the visual field when the locations of the stimuli are close together (Figure 3-8). This high sensitivity has led to an increased detection rate of early glaucomatous defects compared with manual techniques and has enhanced the ability to meaningfully compare successive visual field examinations. Manual perimeters can be used for quantitative threshold measurements but only by highly trained personnel. 3-3-3 Frequency-of-Seeing Curves and Fluctuations. Perimetry is a subjective psychophysical test requiring the patient’s cooperation, effort, and communication. As in any diagnostic test, the response to a specific question has an associated error. The threshold of the differential light sensitivity represents the brightness of the stimulus so dim that is seen only on 50% of repeated presentations. The frequency-of-seeing curve is a useful to emphasize the importance of probability in estimating a location’s threshold (Figure 3-9) and may be generated by repeated testing at a single location. Figure 3-9 indicates that these probabilities can never be 0% or 100% because of the influence of false-positive and false-negative responses, respectively. The slope of the curve is correlated to threshold deviation from age-appropriate normal values at a particular location. Thus, areas of high retinal sensitivity tend to test with high reproducibility, while locations with abnormally reduced sensitivity have a shallower

Figure 3-8. Static measurements of the threshold are relatively sensitive to

localized, relative depressions of the visual field as long as the stimuli are spaced closely enough.

Essentials of Automated Perimetry

93

,

Figure 3-9. Frequency-of-seeing curve. As

the intensity of the stimulus increases (abscissa), the probability of seeing the stimulus increases (ordinate). The threshold is defined as the intensity that is seen 50% of the times it is presented. This implies that thresholds can only be estimated, not measured. The probability of seeing the stimulus is never 0% and never 100% because of false-positive or false-negative answers.

slope of the frequency-of-seeing curve, which is associated with greater uncertainty (Figure 3-10).16,17 Fluctuations in measurements of a physiologic parameter produce variation of a test result. Careful measurements can reveal those fluctuations in visual field thresholds. Bebie et al.18 described several components of this fluctuation: short-term fluctuation (STF) and LTF. STF is the variation of responses that occurs over the performance of a single test and may be caused by a combination of the instability of the threshold being tested and the level of cooperation and attentiveness of the patient.19 STF was calculated for full threshold algorithms (not often used now) of the Humphrey Visual Field Analyzer by measuring threshold values in 10 locations twice during the course of a given test. LTF is the fluctuation between tests, which occurs over days, months, or years. The causes of LTF are not well established. Possible reasons include fluctuations in intraocular pressure and in age and those

Figure 3-10. Normal central regions with high threshold sensitivity tend to have steep frequency-of-seeing curves. Abnormal or peripheral regions with reduced sensitivity demonstrate a broadened curve with greater threshold uncertainty.

Visual Fields

,

94

Figure 3-11. Confidence intervals of 95% of subsequent threshold sensitivities

for test locations with initial sensitivity on the x-axis in a group of clinically stable glaucoma patients. Note the increasing interval with decreasing initial sensitivity. The 5th percentile approaches 0 if initial sensitivity is below 10 dB because of the limitation of an absolute test stimulus.

that occur as the time interval between tests increases. In a group of clinically stable glaucoma patients, LTF was correlated with initial sensitivity and with distance from fixation (Figure 3-11).20 Understanding the LTF helps in comparisons of visual fields for change over time. In the Ocular Hypertension Treatment Study, less than 20% of the initial glaucomatous visual field defects were confirmed on the subsequent testing.21

3-4 TEST SELECTION AND ALGORITHMS Computerized static visual field tests differ for retina locations that are tested and for algorithms used to test those locations. A basic understanding of the test algorithm used by automated perimeters employing threshold strategies is essential to interpreting the results and artifacts seen on visual field printouts. Currently, most commonly used algorithms include Swedish interactive threshold algorithm (SITA) Standard or SITA Fast 24-2 or 30-2 threshold strategies. To be able to reliably compare longitudinal results, the initial algorithm should be used for follow-up tests.

3-4-1 Swedish Interactive Threshold Algorithm (SITA). The relatively new threshold strategy, known as SITA, has become increasingly popular10,22,23 and almost completely replaced full threshold algorithms. SITA is based on the probability

,

Essentials of Automated Perimetry

95

analysis of the patterns of glaucomatous damage and is more time efficient than full threshold strategies, utilizing standard 24-2 or 30-2 patterns to assess the visual field. It significantly minimizes test time without reduction of data quality. Two versions of SITA are currently available: SITA Standard and SITA Fast. SITA Standard takes approximately half the time to complete compared with the full threshold program. SITA Fast takes about half the time of the FASTPAC algorithm and was found to be useful in pediatric population,24 although SITA Fast has somewhat increased variability compared with SITA Standard.25 SITA requires significant computer power, and therefore it is available only on the newer Humphrey Visual Field Analyzers. SITA is based on visual field modeling that uses frequency-of-seeing curves for glaucoma and healthy patients. During the SITA test, a computer also produces an “information index,” which discontinues testing when threshold reaches a preselected level. The SITA algorithm also makes individual adjustments to patient response time, allowing the patient to be in charge of the test. After measurements are complete, the program performs additional calculation of all thresholds measured and produces estimates of false-positive and false-negative errors, displaying results as percentages. Average time reduction by SITA Standard depended on the severity of glaucomatous stage. No significant time difference exists for advanced glaucoma fields, whereas normal fields using SITA are performed in half of the time of full threshold strategy. The reduction of test time reduces the patient fatigue, allowing for more frequent visual field examinations and subsequent early detection of early glaucoma or progressing visual field damage.26

3-4-2 Foveal Threshold. Measurement of the foveal sensitivity is an option that, if selected, occurs at the very beginning of the test. This option should generally be left on, as it takes very few stimulus presentations and provides information about a very valuable portion of the visual field. The patient is asked to maintain gaze on an illuminated diamond that is projected inferior to the standard central fixation target used throughout the remainder of the test. The initial stimulus intensity is 30 dB, and the regular 4-2 bracketing strategy is used to determine foveal sensitivity. Once this portion of the test is completed, the fixation diamond is removed and the patient is asked to fixate on the central target. 3-4-3 Initial Values. Important time savers are used to reduce the numbers of stimuli necessary to estimate the threshold level and somewhat shorten the full threshold test. Starting points for threshold determinations usually depend on results from already tested primary locations, because a location’s threshold result is statistically correlated with its neighboring location’s threshold value. The Humphrey Visual Field Analyzer initially tests four seed locations, one per quadrant located 9° from the horizontal and vertical meridians. The initial stimulus intensity at these four seed locations is 25 dB and the full 4-2 strategy is used. Threshold in each location is measured twice, and the results from these four seed locations are used to determine the starting stimuli in adjacent areas (Figure 3-12). Threshold results from these adjacent areas are, in turn, used to determine their neighbor’s starting locations until the entire test is completed.

Visual Fields

,

96

Figure 3-12. Seed-point and short-term fluctuation locations. The initially thresholded locations are the four primary seed-point locations (circled), one per quadrant 9° from the horizontal and vertical meridians. These four locations are tested twice and used to calculate starting values for threshold measurements in the neighboring threshold. Throughout the course of the fullthreshold test, threshold values in standard locations (circled and squared) are measured twice to calculate the short-term fluctuation. Additional locations (in parentheses) may be thresholded twice if the result of the first threshold differs substantially from age-matched expected values. These additional locations are not used to calculate STF.

Some patients are attentive during the initial four central quadrant location threshold determination process and then rapidly fatigue, with deficient further responses. This results in a characteristic cloverleaf-shaped field (Figure 3-13).

3-4-4 Fixation Monitoring. Throughout the performance of the test, periodic assessment of patient fixation and level of alertness is made to ensure correspondence of projected stimuli to the correct part of the retina. To monitor fixation, the Humphrey Visual Field Analyzer uses the gaze tracker and the Heijl-Krakau blind spot– monitoring technique.27,28 The latter technique assumes that if stimuli projected on the blind spot location (which has a diameter of ≈5-7°) are seen, then the fixation is poor. If patient does not respond to those stimuli, fixation is assumed to be central— which is not always the case. A high number of fixation losses may result from wandering fixation but may also result from a displaced blind spot or from many false-positive responses. High-plus lenses tend to shift the blind spot toward fixation (Figure 3-14), while myopic correction moves the blind spot peripherally (Figure 3-15). These optical effects can be minimized with the use of contact lenses during the test.

Figure 3-13. Cloverleaf field. During the four primary seed-location threshold

process, the patient responded with threshold sensitivities of 25 to 31 db. Throughout the remainder of the test, the patient quickly fatigued or fell asleep, with no additional responses seen through the rest of the test. The falsenegative errors indicate that patient was not responding to the bright light in the original four locations. No response was produced when the blind spot was tested, resulting in the deceptively low rate of fixation losses.

Figure 3-14. High-plus lens artifact. A patient was tested with a +12 spectacle

correction. The high-plus lens shifted the blind spot toward fixation. There also appears to be a peripheral ring scotoma. When the patient was retested with a contact lens, there were no fixation losses and the peripheral scotoma disappeared.

97

Visual Fields

,

98

Figure 3-15. High-minus lens artifact. A highly myopic patient was tested with spectacle correction. The high-minus lenses shifted the blind spot away from fixation.

Gaze tracking is a system that records the patient’s fixation during the test. Deviations from the fixation are displayed on the screen and the printout. The small downward markings on the graph indicate the system’s inability to locate a reflection from the eye, with large downward marking indicating that the patient blinked. The upward markings indicate deviation from the fixation target, with the larger markings showing greater magnitude of the deviation. Another way to reduce the percentage of fixation losses is to instruct the technician to re-plot the blind spot if high fixation losses are detected early in the test.29 The computer then executes a short subprogram that presents densely packed stimuli in the region of the expected blind spot until the actual blind spot is mapped. The technician’s description of patient fixation, based on the observation of the patient’s eye through the video monitor, is also extremely valuable in detecting pseudo–loss of fixation. The absence of a low patient-reliability indicator should not assure the examiner that the test is error-free. Consider the patient who falls asleep at the machine. Clearly, that patient is unlikely to respond to stimuli presented in the blind spot despite poor fixation. False-positive errors are tested by periodic withholding of a stimulus projection, although the faint noise of a stimulus presentation is retained. False-positive responses tend to indicate anxious, “trigger-happy” patients. The rate of falsepositive responses can often be improved if the visual field technician coaches patients to respond only when they are certain that they have seen the stimulus. False-negative errors are tested by projecting a 9-dB suprathreshold stimulus in a region already thresholded. Failure to respond to this markedly suprathreshold stimulus indicates patient fatigue. False-negative errors are less influenced by coaching; however, the visual field technician should ensure that the patient is awake and consider giving the patient a short break. False-negative errors are produced both by patient inattentiveness and by a diseased, easily fatigued visual system. If the threshold results are markedly reduced, the machine may not generate

Essentials of Automated Perimetry

99

,

suprathreshold stimuli. This may lull the inexperienced visual field technician into a false sense of patient responsiveness.

3-4-5 Threshold Testing. Throughout the performance of the Humphrey programs, if a threshold result differs significantly from age-matched normal values, the location will be retested to confirm this deviation. Performance of a complete full threshold program 30-2 in a patient with fairly normal vision required approximately 550 questions, taking about 15 minutes per eye. This was quite fatiguing to the patient, and current SITA algorithms could perform the same test within 4 minutes.

3-5 SINGLE TEST PRINTOUT The single field printout from Programs 30-2 and 24-2 of the Humphrey perimeter contains a large amount of data, with various analyses presented in multiple ways (Figure 3-16). Familiarity with the overall organization and the derivation of the plots and indices greatly facilitates interpretation of the printout, which can be conveniently divided into the following sections: 1. 2. 3. 5. 4. 5. 6. 7. 8. 9.

Test Selection and General Information, located at the top Reliability Indices, located at the top and at the left Numeric Results (dB), located in the second row, at the left Grayscale Results, located in the second row, at the right Total Deviation, a plot located in the third row, at the left Pattern Deviation, a plot located in the third row, in the middle Global Indices, located in the third row, at the right Glaucoma Hemifield Test, located above the Global Indices Probability Symbols and Gaze Graph, located at the bottom of the printout Practice Information, is located on the right side at the bottom of the printout

3-5-1 Test Selection (and General Information). Located at the top of the printout, the general information section displays important data about the individual patient as well as particular test variables. Included here are program name (e.g., Central 24-2 Threshold Test), patient name and patient birth date (removed from the present illustration), stimulus size, background illumination, blind spot check size, threshold strategy (SITA Standard in this example), fixation target type, test time and date, optical correction, pupil diameter (optional), and Snellen acuity (optional). Patient age is calculated from the date of birth and test date information. Many of these variables can significantly affect the raw and calculated data and can be invaluable in interpreting results. For example, a miotic pupil or incorrect refraction can reduce threshold values, while an incorrectly entered birth date will create erroneous age-compared deviations. 3-5-2 Reliability Indices. Located below and at the left of the general information section, the reliability data display the number of fixation losses to characterize stability of the patient’s gaze directed at the fixation target. The false positive errors score is designed to identify the “trigger-happy” patients, who press the response

Figure 3-16. Sample single field printout demonstrating glaucomatous field loss. (A) The important areas of the single field printout are labeled. This patient has a superior arcuate scotoma with a superior fixation loss and a superior nasal step. (B) Because the seventh-most sensitive total deviation value is −1 dB (only slightly depressed), the pattern deviation plot is similar to the total deviation plot.

100

Essentials of Automated Perimetry

101

,

button even when stimulus is not seen. The false-negative errors indicate patient fatigue or inattention. Questions asked can be a rough indicator of the consistency of the patient’s responses. The final items of information displayed in the reliability indices section are test time and the optional foveal sensitivity. If fixation losses reach 20% or if false-positive or false-negative errors reach 33%, a double X is placed next to the percentage to attract the reviewer’s attention. Similarly, the message low patient reliability may be displayed above the reliability indices section.

3-5-3 Numeric Results and Grayscale Results (Raw Data). The two largest images on the printout, located to the right of the reliability indices, are the raw data plots. These images are displayed in numeric and gray tone format, with the actual threshold values in decibels. The testing algorithm would repeat threshold testing in the locations where the initially obtained values significantly deviated from the age-matched normal data. The grayscale plot is extrapolated from the numeric plot, and although it implies uniform sampling of the 30° field, in reality less than 1% of this area is actually tested. The grayscale plot remains useful to alert the examiner to problem areas and is an effective way of demonstrating visual field results to the patient. 3-5-4 Total Deviation. Since the introduction of the Humphrey perimeter, the manufacturer has upgraded the machine with increasingly sophisticated statistical analysis packages. The main package of these, STATPAC, allows comparison of the raw threshold data against age-matched normal values at each location; accompanying probability symbols indicate the significance of any abnormality. These plots are displayed in the lower-left portion of the printout, both numerically and with probability symbols. The P values take into account the wider range of normal values as the distance from fixation increases.30 For example, a value of 23 dB located 9° superiorly and nasally from the fixation (Figure 3-16) demonstrates that this reduced sensitivity is highly significant (P < 5% [.05]) compared with the agematched normal subjects, while the same 23 dB value located 24° superotemporally to fixation may is only significant. The examiner should keep in mind that statistical significance does not always mean clinical significance. 3-5-5 Pattern Deviation. Patterns of visual field loss can be divided into generalized depression, which uniformly affects the entire field by a similar amount, and localized (“scotomatous”) loss, which is frequently more diagnostic. STATPAC modifies the total deviation plots in an attempt to display any superimposed pattern of localized loss that is hidden under generalized depression. This is done by correcting the deviation of the seventh-highest threshold location within the Program 24-2 test grid to zero deviation and “adjusting” the entire field by that value. The threshold value of this seventh-most elevated location has been termed the general height value, although it is not routinely displayed on the printout. In the situation when the entire visual field is diffusely reduced in sensitivity by 7 dB, with an underlying moderate superior arcuate defect of an additional 12 dB, the pattern deviation plot will be reduced by those 7 dB and displayed in the pattern deviation plot, where that relative depression will become more apparent. A probability analysis is again displayed on these adjusted deviation values.

,

102

Visual Fields

3-5-6 Glaucoma Hemifield Test. Another strategy to analyze the result of the visual field test is to compare localized decreased threshold values in corresponding mirror image areas of the superior and inferior hemispheres.31 In the Humphrey Visual Field Analyzer, it is called the Glaucoma Hemifield Test (GHT). This method allows a simple but clinically useful analysis of visual field changes in glaucoma patients. The software produces the GHT result by dividing each of the upper and lower halves of the field into five mirror-image zones; the outer edge loci, temporal loci, and loci around the blind spot are excluded, enabling the GHT to be performed on either Program 30-2 or Program 24-2 (Figure 3-17). Each zone is subsequently scored according to its pattern deviation values, and each upper zone is then compared with the corresponding lower zone. In addition, a general height of the field is determined by analyzing the most normal region of the field. The GHT uses a large normal database to calculate the significance of differences between the two hemispheres31 and has been shown to significantly improve the ability to separate between normal and glaucoma fields.32 The results of the GHT consist of five categories, which are displayed above the global indices, using plain language message: “Outside Normal Limits,” “Borderline,” “General Reduction of Sensitivity,” “Abnormally High Sensitivity,” and “Within Normal Limits.”31 The GHT “Outside Normal Limits,” used together with the pattern deviation probability plot, has been shown to provide high sensitivity and specificity for detecting early glaucomatous visual field changes.33 The five possible GHT messages and their derivations are as follows: 1. Outside Normal Limits This message results if either of these conditions is present: (a) any upper- versus lower-sector pair difference is greater than that found in 1% of the normal population or (b) any upper-lower pair sum differs at the 0.5% normal population level.

Figure 3-17. Glaucoma hemifield test zones.83

,

Essentials of Automated Perimetry

103

2. Borderline This message is displayed if the criteria for outside normal limits are not reached, but an upper-lower difference is present that occurs in less than 3% of the normal population. 3. General Reduction of Sensitivity This message occurs if the criteria for outside normal limits are not met, but the general height calculation finds the most normal region of the field to be below the 0.5% normal population level. Borderline and general reduction of sensitivity messages can result simultaneously (Figure 3-18). 4. Abnormally High Sensitivity This message is displayed when the general height calculation concludes that the best 15% of the field exceeds expected values for 99.5% of the normal population. If these conditions are present, the GHT will not display any of the other four messages (Figure 3-19).

Figure 3-18. Generalized depression. In this patient with cataract, with 20/40 visual acuity, the foveal sensitivity is reduced. The abnormal mean deviation, normal pattern standard deviation, and clean pattern deviation probability plot all indicate uniform field loss.

Figure 3-19. Abnormally high sensitivity. Both the patient and the technician did not seem to understand what was supposed to be happening during this test, resulting in a false nasal defect on the pattern deviation plot.

104

Essentials of Automated Perimetry

105

5. Within Normal Limits This message is displayed if the criteria for the above four conditions are not met.

,

3-5-7 Global Indices. In the lower-right corner of the single field printout, STATPAC displays two global indices, which describe the entire visual field using averaged numeric values: 1. MD (mean deviation) is an average value of overall deviation from the expected results within the same age group of normal visual fields. This value reflects the average height of the entire hill of vision. Negative values represent depression. MD is relatively insensitive to localized defects and is strongly affected by generalized decrease of sensitivity, such as cataracts. 2. PSD (pattern standard deviation) represents the unevenness of the surface of the hill of vision. It is calculated by taking a location-weighted standard deviation of all the threshold values. PSD is insensitive to the overall average height and is strongly affected by localized defects. STF and corrected pattern standard deviation (CPSD) are not used by current SITA algorithms. Of historical interest, STF is the standard deviation of repeated threshold of the 10 predefined locations. STF increases in inconsistent patients. This increase may be due to poor patient cooperation or attention, but STF also tends to increase in scotomatous areas, particularly at their borders. CPSD was used as an attempt to better represent the surface of the hill of vision by accounting for the influence of STF. STATPAC provides probability values for each global index value compared with age-matched normal subjects. For example, if the MD value is accompanied by a P < 5% symbol, the MD of the field is depressed by an amount greater than that found in 99.5% of the same-age normal population. Global indices generally correlate with visual field shape. For example, uniform generalized depression, as seen in cataract, tends to produce an increasingly negative MD. A small localized scotoma may have only slightly decreased MD and a high PSD (Figure 3-20). A dense moderate-size scotoma typically produces an elevation in both MD and PSD (Figure 3-21).

3-6 CUSTOM TESTS Several alternative tests to Program 30-2 with the full threshold strategy exist. These aim to save test time or provide more detailed information regarding specific regions of the visual field. Increasing the number of test locations and the precision with which they are tested does not necessarily provide a more accurate picture of the visual field. Lengthy tests become fatiguing to the patient and may result in greater variability of responses. A number of strategies have been devised in an attempt to shorten the test and reduce the number of tested points while still providing an accurate representation of the visual field. These strategies are described in the sections that follow. Multiple other tests are available with the Humphrey perimeter, and although they are rarely used, they are available in the Humphrey Visual Field Analyzer Main Menu.

Figure 3-20. Localized scotoma. This patient has primary open-angle glaucoma with a superior nasal step. The surrounding uninvolved field is slightly supranormal. When these values are “normalized,” the pattern deviation plot becomes slightly more significant than the total deviation plot. The supranormal uninvolved field, combined with the small size of the scotoma, contributes to only slight decrease in mean deviation. The pattern standard deviation is highly abnormal (99.5%), as is the Glaucoma Hemifield Test.

106

Figure 3-21. Moderate to severe glaucomatous field loss. This patient has

primary open-angle glaucoma with a dense inferior arcuate defect ending in a nasal step. Both the mean deviation and the pattern standard deviation are highly abnormal.

107

,

108

Visual Fields

3-6-1 Grid Size. The Humphrey Visual Field Analyzer 30-2 algorithm samples 76 locations with a uniform 6° grid within 30° from fixation (Figure 3-22). All Humphrey programs ending in -2 (e.g., 30-2, 24-2, 10-2) are offset from the central horizontal and vertical meridians. Programs 30-2 and 24-2, which use 6° spacing, are thus offset by 3° from the central meridians. Before invention of the SITA algorithms, approximately 550 questions would have been asked in a typical full threshold test, which may have taken about 15 minutes per eye. As the distance from fixation increases, the normal threshold values decrease, with a corresponding increase in the test variability, providing diminishing quality of the test. One of the initial approaches to shortening the test was to delete the outer row of locations. Program 24-2 does that by testing only out to 21° except for preservation of the important nasal extent of the 30-2 program (see Figure 3-22). The resulting test contains 54 locations, a 29% reduction compared with the Program 30-2 grid, considerably shortening the test duration. This represents a valuable tradeoff in patients who would be tired with additional testing. 3-6-4 FASTPAC. The FASTPAC algorithm uses an entirely different testing strategy. Instead of the standard 24-2 strategy with a double crossing of threshold, FASTPAC adjusts the stimulus intensity by 3-dB increments until the threshold is crossed once. FASTPAC saves the most time in normal or near-normal fields and provides little time advantage in patients with larger amounts of field abnormality. In comparisons to full threshold strategies, FASTPAC shortens the test time by 35% to 40%. FASTPAC may be suitable for observing ocular hypertensive patients

Figure 3-22. Comparison of the grid pattern, right eye, in Programs 24-2 and

30-2. Both programs test a uniform 6° grid offset 3° from the vertical and horizontal meridians. Program 24-2 tests (excludes outer outlined locations) 54 locations extending to 21° superiorly, inferiorly, and temporally and tests to 27° nasally. Program 30-2 tests an additional 22 locations extending the grid to 27° in all four directions.

Essentials of Automated Perimetry

109

or early glaucoma patients, although little longitudinal data are available on its use in these populations and the effect on LTF is presently unknown. One potential advantage of FASTPAC is the ability to use size V stimulus for central 10° testing, which is not available with current SITA algorithms.

3-6-5 Programs 30-1 and 24-1. Programs 30-1 and 24-1 are not used for clinical testing, because they are not offset from the horizontal and vertical meridians. Because scotomas centered on the meridians are difficult to classify (superior versus inferior, nasal versus temporal), these locations have less diagnostic localizing value. 3-6-6 Program 10-2 and Macula Test. Program 10-2 provides a high-resolution test of the central 10° with a tight 2° grid, offset 1° from the meridians. A total of 68 locations are used (Figure 3-23). Central tests are useful in carefully defining central or paracentral scotomas.34 In patients with advanced damage and small remaining central islands of vision, Program 10-2 can be performed with stimulus size V if the FASTPAC algorithm is used. This strategy provides the advantage of testing more areas with measurable thresholds and seems to increase patient cooperation and reduce patient fatigue. An even more localized test is the macula test, which thresholds 16 locations within the central 5° with 2° spacing. Each location is tested three times to provide better estimates of local STF (Figure 3-24).

Figure 3-23. Program 10-2 tests 68 locations with a 2° grid extending to 9°

from fixation. In this example, only a small central island of vision remains.

110

Visual Fields

Figure 3-24. The macula test thresholds the central 16 locations of the Program

10-2 grid. Each location is tested three times to better assess local short-term fluctuation.

3-6-7 Peripheral 60 and 60-4 Program. These programs allow additional exploration of the peripheral visual field (beyond 30°). Peripheral 60 points of this program extend the test out to 60° with a uniform grid testing 60 additional locations. 3-6-8 Nasal Step Program. Patients with possible nasal steps can be further explored with the nasal step program, which tests 12 locations up to 50° nasally. Two additional locations in the temporal visual field are also included to reduce the predictability of the questions.

Essentials of Automated Perimetry

111

3-6-9 Stimulus Size Option. Most programs are performed with stimulus size III, which subtends a diameter of 0.43° in visual space. This size is derived from the 4 mm2 size III used by the Goldmann perimeter. Testing with size III allows the application of the STATPAC statistical analysis. In fields where the majority of test locations are markedly reduced, it is often preferable to increase the stimulus size to V (1.72° in diameter). This larger size is often preferred by the patient and may reduce fluctuation, although the option of using the glaucoma change analysis is not available.

3-7 FOLLOW-UP PRINTOUT When evaluating a series of automated fields performed over time, the clinician may find the integration of the massive amount of data overwhelming. The Humphrey instrument allows the creation of several serial printouts to minimize confusion and allow the performance of change-over-time statistical analysis.

3-7-1 Overview Printout. The overview printout simply presents a sequential listing of condensed single field printouts chronologically (Figure 3-25). For each test session, four plots are displayed (from left to right): Threshold Graytone plot, Threshold (numeric) dB plot, Total Deviation P value plot, and Pattern Deviation P value plot. Above these four plots are the GHT results, reliability data, pupil size, and Snellen acuity. Below the plots are listed foveal sensitivity and global indices. The examiner familiar with the single field analysis printout will have no difficulty understanding all these terms. 3-7-2 Change Analysis Printout. The change analysis printout represents each field as a box plot depicted graphically over time (Figure 3-26). To create this plot, deviation values at each location are ranked from least to most depressed. As can be seen from the legend at the left side of the printout, each box contains the 15th to 85th percentile deviation values from this ranking, with a central triple line representing the median value. The tails extending from the box cover the full range of deviation throughout the field. In general, a small box with long tails suggests a clumping of values with a few outliers. If a box is small and near 0 dB, most of the field is normal. If the box changes in location over time with a stable size, a generalized change is likely occurring. A change in box dimension usually indicates a more localized process. The change analysis printout also plots the global indices over time, including threshold levels for statistical significance, with STF and PTSD plots being empty, if SITA algorithm was used. If the MD values plot is not demonstrating progressive change, it is likely that the field is not worsening, although a stable MD with an increasing PSD may indicate early progression of a localized scotoma. If the MD value is declining, the other indices and actual fields should be evaluated closely to discern confounding developments such as cataract formation. When five or more fields are available, linear regression analysis of the MD slope over time will be

Figure 3-25. Overview printout. The overview printout displays chronological

series of tests. These include the Graytone, numeric dB, Total Deviation, and Pattern Deviation plots from each test session. Above the plots are the test date and the Glaucoma Hemifield Test (GHT) results. Below the plots are the global and reliability indices, along with foveal sensitivity. Three tests performed over a 15-month period are displayed. Note that the probability symbols displayed in the total deviation and pattern deviation plots compare each location’s threshold with age-matched normal values. Similarly, the probability values next to the global indices are significance values compared with normal data.

112

Figure 3-26. Change analysis printout. The glaucoma change analysis printout

displays each field in box plot form as well as displaying the global indices graphically over time. The legend for the box plot is located to the left. The box plot is created by ranking total deviation values from most depressed (0%) to least depressed (100%). The box itself encompasses the central 70% of the field (between 15th and 85th percentile). The median rank value is displayed as a dark triple stripe inside the box. If the box is elongated toward the bottom of the plot, as in this example, then a large dense scotoma is present.

113

114

Visual Fields

calculated, which describes the slope in decibels per year and assigns a significance level, producing one of the two messages: “MD SLOPE SIGNIFICANT” or “MD SLOPE NOT SIGNIFICANT.” The MD plot and slope calculations are also included in the Glaucoma Progression Analysis printout. Studies suggest that a PSD could be used to monitor glaucoma progression in patients with glaucoma and cataract using SITA Standard to differentiate generalized worsening (cataract) from localized progression (glaucoma).35

3-7-3 Glaucoma Progression Analysis Printout. The Glaucoma Progression Analysis is an optional module in the STATPAC intended to assist with detection of glaucoma progression. At the time this chapter was written, the analysis was available for SITA Standard and SITA Fast algorithms but not for SITA-SWAP tests. Tests using Full Threshold algorithm could be used as baseline. Two initial fields are averaged into a baseline measurement, and then a point-by-point statistical comparison of each subsequent field is performed, indicating the presence or absence of a clinically significant change (Figure 3-27) and demonstrating each point’s change in decibels. The printout first displays the two baseline fields in Graytone and in Total Deviation plots. At the bottom of the page, following these two fields is a plot of Mean Deviation over time (MD Slope), identical to the one seen on Change Analysis Printout. Each follow-up field is displayed on the following pages labeled Follow-up on the top of the page. The follow-up portion of the Glaucoma Progression Analysis printout (Figure 3-28) differs from the Overview printout by replacing the Total Deviation and Pattern Deviation plots at the right with a point-by-point analysis of Deviation From Baseline plot in decibels from the baseline plot and the Progression Analysis probability plot. This latter plot identifies points that have progressively changed on consecutive fields, where crosses indicate changes outside of calculable statistical range. The analysis was found to have high specificity and sensitivity and be a useful test to assist in the detection of glaucomatous progression.36 3-7-4 GPA - Guided Progression Analysis. A new Guided Progression Analysis (GPA)37 may be a somewhat confusing term (see the previously described Glaucoma Progression Analysis). The GPA presents the results of a baseline examination, Visual Field Index Plot, Glaucoma Change Probability Map, and the GPA Alert, indicating “Possible Progression,” “Likely Progression,” or “No Progression Detected” on one page.

3-8 VISUAL FUNCTION SPECIFIC PERIMETRIC TECHNOLOGIES Visual function can also be evaluated by selectively testing specific retinal ganglion cells, potentially allowing for earlier detection of glaucoma. However, no technology has been shown to be superior due to significant variations of visual function in early glaucoma.38

Figure 3-27. Glaucoma progression analysis printout. Baselinee: Th he first

two fields are averaged and the results are considered a basselinee field d. Each subsequent field undergoes a point-by-point change from m the baseliine comparison, displayed with decibel values and accompanyying prob babiliity symbols. The mean deviation (MD) change probability comp paress the MD of each follow-up field to the baseline MD with associated P valu ues. A lineear regression of the MD values over time is displayed next to th he baaselin ne fiellds and, in this case, the resulting MD slope is not significant at a 95% % con nfiden nce level. In this case, note the same level of significance over timee, ind dicatiing stability, despite preexisting significant loss of sensitivity.

115

Figure 3-28. Follow-up demonstrates individual test results with Deviation

from Baseline and Progression Analysis point-by-point.

116

Essentials of Automated Perimetry

117

3-8-1 Short-Wavelength Automated Perimetry (SWAP). Testing a subgroup of ganglion cells, called midget ganglion cells, that are sensitive to blue stimuli may detect loss of visual function at an earlier stage of glaucoma than with standard automated perimetry.26 Short-wavelength automated perimetry (SWAP) presents a large Goldmann size V, short-wavelength blue targets on a bright yellow background.39 SWAP defects appear to represent early glaucomatous damage and may indicate significant change in visual function before it becomes apparent on standard white-on-white visual fields. Several longitudinal studies have demonstrated the ability of blue-on-yellow perimetry to predict the development of glaucoma in ocular hypertensive patients,40,41 as well as which patients with early glaucomatous visual field loss are most likely to progress.42 Other studies have demonstrated a significant relationship between structural optic nerve damage and SWAP visual field defects.43 However, the test is influenced by age and cataracts, and stringent statistical analysis in interpreting the results is necessary.44,45 The SWAP is currently available on the newer Humphrey Visual Field Analyzers, incorporating the efficiency of the SITA algorithm with the 24-2 testing pattern, to decrease testing time from 12 minutes to less than 4 minutes.46,47 SITA-SWAP has been shown to demonstrate a higher prevalence of visual field deficits, suggesting earlier detection of glaucomatous changes, compared with standard automated perimetry.42,48 The printout is very similar to the standard HFA printout. 3-8-2 Frequency Doubling Perimetry. Frequency doubling illusion is the base for the frequency doubling technology (FDT) perimetry.49 Each test stimulus is a series of white-and-black bands flickering at 25 Hz.50 FDT perimetry is thought to be mediated by a subset of the large-diameter ganglion cells that project to the magnocellular visual pathway.51 These cells are presumed to be sensitive to motion and contrast and more vulnerable to glaucomatous damage,52 although this view has been questioned by some.53,54 The FDT is a portable and relatively inexpensive tool1 with a short testing time, making it a very useful screening device.55 However, FDT perimetry was reported to be less sensitive to visual field damage associated with neurologic disorders compared with standard automated perimetry.56 Sensitivity to FDT was found to be reduced in the second tested eye57—this effect is accounted for in the perimeter’s normative database.58 The original FDT perimeter tested only 17 points over the central 20° of the visual field. The second-generation FDT, Matrix, uses 69 smaller stimuli, utilizing 24-2 and 30-2 strategies to allow for early detection of glaucoma.59 The Matrix printout is organized similarly to a printout from the Humphrey Visual Field Analyzer. 3-8-3 High-Pass Resolution Perimetry. High-pass resolution perimetry (HPRP), or ring perimetry, is presumed to selectively test the parvocellular system.60 The stimuli used in HPRP are rings of different sizes with dark borders and bright centers projected at different locations on the screen, creating average luminance of the stimulus equal to the luminance of the background. The results of the test are presumed to correspond to the density of ganglion cells.1 In glaucoma patients, HPRP is comparable to standard perimetry in sensitivity and specificity.61 Some studies suggested that HPRP is a viable test for detection of glaucomatous visual field damage.62

118

Visual Fields

3-8-4 Tendency-Oriented Perimetry. Tendency-oriented perimetry (TOP) is a fast strategy algorithm available on Octopus perimeters.63 It also uses a computational approach to estimate threshold values by extrapolating information from surrounding test points. Compared with SITA Fast, the mean testing time for the TOP strategy was slightly over 2½ minutes, compared with approximately 4 minutes for SITA Fast.64 However, the TOP algorithm may not always accurately estimate visual field defects.65

3-9 LEARNING EFFECT AND ARTIFACTS The results of many psychophysical tests improve as the subject gains more experience performing the test. The learning effect in automated perimetry seems to be small in most patients who have had experience with manual perimetry.66 Some patients, however, may demonstrate a dramatic improvement in the second test compared with the first.67 Occasionally, patients continue to improve over the initial three, four, or (rarely) five automated fields. The variability of test results may decrease significantly with experience. Whenever possible, a patient who is new to perimetry should undergo several test sessions, to establish a baseline for subsequent comparisons. The magnitude of the learning effect can be reduced by an attentive, thoughtful technician who takes the time to explain the examination thoroughly to the patient. Several factors can produce abnormal-appearing field printouts without necessarily affecting the reliability indices. Other causes may contribute to unreliable results yet not be apparent on the printout. Test areas that are noisy, have confounding stray light, or provide uncomfortable chairs may contribute to patient fatigue or inconsistent responses. Some of the clinically important artifacts are discussed in the sections that follow.

3-9-1 Miosis and Mydriasis. Pupil size of less than 3 mm, particularly if combined with lens opacity, tends to produce a diffuse depression of the visual field. For this reason, the pupil size should be recorded on each test, and the influence of pupil size should be considered when a field change is detected. The technician usually enters the pupil size into the general information section, so that it is available to the clinician. Newer versions of automated perimeters can automatically record the size of the pupil. If a pupil in a patient undergoing visual field testing is less than 3 mm, as a result of miotics, neuro-ophthalmologic disease, or age-related miosis, it should be routinely dilated for perimetry. Mydriasis has less influence on the visual field, although pupillary dilatation may reduce peripheral threshold sensitivity with automated perimetry.68,69 Although decreased pupil size should have little effect on a patient’s perception of a stimulus, because background and stimulus are affected equally, significant miosis may depress central and peripheral threshold sensitivities and exaggerate field defects.70 Refractive error can change considerably after pupil dilation, particularly in younger patients, who may require a dilated refraction prior to testing. One study used neutral density filters to reduce the retinal illumination

Essentials of Automated Perimetry

119

the equivalent of decreasing the pupillary diameter in half, which reduced the mean threshold with two automated perimeters by 1.1 to 1.7 dB.71 Another study has also shown that pilocarpine significantly worsened the visual field global indices, such as MD and PSD.72

3-9-2 Media Opacities. Any media opacity, such as a cataract, posterior capsular fibrosis, or corneal scar, can cause a generalized depression of the field result (see Figure 3-17). Prediction of this effect by Snellen acuity is often poor. The patient with 20/25 acuity and 3+ nuclear sclerosis may have 5 dB or more of generalized depression apparent on the MD index. Cataracts produce glare and change the intensity of the stimulus. Therefore, a cataract can cause or exaggerate central or peripheral field defects, which could be mistaken for the development or progression of glaucomatous field loss. Even minimal light scattering, as may be caused by an early cataract that has a relatively insignificant effect on visual acuity, may significantly influence visual field results. After cataract extraction, eyes with glaucoma may have improvement of foveal sensitivity and visual field scores and sometimes even a reversal of a partial or complete scotoma.73 However, cataract surgery can also reveal mild and moderate field defects masked by cataracts.74,75 Reduced clarity of the ocular media from other causes, such as a corneal disturbance, a cloudy posterior lens capsule, or vitreous opacities, may also affect the visual fields. Applanation tonometry prior to automated perimetry was not found to have a detrimental effect on the visual field results.76 3-9-3 Eyelid and Nose Effects. Acquired ptosis of a mild degree is very common in older adults. While it may not affect Snellen acuity, a drooping upper eyelid can produce a superior arcuate–type defect or a superior nasal step. The visual field technician can be instrumental in recognizing this artifact, which can often be improved by taping the eyelid up during the test. Improper head positioning by the visual field technician can also cause the nose to produce an inferonasal defect. 3-9-4 Refractive Errors. Improper refractive correction could cause the projected stimulus to be out of focus on the retina, and the stimulus may not be detected by the patient. Refractive errors primarily influence the central field.77 With a usual size-III stimulus, spherical refractive errors of less than 1 diopter may not need to be corrected, because error in measurement is going to be only slightly more than 1 dB of general reduction of sensitivity.78 Myopia of less than 3 diopters also may not require correction, although high myopic errors can create areas of retinal blur, called refraction scotomas, which appear as a vertical wedge–type defect and may be confused with glaucomatous field loss. Those can usually be eliminated with an appropriate refractive correction. Hyperopia has a greater influence on perimetric results, especially for the central field, and even small hyperopic refractive errors can significantly alter threshold sensitivity.77-79 In general, patients over age 30 require a near correction added to the distance refraction. Spherical add depends on an accommodative amplitude, for which Humphrey Visual Field Analyzer uses age data to aid in determining the appropriate correction. A contact lens provides

120

Visual Fields

the best correction for the aphakic and highly myopic80,81 eyes. Astigmatic error of more than 1 diopter should be corrected in full.

3-9-5 Corrective Lens/Frame Artifacts. A spectacle lens placed too far from the tested eye can produce a ring-type scotoma appearance. Poor lens centering can produce an arcuate-appearing defect. Frame/corrective lens artifacts tend to be perfectly semicircular, or circular and dense (Figure 3-29). They can be minimized with proper spectacle placement or eliminated entirely with contact lens testing.

Figure 3-29. Lens holder artifact. This patient was tested without a corrective

lens, while the lens holder was moved away from the patient face toward the center of the bowl, demonstrating absolute loss in a circular pattern inferiorly. The very sharp falloff in sensitivities above the horizontal meridian with an absent nasal step strongly suggests an artifactual scotoma, which was confirmed when the field appeared normal after the lens holder was moved out of the way.

Essentials of Automated Perimetry

121

3-10 ROLE OF THE VISUAL FIELD TECHNICIAN An attentive, knowledgeable technician is of critical importance to obtaining highquality test results.82 Patients must be instructed on what to expect during the field test, stressing the fact that at least half of the stimuli are not supposed to be seen, and the ones that are seen will barely visible, thereby allowing the proper performance to be achieved. Patients must be given feedback throughout the test on maintaining fixation and proper position, minimizing false-positive and falsenegative responses, and how long the test will last. A good visual field technician is vigilant in providing a comfortable, quiet location for reliable results, as well as recognizing the need for pupil dilation. In addition, the visual field technician should be able to minimize the effects of ptosis and lens artifact. It may also be helpful to let the patient know that the machine will not present an additional stimulus until the button is released, providing the patient with a break to stretch and to become more comfortable and attentive to the test.

REFERENCES 1. Delgado MF, Nguyen NT, Cox TA, et al. Automated perimetry: a report by the American Academy of Ophthalmology. Ophthalmology. 2002;109:2362–2374. 2. Förster C. Vorzeigung des Perimeter. Klin Monatsbl Augenheilk. 1869;7:411–422. 3. Traquair H. An Introduction to Clinical Perimetry. London: Kimpton; 1927. 4. Harms H, Aulhorn E. [Comparative studies on the value of quantitative perimetry, skiaskotometry and fusion frequency for the recognition of beginning visual field disorders in glaucoma.]. Doc Ophthalmol. 1959;13:303–356. 5. Wohlrab TM, Kreinberger H, Erb C, et al. [Threshold-oriented suprathreshold perimetry and threshold value perimetry with the Tubigen CC Automated Perimeter. A comparative study]. Ophthalmologe. 1998;95:92–99. 6. Gloor BP. Franz Fankhauser: the father of the automated perimeter. Surv Ophthalmol. 2009;54:417–425. 7. Fankhauser F. Background illumination and automated perimetry. Arch Ophthalmol. 1986;104:1126. 8. Heijl A, Krakau CE. An automatic perimeter for glaucoma visual field screening and control. Construction and clinical cases. Albrecht Von Graefes Arch Klin Exp Ophthalmol. 1975;197:13–23. 9. Flammer J, Drance SM, Augustiny L, Funkhouser A. Quantification of glaucomatous visual field defects with automated perimetry. Invest Ophthalmol Vis Sci. 1985;26: 176–181. 10. Bengtsson B, Heijl A, Olsson J. Evaluation of a new threshold visual field strategy, SITA, in normal subjects. Swedish Interactive Thresholding Algorithm. Acta Ophthalmol Scand. 1998;76:165–169. 11. Chauhan BC, Garway-Heath DF, Goni FJ, et al. Practical recommendations for measuring rates of visual field change in glaucoma. Br J Ophthalmol. 2008;92: 569–573. 12. Nevalainen J, Paetzold J, Krapp E, et al. The use of semi-automated kinetic perimetry (SKP) to monitor advanced glaucomatous visual field loss. Graefes Arch Clin Exp Ophthalmol. 2008;246:1331–1339.

122

Visual Fields

13. Johnson CA, Keltner JL. Optimal rates of movement for kinetic perimetry. Arch Ophthalmol. 1987;105:73–75. 14. Schiefer U, Strasburger H, Becker ST, et al. Reaction time in automated kinetic perimetry: effects of stimulus luminance, eccentricity, and movement direction. Vision Res. 2001;41:2157–2164. 15. Pineles SL, Volpe NJ, Miller-Ellis E, et al. Automated combined kinetic and static perimetry: an alternative to standard perimetry in patients with neuro-ophthalmic disease and glaucoma. Arch Ophthalmol. 2006;124:363–369. 16. Chauhan BC, Tompkins JD, LeBlanc RP, McCormick TA. Characteristics of frequencyof-seeing curves in normal subjects, patients with suspected glaucoma, and patients with glaucoma. Invest Ophthalmol Vis Sci. 1993;34:3534–3540. 17. Westcott MC, Fitzke FW, Crabb DP, Hitchings RA. Characteristics of frequency-ofseeing curves for a motion stimulus in glaucoma eyes, glaucoma suspect eyes, and normal eyes. Vision Res. 1999;39:631–639. 18. Bebie H, Fankhauser F, Spahr J. Static perimetry: accuracy and fluctuations. Acta Ophthalmol (Copenh). 1976;54:339–348. 19. Chauhan BC, LeBlanc RP, Drance SM, et al. Effect of the number of threshold determinations on short-term fluctuation in automated perimetry. Ophthalmology. 1991;98:1420–1424. 20. Boeglin RJ, Caprioli J, Zulauf M. Long-term fluctuation of the visual field in glaucoma. Am J Ophthalmol. 1992;113:396–400. 21. Keltner JL, Johnson CA, Quigg JM, et al. Confirmation of visual field abnormalities in the Ocular Hypertension Treatment Study. Ocular Hypertension Treatment Study Group. Arch Ophthalmol. 2000;118:1187–1194. 22. Bourne RR, Jahanbakhsh K, Boden C, et al. Reproducibility of visual field end point criteria for standard automated perimetry, full-threshold, and Swedish interactive thresholding algorithm strategies: diagnostic innovations in glaucoma study. Am J Ophthalmol. 2007;144:908–913. 23. Budenz DL, Rhee P, Feuer WJ, et al. Sensitivity and specificity of the Swedish interactive threshold algorithm for glaucomatous visual field defects. Ophthalmology. 2002;109:1052–1058. 24. Akar Y, Yilmaz A, Yucel I. Assessment of an effective visual field testing strategy for a normal pediatric population. Ophthalmologica. 2008;222:329–33. 25. Roggen X, Herman K, Van Malderen L, et al. Different strategies for Humphrey automated perimetry: FASTPAC, SITA standard and SITA fast in normal subjects and glaucoma patients. Bull Soc Belge Ophtalmol. 2001:23–33. 26. Johnson CA. Recent developments in automated perimetry in glaucoma diagnosis and management. Curr Opin Ophthalmol. 2002;13:77–84. 27. Heijl A, Krakau CE. An automatic static perimeter, design and pilot study. Acta Ophthalmol (Copenh). 1975;53:293–310. 28. Heijl A, Krakau CE. A note of fixation during perimetry. Acta Ophthalmol (Copenh). 1977;55:854–861. 29. Sanabria O, Feuer WJ, Anderson DR. Pseudo-loss of fixation in automated perimetry. Ophthalmology. 1991;98:76–78. 30. Heijl A, Lindgren G, Olsson J, Asman P. Visual field interpretation with empiric probability maps. Arch Ophthalmol. 1989;107:204–208. 31. Asman P, Heijl A. Glaucoma Hemifield Test. Automated visual field evaluation. Arch Ophthalmol. 1992;110:812–819. 32. Asman P, Heijl A. Evaluation of methods for automated Hemifield analysis in perimetry. Arch Ophthalmol. 1992;110:820–826.

Essentials of Automated Perimetry

123

33. Johnson CA, Sample PA, Cioffi GA, et al. Structure and function evaluation (SAFE): I. criteria for glaucomatous visual field loss using standard automated perimetry (SAP) and short wavelength automated perimetry (SWAP). Am J Ophthalmol. 2002;134: 177–185. 34. Much JW, Liu C, Piltz-Seymour JR. Long-term survival of central visual field in endstage glaucoma. Ophthalmology. 2008;115:1162–1166. 35. Rehman Siddiqui MA, Khairy HA, Azuara-Blanco A. Effect of cataract extraction on SITA perimetry in patients with glaucoma. J Glaucoma. 2007;16:205–208. 36. Arnalich-Montiel F, Casas-Llera P, Munoz-Negrete FJ, Rebolleda G. Performance of glaucoma progression analysis software in a glaucoma population. Graefes Arch Clin Exp Ophthalmol. 2009;247:391–397. 37. Bengtsson B, Heijl A. A visual field index for calculation of glaucoma rate of progression. Am J Ophthalmol. 2008;145:343–353. 38. Sakata LM, DeLeon-Ortega J, Girkin CA. Selective perimetry in glaucoma diagnosis. Curr Opin Ophthalmol. 2007;18:115–121. 39. Johnson CA. Diagnostic value of short-wavelength automated perimetry. Curr Opin Ophthalmol. 1996;7:54–58. 40. Polo V, Abecia E, Pablo LE, et al. Short-wavelength automated perimetry and retinal nerve fiber layer evaluation in suspected cases of glaucoma. Arch Ophthalmol. 1998;116: 1295–1298. 41. Lewis RA, Johnson CA, Adams AJ. Automated perimetry and short wavelength sensitivity in patients with asymmetric intraocular pressures. Graefes Arch Clin Exp Ophthalmol. 1993;231:274–278. 42. Johnson CA, Adams AJ, Casson EJ, Brandt JD. Progression of early glaucomatous visual field loss as detected by blue-on-yellow and standard white-on-white automated perimetry. Arch Ophthalmol. 1993;111:651–656. 43. Girkin CA, Emdadi A, Sample PA, et al. Short-wavelength automated perimetry and standard perimetry in the detection of progressive optic disc cupping. Arch Ophthalmol. 2000;118:1231–1236. 44. Tafreshi A, Sample PA, Liebmann JM, et al. Visual function-specific perimetry to identify glaucomatous visual loss using three different definitions of visual field abnormality. Invest Ophthalmol Vis Sci. 2009;50:1234–1240. 45. Sample PA, Johnson CA, Haegerstrom-Portnoy G, Adams AJ. Optimum parameters for short-wavelength automated perimetry. J Glaucoma. 1996;5:375–383. 46. Bengtsson B. A new rapid threshold algorithm for short-wavelength automated perimetry. Invest Ophthalmol Vis Sci. 2003;44:1388–1394. 47. Bengtsson B, Heijl A. Diagnostic sensitivity of fast blue-yellow and standard automated perimetry in early glaucoma: a comparison between different test programs. Ophthalmology. 2006;113:1092–1097. 48. Demirel S, Johnson CA. Short wavelength automated perimetry (SWAP) in ophthalmic practice. J Am Optom Assoc. 1996;67:451–456. 49. Johnson CA, Samuels SJ. Screening for glaucomatous visual field loss with frequencydoubling perimetry. Invest Ophthalmol Vis Sci. 1997;38:413–425. 50. Alward WL. Frequency doubling technology perimetry for the detection of glaucomatous visual field loss. Am J Ophthalmol. 2000;129:376–378. 51. Maddess T. Perspectives on the use of frequency doubling and short wavelength perimetry for the diagnosis of glaucoma. Clin Exp Ophthalmol. 2000;28:245–247. 52. Quigley HA. Neuronal death in glaucoma. Prog Retin Eye Res. 1999;18:39–57. 53. Harwerth RS, Carter-Dawson L, Shen F, et al. Ganglion cell losses underlying visual field defects from experimental glaucoma. Invest Ophthalmol Vis Sci. 1999;40:2242–2250.

124

Visual Fields

54. Morgan JE, Uchida H, Caprioli J. Retinal ganglion cell death in experimental glaucoma. Br J Ophthalmol. 2000;84:303–310. 55. Mansberger SL, Edmunds B, Johnson CA, et al. Community visual field screening: prevalence of follow-up and factors associated with follow-up of participants with abnormal frequency doubling perimetry technology results. Ophthalmic Epidemiol. 2007;14:134–140. 56. Wall M, Neahring RK, Woodward KR. Sensitivity and specificity of frequency doubling perimetry in neuro-ophthalmic disorders: a comparison with conventional automated perimetry. Invest Ophthalmol Vis Sci. 2002;43:1277–1283. 57. Tsukamoto H, Mukai S, Iwase A, Mishima HK. Recovery of reliability by retest after a 5-minute interval in frequency doubling technology perimetry. Jpn J Ophthalmol. 2006;50:380–382. 58. Brusini P, Salvetat ML, Zeppieri M, Parisi L. Frequency doubling technology perimetry with the Humphrey Matrix 30–2 test. J Glaucoma. 2006;15:77–83. 59. Johnson CA, Cioffi GA, Van Buskirk EM. Frequency doubling technology perimetry using a 24–2 stimulus presentation pattern. Optom Vis Sci. 1999;76:571–581. 60. Martinez GA, Sample PA, Weinreb RN. Comparison of high-pass resolution perimetry and standard automated perimetry in glaucoma. Am J Ophthalmol. 1995;119: 195–201. 61. Martin L, Wanger P, Vancea L, Gothlin B. Concordance of high-pass resolution perimetry and frequency-doubling technology perimetry results in glaucoma: no support for selective ganglion cell damage. J Glaucoma. 2003;12:40–44. 62. Chauhan BC. The value of high-pass resolution perimetry in glaucoma. Curr Opin Ophthalmol. 2000;11:85–89. 63. Maeda H, Nakaura M, Negi A. New perimetric threshold test algorithm with dynamic strategy and tendency oriented perimetry (TOP) in glaucomatous eyes. Eye. 2000; 14(Pt 5):747–751. 64. King AJ, Taguri A, Wadood AC, Azuara-Blanco A. Comparison of two fast strategies, SITA Fast and TOP, for the assessment of visual fields in glaucoma patients. Graefes Arch Clin Exp Ophthalmol. 2002;240:481–487. 65. Anderson AJ. Spatial resolution of the tendency-oriented perimetry algorithm. Invest Ophthalmol Vis Sci. 2003;44:1962–1968. 66. Werner EB, Krupin T, Adelson A, Feitl ME. Effect of patient experience on the results of automated perimetry in glaucoma suspect patients. Ophthalmology. 1990;97:44–48. 67. Johnson CA, Keltner JL, Cello KE, et al. Baseline visual field characteristics in the ocular hypertension treatment study. Ophthalmology. 2002;109:432–437. 68. Lindenmuth KA, Skuta GL, Rabbani R, et al. Effects of pupillary dilation on automated perimetry in normal patients. Ophthalmology. 1990;97:367–70. 69. Mendivil A. Influence of a dilated pupil on the visual field in glaucoma. J Glaucoma. 1997;6:217–220. 70. Lindenmuth KA, Skuta GL, Rabbani R, Musch DC. Effects of pupillary constriction on automated perimetry in normal eyes. Ophthalmology. 1989;96:1298–1301. 71. Heuer DK, Anderson DR, Feuer WJ, Gressel MG. The influence of decreased retinal illumination on automated perimetric threshold measurements. Am J Ophthalmol. 1989;108:643–650. 72. Edgar DF, Crabb DP, Rudnicka AR, et al. Effects of dipivefrin and pilocarpine on pupil diameter, automated perimetry and LogMAR acuity. Graefes Arch Clin Exp Ophthalmol. 1999;237:117–124. 73. Chen PP, Budenz DL. The effects of cataract extraction on the visual field of eyes with chronic open-angle glaucoma. Am J Ophthalmol. 1998;125:325–333.

Essentials of Automated Perimetry

125

74. Smith SD, Katz J, Quigley HA. Effect of cataract extraction on the results of automated perimetry in glaucoma. Arch Ophthalmol. 1997;115:1515–1519. 75. Hayashi K, Hayashi H, Nakao F, Hayashi F. Influence of cataract surgery on automated perimetry in patients with glaucoma. Am J Ophthalmol. 2001;132:41–46. 76. Ruben JB, Lewis RA, Johnson CA, Adams C. The effect of Goldmann applanation tonometry on automated static threshold perimetry. Ophthalmology. 1988;95: 267–270. 77. Weinreb RN, Perlman JP. The effect of refractive correction on automated perimetric thresholds. Am J Ophthalmol. 1986;101:706–709. 78. Heuer DK, Anderson DR, Feuer WJ, Gressel MG. The influence of refraction accuracy on automated perimetric threshold measurements. Ophthalmology. 1987;94: 1550–1553. 79. Goldstick BJ, Weinreb RN. The effect of refractive error on automated global analysis program G-1. Am J Ophthalmol. 1987;104:229–232. 80. Koller G, Haas A, Zulauf M, et al. Influence of refractive correction on peripheral visual field in static perimetry. Graefes Arch Clin Exp Ophthalmol. 2001;239:759–762. 81. Miller BA, Gelber EC. Aphakic visual fields by automated perimetry. Ann Ophthalmol. 1990;22:419–422. 82. Kutzko KE, Brito CF, Wall M. Effect of instructions on conventional automated perimetry. Invest Ophthalmol Vis Sci. 2000;41:2006–2013. 83. Demirel S, Johnson CA. Incidence and prevalence of short wavelength automated perimetry deficits in ocular hypertensive patients. Am J Ophthalmol. 2001;131: 709–715.

This page intentionally left blank

4 Automated Perimetry in Glaucoma HYLTON R. MAYER, MD, MARC L. WEITZMAN, MD, AND JOSEPH CAPRIOLI, MD

4-1 INTRODUCTION Clinical experience and multiple prospective studies, such as the Collaborative Normal Tension Glaucoma Study and the Los Angeles Latino Eye Study, have demonstrated that the diagnosis of glaucoma is more complex than identifying elevated intraocular pressure.1,2 As a result, increased emphasis has been placed on measurements of the structural and functional abnormalities caused by glaucoma. The refinement and adoption of imaging technologies assist the clinician in the detection of glaucomatous damage and, increasingly, in identifying the progression of structural damage. Because visual field defects in glaucoma patients occur in patterns that correspond to the anatomy of the nerve fiber layer of the retina and its projections to the optic nerve, visual functional tests become a link between structural damage and functional vision loss (Figure 4-1). The identification of glaucomatous damage and management of glaucoma require appropriate, sequential measurements and interpretation of the visual field.

4-2 GLAUCOMATOUS FIELD LOSS Glaucomatous visual field defects usually are of the nerve fiber bundle type, corresponding to the anatomic arrangement of the retinal nerve fiber layer. It is helpful to consider the division of the nasal and temporal retina as the fovea, not the optic nerve head, because this is the location that determines the center of the visual field. The ganglion cell axon bundles that emanate from the nasal side of the retina generally approach the optic nerve head in a radial fashion. The majority of

127

128

Visual Fields

Figure 4-1. The nerve fiber layer anatomy of the retina. Damage to discrete bundles of nerve fibers, usually at the superior and inferior poles of the disc, gives rise to typical patterns of visual field loss in glaucoma. The classic nerve fiber bundle defects are the arcuate scotoma, nasal step, and temporal-sector defect. Diffuse damage to nerve fibers may also cause a diffuse depression of visual field sensitivity.

these fibers enter the nasal half of the optic disc, but fibers that represent the nasal half of the macula form the papillomacular bundle to enter the temporal-most aspect of the optic nerve. In contrast, the temporal retinal fibers, with respect to fixation, arc around the macula to enter the superotemporal and inferotemporal portions of the optic disc. The origin of these arcuate temporal retinal fibers strictly respects the horizontal retinal raphe, temporal to the fovea. As a consequence of this superior-inferior segregation of the temporal retinal fibers, lesions that affect the superotemporal and inferotemporal poles of the optic disc, such as glaucoma, tend to cause arcuateshaped visual field defects extending from the blind spot toward the nasal horizontal meridian. Tremendous variability characterizes the patterns of glaucomatous cupping, with some patients having fairly diffuse, concentric loss of the neuroretinal rim, while others have extremely localized loss. Glaucomatous visual field damage shows a similar variability, with some patients demonstrating early scattered involvement of the visual field, while others display dense localized defects. When only a single hemifield is involved, it is the superior hemifield 60% of the time.3,4 When patients have visual field loss in both the superior and the inferior hemifields, there is still often a detectable difference in the measured threshold on either side of the nasal horizontal meridian. For the most part, clinicians familiar with the typical glaucomatous scotomas detected by manual perimetry recognize these same defects when demonstrated by automated perimetry. Seidel scotomas emenating from the optic nerve, nasal steps, and paracentral scotomas are commonly seen. When more advanced damage occurs, these smaller scotomas may coalesce to form superior and inferior arcuate

Automated Perimetry in Glaucoma

129

scotomas, which may ultimately isolate a spared central and/or temporal island of vision. Temporal wedge defects, representing glaucomatous damage to the nasal nerve fibers, also respect the nerve fiber layer anatomy but are seen less than 5% of the time.5 Scotomas involving fixation (central scotomas) typically occur after advanced cupping occurs. While paracentral scotomas may be seen, classically in normal tension glaucoma, if central scotomas are identified, one should consider alternative diagnoses, such as retinal disease or nonglaucomatous optic neuropathies.6,7 Similarly, defects that respect the vertical meridian are not retinal nerve fiber bundle defects and are often caused by disease at or behind the optic chiasm. One defect well described with manual kinetic perimetry, which is less often seen with static testing, is the Seidel scotoma, an arcuate-shaped elongation of the physiologic blind spot. The hill of vision at the superior and inferior poles of the blind spot tends to be relatively flat. As a consequence, kinetic testing originating from the blind spot can easily mimic a Seidel scotoma if the stimulus speed is slightly too fast. The occurrence of purely generalized field loss in glaucoma is controversial.8-10 Because many common conditions, such as cataract, miosis, refractive error, and patient fatigue, can cause an elevated mean deviation (MD) with a normal pattern standard deviation (PSD) and corrected pattern standard deviation (CPSD), these elements should be carefully considered before attributing generalized depression to glaucoma. Rarely, patients present with asymmetric intraocular pressure, asymmetric cupping, no confounding factors, and asymmetric MD values without localized scotomas (Figure 4-2). As a general rule, normal patients tend to test with MD asymmetry less than 2.0 dB 95% of the time on a single test. A 1.5-dB difference trend over two fields is equally significant, and a difference as small as 1 dB may be meaningful if reproducible over four fields.11 Several studies have suggested that normal-tension glaucoma defects are more localized and are closer to fixation compared with primary open-angle glaucoma defects.12-13 This theory has not been unequivocally proved, with critics alleging bias based on presenting factors because patients with primary open-angle glaucoma are more likely to be diagnosed on a basis of elevated intraocular pressure and may often be detected at an earlier stage of field loss compared with patients with normal-tension glaucoma.14

4-3 AUTOMATED PERIMETRY OPTIONS For more than a decade, the full threshold testing strategy was the most widely used automated perimetry algorithm. Advantages of the full threshold test include numerous stimuli presented at each testing point in the field, theoretically offering a more refined estimation of the threshold value. The full threshold testing algorithm also offers a variety of reliability indicators, including estimations of short-term fluctuation (STF) within the test itself. Finally, many landmark studies regarding glaucoma diagnosis and management used full threshold automated perimetry, establishing it as a gold standard in automated static perimetry. Disadvantages

130

Visual Fields

of the full threshold strategy include longer testing times, which can decrease patient comfort and test reliability. Patient fatigue during full threshold testing may also artificially depress sensitivities and overestimate glaucomatous damage or progression. Evolutions in testing algorithms involved adjustments in the grid size from 30-2 to 24-2, or from manipulation of the starting sensitivity or frequency of points crossing the threshold value, such as in fast threshold, FASTPAC. While FASTPAC can shorten test times 35-40% compared with full threshold strategies, it is thought to underestimate the size and density of scotomas. Humphrey field machines now offer the Swedish interactive threshold algorithm (SITA) that has

A Figure 4-2. Generalized depression on a full threshold test. The intraocular pressure is in the high 20s in the left eye and in the high teens in the right eye, with asymmetric cupping. There is 2.45 dB of asymmetry in the mean deviation between the two eyes with otherwise normal fields. This can be taken as evidence of early glaucomatous generalized depression. (A) Field chart of left eye. (B) Field chart of right eye. (C) Fundus photograph of right eye. (D) Fundus photograph of left eye.

Automated Perimetry in Glaucoma

131

B

C

D

Figure 4-2. (Continued)

largely supplanted previous testing strategies. SITA uses a proprietary software to more rapidly and accurately identify threshold values. In comparison with full threshold tests, SITA tests are 50% shorter, typically lasting about 5 minutes per eye. SITA has been extensively evaluated against full threshold strategies and is widely accepted to be equivalent to the “gold standard.”15 SITA tests have been

132

Visual Fields

applied to large multicenter prospective studies, allowing more direct comparison to clinical applications and improving longitudinal data.16,17 Advanced analytical software is available to assist in identifying glaucomatous defects, progression, and trends (see Section 4-5). SITA is also available in a more streamlined testing algorithm, SITA Fast, which decreases testing time (70% faster than full threshold) by reducing the number of times the threshold limit is checked.18 In comparison with SITA, many believe the advantages in shorter test times are offset by the reduction in the quality of the test data. Advances in testing algorithms, software, and hardware are improving efforts to increase patient comfort by decreasing test duration while maintaining or improving reliability and reproducibility.

4-4 EVALUATION OF A SINGLE TEST Regardless of the method used to obtain them, visual field measurements should not be interpreted in a vacuum. It is almost always preferable to integrate other clinical data, such as refractive status, corneal clarity, condition of the crystalline lens, intraocular pressure, appearance of the optic nerve head, estimation of the patient’s vascular status, and assessment of the patient’s general health. An analysis of specific risk factors and potential side effects should be included in decisions about therapy. Before diagnostic and therapeutic decisions based on visual field information are made, patient reliability, artifacts, and other confounding variables must be considered. The Advanced Glaucoma Intervention Study (AGIS) and The Collaborative Initial Glaucoma Treatment Study (CIGTS) demonstrated that visual field defects on one test often fluctuated on subsequent follow-ups, and patients often demonstrated wide STFs and long-term fluctuations (LTFs).19,20

4-4-1 Patient Reliability. Determination of patient performance is an important initial step in visual field interpretation. A number of routinely measured elements can supply an estimate of patient reliability and are generally displayed with the fields: 1. 2. 3. 4. 5.

Test duration Number of fixation losses Rate of false-positive answers Rate of false-negative answers STF, or the fluctuation of repeated threshold measurements within the same test. This parameter is present on full threshold programs and is not evaluated in the more recent SITA tests. 6. Number of stimuli. This parameter is present on full threshold programs and is not reported in the SITA tests. Also important is the operator’s ability to comment on or score patient cooperation and alertness during the examination. Maintaining accurate fixation and alertness depends on both the patient and the technician.

Automated Perimetry in Glaucoma

133

The standard limits for acceptable false-positive and false-negative answers have been set at 33%, and for fixation losses, at 20%. If these limits are exceeded, the value is highlighted and a warning message is displayed. These messages should not necessarily prompt the examiner to discard the entire field but should result in a review of possible causal factors.

4-4-1-1 Test Duration. It is not uncommon for a normal subject with prior visual field experience to complete an SITA Standard Humphrey Program 24-2 in less than 5 minutes. Both lack of experience and localized scotomas increase test time. Nevertheless, it is unusual for reliable patients to require more than 10 minutes to complete Program 24-2. (Full threshold test times for 24-2 and 30-2 should take about 9 and 12 minutes, respectively.) 4-4-1-2 Fixation Losses. Fixation losses, as measured by blind spot checking, can occur with wandering gaze but can also result from a displaced blind spot or from a high rate of false-positive answers (see Section 3-4-4). Wandering gaze may result in peripheral threshold values as high as foveal sensitivity values. Further, an acceptable rate of fixation losses does not ensure that fixation was good (see Section 3-4-4 for an explanation of this apparent paradox). Evaluation of the quality of fixation can be best determined by a dedicated and attentive operator, but gaze tracking software enables a graphic record of fixation at the bottom of the Humphrey Visual Field printout (Figure 4-3). 4-4-1-3 False-Positive Responses. High false-positive values (Figure 4-4) tend to occur if the patient repeatedly responds when no stimulus is presented and correlate with patient response to subthreshold stimuli. If a false-positive response occurs during a blind spot check, a fixation loss is recorded. False-positive responses tend to create the false impression that the field contains high sensitivity. The graytone display may reveal regions without any gray stippling (decibel values higher than 41 dB), which are clearly nonphysiologic. These are known as white scotomas or a moth-eaten visual field.

Figure 4-3. Humphrey visual field analyzer software graphically displays variations in gaze for a semiquantitative record of visual fixation stability; a smooth line indicates steady fixation. Patients A and B both scored 3/14 fixation losses, but Patient B had more and larger eye movements.

Figure 4-4. High false-positive errors. Although the testing parameters identified only 6% false-positive results, note the frequent midperipheral threshold values greater than 41 dB, which produce white scotomas. The general height correction of the total deviation by approximately 17 dB in this case produces a markedly abnormal pattern deviation plot. Because more than 15% of the test locations contain abnormally high sensitivity values, the glaucoma hemifield test (GHT) displays abnormally high sensitivity. In addition, the mean deviation is positive.

134

Automated Perimetry in Glaucoma

135

Frequent false-positive responses are seen in the total deviation plot as scattered positive numbers. These are not flagged on the total deviation probability plot, because they are not reduced below the 5th percentile. The pattern deviation values, however, are affected differently. Because the instrument adjusts the entire field by a value that sets the seventh-highest locus on the total deviation plot to 0, loci where the patient responded appropriately may become markedly reduced. These may reach statistically significant levels and be highlighted by probability symbols. Frequent false-positive responses tend to affect the global indices as well. MD becomes positive and the GHT alerts the interpreter that more than 15% of the test locations contain abnormally high sensitivity values (Figure 4-5).

Figure 4-5. High false-positive errors superimposed on glaucomatous defects in a full threshold test. This field, although identified as low patient reliability, still provides evidence of nerve fiber bundle defects. Despite the high rate of fixation losses and falsepositive errors, there is a recognizable superior arcuate defect ending in a superior nasal step. Because less than 15% of the locations in the total deviation plot are elevated above the 99.5% percentile, the GHT does not display abnormal high sensitivity and correctly identifies the field as outside normal limits. The inferior arcuate defect suggested on the pattern deviation plot may be partly an artifact based on the 6-dB correction of the total deviation plot by the general height index.

136

Visual Fields

In general, high false-positive responses are not physiologic and greatly affect the reliability of the field. Fortunately, this problem can often be mitigated with instruction and encouragement of the patient by the perimetrist.

4-4-1-4 False-Negative Responses. In most respects, the patient with high falsenegative responses is the opposite of the patient with high false-positive responses. Instead of being “trigger-happy,” the former patients do not respond to stimuli that are much brighter than already measured thresholds. This situation correlates with technician comments such as “patient tired” or “patient falling asleep.” These patients often become increasingly fatigued during the test, with a tendency toward clusters of reduction in sensitivity toward the edge of the field, the final area tested. The total deviation plots become falsely abnormal, while the pattern deviation plot may be artifactually improved. MD is shifted more negatively. GHT may produce the message outside normal limits or on the borderline or with a general reduction in sensitivity. An extreme example is the cloverleaf field, in which the patient responds to the four primary seed location determinations and then becomes progressively less responsive (Figure 4-6A and B). A major interpretation problem can occur because bona fide field loss secondary to disease tends to increase the remaining field fatigability. High false-negative responses can thus be physiologic and the field results reliable. Long programs can cause fatigue and reduce reliability. Critics of a fatigue-reducing program like FASTPAC or SITA Fast claim that it underestimates true defects, although perhaps it is the full threshold strategy that is overestimating defects by producing a perimetric “stress test.” Some clues may help the clinician distinguish between a normal field that looks abnormal because of patient fatigue or frustration and a truly abnormal field, with high false-negative responses as one of its features. True defects are more reproducible and tend to respect anatomic distributions (e.g., arcuate, paracentral), while pseudodefects tend to occur variably in the periphery. Either defect type can produce dense constriction with only a central island of vision remaining. Performance of a 10° program on a subsequent day or after the patient has had a break often disproves or confirms the defect’s presence (Figure 4-7A and 7B). Similarly, the patient can be retested with a less fatiguing test such SITA Fast or with manual kinetic perimetry. Finally, correlation with the clinical examination of the optic nerve head, nerve fiber layer, and nerve fiber layer analysis cannot be overemphasized. 4-4-1-5 STF. STF on the full threshold algorithm, or intratest fluctuation as measured on the full threshold algorithm, can be caused by an inconsistent patient or by true visual field loss that causes the patient to tire easily. If the STF value in the global indices is flagged with a probability symbol, a careful inspection of the actual threshold data may reveal the cause (see Figure 4-8A and B) 4-4-1-6 Stimuli Number. The number of stimuli needed to complete an examination can be used to estimate patient consistency. An abnormal field may require an

A Figure 4-6. Apparent dense peripheral constriction, actually due to patient fatigue (A, B). The cloverleaf pattern develops in some patients because the testing algorithm begins by evaluating the center of each of the four quadrants of the field. Patients often initially pay attention and thresholds are high in the center of each quadrant but later, as fatigue prevails, thresholds in each quadrant decline and the field develops a nonphysiologic pattern of depression.

137

138

Visual Fields

B Figure 4-6. (Continued)

increased number of questions to complete the examination. If the patient is inconsistent, the testing strategy requires the presentation of more stimuli to arrive at a final threshold value. Even with markedly abnormal fields, it is unusual for a reliable patient to complete a full threshold central 30° (24°) examination with more than a total of 550 (450) questions.

A Figure 4-7. True dense peripheral constriction. (A) The results of this patient’s Program 24-2 suggested marked constriction. (B) When the patient was retested with Program 10-2, the defect was confirmed; there was only a small remaining central island of vision.

139

140

Visual Fields

B Figure 4-7. (Continued)

4-4-2 Criteria for Abnormality. Once armed with a solid understanding of the methods by which the automated perimeter tests and analyzes the field, as well as familiarity with artifacts and other confounding factors, the clinician is much better equipped to interpret the results. Often, interpretation is relatively easy, as in an ocular hypertensive patient with mildly elevated intraocular pressure, a normal-appearing

Automated Perimetry in Glaucoma

141

A Figure 4-8. Learning effect recognized. (A) In this series of fields, significant learning occurs between the first and second fields. (B) This improvement is seen as an upward shift in box location on the change analysis printout. The mean deviation also improves greatly from −10.69 dB to −5.07 dB between the first and second tests. (C) The glaucoma change probability function of Auto STATPAC averages the second and third field for a baseline. Because the first field is deleted, the mean deviation (MD) slope is reported as a modified MD slope.

optic disc, normal retinal nerve fiber layer analysis, and a completely unremarkable, reliably performed test. Interpreting moderately abnormal fields is also fairly simple, especially when they conform to classic patterns of loss, such as a sharply demarcated arcuate scotoma ending in a nasal step. More difficult is the patient who exhibits an abnormal disc and an abnormal field, with evidence of reduced reliability and other confounding factors (e.g., cataract,

142

Visual Fields

B

C Figure 4-8. (Continued)

retinovascular abnormality, prior retinal laser treatment). Although recognizing that a field is not normal may be reasonably straightforward, classifying the amount of damage attributable to a specific disorder may pose considerable difficulty. Equally difficult can be determining a subtle abnormality in a borderline clinical case. When attempting to classify precisely the degree of abnormality, clinicians should

Automated Perimetry in Glaucoma

143

TABLE 4-1. Minimal Criteria for Grading Abnormality (Central 30°) Strict ≥ 4 adjacent points of ≥ 5 dB loss* each ≥ 3 adjacent points of ≥ 10 dB loss* each Difference of ≥ 10 dB across nasal horizontal meridian at ≥ 3 adjacent points Exclusions: physiologic blind spot; superior and inferior rows Moderate ≥ 3 adjacent points of ≥ 5 dB loss* each ≥ 2 adjacent points of ≥ 10 dB loss* each Difference of ≥ 10 dB across nasal horizontal meridian at ≥ 2 adjacent points Exclusions: physiologic blind spot; superior and inferior rows Liberal ≥ 2 adjacent points of ≥ 5 dB loss* each ≥ 1 adjacent points of ≥ 10 dB loss* each Difference of ≥ 5 dB across nasal horizontal meridian at ≥ 2 adjacent points Exclusions: physiologic blind spot; superior and inferior rows *Loss is relative to normal values or to values of surrounding points. For probability maps that compare measured thresholds to normal values. Substitute P