Ophthalmology, Genetics

Essentials in Ophthalmology Pediatric Ophthalmology, Neuro-Ophthalmology, Genetics

B. Lorenz M. C. Brodsky Editors Essentials in Ophthalmology Glaucoma

G. K. Krieglstein R. N. Weinreb Cataract and Refractive Surgery Series Editors Uveitis and Immunological Disorders

Vitreo-retinal Surgery

Medical Retina

Oculoplastics and Orbit

Pediatric Ophthalmology, Neuro-Ophthalmology, Genetics

Cornea and External Eye Disease Editors Birgit Lorenz Michael C. Brodsky

Pediatric Ophthalmology, Neuro- Ophthalmology, Genetics

Strabismus - New Concepts in Pathophysiology, Diagnosis, and Treatment Series Editors Volume Editors Günter K. Krieglstein, MD Birgit Lorenz, MD Professor and Chairman Professor of Ophthalmology Department of Ophthalmology Klinik und Poliklinik für University of Cologne Augenheilkunde Joseph-Stelzmann-Straße 9 Justus-Liebig-University 50931 Köln UKGM GmbH Giessen Campus Germany Friedrichstraβe 18 35392 Giessen Robert N. Weinreb, MD Germany Professor and Director Hamilton Glaucoma Center Michael C. Brodsky, MD Department of Ophthalmology Professor of Ophthalmology and Neurology University of California at San Diego Mayo Clinic 9500 Gilman Drive Department of Ophthalmology La Jolla, CA 92093-0946 200 First Street SW USA Rochester, MN 55905 USA

ISBN: 978-3-540-85850-8 e-ISBN: 978-3-540-85851-5

DOI: 10.1007/978-3-540-85851-5

Library of Congress Control Number: 2009938957

Th is work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, speciﬁ cally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microﬁ lm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law.

Th e use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a speciﬁ c statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Product liability: Th e publishers cannot guarantee the accuracy of any information about dosage and application contained in this book. In every individual case the user must check such information by consulting the relevant literature.

Cover design: wmx-Design, Heidelberg

Printed on acid-free paper

9 8 7 6 5 4 3 2 1

(www.springer.com) Foreword

Th e Essentials in Ophthalmology series represents an ership acceptance of the fi rst two series, each of eight unique updating publication on the progress in all sub- volumes. Th is is a success that was made possible pre- specialties of ophthalmology. dominantly by the numerous opinion-leading authors In a quarterly rhythm, eight issues are published cov- and the outstanding section editors, as well as with the ering clinically relevant achievements in the whole fi eld constructive support of the publisher. Th ere are many of ophthalmology. Th is timely transfer of advancements good reasons to continue and still improve the dissemina- for the best possible care of our eye patients has proven to tion of this didactic and clinically relevant information. be eff ective. Th e initial working hypothesis of providing new knowledge immediately following publication in the peer-reviewed journal and not waiting for the textbook appears to be highly workable. G.K. Krieglstein We are now in the third cycle of the Essentials in R.N. Weinreb Ophthalmology series, having been encouraged by read- Series Editors Preface

Th e fi eld of strabismology has long suff ered from a dis- philosophy of strabismus that integrates new concepts of crepancy between its levels of sophistication in practice pathogenesis into the clinic. and theory. Although its diagnostic and therapeutic arma- Th is book provides a compendium of chapters that mentarium has become quite advanced, the scientifi c highlight new ideas in the fi eld of strabismus. We have understanding of disease pathogenesis has remained rudi- assembled an international panel of contributors who mentary. Consequently, educational training in strabismus have advanced our understanding of strabismus patho- diagnosis and treatment has become a didactic exercise in genesis. Some chapters are new while others are derived “learning the rules.” from recent seminal articles that have challenged our Recent advances in epidemiology, neuroimaging, understanding of strabismus diagnosis and treatment. genetics, and neurobiology have revolutionized our Original sources for these chapters are appropriately understanding of strabismus. Conceptualizing strabis- acknowledged. We thank our innovative authors for their mus within an evolutionary framework has advanced our important contributions, and hope that the reader fi nds understanding of why it arises and provided new clues to this edition both stimulating and enlightening. its neurological underpinnings. As new information is Birgit Lorenz consolidated, we are beginning to formulate a unifi ed Michael C. Brodsky Contents

Chapter 1 2.1.3 Muscle Length Adaptation ...... 12 Epidemiology of Pediatric Strabismus 2.2 Modeling the Binocular Amy E. Green-Simms and Brian G. Mohney Alignment Control System...... 13 2.2.1 Breakdown of the Binocular 1.1 Introduction ...... 1 Alignment Control System...... 14 1.2 Forms of Pediatric Strabismus ...... 1 2.2.2 Clariﬁ cation of Unanswered 1.2.1 Esodeviations ...... 1 Questions Regarding the 1.2.1.1 Congenital Esotropia ...... 2 Long-Term Binocular Alignment 1.2.1.2 Accommodative Esotropia...... 2 Control System...... 14 1.2.1.3 Acquired Nonaccommodative 2.2.3 Changes in Strabismus as Esotropia ...... 2 Bilateral Phenomena ...... 14 1.2.1.4 Abnormal Central Nervous System 2.2.4 Changes in Basic Muscle Length ...... 15 Esotropia ...... 2 2.2.5 Version Stimulation and 1.2.1.5 Sensory Esotropia ...... 2 Vergence Stimulation ...... 16 1.2.2 Exodeviations ...... 3 2.2.6 Evidence Against the “Final 1.2.2.1 Intermittent Exotropia...... 3 Common Pathway”...... 17 1.2.2.2 Congenital Exotropia ...... 3 2.3 Changes in Strabismus ...... 18 1.2.2.3 Convergence Insuﬃ ciency...... 3 2.3.1 Diagnostic Occlusion: And the 1.2.2.4 Abnormal Central Nervous System Hazard of Prolonged Occlusion ...... 19 Exotropia ...... 3 2.3.2 Unilateral Changes in Strabismus ...... 19 1.2.2.5 Sensory Exotropia...... 3 2.3.2.1 Supporting Evidence for Bilateral 1.2.3 Hyperdeviations ...... 3 Feedback Control of Muscle Lengths. . . . 19 1.3 Strabismus and Associated 2.4 Applications of Bilateral Feedback Conditions ...... 4 Control to Clinical Practice and 1.4 Changing Trends in Strabismus to Future Research ...... 21 Epidemiology ...... 4 References ...... 22 1.4.1 Changes in Strabismus Prevalence ...... 4 1.4.2 Changes in Strabismus Surgery Rates . . . 4 Chapter 3 1.5 Worldwide Incidence and Prevalence A Dissociated Pathogenesis of Childhood Strabismus ...... 4 for Infantile Esotropia 1.6 Incidence of Adult Strabismus ...... 7 Michael C. Brodsky References ...... 7 3.1 Dissociated Eye Movements ...... 25 Chapter 2 3.2 Tonus and its relationship Changes in Strabismus Over Time: The Roles of to infantile esotropia ...... 25 Vergence Tonus and Muscle Length Adaptation 3.3 Esotropia and Exotropia as David L. Guyton a Continuum...... 26 3.4 Distinguishing Esotonus 2.1 Binocular Alignment System...... 11 from Convergence ...... 28 2.1.1 Long-Term Maintenance 3.5 Pathogenetic Role of Dissociated of Binocular Alignment...... 11 Eye Movements in Infantile Esotropia . . . 29 2.1.2 Vergence Adaptation...... 12 References ...... 30 x Contents

Chapter 4 5.1.10 Binocular Connections Join The Monoﬁ xation Syndrome: New Monocular Compartments Within Considerations on Pathophysiology Area V1 (Striate Cortex)...... 44 Kyle Arnoldi 5.1.11 Too Few Cortical Binocular Connections in Strabismic Primate...... 46 4.1 Introduction ...... 33 5.1.12 Projections from Striate Cortex 4.2 Normal and Anomalous (Area V1) to Extrastriate Cortex Binocular Vision ...... 33 (Areas MT/MST) ...... 46 4.2.1 Binocular Correspondence: 5.1.13 Inter-Ocular Suppression Anomalous, Normal, or Both?...... 34 Rather than Cooperation 4.3 MFS with Manifest Strabismus ...... 35 in Strabismic Cortex ...... 46 4.3.1 Esotropia is the Most Common 5.1.14 Naso-Temporal Inequalities form of MFS...... 35 of Cortical Suppression...... 47 4.3.2 Esotropia Allows for Better 5.1.15 Persistent Nasalward Binocular Vision ...... 35 Visuomotor Biases in Strabismic 4.3.3 Esotropia is the Most Stable Form...... 36 Primate ...... 47 4.4 Repairing and Producing MFS ...... 36 5.1.16 Repair of Strabismic Human 4.4.1 Animal Models for the Study Infants: The Historical Controversy ...... 50 of MFS ...... 37 5.1.17 Repair of High-grade Fusion 4.5 Primary MFS (Sensory Signs of is Possible...... 50 Infantile-Onset Image Decorrelation) . . . 38 5.1.18 Timely Restoraion of Correclated 4.5.1 Motor Signs of Infantile-Onset Binocular Input: The Key to Repair ...... 50 Image Decorrelation ...... 38 5.2 Visual Cortex Mechanisms References ...... 39 in Micro-Esotropia (Monoﬁ xation Syndrome) ...... 51 5.2.1 Neuroanatomic Findings in Chapter 5 Area V1 of Micro-Esotropic Primates . . . . 52 Visual Cortex Mechanisms of Strabismus: 5.2.2 Extrastriate Cortex in Development and Maldevelopment Micro-Esotropa...... 52 Lawrence Tychsen References ...... 54

5.1 Esotropia as the Major Type Chapter 6 of Developmental Strabismus ...... 41 Neuroanatomical Strabismus 5.1.1 Early-Onset (Infantile) Esotropia ...... 41 Joseph L. Demer 5.1.2 Early Cerebral Damage as the Major Risk Factor ...... 41 6.1 General Etiologies of Strabismus...... 59 5.1.3 Cytotoxic Insults to Cerebral Fibers...... 42 6.2 Extraocular Myopathy ...... 59 5.1.4 Genetic Inﬂ uences on 6.2.1 Primary EOM Myopathy ...... 59 Formation of Cerebral Connections . . . . . 42 6.2.2 Immune Myopathy...... 60 5.1.5 Development of Binocular 6.2.3 Inﬂ ammatory Myositis...... 61 Visuomotor Behavior 6.2.4 Neoplastic Myositis...... 61 in Normal Infants...... 42 6.2.5 Traumatic Myopathy ...... 61 5.1.6 Development of Sensorial 6.3 Congenital Pulley Heterotopy ...... 62 Fusion and Stereopsis ...... 43 6.4 Acquired Pulley Heterotopy ...... 63 5.1.7 Development of Fusional 6.5 “Divergence Paralysis” Esotropia ...... 64 Vergence and an Innate 6.5.1 Vertical Strabismus Due to Convergence Bias ...... 44 Sagging Eye Syndrome ...... 65 5.1.8 Development of Motion 6.5.2 Postsurgical and Traumatic Pulley Sensitivity and Conjugate Heterotopy ...... 65 Eye Tracking (Pursuit/OKN) ...... 44 6.5.3 Axial High Myopia...... 65 5.1.9 Development and 6.6 Congenital Peripheral Neuropathy: Maldevelopment of Cortical The Congenital Cranial Binocular Connections ...... 44 Dysinnervation Disorders (CCDDs) ...... 66 Contents xi

6.6.1 Congenital Oculomotor (CN3) Palsy. . . . . 67 Chapter 8 6.6.2 Congenital Fibrosis of the The Value of Screening for Amblyopia Revisited Extraocular Muscles (CFEOM) ...... 67 Jill Carlton and Carolyn Czoski-Murray 6.6.3 Congenital Trochlear (CN4) Palsy...... 69 6.6.4 Duane’s Retraction 8.1 Amblyopia ...... 95 Syndrome (DRS)...... 69 8.2 What Is Screening? ...... 96 6.6.5 Moebius Syndrome ...... 70 8.2.1 Screening for Amblyopia, 6.7 Acquired Motor Neuropathy...... 71 Strabismus, and/or 6.7.1 Oculomotor Palsy ...... 71 Refractive Errors...... 96 6.7.2 Trochlear Palsy ...... 71 8.2.1.1 Screening for Amblyopia ...... 97 6.7.3 Abducens Palsy ...... 71 8.2.1.2 Screening for Strabismus ...... 97 6.7.4 Inferior Oblique (IO) Palsy ...... 71 8.2.1.3 Screening for Refractive Error...... 97 6.8 Central Abnormalities 8.2.1.4 Screening for Other Ocular Conditions . . 97 of Vergence and Gaze ...... 72 8.2.2 Diff erence Between a Screening 6.8.1 Developmental Esotropia and Diagnostic Test ...... 97 and Exotropia ...... 72 8.2.3 Justifi cation for Screening for 6.8.2 Cerebellar Disease...... 72 Amblyopia and/or Strabismus ...... 98 6.8.3 Horizontal Gaze Palsy and 8.2.4 Recent Reports Examining Progressive Scoliosis ...... 72 Pre-School Vision Screening ...... 98 References ...... 72 8.3 Screening Tests for Amblyopia, Strabismus, and/or Refractive Error...... 100 Chapter 7 8.3.1 Vision Tests ...... 100 Congenital Cranial Dysinnervation Disorders: 8.3.2 Cover-Uncover Test...... 100 Facts and Perspectives to Understand Ocular 8.3.3 Stereoacuity ...... 101 Motility Disorders 8.3.4 Photoscreening and/or Antje Neugebauer and Julia Fricke Autorefraction ...... 101 8.3.5 What to Do with Those Who 7.1 Congenital Cranial Dysinnervation Are Unable to Perform Disorders: Facts About Ocular Screening Tests?...... 102 Motility Disorders ...... 77 8.3.6 Who Should Administer 7.1.1 The Concept of CCDDs: the Screening Program? ...... 102 Ocular Motility Disorders as 8.4 Treatment of Amblyopia...... 103 Neurodevelopmental Defects ...... 77 8.4.1 Type of Treatment...... 103 7.1.1.1 Brainstem and Cranial 8.4.2 Refractive Adaptation ...... 103 Nerve Development...... 78 8.4.3 Conventional Occlusion ...... 104 7.1.1.2 Single Disorders 8.4.4 Pharmacological Occlusion ...... 104 Representing CCDDs ...... 78 8.4.5 Optical Penalization ...... 104 7.1.1.3 Disorders Understood as CCDDs ...... 81 8.4.6 Eff ective Treatment of 7.2 Congenital Cranial Dysinnervation Amblyopia in Older Children Disorders: Perspectives to Understand (Over the Age of 7 Years)...... 104 Ocular Motility Disorders ...... 83 8.4.7 Treatment Compliance ...... 105 7.2.1 Congenital Ocular 8.4.8 Other Treatment Options Elevation Defi ciencies: A for Amblyopia...... 105 Neurodevelopmental View ...... 83 8.4.9 Recurrence of Amblyopia 7.2.1.1 Brown Syndrome...... 83 Following Therapy ...... 105 7.2.1.2 Congenital Monocular 8.5 Quality of Life ...... 106 Elevation Defi ciency and 8.5.1 The Impact of Amblyopia Vertical Retraction Syndrome...... 87 Upon HRQoL...... 106 7.2.2 A Model of some Congenital 8.5.2 Stereoacuity and Motor Skills Elevation Defi ciencies as in Children with Amblyopia...... 106 Neurodevelopmental Diseases ...... 89 8.5.3 Reading Speed and Reading References ...... 91 Ability in Children with Amblyopia...... 106 xii Contents

8.5.4 Impact of Amblyopia 10.2.4 Pharmacological Therapy Upon Education...... 106 Combined with a Plano Lens...... 130 8.5.5 Emotional Well-Being and 10.3 Other Treatment Issues ...... 131 Amblyopia ...... 107 10.3.1 Bilateral Refractive Amblyopia ...... 131 8.5.6 The Impact of Strabismus 10.3.2 Age Eff ect...... 131 Upon HRQoL...... 107 10.3.3 Maintenance Therapy ...... 131 8.5.7 Critique of HRQoL Issues 10.3.4 Long-Term Persistence of in Amblyopia ...... 108 an Amblyopia Treatment Benefi t...... 132 8.5.8 The Impact of the Condition 10.4 Other Treatments ...... 132 or the Impact of Treatment? ...... 108 10.4.1 Filters...... 132 References ...... 109 10.4.2 Levodopa/Carbidopa Adjunctive Therapy ...... 133 Chapter 9 10.5 Controversy...... 133 The Brückner Test Revisited 10.5.1 Optic Neuropathy Rather than Amblyopia ...... 133 Michael Gräf References ...... 134 9.1 Amblyopia and Amblyogenic Disorders ...... 113 Chapter 11 Best Age for Surgery for Infantile Esotropia: 9.1.1 Early Detection of Amblyopia...... 113 Lessons from the Early vs. Late Infantile 9.1.2 Brückner’s Original Description ...... 114 Strabismus Surgery Study 9.2 Corneal Light Refl exes (First Purkinje Images)...... 114 H. J. Simonsz and G. H. Kolling 9.2.1 Physiology ...... 114 11.1 Introduction ...... 137 9.2.2 Performance ...... 115 11.1.1 Defi nition and Prevalence ...... 137 9.2.3 Shortcomings and Pitfalls ...... 115 11.1.2 Sensory or Motor Etiology ...... 137 9.3 Fundus Red Refl ex (Brückner Refl ex) . . . . 115 11.1.3 Pathogenesis: Lack of 9.3.1 Physiology ...... 116 Binocular Horizontal Connections 9.3.2 Performance ...... 119 in the Visual Cortex...... 138 9.3.3 Possibilities and Limitations ...... 120 11.1.4 History...... 138 9.4 Pupillary Light Refl exes...... 120 11.1.5 Outcome Parameters...... 138 9.4.1 Physiology ...... 121 11.2 Outcome of Surgery in the ELISSS...... 139 9.4.2 Performance ...... 121 11.2.1 Reasons for the ELISSS...... 139 9.4.3 Possibilities and Limitations ...... 121 11.2.2 Summarized Methods of the ELISSS. . . . . 139 9.5 Eye Movements with Alternating 11.2.3 Summarized Results of the ELISSS ...... 140 Illumination of the Pupils ...... 122 11.2.4 Binocular Vision at Age Six...... 140 References ...... 122 11.2.5 Horizontal Angle of Strabismus at Age Six ...... 140 Chapter 10 11.2.6 Alignment is Associated Amblyopia Treatment 2009 with Binocular Vision ...... 141 Michael X. Repka 11.3 Number of Operations and Spontaneous Reduction into 10.1 Amblyopia Treatment 2009...... 125 Microstrabismus Without Surgery...... 142 10.1.1 Introduction ...... 125 11.3.1 The Number of Operations Per Child and 10.1.2 Epidemiology ...... 125 the Reoperation Rate in the ELISSS...... 142 10.1.3 Clinical Features of Amblyopia...... 126 11.3.2 Reported Reoperation Rates...... 142 10.1.4 Diagnosis of Amblyopia ...... 126 11.3.3 Test-Retest Reliability Studies...... 144 10.1.5 Natural History...... 127 11.3.4 Relation Between the Postoperative 10.2 Amblyopia Management ...... 127 Angle of Strabismus and the 10.2.1 Refractive Correction ...... 127 Reoperation Rate...... 145 10.2.2 Occlusion by Patching...... 128 11.3.5 Scheduled for Surgery, but no 10.2.3 Pharmacological Treatment Surgery Done at the End of the with Atropine ...... 129 Study at the Age of Six Years...... 145 Contents xiii

11.3.6 Spontaneous Reduction 12.3.6.2 Management of Vertical AHP ...... 166 of the Angle...... 146 12.3.6.3 Management of Head Tilt...... 167 11.3.7 Predictors of Spontaneous 12.3.6.4 Artifi cial Divergence Surgery ...... 167 Reduction into Microstrabismus ...... 146 12.3.6.5 Surgery to Decrease the 11.3.8 Random-Eff ects Model Intensity of Nystagmus ...... 168 Predicting the Angle and References ...... 169 its Variation ...... 146 Appendix ...... 149 Chapter 13 References ...... 149 Surgical Management of Dissociated Deviations Susana Gamio Chapter 12 Management of Congenital Nystagmus 13.1 Dissociated Deviations ...... 174 with and without Strabismus 13.2 Surgical Alternatives to Treat Anil Kumar, Frank A. Proudlock, and Irene Gottlob Patients with DVD ...... 175 13.2.1 Symmetric DVD with Good Bilateral 12.1 Overview ...... 154 Visual Acuity, with No Oblique 12.1.1 Congenital Nystagmus with Muscles Dysfunction ...... 175 and Without Sensory Defi cits ...... 154 13.2.2 Bilateral DVD with Deep 12.1.1.1 The Clinical Characteristics Unilateral Amblyopia...... 175 of Congenital Nystagmus...... 156 13.2.3 DVD with Inferior Oblique 12.1.2 Manifest Latent Nystagmus (MLN) ...... 157 Overaction (IOOA) and V Pattern...... 176 12.1.2.1 Clinical Characteristics 13.2.4 DVD with Superior Oblique of Manifest Latent Nystagmus (MLN). . . . 157 Overaction (SOOA) and A Pattern ...... 177 12.1.3 Congenital Periodic Alternating 13.2.5 Symmetric vs. Asymmetric Nystagmus (PAN)...... 158 Surgeries for DVD ...... 178 12.1.3.1 Clinical characteristics 13.2.6 DVD with Hypotropia of the of congenital periodic Nonfi xating Eye ...... 178 alternating nystagmus ...... 159 13.3 Dissociated Horizontal Deviation ...... 179 12.2 Compensatory Mechanisms ...... 160 13.4 Dissociated Torsional Deviation. 12.2.1 Dampening by Versions ...... 160 Head tilts in patients with 12.2.2 Dampening by Vergence ...... 160 Dissociated Strabismus...... 180 12.2.3 Anomalous Head Posture (AHP) ...... 160 13.5 Conclusions...... 182 12.2.3.4 Measurement of AHP...... 160 References ...... 182 12.2.3.5 Eff ect of Monocular and Binocular Visual Acuity Chapter 14 Testing on AHP...... 161 Surgical Implications of the 12.2.3.6 Testing AHP at Near ...... 162 Superior Oblique Frenulum 12.2.3.7 The Eff ect of Straightening Burton J. Kushner and Megumi Iizuka the Head in Patients with AHP ...... 162 12.3 Treatment ...... 162 14.1 Introduction ...... 185 12.3.1 Optical Treatment ...... 162 14.2 Clinical and Theoretical 12.3.1.1 Refractive Correction ...... 162 Investigations...... 186 12.3.1.2 Spectacles and Contact 14.2.1 The Eff ect of Superior Rectus Muscle Lenses (CL)...... 162 Recession on the Location of the 12.3.1.3 Prisms ...... 163 Superior Oblique Tendon Before 12.3.1.4 Low Visual Aids...... 163 and After Cutting the Frenulum...... 186 12.3.2 Medication...... 163 14.2.2 The Eff ect of the Frenulum 12.3.3 Acupuncture...... 164 on Superior Oblique Recession 12.3.4 Biofeedback ...... 164 Using a Suspension Technique...... 188 12.3.5 Botulinum Toxin-A (Botox)...... 164 14.2.3 The Theoretical Eff ect of the Superior 12.3.6 Surgical Treatment of Congenital Oblique Frenulum on the Posterior Partial Nystagmus...... 164 Tenectomy of the Superior Oblique ...... 189 12.3.6.1 Management of Horizontal AHP ...... 165 References ...... 192 xiv Contents

Chapter 15 16.2 Natural History...... 212 Pearls and Pitfalls in Surgical 16.3 Treatment of GO ...... 213 Management of Paralytic Strabismus 16.3.1 Active Infl ammatory Phase ...... 213 Seyhan B. Özkan 16.3.1.1 Glucocorticoid Treatment ...... 213 16.3.1.2 Orbital Radiotherapy ...... 213 15.1 General Principles of Surgical 16.3.1.3 Combined Therapy: Glucocorticoids Treatment in Paralytic and Orbital Radiotherapy...... 213 Strabismus...... 195 16.3.1.4 Other Immunosuppressive Treatments 15.1.1 Aims of Treatment...... 195 and New Developments...... 213 15.1.2 Timing of Surgery ...... 195 16.3.1.5 Therapy of Dysthyroid Optic 15.1.3 Preoperative Assessment ...... 196 Neuropathy [DON] and 15.1.4 Methods of Surgical Treatment ...... 197 Sight-Threatening Corneal 15.2 Third Nerve Palsy...... 198 Breakdown ...... 214 15.2.1 Complete Third Nerve Palsy ...... 198 16.3.1.6 Other Simple Measures that 15.2.2 Incomplete Third Nerve Palsy ...... 199 may Alleviate Symptoms ...... 214 15.3 Fourth Nerve Palsy ...... 200 16.3.2 Inactive Disease Stages...... 215 15.4 Sixth Nerve Palsy...... 204 16.3.2.1 Orbital Decompression...... 215 References ...... 205 16.3.2.2 Extraocular Muscle Surgery...... 216 16.3.2.3 Lid Surgery ...... 217 Chapter 16 Modern Treatment Concepts 16.4 Thyroid Dysfunction and GO...... 220 in Graves Disease 16.4.1 Association Between Treatment of Hyperthyroidism and Course of GO . . . . . 220 Anja Eckstein and Joachim Esser 16.4.2 Relationship Between 16.1 Graves Orbitopathy (GO): TSH-Receptor-Antibody (TRAb) Pathogenesis and Clinical Signs...... 207 Levels and Orbitopathy...... 220 16.1.1 Graves Orbitopathy is Part of a Systemic 16.5 Environmental and Genetic Disease: Graves Disease (GD) ...... 207 Infl uence on the Course of GO ...... 221 16.1.2 Graves Orbitopathy−Clinical Signs ...... 208 16.5.1 Relationship Between Cigarette 16.1.2.1 Clinical Changes Result in Smoking and Graves Orbitopathy...... 221 Typical Symptoms...... 208 16.5.2 Genetic Susceptibility ...... 221 16.1.3 Clinical Examination of GO ...... 208 16.6 Special Situations ...... 222 16.1.3.1 Signs of Activity ...... 208 16.6.1 Euthyroid GO ...... 222 16.1.3.2 Assessing Severity of GO ...... 209 16.6.2 Childhood GO...... 222 16.1.3.3 Imaging ...... 211 16.6.3 GO and Diabetes ...... 222 16.1.4 Classifi cation of GO...... 211 References ...... 223 Contributors

Kyle Arnoldi Susana Gamio Ross Eye Institute Gallo 1330, Ricardo Gutierrez Children’s Hospital, Department of Ophthalmology, Matienzo 1731 First Floor E, University at Buﬀ alo, Ross Eye Institute, Buenos Aires, Captial Fedral 1426, 1176 Main Street, NY, 14209, USA Argentina, South America

Michael C. Brodsky Irene Gottlob Departments of Ophthalmology and Neurology, Department of Ophthalmology, Mayo Clinic 200 First Street, SW Rochester, Ricardo Gutiérrez Children’s Hospital, Buenos Aires, MN 55905, USA Argentina Jill Carlton Health Economics and Decision Science, Michael Gräf CHARR, University of Sheﬃ eld, Regent Court, Department of Ophthalmology, Justus-Liebig-University 30 Regent Street, Sheﬃ eld, Giessen, Giessen Campus, Friedrichstraβe 18, 35385 S1 4DA, UK Giessen, Germany

Carolyn Czoski-Murray Amy E. Greenberg Leeds Institute of Health Sciences, Department of Ophthalmology, Mayo Clinic, University of Leeds, 200 First Street Southwest, Rochester, MN 55905, USA Room 1.26, 6 Charles Th ackrah Building, 101 Clarendon Road, David L. Guyton Leeds LS2 9LJ, UK Th e Krieger Children’s Eye Center at the Wilmer Institute, Joseph L. Demer Th e Johns Hopkins University School of Medicine, Baltimore, MD 21287-9028, USA Jules Stein Eye Institute, 100 Stein Plaza, UCLA, Box 957002, Megumi Iizuka Los Angeles, CA 90095-7002, USA University of Toronto, St. Michael’s Hospital, 61 Queen Street East, 8th Floor, Care of the Eye Clinic, Anja Eckstein Toronto, ON, Canada M5C 2T2 University Eye Hospital, Hufelandstraβe 55, 45122 Essen, Germany Gerold H. Kolling Marinus J.C. Eijkemans Department of Ophthalmology, University Clinic Department of Public Health, Erasmus Medical Center, Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, PO Box 2040, 3000 CA, Rotterdam, Germany Th e Netherlands A. S. Anil Kumar Julia Fricke Department of Ophthalmology, Department of Ophthalmology, Kerpener Straβe 62, University of Leicester, 50937 Köln, Germany UK xvi Contributors

Burton J. Kushner Frank A. Proudlock Department of Ophthalmology and Visual Sciences, Department of Ophthalmology, 2870 University Avenue, Suite 206, Madison, WI 53705, Ricardo Gutiérrez Children’s Hospital, Buenos Aires, USA Argentina

Birgit Lorenz Michael X. Repka Department of Ophthalmology, Johns Hopkins University School of Medicine, Justus-Liebig-University Giessen Wilmer 233, Johns Hopkins Hospital, Giessen Campus 600 North Wolfe Street, Baltimore, MD 21287-9028, USA Friedrichstraβe 18, 35392 Giessen Huibert J. Simonsz Germany Department of Ophthalmology, Erasmus Medical Center, PO Box 2040, 3000 CA Rotterdam, Th e Netherlands Brian G. Mohney Department of Ophthalmology, Mayo Clinic, Lawrence Tychsen 200 First Street Southwest, Rochester, MN 55905, USA St Louis Children’s Hospital at Washington University Medical Center, 1 Children’s Place, St Louis, MO 63110, Antje Neugebauer USA Department of Ophthalmology, Kerpener Straβe 62, 50937 Köln, Germany

Seyhan B. Özkan Guzelhisar Mah. 35. sok. No: 8/A, 09010 Aydin, Turkey Chapter 1

Epidemiology of Pediatric Strabismus 1 Amy E. Green-Simms and Brian G. Mohney

Core Messages ■ Recognition and diagnosis of the individual and the most commonly diagnosed form of forms of childhood strabismus are important for exodeviation worldwide. the best preservation of visual function. ■ Hyperdeviations are uncommon, with fourth ■ Esotropia is the most common form of pediatric cranial nerve palsy being the most prevalent ocular deviation in the West, whereas exotropia etiology. predominates in the East. ■ Major independent risk factors associated with ■ Accommodative esotropia is the most prevalent strabismus development include: prematurity, form of strabismus in the West, comprising half central nervous system (CNS) impairment, low of all esodeviations. birth weight, family history, and refractive error. ■ Congenital, or infantile, esotropia accounts for ■ Recent studies have reported a decline in the less than 10% of all pediatric esotropia, a ﬁ gure number of surgeries performed for strabismus; much smaller than once widely believed. however, population-based data of congenital ■ Intermittent exotropia is the second most com- esotropia in the United States conﬁ rms a more mon form of childhood strabismus in the West stable rate.

solely on tropic deviations rather than phorias and will 1.1 Introduction encompass worldwide incidence and prevalence as well Strabismus, or squint, is a disorder of ocular alignment. Th is as clinical characteristics of the various strabismus overarching term may be further characterized by the direc- subtypes. tion of the misalignment: the prefi x eso- describes an inward ocular deviation; exo-, an outward deviation; and hyper-, a vertical deviation. Descriptive suffi xes include -tropia, a 1.2 Forms of Pediatric Strabismus manifest deviation in which fusional control is not present, and -phoria, a latent deviation that is controlled by fusion. 1.2.1 Esodeviations Strabismus detection, classifi cation, and treatment are especially important in pediatric populations as strabis- Esodeviations are characterized by an intermittent or mus is a leading factor in the development of amblyopia, constant inward deviation of the eye or eyes (Fig. 1.1). or a loss in visual function resulting from inadequate or Esotropia comprises approximately 60% of all strabismus abnormal visual system stimulation. Th is strong connec- in the West [1] whereas only about 30% in the East [2]. In tion with amblyopia diff erentiates pediatric from adult- the United States, children are diagnosed with esotropia onset strabismus, wherein vision and stereopsis are less at a mean age of 3.1 years [3], and 90% of esodeviations likely to be irreversibly harmed. In children, strabismus occur by 5 years of age [4]. Esotropia is more commonly should be corrected to decrease the occurrence of ambly- associated with amblyopia than either exo- or hypertro- opia, to maximize the potential for stereopsis, and to pia, occurring in one of three esotropic children vs. 1 of straighten the visual axes of the eyes. 12 exo- or hypertropic children [5]. Th ere is no signifi - Th is chapter will review recent data on the epidemiol- cant gender predilection among any of the following sub- ogy of pediatric strabismus. Th e information will focus types of childhood esotropia. 2 1 Epidemiology of Pediatric Strabismus

1.2.1.3 Acquired Nonaccommodative Esotropia Acquired nonaccommodative esotropia deﬁ nes children 1 whose deviation develops aft er 6 months of age and is not associated with accommodative eﬀ ort. Th is subtype has typically been thought of as uncommon and as portend- ing underlying neurological disease. However, a recent population-based study showed that it is the second most common form of childhood esotropia [3], with an incidence of 1 in 257 children and is rarely the result of neurologic disease [8].

Fig. 1.1 A child with esotropia 1.2.1.4 Abnormal Central Nervous System Esotropia 1.2.1.1 Congenital Esotropia Esotropic children with a developmental or neurologic Congenital esotropia, also known as infantile or essential disorder may be classiﬁ ed under central nervous system infantile esotropia, is generally deﬁ ned as a neurologically (CNS) defects regardless of the age at onset or form of intact child with a constant nonaccommodative esotropia esotropia. Th e most commonly associated conditions that develops by 6 months of age. Th is term is oft en con- include cerebral palsy, developmental delay, Down syn- fusing as children do not typically present at birth with drome, and seizure disorder. CNS-associated esotropia their deviation. Moreover, esotropia measuring up to 40 makes up approximately 10% of all diagnosed esodevia- prism diopters (PD) between weeks 4 and 20 of life has tions [3]. been reported to resolve in 27% of children [6]. Congenital esotropia has, for decades, been considered the most common form of strabismus. However, more recent reports have demonstrated that congenital esotro- 1.2.1.5 Sensory Esotropia pia is much less common than once believed. In a recent Sensory esotropia includes patients with a unilateral or incidence study among children born over a 30-year time bilateral ocular condition that prevents normal fusion. period in the US, 1 in 403 live births developed congenital Th is form of esodeviation is commonly associated with esotropia [7]. Other recent reports from the same popula- anisometropic amblyopia as well as with disorders of tion reported similar results, with infantile esotropia mak- deprivation such as cataract, corneal scarring, and retinal ing up only 8.1% of all forms of esotropia [3]. or optic nerve disorders [3].

1.2.1.2 Accommodative Esotropia Summary for the Clinician Accommodative esotropia is characterized by an acquired constant or intermittent deviation that is cor- ■ Accommodative esotropia comprises approxi- rected or reduced 10 PD or more aft er wearing hyper- mately half of all pediatric esotropia. opic spectacles full time for at least 3 weeks. Patients can ■ Acquired nonaccommodative esotropia is the further be classiﬁ ed as having fully accommodative second most common form of esodeviation in esotropia, in which the deviation is reduced to ≤8 PD, or the West and is rarely associated with neurologic partially accommodative esotropia, in which there is a disease. residual deviation of 10 or more PD. Accommodative ■ Congenital esotropia, once thought to be the esotropia, including both the partially and fully accom- most common esodeviation, makes up less than modative forms, comprises approximately one half of all 10% of all esotropia diagnosed in childhood. pediatric esotropia in the United States and is the most ■ Amblyopia occurs in one of three children with prevalent form of childhood strabismus in the West [3]. esotropia, a rate signiﬁ cantly higher than in chil- Th is form of esodeviation has been reported to occur in dren with either exotropia or hypertropia. 1 in 92 children [3]. 1.2 Forms of Pediatric Strabismus 3

exodeviation at near. It is the second most commonly 1.2.2 Exodeviations diagnosed type of exodeviation and comprises approxi- Exotropia is a disorder of ocular alignment characterized by mately one in ﬁ ve children with exotropia [9] with an an outward deviation of the eye or eyes (Fig. 1.2). Exotropia incidence of 1 in 411 children [9]. However, this disorder is less common than esotropia among Western populations is likely to be under-diagnosed given the obscure symp- [1]; however, it is the predominant form of strabismus in toms and relatively imperceptible nature of the deviation the East [2]. Regardless of the relative prevalence, the age at to outside observers. presentation for children with exotropia tends to be older than for those with esotropia [4]. Amblyopia is less commonly associated with exotropia than esotropia [5]. 1.2.2.4 Abnormal Central Nervous System Exotropia

1.2.2.1 Intermittent Exotropia Exotropic children with a congenital or acquired developmental or neurological disorder may be grouped under Intermittent exotropia is an acquired, intermittent devia- CNS defects regardless of the age at onset. Approximately, tion of 10 or more PD unassociated with other ocular, 15% of children with exotropia may have neurologic paralytic, or neurologic disorders. It is the second most abnormalities, most commonly cerebral palsy and devel- commonly diagnosed form of strabismus (at approxi- opmental delay [9]. mately 17%) in the United States [1] and the most commonly diagnosed subtype of exodeviation with an incidence of 1 in 155 children [9]. In a recent population- 1.2.2.5 Sensory Exotropia based study, it was reported to occur nearly twice as oft en in girls compared with boys [10]. Sensory exotropia includes children with a unilateral or bilateral ocular condition that prevents normal fusion, most commonly anisometropic amblyopia or cataract [9]. 1.2.2.2 Congenital Exotropia Children with sensory disturbances are more likely to develop exotropia (24 of 235 children, or 10.2%) than Congenital exotropia includes children with a constant esotropia (15 of 221 children, or 6.8%) [12]. Th is diff er- exodeviation that develops by 6 months of age. Although this ence may be in part due to the age at onset of visual condition is rare, many children will have associated neuro- impairment. Havertape and coauthors have shown that logic or other disorders and should undergo CNS imaging children with a unilateral or bilateral visual loss by 6 [11]. Th is form of exotropia results in amblyopia much more months of age are more likely to develop sensory esotro- oft en than other subtypes of divergent strabismus. pia, whereas those with an acquired visual loss are much more likely to develop sensory exotropia [13]. 1.2.2.3 Convergence Insuffi ciency Convergence insuffi ciency describes children who are Summary for the Clinician generally orthotropic at distance fi xation but whose eyes ■ Exotropia is the predominant form of strabismus do not converge suffi ciently at near fi xation, leaving an among Asian populations; however, it is less common than esotropia in the West. ■ Intermittent exotropia is the most commonly diagnosed form of exodeviation. ■ Amblyopia is less commonly associated with exotropia than esotropia.

1.2.3 Hyperdeviations Hypertropia, or a vertical displacement of one eye relative to the other, is the least diagnosed form of strabismus [1]. Nearly one-third of all cases are associated with fourth cranial nerve palsy (Fig. 1.3), corresponding to an Fig. 1.2 A child with exotropia incidence of 1 in 1,264 children [14]. Other causes of 4 1 Epidemiology of Pediatric Strabismus

a included children with CNS disorders or acquired nonaccommodative esotropia, distinct forms of early-onset esotropia that have been shown to occur more frequently than infantile esotropia. Acquired nonaccommodative 1 esotropia, on the other hand, appears to be relatively prevalent and is a form of esotropia that is much more likely to develop fusion and normal stereopsis with treatment [8]. Intermittent exotropia, the most common form of exodeviation, is more prevalent than any other form of strabismus in Asia and, as a result, may be the most prevalent form of strabismus worldwide. b

1.4.2 Changes in Strabismus Surgery Rates Th ere have been several reports from the United Kingdom describing a decrease in the incidence of strabismus or strabismus surgery in recent years [21–24]. Explanations for this decline have included the implementation of childhood vision screening programs and the more frequent correction of the full hyperopic refractive error. Contrasting data, however, has come from Louwagie et al.’s population-based cohort study reporting on the incidence of infantile esotropia as well as the incidence of Fig. 1.3 A child with left fourth nerve palsy showing, (a) right surgery for infantile esotropia in Rochester, Minnesota, head tilt and (b) left hypertropia with left head tilt US [7]. From 1965 through 1994, there was no signifi cant change in the numbers of children diagnosed with infantile esotropia, and there was no signifi cant change in the hypertropia include primary inferior oblique overaction, number of surgeries performed on these children. Brown syndrome, and CNS-associated hypertropia [14]. Summary for the Clinician 1.3 Strabismus and Associated Conditions ■ Congenital esotropia appears to be less prevalent A number of studies have demonstrated an association than previously believed, whereas other forms between prenatal and environmental factors and the devel- such as acquired nonaccommodative esotropia opment of strabismus. Signifi cant independent risk factors are relatively common. for strabismus include: family history, prematurity, low ■ Intermittent exotropia may be the most preva- birth weight, low Apgar scores (at 1 and 5 min), maternal lent form of strabismus worldwide. cigarette smoking, increasing maternal age, retinopathy of ■ Th e rate of pediatric strabismus surgery has recently prematurity, refractive error, and anisometropia [15–20]. been reported to be in decline; however, data from a population-based cohort of children with congenital esotropia in the United States found no 1.4 Changing Trends in Strabismus change in strabismus incidence or surgical rate over Epidemiology a 30-year period (1965–1994).

1.4.1 Changes in Strabismus Prevalence Our understanding of the prevalence of childhood stra- 1.5 Worldwide Incidence and Prevalence bismus continues to change. As discussed earlier, con- of Childhood Strabismus genital esotropia has recently been reported to occur less commonly than once widely believed, comprising only Recent reports describe the prevalence of pediatric stra- 8.1% of all diagnosed esodeviations [3]. Th e previously bismus as ranging from 0.12% in 1.5-year-old Japanese reported higher incidence of infantile esotropia may have children [25] to 20.1% in a cohort of low birth weight 1.5 Worldwide Incidence and Prevalence of Childhood Strabismus 5

Table 1.1. Pediatric strabismus prevalence rates by regions of the world

Reference Categorical Number Age of Strabismus Esotropia Exotropia Hypertropia descriptions of subjects prevalence prevalence prevalence prevalence within the children (years) (%) (%) (%) (%) study examined

North America

Canada [31] 946 1.6–11.6 4.3 [32] 1,074 <3 3.2 2.0 1.0 0.09 [33] 2,619 6 4.5 2.7 1.7 0.08 USA [34] Caucasian 306 6–7 1.6 Hispanic 548 6–7 0.9 [15] All races 39,227 0–7 4.5 3.2 1.2 Caucasian 17,931 0–7 5.4 4.1 1.3 African American 19,619 0–7 3.6 2.3 1.3 [35] Caucasian 119 8–16 3.4 Asian 310 8–16 2.9 Hispanic 1,781 8–16 1.8 Black 9 8–16 1/9 [27] Hispanic 3,003 0.5–6 2.4 0.9 1.5 African American 3,005 0.5–6 2.5 1.1 1.4 ([3]a, [9]a, [14]a) Population-based 0–19 3.9 2.3 1.3 0.3 Mexico [36] 1,035 12–13 2.3 1.2 0.8 0.4 [37] 343 3–6 1.2 0.6 0.6 Europe England [38] 4,784 5–6 4.4 3.6 0.8 [39] 6,634 2 1.5 1.1 0.4 [40] 7,538 7 2.3 1.7 0.5 0.1 Ireland [41] 1,582 8–9 4.0 3.4 0.6 Denmark [42] 14,107 0–19 4.5 3.5 0.9 0.1 Sweden [43] 6,004 0–7 3.9 3.4 0.4 0.05 [44] 1,046 12–13 2.7 1.4 0.7 0.6 [45] 3,126 ≤10 2.7b 1.5 0.6 [46] 143 4–15 3.5 2.8 0.7 Croatia [47] All children 20,045 Unspecifi ed 4.0 2.1 1.8 Term 17,163 Unspecifi ed 3.3 1.7 1.6 Preterm 2,882 Unspecifi ed 8.0 4.7 3.3 Australia [20] 1,739 6 2.8 1.6 1.2 0

[48] 2,353 12 2.7c 0.9 1.1 Asia Malaysia [49] 650 8 2.2 0.2 1.8 0.2 [50] Near ﬁ xation 4,634 7–15 0.7 0.5 Distance ﬁ xation 4,634 7–15 0.7 0.6

(continued) 6 1 Epidemiology of Pediatric Strabismus

Table 1.1. (continued)

China [51] Near ﬁ xation 4,364 5–15 1.9 1.6 Distance ﬁ xation 4,364 5–15 3.0 2.6 [52] 1,084 6–14 2.5 0.4 2.1 Japan [53] Study year 2003 86,531 6–12 1.3d 0.3 0.7 [54] Study year 2005 84,619 6–12 1.0e 0.2 0.6 [25] Five consecutive 33,929 total 1.5 0.01–0.1 0–0.03 0–0.07 measurements between years 2000 and 2004 Five consecutive 33,193 total 3 0.2–0.3 0.02–0.1 0.2–0.3 measurements between years 2000 and 2004 Taiwan [55] 862 6, 8, 11 1.6f 0.5 0.9 Th ailand [56] 3,898 1 0.6 India [57] 6,447 5–15 0.5 0.3 0.2 0.02 [58] 10,605 ≤15 0.4 Nepal [59] 1,100 5–16 1.6 0.09 1.5 [60] 1,816 5–16 1.3 Middle East Israel [61] 38,000 1–2.5 1.3 0.9 0.3 0.06 Oman [62] 6,292 6–7, 11–12 0.6 0.4 0.2 [63] 143,112 6–7 0.5 Africa Cameroon [64] 11,230 ≤26 1.2 0.5 0.8 Nigeria [65] 1,144 4–24 0.3 Ghana [66] 957 6–22 0.2 Tanzania [67] 1,386 7–19 0.5 Madagascar [68] 1,081 8–14 0.7 0.5 0.3

aStudy of incidence rather than prevalence bStrabismus prevalence includes 19 cases of microtropia cStrabismus prevalence includes 16 cases of microtropia dStrabismus prevalence includes 245 cases of “unknown” and 20 cases of “other” types of strabismus eStrabismus prevalence includes 110 cases of “unknown” and 23 cases of “other” types of strabismus fStrabismus prevalence includes two cases of “other” strabismus

English children [26]. Prevalence studies, reporting on speciﬁ c period of time, may survey any number of char- the number of people with a speciﬁ c disease at a pre- acteristics and their changes over time. scribed point in time, are found most commonly in the Table 1.1 includes recent strabismus prevalence and pediatric strabismus literature. However, this type of incidence data organized by regions of the world. One study may only capture a snapshot of childhood ocular overarching trend is that strabismus prevalence rates dif- deviations. Incidence reports, on the other hand, by fer based on racial and ethnic background. Esodeviations including the number of new cases diagnosed during a are found with a relatively higher prevalence among References 7

Caucasian populations, while exodeviations are more 5. Greenberg AE, Mohney BG, Diehl NN (2007) Prevalence commonly reported among Asian and African children. of amblyopia in an incidence cohort of childhood strabis- As shown, North American, European, and Australian mus. In: Transactions of the 31st meeting of the european data contrast with epidemiologic information gathered strabismological association. Editor: Rosario Gomez de from multiple studies in Asia and Africa. Th is trend is Liano. European Strabismological Association. DP- M-8797- additionally evident among non-Caucasians in the US. In 2008. Madrid. Spain. Pg 51-54 the Multi-Ethnic Pediatric Eye Disease Study group’s 6. Pediatric eye disease investigator group (2002) Spontaneous work describing strabismus prevalence among Hispanic resolution of early-onset esotropia: experience of the and African-American children, for instance, exotropia Congenital Esotropia Observational Study. Am J Ophthalmol was diagnosed more commonly than esotropia [27]. Th e 133(1):109–118 basis of this diff erence may be in part linked with popula- 7. Louwagie CR, Diehl NN, Greenberg AE, Mohney BG, et al tion-based diff erences in refractive error. Esotropia is (2009) Is the incidence of congenital esotropia declining? commonly associated with hyperopia, whereas exotropia A population-based study from Olmsted County, is more oft en diagnosed in children with myopia [28]. Minnesota, 1965–1994. Arch Ophthalomol 127:200–203 8. Mohney BG (2001) Acquired nonaccommodative esotropia in childhood. JAAPOS 5(2):85–89 Summary for the Clinician 9. Govindan M, Mohney BG, Diehl NN, et al (2005) Incidence ■ Th e prevalence of strabismus subtypes varies and types of childhood exotropia: a population-based based on racial and ethnic background; Asians study. Ophthalmology 112:1046–108 are primarily diagnosed with exotropia whereas 10. Nusz KJ, Mohney BG, Diehl NN (2005) Female predominance Europeans, Australians, and Americans are pre- in intermittent exotropia. Am J Ophthalmol 140: 546–547 dominantly diagnosed with esotropia. 11. Hunter DG, Ellis FJ (1999) Prevalence of systemic and ocular disease in infantile exotropia: comparison with infantile esotropia. Ophthalmology 106:1951–1956 12. Mohney BG, Huff aker RK (2003) Common forms of child- 1.6 Incidence of Adult Strabismus hood exotropia. Ophthalmology 110:2093–2096 Although there is substantially less epidemiological infor- 13. Havertape SA, Cruz OA, Chu FC (2001) Sensory strabismus mation regarding adult strabismus, its prevalence has – eso or exo? J Pediatr Ophthalmol Strabismus 38: 327–330 been reported as approximately 4% in the United States 14. Tollefson, MM, Mohney BG, Diehl NN, et al (2006) [29]. In a study of strabismus patients over 60 years of Incidence and types of childhood hypertropia: a popula- age, 29% developed their ocular deviation in childhood tion-based study. Ophthalmology 113:1142–1145 [30]. Beauchamp and colleagues similarly found that a 15. Chew E, Remaley NA, Tamboli A, et al (1994) Risk minority, or 38%, of strabismus patients between 17 and factors for esotropia and exotropia. Arch Ophthalmol 112: 92 years of age developed their deviation before visual 1349–1354 maturation [29]. Common causes of adult strabismus in 16. Hakim RB, Tielsch JM (1992) Maternal cigarette smoking descending order include neuroparalytic, restrictive, and during pregnancy: a risk factor for childhood strabismus. sensory factors [30]. Arch Ophthalmol 110:1459–1462 17. Holmstrom G, Rydberg A, Larsson E (2006) Prevalence and development of strabismus in 10-year-old premature References children: a population-based study. J Pediatr Ophthalmol Strabismus 43:346–352 1. Mohney BG (2007) Common forms of childhood strabis- 18. Mohney BG, Erie JC, Hodge DO, et al (1998) Congenital mus in an incidence cohort. Am J Ophthalmol 144:465–467 esotropia in Olmsted County, Minnesota. Ophthalmology 2. Yu CB, Fan DS, Wong VW, et al (2002) Changing patterns 105:846–850 of strabismus: a decade of experience in Hong Kong. Br J 19. Pennefather PM, Clarke MP, Strong NP, et al (1999) Risk Ophthalmol 86:854–856 factors for strabismus in children born before 32 weeks’ 3. Greenberg, AE, Mohney BG, Diehl NN, et al (2007) gestation. Br J Ophthalmol 83:514–518 Incidence and types of childhood esotropia: a population- 20. Robaei D, Rose KA, Kifl ey A, et al (2006) Factors associated based study. Ophthalmology 114:170–174 with childhood strabismus: fi ndings from a population- 4. Mohney BG, Greenberg AE, Diehl NN (2007) Age at stra- based study. Ophthalmology 113:1146–1153 bismus diagnosis in an incidence cohort of children. Am J 21. Arora A, Williams B, Arora AK, et al (2005) Decreasing Ophthalmol 144:467–469 strabismus surgery. Br J Ophthalmol 89:409–412 8 1 Epidemiology of Pediatric Strabismus

22. Carney CV, Lysons DA, Tapley JV (1995) Is the incidence 38. Graham PA (1974) Epidemiology of strabismus. Br J of constant esotropia in childhood reducing? Eye 9(Suppl): Ophthalmol 58:224–231 40–41 39. Stayte M, Johnson A, Wortham C (1990) Ocular and visual 23. Ferguson JA, Goldacre MJ, Henderson J, et al (1991) defects in a geographically defi ned population of 2-year- 1 Ophthalmology in the Oxford region: analysis of time old children. Br J Ophthalmol 74:465–468 trends from linked statistics. Eye 5(Pt 3):379–384 40. Williams C, Northstone K, Howard M, et al (2008) Prevalence 24. MacEwen CJ, Chakrabarti HS (2004) Why is squint sur- and risk factors for common vision problems in children: gery in children in decline? Br J Ophthalmol 88:509–511 data from the ALSPAC study. Br J Ophthalmol 92:959–964 25. Matuso T, Matsuo C, Matsuoka H, et al (2007) Detection of 41. Donnelly UM, Stewart NM, Hollinger M (2005) Prevalence strabismus and amblyopia in 1.5- and 3-year-old children and outcomes of childhood visual disorders. Ophthalmic by a preschool vision-screening program in Japan. Acta Epidemiol 12:243–250 Med Okayama 61(1):9–16 42. Frandsen AD (1960) Occurrence of squint: a clinical- 26. O’Connor AR, Stephenson TJ, Johnson A, et al (2002) statistical study on the prevalence of squint and associated Strabismus in children of birth weight less than 1701 g. signs in diff erent groups and ages of the Danish popula- Arch Ophthalmol 120:767–773 tion [dissertation] Acta Ophthalmol 62(Suppl):1 27. Multi-ethnic pediatric eye disease study group (2008) 43. Nordlow W (1964) Squint – the frequency of onset at dif- Prevalence of amblyopia and strabismus in African ferent ages, and the incidence of some associated defects in American and Hispanic children ages 6 to 72 months. a Swedish population. Acta Ophthalmol (Copenh) 42: Ophthalmology 115:1229–1236 1015–1037 28. Lambert SR (2002) Are there more exotropes than eso- 44. Ohlsson J, Villarreal G, Sjostrom A, et al (2001) Visual acu- tropes in Hong Kong? Br J Ophthalmol 86:835–836 ity, residual amblyopia and ocular pathology in a screened 29. Beauchamp GR, Black BC, Coats DK, et al (2003) Th e population of 12–13-year-old children in Sweden. Acta management of strabismus in adults – I. clinical character- Ophthalmol Scand 79:589–595 istics and treatment. J AAPOS 7:233–240 45. Kvarnstrom G, Jakobsson P, Lennerstrand G (2001) Visual 30. Magramm I, Schlossman A (1991) Strabismus in patients screening of Swedish children: an opthalmological evalua- over the age of 60 years. J Pediatr Ophthalmol Strabismus tion. Acta Ophthalmol Scand 79:240–244 28:28–31 46. Gronlund MA, Andersson S, Aring E, et al (2006) 31. Drover JR, Kean PG, Courage ML, et al (2008) Prevalence Ophthalmological fi ndings in a sample of Swedish children of amblyopia and other vision disorders in young aged 4–15 years. Acta Ophthalmol Scand 84:169–176 Newfoundland and Labrador children. Can J Ophthalmol 47. Karlica D, Galetovic D, Znaor L, et al (2008) Strabismus 43:89–94 incidence in infants born in Split-Dalmatia county 2002– 32. Kornder LD, Nursey JN, Pratt-Johnson AJ, et al (1974) 2005. Acta Clin Croat 47:5–8 Detection of manifest strabismus in young children 1. 48. Robaei D, Kifl ey A, Mitchell P (2006) Factors associated A prospective study. Am J Ophthalmol 77:207–210 with a previous diagnosis of strabismus in a population- 33. Kornder LD, Nursey JN, Pratt-Johnson AJ, et al (1974) based sample of 12-year-old Australian children. Am J Detection of manifest strabismus in young children 2. Ophthalmol 142:1085–1087 A retrospective study. Am J Ophthalmol 77:211–214 49. Teoh GH, Yow CS (1982) Prevalence of squints and visual 34. Fischbach LA, Lee DA, Englehardt RF, et al (1993) Th e defects in Malaysian primary one school children. Med J prevalence of ocular disorders among Hispanic and Malaysia 37(4):336–337 Caucasian children screened by the UCLA mobile eye 50. Goh P-P, Abqariyah Y, Pokharel GP, et al (2005) Refractive clinic. J Community Health 18(4): 201–211 error and visual impairment in school-age children 35. Voo I, Lee DA, Oelrich FO (1998) Prevalences of ocular in Gombak district, Malaysia. Ophthalmology 112: conditions among Hispanic, white, Asian, and black immi- 678–685 grant students examined by the UCLA mobile eye clinic. 51. He M, Zeng J, Liu Y, et al (2004) Refractive error and visual J Am Optom Assoc 69:255–261 impairment in urban children in southern China. Invest 36. Ohlsson, J, Villarreal G, Sjostrom A, et al (2003) Visual Ophthalmol Vis Sci 45:793–799 acuity, amblyopia, and ocular pathology in 12- to 13-year- 52. Lu, P, Chen X, Zhang W, et al (2008) Prevalence of ocular old children in northern Mexico. J AAPOS 7:47–53 disease in Tibetan primary school children. Can J 37. Juarez-Munoz, IE, Rodriguez-Godoy ME, Guadarrama- Ophthalmol 43:95–99 Sotelo ME, et al (1996) Frecuencia de trastornos oft almo- 53. Matsuo T, Matsuo C (2005) Th e prevalence of strabismus logicos comunes en poblacion preescolar de una delegacion and amblyopia in Japanese elementary school children. de la Ciudad de Mexico. Salud Publica Mex 38:212–216 Ophthalmic Epidemiology 12:31–36 References 9

54. Matsuo T, Matsuo C (2007) Comparison of prevalence 61. Friedman Z, Neumann E, Hyams SW, et al (1980) rates of strabismus and amblyopia in Japanese elementary Ophthalmic screening of 38,000 children, age 1 to 2 ½ years, school children between the years 2003 and 2005. Acta in child welfare clinics. J Pediatr Ophthalmol Strabismus Med Okayama 61(6):329–334 17:261–267 55. See L-C, Song H-S, Ku W-C, et al (1996) Neglect of child- 62. Lithander J (1998) Prevalence of amblyopia with ani- hood strabismus: Keelung Ann-Lo community ocular sur- sometropia or strabismus among schoolchildren in the vey 1993–1995. Chang Gung Med J 19: 217–224 Sultanate of Oman. Acta Ophthalmol Scand 76:658–662 56. Tengtrisorn S, Singha P, Chuprapawan C (2005) Prevalence 63. Khandekar RB, Abdu-Helmi S (2004) Magnitude and of abnormal vision in one-year-old Th ai children, based on determinants of refractive error in Omani school children. a prospective cohort study of Th ai children. J Med Assoc Saudi Med J 25(10):1388–1393 Th ai 88(Suppl 9):S114–S120 64. Ebana Mvogo C, Bella-Hiag AI, Epesse E (1996) Le stra- 57. Murthy GVS, Gupta SK, Ellwein LB, et al (2002) Refractive bisme au Cameroun. J Fr Ophtalmol 19:705–709 error in children in an urban population in New Delhi. 65. Ajaiyeoba AI, Isawumi MA, Adeoye AO, et al (2007) Pattern Invest Ophthalmol Vis Sci 43:623–631 of eye diseases and visual impairment among students in 58. Nirmalan PK, Vijayalakshmi P, Sheeladevi S, et al southwestern Nigeria. Int Ophthalmol 27:287–292 (2003) The Kariapatti pediatric eye evaluation project: 66. Ntim-Amponsah CT, Ofosu-Amaah S (2007) Prevalence baseline ophthalmic data of children aged 15 years or of refractive error and other eye diseases in schoolchildren younger in southern India. Am J Ophthalmol 136: in the greater Accra region of Ghana. J Pediatr Ophthalmol 703–709 Strabismus 44:294–297 59. Nepal BP, Koirala S, Adhikary S, et al (2003) Ocular mor- 67. Wedner SH, Ross DA, Balira R, et al (2000) Prevalence of bidity in schoolchildren in Kathmandu. Br J Ophthalmol eye diseases in primary school children in a rural area of 87:531–534 Tanzania. Br J Ophthalmol 84:1291–1297 60. Shrestha RK, Joshi MR, Ghising R, et al (2006) Ocular 68. Auzemery A, Andriamanamihaja R, Boisier P (1995) morbidity among children studying in private schools of Enquete sur la prevalence et les causes des aﬀ ections ocu- Kathmandu valley: a prospective cross sectional study. laires chez les enfants des ecoles primaries d’Antananarivo. Nepal Medical College Journal 8(1):43–46 Cahiers Sante 5:163–166 Chapter 2 Changes in Strabismus Over Time: The Roles of Vergence Tonus and Muscle Length Adaptation1 2 David L. Guyton

Core Messages ■ Patients with long-standing unilateral strabismus, which changes the lengths of the extraocular such as “sensory” exotropia in the absence of muscles bilaterally, is largely responsible for fusion or esotropia with unilateral amblyopia, changes in the angle of strabismus over time. typically show bilateral deviations under anesthe- ■ Th is mechanism helps explain the development of sia, oft en symmetric. (1) increasing “basic” deviations in accommoda- ■ Forced ductions usually show symmetric muscle tive esotropia, (2) torsional deviations with appar- tightness. Changes in extraocular muscle lengths ent oblique muscle “overaction/underaction” and thus appear to occur primarily bilaterally, whether A and V patterns, (3) recurrent esotropia with or not fusion is present. early presbyopia, (4) occasional divergence insuf- ■ With skeletal muscles responding to changes in ﬁ ciency in presbyopes, and (5) basic cyclovertical stimulation by the gain or loss of sarcomeres, it is deviations that mimic superior oblique muscle likely that abnormal or unguided vergence tonus, paresis.

in terms of the fi xation pattern, but rather in terms of the 2.1 Binocular Alignment System relative basic lengths of the extraocular muscles and the A vexing problem in the fi eld of strabismus is what tonus of their vergence innervation. Before discussing causes strabismus to change over time. For example, why the bilateral nature of strabismus changes, the two basic do patients with accommodative esotropia develop a mechanisms are reviewed that regulate long-term bin- basic component over time [2, 3]? Why do torsional ocular alignment. deviations develop, with accompanying A and V patterns [4]? Why does superior oblique paresis change in its pattern of deviation over time? When vision is lost in 2.1.1 Long-Term Maintenance one eye, or simply when fusion is lost, why does sensory of Binocular Alignment exotropia develop? If we can get a handle on the underlying mechanism involved in these changes, we may be In the normal situation, sensorimotor fusion maintains able to better guide our research and improve the care binocular alignment on a moment-by-moment basis, but we give to our patients. Th is chapter is intended to pro- there are two further mechanisms that maintain binocu- vide some further insight to this predominant underly- lar alignment in the long term. Th e fi rst is a neurologic ing mechanism that induces changes in strabismus, to a one, “vergence adaptation,” and the second is a muscular large extent, bilaterally. Th is does not refer to strabismus one, “muscle length adaptation.”

1Adapted from [1]. Reprinted with permission of the publisher. 12 2 Changes in Strabismus Over Time: The Roles of Vergence Tonus and Muscle Length Adaptation

of gaze, with increased comitance of the overall pattern of 2.1.2 Vergence Adaptation deviation [13]. Neurologically, retinal image disparity invokes a fusional However, maximum neuronal fi ring rates impose lim- vergence response which moves the eyes in opposite its on how much misalignment can be compensated for 2 directions to eliminate the retinal image disparity, accu- by vergence adaptation. In particular, orbital changes rate to within a few minutes of arc, both horizontally and with skeletal growth require not only lengthening of the vertically. Th is is sometimes called “fast” fusional ver- extraocular muscles, but also require relative changes in gence. It responds to retinal image disparity in less than a functional muscle length that are far beyond the capabili- second, and if one eye is suddenly covered, it decays in ties of neurologic adaptation. It is the process of muscle 10–15 s or less [5]. length adaptation that comes to the rescue. It is feedback from fast fusional vergence that stimulates changes in tonic vergence, or vergence tonus, over time [6]. Th is process is sometimes called “slow” ver- 2.1.3 Muscle Length Adaptation gence, or vergence adaptation. Vergence adaptation occurs selectively for diff erent directions of gaze and for Th e topic of muscle length adaptation does not appear in diff erent distances, as if the brain establishes a table of most texts on strabismus. Th e historic assumptions have how much innervational tonus to provide to each been that extraocular muscle lengths are determined extraocular muscle to keep the eyes aligned in each genetically, and that the basic forms of strabismus are due direction of gaze and at each distance – horizontally, ver- to primary abnormalities in muscle anatomy, in innerva- tically, and torsionally [7]. Th e eff ects of vergence adaption, or in neurologic tonus. However, there must be tation can persist for minutes to hours and perhaps much dynamic mechanisms involved in the regulation of basic longer. Vergence adaptation wears off slowly when one muscle length which normally play a critical role in the eye is occluded or during sleep, but much faster in the long-term maintenance of binocular alignment. presence of a competing vergence [6]. Th is mechanism Tracer studies have shown that skeletal muscles was phenomenologically described as long ago as 1868 throughout the body undergo continuous remodeling by Hering (cited in [8]), and in 1893 by Maddox (cited in throughout life. In fact, the half-life of the contractile [9]). Alfred Bielschowsky actually studied this early in proteins in adult skeletal muscles is only 7–15 days his career, reporting with Hofmann in 1900 that ver- [14]. Muscle physiologists in France and England [14– gence adaptation decays slowly, and with an exponential 16] discovered in the 1970s and 1980s that skeletal time course (cited in [10]). It has been studied exten- muscles intrinsically adapt their lengths, by serial addi- sively by Ellerbrock [8], Ogle and Prangen [11], Carter tion or subtraction of sarcomeres at the ends of the [6], Crone [12], Schor [10], and many others [9]. Clearly, myofi brils, to maintain the proper overlap of the actin by supplying learned tonus levels to keep the eyes roughly and myosin myofi laments so as to obtain optimal force aligned in various direction of gaze, vergence adaptation generation, velocity, and power output over the range signifi cantly eases the burden on sensorimotor fusion, of motion through which the muscle is most used [17]. leaving sensorimotor fusion free to fi ne-tune the align- Th e exact biologic mechanism that accomplishes this is ment of the eyes [6]. still unknown. Vergence adaptation provides a tonic neural compen- In 1994, Alan Scott [18] showed that the extraocular sation for ocular deviations. It eliminates the anisophoria muscles can adapt their lengths in the same way as the produced by new anisometropic spectacle lenses. It begins other skeletal muscles throughout the body. He sutured to decay slowly when one eye is covered, as evidenced by one eye of a monkey to the lateral orbital wall in an exo- the “screening-up” of ocular deviations when measuring tropic position of approximately 30 prism diopters. Aft er with the prism and alternate cover test. In the longer 2 months, when the basic lengths of the extraocular mus- term, it is responsible for the “eating up” of prisms over cles were examined, the medial rectus muscle had gained minutes to days in the process called prism adaptation. sarcomeres, and the lateral rectus muscle had lost sar- Clinically, we oft en try to uncover the underlying devia- comeres in the experimental animal, compared with con- tion by occluding one eye. For example, Lancaster red- trol animals operated in the same manner and sacrifi ced green plots of incomitant strabismus with partial fusion immediately. oft en show best alignment in primary gaze, and in the Change in skeletal muscle length is not only respon- reading position, those directions of gaze that are most sive to the position in which the muscle is held, but also, used and, therefore, best adapted to. Aft er a 30-min patch and most importantly, in the case of the extraocular mus- test, the plotted tropia oft en increases in these directions cles, to the stimulation that it receives. If a muscle is not 2.2 Modeling the Binocular Alignment Control System 13 held in a stretched position, increased stimulation causes 2.2 Modeling the Binocular Alignment actual loss of sarcomeres with shortening of the basic Control System muscle length [15, 16, 19]. Th is change in basic muscle length in response to the level of stimulation is precisely Th e basic components are now in place to model the bin- in the right direction to help maintain binocular align- ocular alignment control system (Fig. 2.1), beginning ment. In fact, it is probably the chronic average level of with the existing basic muscle length of each muscle, vergence tonus, as maintained by vergence adaptation determined by the number of sarcomeres. Each muscle is and contributed to by the current level of fast fusional stimulated by the current level of vergence tonus to result vergence, which provides the primary input to extraocu- in the approximate functional muscle length (the physical lar muscle length adaptation. length) to yield aligned eyes. Acute vergence stimulation Th is further feedback mechanism, that is, vergence supplied by fast fusional vergence completes the binocu- tonus regulating muscle length adaptation, completes the lar alignment process. dynamic feedback system for maintenance of long-term However, a perturbation suddenly occurs, such as a binocular alignment (Fig. 2.1). Retinal image disparity hormonal growth spurt with a change in the divergence elicits fast fusional vergence, which leads in the short of the orbits, new glasses with a small change in prism term to vergence adaptation, producing a change in ver- eff ect, or simply a switch of the object of regard from gence tonus, which stimulates muscle length adaptation the computer screen to the bird out the window. Such a over a longer term, all of which reduce the retinal image perturbation requires diff erent eye alignment and will disparity. Each level of this marvelous three-level feed- thus result in misaligned eyes for the new task if no back process also works in the direction to ease the bur- compensation is made. Nevertheless, misaligned eyes den on the level that precedes it. Vergence adaptation cause retinal image disparity, with a double image of frees up fast fusional vergence to be able to respond accu- the bird out the window, which the brain does not rately to rapid changes in retinal image disparity. Muscle like. length adaptation relieves vergence adaptation of exces- Hence, the brain responds with fast fusional vergence, sive demands, which would otherwise saturate neuronal changing the acute stimulation levels to the muscles. Th is fi ring rates, and thereby eff ectively resets vergence adap- yields new functional muscle lengths in the proper direc- tation so that it can continue to function optimally in tion to compensate for the original perturbation, and response to input from fast fusional vergence. realigns the eyes. Something else now happens. Sustained fast fusional vergence leads to vergence adaptation, which adjusts the basic level of vergence tonus to ease the burden on fast fusional vergence, freeing it to be able to respond to the Basic muscle lengths next perturbation. (vergence tonus) However, there is a limit to the amount of vergence Approx. functional muscle lengths tonus that can be sustained, so something further happens. In response to the amount of overall vergence tonus, ( acute stimulation) the muscle lengths slowly adapt to new basic lengths in Exact functional muscle lengths the proper direction to reduce the original retinal image [perturbation] disparity. Once the basic muscle lengths have adapted, the neurologic feedback mechanisms that the original Retinal image disparity (diplopia) perturbation brought into play can subside, with the eyes Fast fusional vergence aligned once again. Furthermore, the neurologic mechanisms can now be maximally responsive to the next Vergence adaptation perturbation. Th is is the normal functioning of the long-term (as Vergence tonus well as short-term) binocular alignment control system. Th is is the feedback scheme that keeps the eyes aligned Muscle length adaptation during the growth of the skull in early life, throughout the development of hand–eye coordination in oblique directions of gaze, and throughout the development of presby- Fig. 2.1 Th ree-level dynamic feedback system for the mainte- opia, which would otherwise cause a signifi cant disruption nance of binocular alignment of near vs. distance alignment. 14 2 Changes in Strabismus Over Time: The Roles of Vergence Tonus and Muscle Length Adaptation

literature that muscle length adaptation can be responsive 2.2.1 Breakdown of the Binocular Alignment Control System to stimulation, the above-mentioned model was fi rst described by the author in a paper in Binocular Vision and However, what happens when something goes wrong Eye Muscle Surgery Quarterly in 1994 [4], with further 2 with this feedback system? Surely it is possible that abnor- elaboration in 2005 [21]. Th e model explained how defects malities can be present, or can develop, at various levels in fusion, or the loss of fusion, which for this purpose within this system, any of which will lead to misalign- were considered the same as loss of vision in one eye, ment of the eyes. Th e most common abnormality is prob- could lead to “sensory”-type changes in strabismus. In ably the absence of, or loss of, fast fusional vergence, particular, in the torsional dimension, lack of proper feed- which is simply referred to as fusion. Fusion is at a most back to the torsional control mechanism would be critical position in this feedback pathway system. expected to produce what we dubbed “sensory torsion,” If fusion does not occur in response to retinal image leading to the development of what is probably errone- disparity, stimulation levels do not change appropriately, ously called primary oblique muscle overaction, or under- and the entire system breaks down. With loss of input action, with accompanying A- or V-pattern strabismus. from the fast fusional vergence system, the longer-term It was not clear in 1994, however, whether extraocular mechanisms for binocular alignment, vergence adapta- muscle length adaptation responds to version stimulation [20], and muscle length adaptation [4] become free- tion. Th at is, will an extraocular muscle adapt its length wheeling – in other words, without guidance. for optimal function in the position in which it is held Neurologic feedback mechanisms do not necessarily most of the time by version stimulation? If so, what are shut off when their input disappears. Th ey will oft en con- the relative roles of version and vergence stimulation in tinue to function at a basal level, with a low level of out- the regulation of extraocular muscle length? New obser- put being generated. Th is basal output level can be biased vations have clarifi ed these questions. Th ese observations, in one direction or the other, and therefore, in this case, the resulting clarifi cation, and the consequences to our can continue to drive the muscle length adaptation understanding of strabismus are expected benefi ts from mechanism slowly in one direction or the other, produc- this chapter. ing strabismus that was not there in the fi rst place, or causing progressive misalignment if strabismus was already present. A prime example of this mechanism is the phenome- 2.2.3 Changes in Strabismus as a Bilateral Phenomenon non we call “sensory” exotropia. With loss of vision in one eye, fusion is lost, and as we have assumed in the past, Th e primary new observation of the author is that changes the eye simply passively drift s outward over time. From in strabismus occur, to a large extent, bilaterally. Th is is this feedback mechanism, we can begin to understand not speaking of strabismus in terms of the fi xation pat- that if one eye develops poor vision, and therefore, if the tern, but rather in terms of the relative basic lengths of the eyes have no need for convergence, the average vergence extraocular muscles and the tonus of their innervation. stimulation to the extraocular muscles (which had pre- In the case of sensory exotropia, one eye is always fi x- viously maintained alignment equilibrium) will shift ing, and the other eye gradually turns outward over time. slightly to less convergence and more divergence, actively However, there is usually mild limitation of adduction of driving the eyes into a position of exotropia. Th is sensory both eyes, and when that patient is put to sleep, very oft en exotropia can thus be seen to be not a passive process both the eyes turn out. Figures 2.2–2.4 show examples aft er all, but an active driving of the eyes outward by the of this bilateral phenomenon in patients with sensory otherwise normal alignment mechanisms that have lost exotropia. proper guidance. Th is observation was fi rst made by the author 25 years ago aft er a recess-resect procedure on a patient with sensory exotropia. Th e sensory exotropia recurred. When the patient was put back to sleep for a repeat recess-resect 2.2.2 Clarifi cation of Unanswered Questions procedure on the same eye, the previously operated eye Regarding the Long-Term Binocular was straight. It was the sound eye that was turning out Alignment Control System signifi cantly. Th e muscle changes that caused the original Th e description of the above-mentioned three-stage feed- sensory exotropia had occurred bilaterally. Arthur back model of the long-term binocular alignment control Jampolsky [22] reported this phenomenon in 1986, but system is not new. Upon appreciating the evidence in the he off ered no explanation for it. 2.2 Modeling the Binocular Alignment Control System 15

Fig. 2.2 Eighty-year-old woman with dense amblyopia in her Fig. 2.3 Twenty-one-year-old man with left sensory exotropia left eye since childhood, fi xing with her right eye only, all her life. (top), from a left macular scar since birth, with counting fi ngers Note the left sensory exotropia (top). Under general anesthesia vision in his left eye. His eyes also turn out essentially equally (bottom), both eyes turn out, equally – and signifi cantly farther under anesthesia (bottom) than the usual divergence seen under anesthesia

Th ere is more evidence that changes in strabismus 2.2.4 Changes in Basic Muscle Length occur bilaterally over time. Infants with esotropia and amblyopia, where the amblyopic eye is practically con- Th ese changes in strabismus occur because the muscles stantly adducted during waking hours, usually show some change their basic length, i.e., the number of sarcomeres. limited abduction bilaterally and symmetric positions of A basically short muscle has fewer sarcomeres than nor- the eyes under anesthesia. Furthermore, during surgery, mal, and a basically long muscle has more sarcomeres both medial rectus muscles are usually equally and abnor- than normal. As noted before, skeletal muscles are con- mally tight. Th ey are both abnormally short. Th ese children tinually changing their basic lengths throughout life, by sometimes show a small head turn, fi xing with the sound the serial addition or subtraction of sarcomeres, for opti- eye in slight adduction [23], consistent with a short medial mal function in the position where they are usually held. rectus muscle in the sound eye as well as in the amblyopic However, if this were the only mechanism by which eye. Figure 2.5 shows the same phenomenon in an adult extraocular muscle basic lengths are regulated, we should with esotropia and long-standing unilateral fi xation. expect the patient with sensory exotropia to show only Th ere is still further evidence that changes in strabis- the poor vision eye turning out under anesthesia, because mus occur bilaterally. Th e torsional changes that are asso- the exodeviated eye would have adapted its muscle lengths ciated with primary A and V patterns are practically for optimal function centered in far abduction. But this is always bilateral, although sometimes asymmetric. If the not what we observe. Usually, both eyes in sensory exotro- eye with greater elevation in adduction is operated upon pia turn out under general anesthesia, signifi cantly more with an inferior oblique weakening procedure, the other than the usual divergence seen under anesthesia. Th ere eye soon shows as much or more elevation in adduction. must be another mechanism that causes basic muscle 16 2 Changes in Strabismus Over Time: The Roles of Vergence Tonus and Muscle Length Adaptation

Fig. 2.5 Th irty-four-year-old woman with esotropia since Fig. 2.4 Th irty-eight-year-old man aft er a right optic nerve childhood with fi xation with her left eye only (top), for many injury 15 years before, with resulting blindness in his right eye. years. Both eyes turn in signifi cantly under anesthesia (bottom) Th e fi xing left eye (top) turns out abnormally under anesthesia (bottom), but not as much as the blind right eye. Not every patient turns out equally

lengths to change bilaterally, and that mechanism is most response which moves the eyes in opposite directions to surely related to stimulation, given the fact that chronic eliminate image disparity, accurate to within a few min- electrical stimulation has been shown to shorten muscles utes of arc, both horizontally and vertically. by causing the loss of sarcomeres [15]. Might one of these types of stimulation, version stimulation or vergence stimulation, be involved in the regulation of basic muscle lengths for long-term alignment of the two eyes? Clearly, version stimulation would not be 2.2.5 Version Stimulation expected to be useful in such regulation, because version and Vergence Stimulation stimulation moves both the eyes in the same direction. If What type of stimulation do the extraocular muscles the extraocular muscles do change their basic lengths in normally receive? If one thinks about it, the extraocular response to version stimulation, then in the normal state, muscles between the two eyes are yoked as much as, or the eff ect would average to zero over time as the eyes look more than, any other muscles in the body. Th ey are heav- about in various directions. ily bilaterally innervated. Th ey are linked in versions, Vergence stimulation, on the other hand, is precisely movements of the two eyes in the same directions, and in the type of bilateral stimulation which could play a role in vergences, movements of the two eyes in opposite direc- muscle length adaptation. If the basic muscle lengths of tions. Versions allow us to look in diff erent directions, the extraocular muscles are altered for any reason from while vergences allow us to change our gaze from dis- their current lengths, image disparity will be sensed by tance to near. However, vergences also, and most impor- the brain, and fusional vergence will occur to restore bin- tantly, fi ne-tune both eyes to be aligned with the object of ocular alignment. Th e same fusional vergence that realigns regard, in any direction of gaze and at any distance, as the eyes momentarily, leads via vergence adaptation to part of the process of sensorimotor fusion. Disparity changes in vergence tonus. Changes in vergence tonus, between the two eyes’ images invokes a fusional vergence representing chronic changes in the levels of stimulation, 2.2 Modeling the Binocular Alignment Control System 17 can indeed serve as the necessary and suffi cient stimuli for chronic muscle length adaptation to adjust the basic muscle lengths. In the normal situation it is not necessary to postulate that basic extraocular muscle lengths respond only to vergence stimulation and not to version stimulation. As both vergence stimulation and version stimulation occur, both could be slowly stimulating muscle length adaptation. However, the eff ect of the version stimulation would average out to zero over time. Th e vergence stimulation, on the other hand, would exert a net eff ect, changing the basic muscle lengths in the directions necessary to reduce the need for the vergence stimulation in the fi rst place – a marvelous negative-feedback servomechanism, as pointed out previously. Th e mechanism just proposed would work in the normal situation, but there is strong evidence from what happens in strabismic states that extraocular muscle length adaptation responds to vergence stimulation primarily, and only minimally to version stimulation. And that is a fundamental diff erence between extraocular muscles and the other skeletal muscles. Th e evidence is the same as that noted earlier simply the observation that chronic monocular deviations of the eyes, as in sensory exotropia or in esotropia with unilateral amblyopia, practically always become binocular deviations under anesthesia, with bilaterally abnormal basic muscle lengths. Fig. 2.6 Th irty-three-year-old woman with esotropia since Th e argument is this: In constant strabismic states birth. Only her right eye was operated for the esotropia at the age where there is no fusion, there is no signifi cant fusional of 2½ years. She has fi xed with her LE only (top), as long as she vergence stimulation, but version stimulation still exists. can remember, because of mild hyperopia and amblyopia in her right eye. Neither eye has adapted to these positions, because If the extraocular muscles should adapt their lengths when she is placed under deep anesthesia (bottom), both the according to version stimulation, then the muscle lengths eyes deviate rightward. Th e muscle lengths clearly did not adapt in the deviating eye in the patient with sensory exotropia in response to chronic everyday version stimulation would totally adapt to the deviated position. Th e sound eye, spending its average time in straight Th erefore we must conclude that the stimulation from ahead gaze, would have normal muscle lengths. However, vergence tonus is the primary regulator of extraocular this is clearly not the case, because in most cases of sen- muscle length adaptation, and that its eff ects are bilateral. sory exotropia, both eyes turn out under anesthesia, and In this regard, the regulation of the extraocular muscle in most cases of esotropia with unilateral amblyopia, the lengths appears to be fundamentally diff erent from the two eyes are essentially symmetric under anesthesia. By regulation of the lengths of other skeletal muscles. Only forced duction testing, especially in the cases of esotropia, the extraocular muscles experience this bilateral vergence the basic muscle lengths are clearly bilaterally abnormal. stimulation. Th e other skeletal muscles receive primarily Th e position of the eyes when asleep probably has little unilateral stimulation, or bilateral stimulation akin to or no eff ect on muscle length adaptation, because Breinin version stimulation, and their lengths are responsive to has shown that electrical activity in the extraocular these forms of stimulation as well as to stretching or muscles essentially disappears in deep sleep [24], and slackening of the muscles depending on use. decreased stimulation of skeletal muscles signifi cantly slows down muscle length adaptation, as shown by denervation experiments [19]. 2.2.6 Evidence Against the Figure 2.6 shows a patient illustrating the ineff ective- “Final Common Pathway” ness of version stimulation. Th e muscle lengths clearly did not adapt to the positions in which the eyes were held Th ere is a potential problem with the conclusion that by chronic everyday version stimulation. vergence tonus is the primary regulator of extraocular 18 2 Changes in Strabismus Over Time: The Roles of Vergence Tonus and Muscle Length Adaptation

muscle length adaptation, and that its eff ects are bilateral. ocular torsion, with associated A and V patterns, are Neurophysiologists, with few exceptions [25], have long forms of sensory deviations developing over time when believed that version and vergence stimulation, while fusion is faulty or absent [4]. Clearly, the simple decreased arising in diff erent centers in the brainstem, are com- need to converge that occurs when vision is lost in one 2 bined into a “fi nal common pathway” at the motoneu- eye cannot explain the development of esotropia, verti- rons whose axons constitute the motor nerves to the cal deviations, or torsional deviations. Th e many diff er- extraocular muscles [26, 27]. In other words, it has been ent ways that strabismus can change over time, if linked believed that version and vergence stimulation are indis- to changes in vergence tonus, require a more general tinguishable by the time the impulses reach the extraoc- explanation. ular muscles. If that were the case, extraocular muscle Th e explanation, as noted earlier, probably lies in the length adaptation could not be preferentially responsive very nature of biologic control systems. When input to to vergence stimulation. Recent evidence suggests, how- such control systems shuts down, the output rarely goes ever, that version and vergence signals may indeed to zero, but rather goes to a baseline state that may be remain segregated in the motor nerves and stimulate dif- biased on either side of zero output. In the case of the ferent fi ber types in the extraocular muscles [28, 29]. It is ocular motor control systems, when the eyes become tempting to speculate that those fi ber types receiving misaligned enough that fusional vergence cannot oper- vergence stimulation are those primarily responsible for ate, retinal image disparities do not result in corrective muscle length adaptation, but such details have not yet vergences. In this case, the fusional vergence control been worked out. mechanisms for horizontal, vertical, and torsional align- Recent experiments by Joel Miller support the notion ment probably do not shut down entirely, but rather of segregation of version and vergence signals by demon- decrease their outputs to small nonzero levels, with per- strating that measured extraocular muscle tension shows sistent weak vergence signals biased in one direction or discrepancies with electrical activity [30]. Th ese observa- the other, with the direction of this bias depending upon tions argue against the fi nal common pathway concept numerous factors. and at least allow the thesis that vergence tonus is primar- For example, young children oft en have a stronger ily responsible for muscle length adaptation. convergence bias than divergence bias, as evidenced by the relative frequency of esotropia vs. exotropia in infancy. Th is may simply be a manifestation of more hyperopia in childhood, with the attendant increased convergence 2.3 Changes in Strabismus tonus from accommodative convergence. If vision is lost However, if the basic muscle lengths change primarily in in one eye in early infancy, it is not surprising that a non- response to vergence stimulation, how does constant zero convergence bias in the horizontal alignment control strabismus change over time, when there is presumably system could shorten the medial rectus muscles over no fusional vergence stimulation occurring? It is easy to time, resulting in sensory esotropia. answer this question in the case of sensory exotropia, Likewise, when fusion is faulty or absent, either pri- because other forms of vergence are occurring. With poor marily or from horizontal misalignment early in life, a vision in one eye, there is no advantage or incentive to baseline output bias in the torsional alignment mecha- actively align the eyes, or even to converge them when nism can drive the eyes into torsional misalignment with looking up close. With less convergence occurring than apparent oblique muscle dysfunction and accompanying before vision was lost in one eye, and at least in older A and V patterns. Th e torsion is oft en seen at fi rst only individuals, the normal balance between convergence when awake, disappearing when under anesthesia [31]. and divergence is upset in favor of a slight divergence Later, as the oblique muscle lengths change, the fundus bias, and this divergence bias slowly but actively shortens torsion persists under anesthesia [32]. Still later, aft er soft both lateral rectus muscles and lengthens both medial tissue remodeling occurs in response to the chronic ocu- rectus muscles over time, resulting in increasing exotro- lar torsion (the author’s interpretation), the eyes move pia. Th e deviation, of course, shows up only in the eye more along the torted planes defi ned by the muscle inser- with poor vision, until the patient is put under anesthesia, tions, showing clinical oblique muscle “overaction” (ele- when both the eyes turn out. vation or depression in adduction), and on MRI studies, Some patients with loss of vision or fusion develop the connective tissue “pulleys” may be seen to have shift ed esotropia, especially when vision is lost in early infancy. [33] (the author’s interpretation). Vertical misalignment can also develop when vision is Furthermore, a baseline output bias in the cycloverti- lost in one eye. It has been argued before that abnormal cal alignment mechanism can drive the eyes into a basic 2.3 Changes in Strabismus 19 cyclovertical misalignment, a cyclovertical misalignment deviations [39–41]. By careful study of Marlow’s published which we oft en call congenital superior oblique paresis, graphs [40], it is apparent that aft er 3–5 days of monocular probably mistakenly, because we have no other term for occlusion, signifi cant changes in the monitored deviations it. Most cases of esotropia are not attributed to sixth nerve oft en began to appear, and worsen. For example, hyperde- palsy, but we persist in attributing many cyclovertical viations and torsional deviations began to appear when deviations of unknown cause to fourth nerve palsy. there had been none previously. Also, the occluded eye Problems at other points in these control mechanisms most oft en developed a hyperdeviation, regardless of can perhaps lead to strabismus in the fi rst place. An which eye was covered, speaking against the uncovering of abnormality in vergence adaptation has been proposed to a latent hyperdeviation [42–44]. Rather than the uncover- cause divergence insuffi ciency or convergence excess ing of latent deviations, “Marlow occlusion” may indeed [34]. Poor or absent fusion from birth, in combination have promoted the onset of unguided vergence adaptation with a robust AC/A ratio, could lead to imbalance of and even the onset of muscle length adaptation, with new muscle length adaptation on the eso side, with progres- deviations beginning to occur. Th e same may be the case sive esotropia, which we would call congenital esotropia. in more recent studies by Viirre et al. [45] in monkeys, and Alternatively, a higher than normal AC/A ratio [35] could by Liesch and Simonsz [46] in normal human subjects. In strain fusion suffi ciently to cause intermittent esotropia, these studies, new vertical and torsional deviations were which would then progress to a constant esotropia [2, 3] noted aft er 7 days of monocular occlusion of the monkeys by the feedback mechanisms just noted. In intermittent and aft er 3 days of monocular occlusion of the human exotropia, only a minor defect in fusion could be the ini- subjects. tial problem, but as fusion deteriorates, the feedback- deprived muscle length adaptation mechanism will cause progressive worsening. 2.3.2 Unilateral Changes in Strabismus Convergence brought into play to damp some forms of nystagmus clearly disrupts the normal alignment control Clearly, not all changes in strabismus are bilateral. mechanism, leading directly to shortened medial rectus Patients with loss of fusion from sixth nerve palsy muscles and esotropia. Th is is the “nystagmus blockage” develop an increasingly short and tight ipsilateral or “nystagmus compensation” mechanism originally medial rectus muscle. Th e contralateral rectus muscle described by Adelstein and Cüppers (cited in [36]). And does not shorten concomitantly. Th is represents unilat- now that we know that manifest latent nystagmus as well eral muscle length adaptation, but from a diff erent as congenital nystagmus can be damped by convergence mechanism. When a skeletal muscle continues to be [37], this mechanism may be involved in Ciancia’s syn- stimulated but is not stretched out from time to time, it drome as well [38]. progressively shortens via the active loss of sarcomeres [16]. Th is is the mechanism demonstrated by Alan Scott by suturing his monkey’s eye temporally [18], and is the mechanism determining changes in the medial 2.3.1 Diagnostic Occlusion: And the Hazard and/or lateral rectus muscles in various types of Duane’s of Prolonged Occlusion syndrome as documented by Collins, Jampolsky, and Diagnostic occlusion of one eye has long been used as a Howe [47] and by Castañera de Molina and Giñer valuable method to break down vergence adaptation to Muñoz [48]. uncover the underlying deviation. Such occlusion will not reverse the eff ects of muscle length adaptation in the short term, but will simply reduce the eff ects of vergence 2.3.2.1 Supporting Evidence for Bilateral Feedback Control of Muscle Lengths adaptation over an exponential time course. Th irty to forty-fi ve minutes of monocular occlusion are usually What further evidence is there for bilateral feedback con- long enough to eliminate most vergence adaptation [13], trol of muscle lengths? We have previously demonstrated although diagnostic monocular occlusion for up to 1–2 that patients with consecutive esotropia following surgery weeks has been reported. for intermittent exotropia oft en develop intorsion or If diagnostic occlusion is continued for days, eliminat- extorsion of the eyes, with accompanying oblique muscle ing fusion, there is a very real possibility of creating new overaction and A or V patterns, aft er having lost fusion deviations by the stimulation of new extraocular muscle for only 1 month [4, 49]. We attribute this to a type of length adaptation. In the 1920 and 1930s, Marlow advo- “sensory torsional” deviation due to muscle length adap- cated occlusion for 7–10 days to fully uncover latent tation in the torsional dimension. 20 2 Changes in Strabismus Over Time: The Roles of Vergence Tonus and Muscle Length Adaptation

Weldon Wright, Katie Gotzler, and the author have studies have shown that many patients with these deviations recently collected a large series of patients with early pres- have superior oblique muscles with normal cross-sectional byopia, mostly with defi cient or absent fusion, who have area and normal contractility [59, 60]. Demer et al. wrote in developed progressive esotropia probably from the 1995 [59], “Of 19 SO muscles diagnosed to be palsied based 2 increased convergence tonus accompanying the increas- on clinical criteria, MRI demonstrated that about half ing eff ort to accommodate. Seeking evidence that such exhibited normal cross-sectional size and contractile char- patients are fairly common, we tabulated all the patients acteristics.” Might there be no superior oblique paresis at all that the author had operated on for esotropia over a in these patients? Aft er all, we do not speak of patients with 17-year period where a reliable onset of the esotropia congenital esotropia as having sixth nerve paresis! could be established. Compared with a similar number of Howard Ying, Nicholas Ramey, and the author are patients operated on for exotropia, the esotropia popula- currently investigating the patterns of cyclovertical stra- tion showed a signifi cantly increased onset of esotropia in bismus that they can create in normal subjects. Th ey have their 30s and 40s, as expected [21]. Th is mechanism, constructed a special haploscope that allows adaptation involving muscle length adaptation, is probably to increasing vertical, torsional, or horizontal disparities, responsible for other reports of esotropia developing in with near fi xation, with fi elds of view of over 50°, utilizing adulthood [50, 51] and is similar to the mechanism of video-oculography for recording. Th e entire apparatus hypoaccommodative esotropia occurring in children, as can tilt, up to 45°, to the right or left . fi rst described by Costenbader [52]. To confi rm the capability of this apparatus, Fig. 2.7 Elizabeth Bell, Adam Bowen, and the author have shows the expected counter roll with head tilt to the right also identifi ed a series of presbyopic patients, aged 50 and left before any adaptation. years and older, who either had a small amount of uncor- So far, we have adapted normal subjects to vertical dis- rected hyperopia, or who oft en tried to function without parities increasing to 6° for 30–45 min. With adaptation, needed correction for near, and developed divergence we expect to fi nd that the hyperdeviations induced are insuffi ciency in the later decades of life. Th ey had inter- accompanied by torsional changes, and that the patterns mittent or constant esotropia in the distance with diplo- of misalignment induced, especially with forced head tilt- pia, but could still fuse at near. Th ey are best corrected by ing, will help explain the patterns that heretofore have bilateral medial rectus muscle recessions [53, 54], with been associated with what is called congenital superior the fi nding that both medial rectus muscles tend to be oblique paresis. tighter than normal by forced ductions at the beginning Th e fi rst results appear promising. A normal subject of surgery. In these patients, we suspect that chronic with head straight was slowly adapted over 45 min, activation of the near triad [55], which can provide maintaining fusion, to an increasing left -over-right improved visual acuity via slight pupillary constriction, causes increased convergence tonus, leading to short- Ocular Counter Roll ened medial rectus muscles and the characteristic pat- 10 tern of divergence insuffi ciency. Of interest is that the Right Eye presbyopic patients identifi ed with uncorrected or Left Eye undercorrected hyperopia showed a somewhat linear 5 increase of distance esotropia with the amount of hyperopia (Bell, Bowen, and Guyton, unpublished). Clockwise[deg] In the cyclovertical “plane,” which is not really a plane 0 aft er all, we have long suspected that there should be a thing such as a basic cyclovertical deviation, an analog of straight- –5 forward esotropia in the horizontal plane. Recent evidence suggests that the oblique muscles play a much larger role in STR RHT STR LHT cyclovertical fusion than previously expected [56–58]. A Counterclockwise –10 chronic level of cyclovertical vergence might indeed drive 50 100 150 the eyes into a basic cyclovertical deviation, one involving time [s] both the vertical rectus muscles and the oblique muscles. But what is this basic cyclovertical deviation? We do not Fig. 2.7 Plot of torsional position for each eye shows ocular counter roll with 45° head tilt. A normal subject is continuously have a name for it. Th e vast majority of idiopathic cyclover- recorded with head straight (STR), right head tilt (RHT), and left tical deviations are termed congenital superior oblique head tilt (LHT) of 45°. Traces show counter rolling of both the paresis, or congenital superior oblique palsy. Yet, recent eyes of 4–7° 2.4 Applications of Bilateral Feedback Control to Clinical Practice and to Future Research 21

Vertical Difference R-L 10 2.4 Applications of Bilateral Feedback Control to Clinical Practice and to Future Research

5 Practically speaking, the consequences of muscle length Right tilt Left tilt adaptation are oft en best appreciated under deep general anesthesia, when the anatomic positions of the eyes can 0 be seen and careful forced ductions can be performed. Th e decision about which eye or eyes to operate on, and which muscles, may best be postponed until obtaining these intraoperative fi ndings. Th is has been advocated by –5 many, including Roth in Switzerland [61], Jampolsky in Down Up[deg] the United States [22], and the author [62]. Because version stimulation is only minimally eff ec- –10 51015 tive in changing extraocular muscle length, surgery time [s] designed to eliminate or minimize extraocular muscle contracture by creating chronic version stimulation, for Fig. 2.8 Vertical recordings, with head straight and tilted 45° to example, by recessing the contralateral medial rectus either side, aft er 45 min of adaptation, with head straight, to a muscle, on the sound eye, in cases of sixth nerve palsy left -over-right vertical disparity of 50+° fi elds of concentric cir- [63], may not work as well as expected. cles. Th e relative positions of the two eyes are shown in the It has long been the teaching in the fi eld of strabismus fusion-free, dissociated state. Th e negative values correspond to the induced right hypodeviation to wait for stabilization of the angle of deviation before intervening surgically. However, the consequences of unguided vergence adaptation and muscle length adapta- vertical disparity. Th is arrangement simulated a relative tion suggest a revision of this teaching. If there is poten- right hyperdeviation, because the right eye had to move tial for fusion, it now appears that every eff ort should be downward to fuse, and the left eye had to move upward. made to realign the eyes without delay, using glasses, Aft er adaptation, the relative positions of the eyes were prisms, and surgery when necessary, and not wait for sta- measured in the fusion-free, dissociated state. Th e eyes bilization. Waiting for stabilization may actually be harm- had partially adapted to the simulated relative right ful if there is fusion potential, for it is now known [64] hyperdeviation by developing a measured right hypode- that the chances for successful restoration of binocular viation. Th e relative shift of the right eye downward of vision decrease with each month that misalignment per- 3° with head straight increased to 5° with right head tilt sists. On the other hand, if fusion potential is truly not (RHT) and decreased to 0° with left head tilt (LHT) present, early surgery may best be postponed. Th e biases (see Fig. 2.8). Th ese changes with forced head tilt are in that exist in the unguided vergence and muscle length the directions that are expected from increased tonus adaptation mechanisms may themselves change over to the normal right superior oblique muscle and to the time, altering the angle of strabismus naturally. Waiting normal left inferior oblique muscle. Th is increased for stabilization of these biases, as refl ected by stability of tonus was produced by vergence adaptation to the rela- the deviation, may indeed be warranted in such cases. tive right hyperdeviation. Th e deviations recorded sim- Th e challenge, therefore, lies in the accurate determina- ply represent a basic cyclovertical deviation induced in tion of fusion potential. a normal subject by vergence adaptation to a vertical Whenever strabismus is corrected, by whatever means, disparity. any fusion that develops will need to compete with any Th e demonstration of such head-tilt changes accom- biases in the vergence and muscle length adaptation panying the induced cyclovertical deviation is in favor mechanisms in order for the eyes to remain straight. It is of the belief that many deviations currently called con- very possible that we shall learn in the future how to mea- genital superior oblique paresis are nothing more than sure such destabilizing biases and learn how to minimize basic cyclovertical deviations of the eyes. To explore this or counteract them by pharmacologic, surgical, or other thesis, these adaptation techniques will be used to study interventional means, in order to help maintain good not only normal subjects but also patients with congeni- binocular alignment aft er we have achieved it. tal and acquired forms of apparent superior oblique For example, selective activation of vergence should paresis. be able to change not only vergence adaptation, but also 22 2 Changes in Strabismus Over Time: The Roles of Vergence Tonus and Muscle Length Adaptation

muscle lengths over time. Th is of course is currently the 3. Baker JD, Parks MM (1980) Early-onset accommodative goal of fusional vergence exercises as part of orthoptic esotropia. Am J Ophthalmol 90:11–18 training. However, eventually we may be able to supply 4. Guyton DL, Weingarten PE (1994) Sensory torsion as the vergence stimulation from external sources, such as is cause of primary oblique muscle overaction/underaction 2 currently done with the transcutaneous electrical stimu- and A- and V-pattern strabismus. Binocul Vis Eye Muscle lation used in orthopedic applications to correct or pre- Surg Q 9:209–236 vent scoliosis as well as contractures in cases of hemiplegia 5. Ludvigh E, McKinnon P, Zaitzeff L (1964) Temporal course or cerebral palsy [65]. To do this, we shall need to dis- of the relaxation of binocular duction (fusion) movements. cover the diff erences between version and vergence stim- Arch Ophthalmol 71:389–399 ulation of the extraocular muscles so as to be able to 6. Carter DB (1965) Fixation disparity and heterophoria fol- supply vergence stimulation selectively. To be sure, cor- lowing prolonged wearing of prisms. Am J Optom Arch rection of strabismus in the future may possibly be by Am Acad Optom 42:141–152 selective electrical stimulation rather than by surgery. 7. Taylor MJ, Roberts DC, Zee DS (2000) Eff ect of sustained cyclovergence on eye alignment: Rapid torsional phoria adaptation. Invest Ophthalmol Vis Sci 41:1076–1083 Summary for the Clinician 8. Ellerbrock VJ (1950) Tonicity induced by fusional move- ■ At least a three-level feedback control system ments. Am J Optom Arch Am Acad Optom 27:8–20 exists for the maintenance of binocular align- 9. Cooper J (1992) Clinical implications of vergence adapta- ment. Of particular interest is the unique regulation. Optom Vis Sci 69:300–307 tion of extraocular muscle lengths by vergence 10. Schor CM (1979) Th e relationship between fusional ver- stimulation as opposed to version stimulation. gence eye movements and fi xation disparity. Vis Res ■ Even though we may treat these mechanisms in 19:1359–1367 a black-box fashion in the beginning, we can use 11. Ogle KN, Prangen Ade H (1953) Observations on vertical this understanding to explain currently observed divergences and hyperphorias. Arch Ophthalmol 49: phenomena such as the development of so- 313–324 called oblique muscle dysfunction with the 12. Crone RA, Hardjowijoto S (1979) What is normal binocu- development of A and V patterns. We also can lar vision? Doc Ophthalmol 47(1):163–199 use this understanding to appreciate previously 13. Hwang J-M, Guyton DL (1999) Th e Lancaster red-green unrecognized patterns of misalignment such as test before and aft er occlusion in the evaluation of incomi- the basic cyclovertical deviation that mimics tant strabismus. J AAPOS 3:151–156 superior oblique muscle paresis. 14. Goldspink G, Williams P (1992) Cellular mechanisms ■ Not all answers are yet known, and some of the involved in the determination of muscle length and mass mechanisms proposed in this chapter are still during growth; problems arising from imbalance between quite speculative. However, from such specula- antagonists muscle groups. In: Proceedings of the mechan- tion, models such as those formulated here can ics of strabismus symposium. Th e Smith-Kettlewell Eye help in the understanding of not only how stra- Research Institute, San Francisco, pp 195–206 bismus changes over time, but also the causes of 15. Tabary J-C, Tardieu C, Tardieu G, Tabary C (1981) the many forms of strabismus, facilitating the Experimental rapid sarcomere loss with concomitant development of preventive measures as well as hypoextensibility. Muscle Nerve 4:198–203 better and longer-lasting treatment methods for 16. Williams PE, Catanese T, Lucey EG, Goldspink G (1988) the future. Th e importance of stretch and contractile activity in the prevention of connective tissue accumulation in muscle. J Anat 158:109–114 17. Goldspink G, Williams P, Simpson H (2002) Gene expression in response to muscle stretch. In: Clinical orthopae- References dics and related research. Lippincott Williams and Wilkins, 1. Guyton DL (2006) Th e 10th Bielschowsky lecture: changes Philadelphia No. 403S, pp S146–S152 in strabismus over time: the roles of vergence tonus and 18. Scott AB (1994) Change of eye muscle sarcomeres accord- muscle length adaptation. Binocular Vis Strabismus Quart ing to eye position. J Pediatr Ophthalmol Strabismus 21:81–92 31:85–88 2. Parks MM (1975) Ocular motility and strabismus. Harper 19. Hayat A, Tardieu C, Tabary J-C, Tabary C (1978) Eff fects of and Row, Hagerstown, Maryland, pp 101 denervation on the reduction of sarcomere number in cat References 23

soleus muscle immobilized in shortened position during 38. Ciancia AO (1995) On infantile esotropia with nystag- seven days. J Physiol (Paris) 74:563–567 mus in abduction. J Pediatr Ophthalmol Strabismus 32: 20. Schor CM, Maxwell JS, Graf EW (2001) Plasticity of con- 280–288 vergence-dependent variations of cyclovergence with ver- 39. Marlow FW (1921) Prolonged monocular occlusion as a tical gaze. Vis Res 41:3353–3369 test for the muscle balance. Am J Ophthalmol 4:238–250 21. Wright WW, Gotzler KC, Guyton DL (2005) Esotropia 40. Marlow FW (1927) Observations on the prolonged occlu- associated with early presbyopia caused by inappropriate sion test. Am J Ophthalmol 10:567–574 muscle length adaptation. J AAPOS 9:563–566 41. Marlow FW (1938) A tentative interpretation of the fi nd- 22. Jampolsky A (1986) Treatment of exodeviations. In: ings of the prolonged occlusion test on an evolutionary Pediatric ophthalmology and strabismus, trans new orleans basis. Arch Ophthalmol 19:194–204 acad ophthalmol. Raven, New York, pp 201–234 42. Abraham SV (1931) Bell’s phenomenon and the fallacy of 23. Gonzalez C, Jaros PA (1988) Strabismus surgery on the the occlusion test. Am J Ophthalmol 14:656–664 nonamblyopic eye. Graefe’s Arch Clin Exp Ophthalmol 43. Beisbarth C (1932) Hyperphoria and the prolonged occlu- 226:304–308 sion test. Am J Ophthalmol 15:1013–1015 24. Breinin GM (1957) Th e position of rest during anesthesia 44. Holmes JM, Kaz KM (1994) Recovery of phorias following and sleep. AMA Arch Ophthalmol 57:323–326 monocular occlusion. J Pediatr Ophthalmol Strabismus 25. Jampel RS (1967) Multiple motor systems in the extraocu- 31:110–113 lar muscles of man. Invest Ophthalmol 6:288–293 45. Viirre E, Cadera C, Vilis T (1987) Th e pattern of changes 26. Keller EL, Robinson DA (1971) Absence of a stretch refl ex produced in the saccadic system and vestibuloocular in extraocular muscles of the monkey. J Neurophysiol refl ex by visually patching one eye. J Neurophysiol 57: 34:908–919 92–103 27. Scott AB, Collins CC (1973) Division of labor in human 46. Liesch A, Simonsz HJ (1993) Up-and downshoot in adduc- extraocular muscle. Arch Ophthalmol 90:319–322 tion aft er monocular patching in normal volunteers. 28. Eberhorn AC, Büttner-Ennever JA, Horn AKE (2005) Strabismus 1:25–36 Identifi cation of motoneurons supplying multiply- or sin- 47. Collins CC, Jampolsky A, Howe PS (1992) Mechanical gly-innervated extraocular muscle fi bers in the rat. limitations of rotation. In: Proceedings of the mechanics of Neuroscience 137:891–903 strabismus symposium. Th e Smith-Kettlewell Eye Research 29. Büttner-Ennever JA (2006) Th e extraocular motor nuclei: Institute, San Francisco, pp 19–40 organization and functional neuroanatomy. Prog Brain Res 48. Castañera de Molina A, Giñer Muñoz ML (1997) Short- 151:95–125 stiff extraocular muscles: Mechanisms involved in EOM 30. Miller JM (2003) No oculomotor plant, no fi nal common adaptations to squint. In: Prieto-Diaz J, Hauviller V (eds) path. Strabismus 11:205–211 XII Congreso del Consejo Latinoamericano de Estrabismo 31. Guyton DL (1984) Discussion of Paez JH, Isenberg S, Apt (CLADE). Grafi ca Lifra, La Plata, pp 503–508 L: torsion and elevation under general anesthesia and dur- 49. Miller MM, Guyton DL (1994) Loss of fusion and the ing voluntary eyelid closure (Bell phenomenon). J Pediatr development of A or V patterns. J Pediatr Ophthalmol Ophthalmol Strabismus 21:78 Strabismus 31:220–224 32. Eustis HS, Nussdorf JD (1966) Inferior oblique overaction 50. Olitsky SE, Juneja RA (1997) Adult onset esotropia with in infantile esotropia: Fundus extorsion as a predictive distance near disparity: a report of two cases. Binocul Vis sign. J Pediatr Ophthalmol Strabismus 33:85–88 Strabismus Q 12:265–267 33. Clark RA, Miller JM, Rosenbaum AL, Demer JL (1998) 51. Simon AL, Borchert M (1997) Etiology and prognosis Heterotopic muscle pulleys or oblique muscle dysfunc- of acute, late-onset esotropia. Ophthalmology 104: tion? J AAPOS 2:17–25 1348–1352 34. Schor C, Horner D (1989) Adaptive disorders of accom- 52. Costenbader FD (1958) Clinical course and management modation and vergence in binocular dysfunction. Ophthal of esotropia. In: Allen JH (ed) Strabismus ophthalmic Physiol Opt 9:264–268 symp II. Mosby, St Louis, pp 325–353 35. von Noorden GK (1988) Current concepts of infantile 53. Th omas AH (2000) Divergence insuffi ciency. J AAPOS esotropia. Eye 2:343–357 4:359–361 36. von Noorden GK (1976) Th e nystagmus compensation 54. Bothun ED, Archer SM (2005) Bilateral medial rectus mus- (blockage) syndrome. Am J Ophthalmol 82:283–290 cle recession for divergence insuffi ciency pattern esotropia. 37. Guyton DL (2000) Dissociated vertical deviation: etiol- J AAPOS 9:3–6 ogy, mechanism, and associated phenomena. J AAPOS 55. Sheedy JE, Saladin JJ (1975) Exophoria at near in presby- 4:131–144 opia. Am J Optom Physiol Optics 53:474–481 24 2 Changes in Strabismus Over Time: The Roles of Vergence Tonus and Muscle Length Adaptation

56. Enright JT (1992) Unexpected role of the oblique muscles 61. Roth A (1983) Oculomotor asymmetry in concomitant in the human vertical fusion reﬂ ex. J Physiol 451: strabismus and its consequences for the choice of surgical 279–293 intervention. In: Castañera de Molina A (ed) Congenital 57. van Rijn LJ, Collewijn H (1994) Eye torsion associated with disorders of ocular motility. Editorial JIMS, Barcelona, 2 disparity-induced vertical vergence in humans. Vis Res pp 89–97 34:2307–2316 62. Castelbuono AC, White JE, Guyton DL (1999) Th e use of 58. Cheeseman EW Jr, Guyton DL (1999) Vertical fusional (a)symmetry of the rest position of the eyes under general vergence: the key to dissociated vertical deviation. Arch anesthesia or sedation-hypnosis in the design of strabis- Ophthalmol 117:1188–1191 mus surgery: a favorable pilot study in 51 exotropia cases. 59. Demer JL, Miller JM, Koo EY, Rosenbaum AL, Bateman Binocul Vis Strabismus Q 14:285–290 JB (1995) True versus masquerading superior oblique 63. Gonzalez C, Chen H.H, Ahmadi MA (2005) Sherrington palsies: muscle mechanisms revealed by magnetic reso- innervational surgery in the treatment of chronic sixth nance imaging. In: Lennerstrand G (ed) Update on stra- nerve paresis. Binocul Vis Strabismus Q 209:159–166 bismus and pediatric ophthalmology. CRC, Boca Raton, 64. Fawcett SL, Birch EE (2003) Risk factors for abnormal bin- pp 303–306 ocular vision aft er successful alignment of accommodative 60. Sato M, Amano E (2003) Clinical ﬁ ndings and surgical esotropia. J AAPOS 7:256–262 results of true and masquerading congenital superior oblique 65. Farmer SE, James M (2001) Contractures in orthopaedic palsy. In: de Faber J-T (ed) Progress in strabismology. Swets and neurological conditions: a review of causes and treat- and Zeitlinger, Lisse, Th e Netherlands, pp 211–214 ment. Disabil Rehabil 23:549–558 Chapter 3 A Dissociated Pathogenesis for Infantile Esotropia 3 Michael C. Brodsky

Core Messages ■ Binocular movements that result from unequal ■ Because dissociated eye movements arise in the visual input to the two eyes are deﬁ ned as setting of infantile strabismus, they have tradi- dissociated. tionally been considered to be the result of dis- ■ Dissociated esotonus, an unrecognized form of rupted binocular vision. binocular dissociation, underlies dissociated hor- ■ Dissociated eye movements may be the cause, izontal deviation. rather than the eﬀ ect, of infantile esotropia.

3.1 Dissociated Eye Movements 3.2 Tonus and its relationship to infantile esotropia Although the term dissociated has historically been restricted to the description of vergence eye movements Tonus refers to the eff ects of baseline innervation on [1–3], in a more general sense it describes any ocular musculature in the awake, alert state. Since the normal movements that result from a change in the relative bal- anatomic resting position of the eyes is an exodeviated ance of visual input to the two eyes [4]. Th ese movements position, extraocular muscle tonus plays a vital physio- arise almost exclusively in the setting of infantile strabis- logic role in establishing ocular alignment [9]. Under mus [5], which has a strong predilection for esotropia normal conditions, binocular esotonus is superimposed over exotropia. Dissociated vertical divergence, latent upon the normal anatomic position of rest to maintain nystagmus, and dissociated horizontal deviation repre- approximate ocular alignment, save for a minimal exo- sent the conditions in which dissociated visual input alter phoria that is easily overcome by active convergence. the position of the eyes [6–8]. It is held that infantile When binocular visual input is preempted early in life, esotropia disrupts binocular control mechanisms and dissociated esotonus gradually drives the two eyes in a thereby engenders these dissociated eye movements [5]. “convergent” position, resulting in infantile esotropia. Th is time-honored notion assumes a distinct and unre- Th us, while convergence functions to actively alter hori- lated pathogenesis for infantile esotropia. It is equally zontal eye position, tonus eff ectively resets the baseline possible, however, that infantile esotropia arises from an eye position. unrecognized form of dissociated deviation known as When superimposed upon a baseline orthoposition, dissociated esotonus. dissociated esotonus manifests as an intermittent esotropia that is asymmetrical or unilateral (Fig. 3.1) [10]. More commonly, dissociated esotonus is superimposed upon a baseline exodeviation, producing an intermittent Summary for the Clinician exodeviation that is asymmetrical, unilateral, or associ- ■ Dissociated eye movements include dissociated ated with a paradoxical esodeviation when the nonpre- vertical divergence, latent nystagmus, and disso- ferred eye is used for fi xation (Figs. 3.2 and 3.3) [11–17]. ciated horizontal deviation. Th ese variants of intermittent exotropia are known as dissociated horizontal deviation. Th e clinical features 26 3 A Dissociated Pathogenesis for Infantile Esotropia

Fig. 3.1 Dissociated horizontal deviation manifesting as a large unilateral intermittent esodeviation (from ref [6], with permission)

Fig. 3.2 Dissociated horizontal deviation with greater exodeviation in the left eye than the right eye (from ref [6], with permission)

distinguishing dissociated horizontal deviation from the intermittent exotropia? Although we use the term inter- nondissociated form of intermittent exotropia are sum- mittent exotropia diagnostically, it is ultimately a descrip- marized in Table 3.1. tive term that includes a variety of diﬀ erent conditions with speciﬁ c diagnostic implications. Th e intermittent exodeviation caused by dissociated horizontal deviation Summary for the Clinician simply constitutes one distinct form of intermittent ■ Tonus determines the contractile state of extraoc- exotropia with its own unique pathophysiology. ular musculature under baseline conditions. Many clinicians apply the hybrid term “intermittent ■ Physiologic tonus maintains normal binocular exotropia/dissociated horizontal deviation” implying that alignment. the two conditions oft en coexist, and perhaps acknowl- edging some diagnostic ambiguity [13, 15–17, 18, 19]. So what are the innervational substrates for these distinct but overlapping categories of intermittent exotropia? Although Burian believed intermittent exotropia to be caused by an 3.3 Esotropia and Exotropia as a Continuum active divergence mechanism [20], independent studies If the dissociated esotonus that manifests as dissociated have found that these patients are approximately 30 PD horizontal deviation gives rise to infantile esotropia, why more exotropic when deeply anesthetized than in the does dissociated horizontal deviation manifest as an awake state [21, 22], suggesting that intermittent exotropia 3.3 Esotropia and Exotropia as a Continuum 27

Fig. 3.3 Dissociated horizontal deviation manifesting as a large left exodeviation when the patient fi xates with the preferred right eye (top and left ) and converting to a right esodeviation with dissociated vertical divergence when the patient fi xates with the nonpreferred left eye (bottom). (All photographs courtesy of Michael Gräf, M.D and from ref [6], with permission) actually results from intermittent fusional control of a exodeviation (Figs. 3.2 and 3.3) [6–8]. Th e distinction large baseline exodeviation [23, 24]. between intermittent exotropia and dissociated When intermittent exotropia is associated with dis- horizontal deviation lies primarily in the relative sociated horizontal deviation, fi xation with either activation of binocular fusion (which behaves as an eye superimposes dissociated esotonus on the base- all-or-nothing phenomenon in most forms of inter- line exodeviation to produce a variable intermittent mittent exotropia), vs. dissociated esotonus (which

Table 3.1. Clinical signs distinguishing dissociated horizontal deviation from other forms of intermittent exotropia [6, 7]

Dissociated horizontal deviation Nondissociated intermittent exotropia

Amplitude of exodeviation is dependent on the Amplitude of exodeviation is independent fi xating eye (i.e., asymmetrical) of the fi xating eye (i.e., symmetrical) Slow velocity of spontaneous exodeviation Rapid velocity of spontaneous exodeviation Variable amplitude of spontaneous exodeviation Constant amplitude of spontaneous exodeviation Positive Bielschowsky phenomenon Negative Bielschowsky phenomenon Associated latent nystagmus and torsional No associated latent nystagmus or torsional ocular ocular rotations, prominent dissociated rotations, little if any dissociated vertical vertical divergence divergence Positive reversed fi xation test Negative reversed fi xation test 28 3 A Dissociated Pathogenesis for Infantile Esotropia

functions as an open-loop process without reference to retain binocular fusion, it can produce a combined clin- ultimate binocular alignment in dissociated horizontal ical picture of intermittent exotropia (with intermittent deviation). Because fi xation with the nonpreferred eye fusion), an asymmetrical exodeviation of the two eyes, exerts greater esotonus [6–8], the baseline exodevia- or an exodeviation of the nonpreferred eye with a para- 3 tion can be unilateral, asymmetrical, or associated with doxical esodeviation of the preferred eye. In classifying a paradoxical esotropia when the nonpreferred eye is these disorders pathogenetically, it becomes critically used for fi xation. important to distinguish sensory motor factors from Infantile esotropia and intermittent exotropia are uni- the diff erent forms of ocular misalignment that they versally regarded as distinct forms of strabismus that ultimately produce. Dissociated horizontal deviation occupy opposite points on a clinical spectrum. In con- shows us how it is only the resultant horizontal devia- trast to infantile esotropia, intermittent exotropia usually tions, and not the underlying conditions, that are dia- has a later onset and is rarely associated with prominent metrically opposed. dissociated eye movements (although small degrees of dissociated vertical divergence can be detected) [25]. At fi rst glance, it is diffi cult to imagine how these diametrical Summary for the Clinician forms of horizontal misalignment are not mutually ■ Dissociated esotonus can be superimposed upon exclusive. the baseline position of the eyes to produce Th e beauty of dissociated horizontal deviation is that intermittent esotropia or intermittent exotropia. it allows us to recast horizontal strabismus as the relative balance of mechanical and innervational forces, without regard to fi nal eye position. Dissociated esotonus can still 3.4 Distinguishing Esotonus be expressed from an exodeviated position, because it is from Convergence generated by unbalanced binocular input that exerts its infl uence upon any baseline deviation. Consequently, Th ere remains the unfortunate tendency in the strabis- intermittent exotropia is a common clinical manifesta- mus literature to confl ate esotonus of the eyes as a base- tion of dissociated esotonus. Mechanistically, there is line innervation with convergence of the eyes as an active nothing sacred about orthotropia as a clinical demarca- function. Jampolsky has emphasized the mechanistic tion, and nothing signatory about the direction of hori- importance of distinguishing between convergence as an zontal misalignment. active binocular function and esotonus as a baseline In this light, dissociated horizontal deviation is trans- innervational state that is centrally driven by unequal formed from a clinical curiosity to a fundamental piece of visual input to the two eyes [21, 29]. Th e importance of the puzzle for understanding horizontal strabismus. Th e this distinction lies in the understanding that conver- exotropic form of dissociated horizontal deviation gence implies a deviation from baseline under normal uniquely embodies the coexistence of the mechanical conditions of sensory input, whereas tonus implies a exodeviating forces that give rise to intermittent exotropia, return to baseline under altered conditions of sensory and the dissociated esotonus that may give rise to infantile input. Th e distinction between convergence (the eff ect) esotropia. For example, infantile exotropia is oft en accom- and monocular esotonus (the cause) lies at the heart of panied by dissociated eye movements such as latent nys- understanding infantile esotropia. Horwood and col- tagmus and dissociated vertical divergence [26, 27]. Some leagues have recently shown that normal infants display infants exhibit an intermittent form of exotropia with fl eeting, large-angle convergent eye movements during other dissociated eye movements [28], suggesting a com- the fi rst 2 months of life, and that these convergent ponent of dissociated horizontal deviation. Patients with movements are ultimately predictive of normal binocu- primary dissociated horizontal deviation also display an lar alignment [30]. By contrast, infantile esotropia tends intermittent exodeviation of one or both eyes with disso- to increase over the period when this excessive conver- ciated ocular signs [13]. gence is disappearing in normal infants [31]. Th is time All of these conditions share a common pathophysi- course challenges the dubious assumption that infantile ology wherein dissociated esotonus is superimposed esotropia arises from excessive convergence output. Our upon a baseline exodeviation to produce an intermit- fi nding of dissociated esotonus shows how we retain a tent exodeviation, which varies in size depending upon primitive tonus system, independent of convergence which eye is used for fi xation. In patients without bin- output, which can operate under conditions of unequal ocular fusion, dissociated esotonus can cause a constant visual input to reset eye position to a new baseline “con- exodeviation to appear intermittent. In patients who vergent” position. 3.5 Pathogenetic Role of Dissociated Eye Movements in Infantile Esotropia 29

and central (fi xational) refl exes augment dissociated eso- Summary for the Clinician tonus, and lead over time to infantile esotropia. Subcortical ■ Since large convergent movements in early visual refl exes would provide the default system through infancy are predictive of normal binocular align- which dissociated esotonus operates to re-establish the ment, infantile esotropia does not result from baseline horizontal eye position. Th is process can ulti- excessive convergence. mately lead to loss of sarcomeres and secondary shortening of the medial rectus muscles. Th e fact that the eyes straighten considerably under general anesthesia [18, 22, 29, 38, 39], however, suggests that esotonus is the driving force for infantile esotropia, and that mechanical eff ects 3.5 Pathogenetic Role of Dissociated Eye Movements in Infantile Esotropia play a secondary role in its pathogenesis. It is possible that stable, large-angle esodeviation that we recognize as infan- Contrary to the stereotype of “congenital” esotropia as a tile esotropia simply represents the fi nal stage of dissoci- large-angle deviation that is present at birth, most cases of ated esotonus. As with many other forms of ocular “congenital” esotropia are acquired (i.e., “infantile” in ori- misalignment, the constant esodeviation that develops gin) [25, 32]. Furthermore, the eyes do not simply snap in over time may eventually obscure the pathogenesis. to their fi nal esotropic position. Before 12 weeks of age, Early monocular visual loss is known to generate nascent infantile esotropia is an intermittent, variable esotonus and reproduce the same constellation of dis- esodeviation that gradually becomes constant aft er build- sociated eye movements that accompany infantile ing in intensity to a large, fi xed-angle of horizontal mis- esotropia [18]. Patients with unilateral congenital cata- alignment [32, 33]. Ing has noted that 50% of patients ract oft en develop large-angle esotropia, latent nystag- with infantile esotropia show an increase in the measured mus, dissociated vertical divergence, and a head turn to angle between the time of fi rst examination and the date fi xate in adduction with the preferred eye [18]. By con- of surgery [34]. Clearly, unequal visual input in infancy trast, early infantile esotropia is oft en characterized by must produce a gradual and progressive increase in the similar visual acuity in the two eyes, and with alternat- angle of esotropia. Th at this esodeviation appears during ing suppression of the nonfi xating eye. So perhaps dis- the early period when stereopsis is developing, but before sociated horizontal deviation is not an epiphenomenon macular anatomy has matured suffi ciently to provide high of infantile esotropia, but a “footprint in the snow” of resolution acuity [35] suggests that it is actively driven the horizontal tonus imbalance that is actually respon- primarily by an imbalance in peripheral visual input. sible for its inception. In a recent hypothesis, Guyton has invoked vergence adaptation and muscle length adaptation to explain how a small innervational bias (such as the convergence pro- Summary for the Clinician duced by increased accommodative eff ort in the presby- ope) can build slowly over time into a large constant ■ Dissociated esotonus may provide the physio- deviation [36]. Vergence adaptation refers to the tonus logic substrate for vergence adaption in infancy. levels that normally operate to maintain a baseline ocular ■ If so, then dissociated esotonus is the cause, alignment and thereby minimize retinal image disparity. rather than the eff ect, of infantile esotropia. According to Guyton, vergence adaptation can allow ■ Th e prevailing concept of infantile esotropia as primitive ocular motor biases to gradually amplify and the proximate cause of dissociated deviations create strabismic deviations under pathological condi- may need to be revised. tions [36]. Muscle length adaptation refers to the change in extraocular muscle length due to gain or loss of sarcomeres. Muscle length adaption is driven in part by the physiologic eff ects of vergence adaptation. Acknowledgment Portions of this chapter have previ- Dissociated esotonus may provide the sensorimotor ously been published in a thesis for the American substrate for vergence adaptation when binocular cortical Ophthalmological Society [6] and in its two derivative control mechanisms fail to take hold. Th e fi nding of a pos- papers published in the Archives of Ophthalmology [7, 8]. itive Bielschowsky phenomenon in dissociated horizontal All fi gures in this chapter are used with permission from deviation [15, 17] shows that peripheral luminance refl exes the American Medical Association and the American are retained, as in dissociated vertical divergence [37]. In Ophthalmological Society. This chapter is abstracted from this setting, both peripheral (luminance and optokinetic) an American Ophthalmological Society thesis [6]. 30 3 A Dissociated Pathogenesis for Infantile Esotropia

18. Th ouvenin D, Nogue S, Fontes L, Norbert O (2004) References Strabismus aft er treatment of unilateral congenital cata- 1. Bielschowsky A (1904) Über die Genese einseitiger racts. A clinical model for strabismus physiopathogenesis? Vertikalbewegungen der Augen. Z Augenheilkd 12: 545–557 In: de Faber (ed) Transactions 28th European strabismologi- 3 2. Bielschowsky A (1930) Die einseitigen und gegensinnigen cal association meeting, Bergen, Norway, 2003. London, (“dissoziierten”) Vertikalbewegungen der Augen. Albrecht Taylor and Francis, pp 147–152 Von Graefes Arch Ophthalmol 125:493–553 19. Wilson ME, Hutchinson AK, Saunders RA (2000) 3. Bielschowsky A (1938) Disturbances of the vertical motor Outcomes from surgical treatment for dissociated hori- muscles of the eye. Arch Ophthalmol 20:175–200 zontal deviation. J AAPOS 4:94–101 4. Lyle TK (1950) Worth and Chavasse’s Squint. Th e binocu- 20. Burian HM (1971) Pathophysiology of exodeviations. In: lar refl exes and treatment of strabismus. Blakiston, Manley DR (ed) Symposium on horizontal ocular devia- Philadelphia, pp 40–41 tions. CV Mosby, St Louis, pp 119–127 5. Brodsky MC (2005) Visuo-vestibular eye movements. 21. Jampolsky A (1970) Ocular divergence mechanisms. Trans Infantile Strabismus in Th ree Dimensions. Arch Oph- Am Acad Ophthalmol 68:808 thalmol 123:837–842 22. Romano P, Gabriel L, Bennett W, et al (1988) Stage I intra- 6. Brodsky MC (2007) Dissociated horizontal deviation: clin- operative adjustment of eye muscle surgery under general ical spectrum, pathogenesis, evolutionary underpinnings, anesthesia: consideration of graduated adjustment. Graefes diagnosis, treatment, and potential role in the development Arch Clin Exp Ophthalmol 226:235–240 of infantile esotropia. Transact Am Acad Ophthalmol 105: 23. Kushner BK (1992) Exotropic deviations: a functional clas- 272–293 sifi cation and approach to treatment. Am Orthop J 38: 7. Brodsky MC, Fray KJ (2007) Dissociated horizontal devia- 81–93 tion aft er surgery for infantile esotropia – clinical charac- 24. Kushner BJ, Morton GV (1998) Distance/near diff erences teristics and proposed pathophysiologic mechanisms. Arch in intermittent exotropia. Arch Ophthalmol 116:478–486 Ophthalmol 125:1683–1692 25. Pritchard C (1998) Incidence of dissociated vertical 8. Brodsky MC, Fray KJ (2007) Does infantile esotropia arise deviation in intermittent exotropia. Am Orthop J 48: from a dissociated deviation? Arch Ophthalmol 125: 90–93 1683–1692. 26. Moore S, Cohen RL (1985) Congenital exotropia. Am 9. Proceedings of the Smith-Kettlewell Ocular Motor Tonus Orthop J 35:68–70 Symposium. Tiberon, California, (June 2–4, 2006) pp 2–4 27. Rubin SE, Nelson LB, Wagner RS, et al (1988) Infantile 10. Spielmann A (1990) Déséquilibres verticaux et torsionnels exotropia in healthy infants. Ophthalmic Surg 19: dans le strabisme précoce. Bull Soc Ophtalmol Fr 792–794 4: 373–384 28. Hunter DG, Kelly JB, Ellis FJ (2001) Long-term outcome 11. Quintana-Pali L (1990) Desviacion horizontal disociada. of uncomplicated infantile exotropia. J AAPOS 5: Bol Oft almol Hosp de la Luz 42:91–94 352–356 12. Romero-Apis D, Castellanos-Bracamontes A (1990) 29. Jampolsky A (2005) Strabismus and its management? In: Desviacion horizontal disociada (DHD). Rev Mex Oft almol Taylor DS, Hoyt CS (eds) Pediatric ophthalmology and 64:169–173 strabismus, 3rd edn. Elsevier Saunders, London, New York, 13. Romero-Apis D, Castellanos-Bracamontes A (1992) pp 1001–1010 Dissociated Horizontal deviation: clinical fi ndings and 30. Horwood A (2003) Too much or too little: neonatal ocular surgical results in 20 patients. Binoc Vis Q 7:173–178 misalignment frequency can predict lateral abnormality. 14. Spielmann AC, Spielmann A (2004) Antinomic deviations: Br J Ophthalmol 87:1142–1145 esodeviation associated with exodeviation. In: Faber TJ 31. Horwood AM, Riddell PM (2004) Can misalignments in (ed) Transactions 28th meeting European strabismological typical infants be used as a model for infantile esotropia? association, Bergen, Norway, June 2003. Taylor and Francis, Invest Ophthalmol Vis Sci 45:714–720 London, pp 173–176 32. Pediatric eye disease investigator group. (2002) Spontaneous 15. Wilson ME, McClatchey SK (1991) Dissociated horizontal resolution of early-onset esotropia: Experience of the con- deviation. J Pediatr Ophthalmol Strabismus 28:90–95 genital esotropia observational study. Am J Ophthalmol 16. Wilson ME (1993) Th e dissociated strabismus complex. 133:109–118 Binocul Vis Strabismus Q 8:45–46 33. Pediatric eye disease investigator group. (2002) Th e clinical 17. Zabalo S, Girett C, Domínguez D, Ciancia A (1993) spectrum of early-onset esotropia: Experience of the con- Exotropia intermitente con desviación vertical discodiada. genital esotropia observational study. Am J Ophthalmol Arch Oft almol B Aires 68:11–20 133:102–108 References 31

34. Ing MR (1994) Progressive increase in the quantity of devi- 37. Brodsky MC (1999) Dissociated vertical divergence. A ation in congenital esotropia. Trans Am Ophthalmol Soc righting reﬂ ex gone wrong. Arch Ophthalmol 117: 92:117–131 1215–1222 35. Fawcett SL, Wang YZ, Birch EE (2005) Th e critical period 38. Apt L, Isenberg S (1977) Eye position of strabismic for susceptibility of human stereopsis. Invest Ophthalmol patients under general anesthesia. Am J Ophthalmol 84: Vis Sci 46:521–525 574–579 36. Guyton DL (2006) Changes in strabismus over time: the 39. Roth A, Speeg-Schatz C (1995) Eye muscle surgery. Basic roles of vergence tonus and muscle length adaptation. data, operative techniques, surgical strategy. Swets and Binocul Vis Strabismus Q 21:81–92 Zeitlinger, Masson, Paris, pp 283–324 Chapter 4 The Monoﬁ xation Syndrome: New Considerations on Pathophysiology 4 Kyle Arnoldi

Core Messages ■ Parks’ monofi xation syndrome (MFS) is an ■ MFS associated with small angle esotropia is abnormality of binocular vision consisting of a the most common form, the most stable, and foveal suppression scotoma, peripheral sensory the form that allows for the best binocular fusion, fusional vergence, and stereopsis. A vision. Th is may be due to the natural superior- majority of cases also demonstrate small angle ity of the nasal retina and its input to the visual strabismus or amblyopia, but these are secondary cortex. to the monofi xation and not characteristics of the ■ Monofi xation is a desirable state when bifi xation syndrome. is not possible. Nothing is gained, and much can ■ Animal studies have begun to clarify the path- be lost, if a cure is attempted. ways for normal binocular vision, and anatomic ■ Very early repair of strabismus or anisometropia and metabolic adaptations which may result in may prevent the development of monofi xation in monofi xation. favor of bifi xation.

raised in his original manuscript. Why would an ortho- 4.1 Introduction tropic patient with no history of strabismus or ani- In 1969, in his American Ophthalmological Society the- sometropia have primary MFS? Is the foveal suppression sis, Marshall Parks described 100 patients with a specifi c the cause or the result of MFS? Why do some cases mani- set of sensory fi ndings: a foveal suppression scotoma; fest a small tropia in the presence of motor fusion ampli- peripheral sensory fusion; motor fusion amplitudes tudes that are more than suffi cient to overcome the (fusional vergence); and gross stereopsis. He termed this deviation? What is the state of binocular correspondence constellation of fi ndings the monofi xation syndrome in the strabismic and nonstrabismic cases of MFS? Can (MFS) to distinguish it from bifi xation (or bi-foveal fi xa- monofi xation be prevented? Can it be cured? And if so, tion) [1]. Parks outlined four principle causes of MFS: (1) should a cure be attempted? Recent clinical and labora- anisometropia (found in 6% of his cases); (2) corrected tory studies have shed some light on the features and strabismus (66%); (3) an organic macular lesion (1%); pathophysiology of MFS which may help us begin to and (4) primary MFS (19%). Another 8% had both ani- answer some of these questions. sometropia and a history of strabismus. Although 66% of his cases had a small angle, manifest, horizontal ocular deviation, strabismus is not included as 4.2 Normal and Anomalous Binocular Vision a characteristic of MFS, emphasizing that this is a sensory disorder. Similarly, Parks considered amblyopia a variable Th e MFS is an abnormality of binocular vision. In normal feature rather than a characteristic of the syndrome, binocular vision, bilateral retinal input from overlapping occurring as a result of MFS in 77%. Like the small angle visual fi elds is projected to the same general location in manifest strabismus, he felt the presence or absence of the visual cortex, stimulating adjacent ocular dominance amblyopia was dependent on associated factors such as a columns of opposite ocularity [2]. Th is close proximity of history of infantile strabismus or anisometropia. input from the two eyes corresponding to the same point Since its original description, there has been much in space facilitates the communication necessary for bin- study and debate regarding questions that Parks himself ocular single vision. Th is communication appears to take 34 4 The Monofi xation Syndrome: New Considerations on Pathophysiology

place within a population of binocular cells, neurons that theoretically capable of joining visual receptive fi elds up receive input from both eyes and are sensitive to image to 2.5° (4.4D) distant [6]. In Parks’ original description, disparity. Th ese cells are prevalent throughout the super- manifest deviations no larger than 8D were consistent fi cial and deep layers of area V1, as well as several areas with MFS. A two-neuron chain could allow the fovea to 4 outside the striate cortex such as areas V2, MT (middle eff ectively communicate with a peripheral retinal ele- temporal visual area or area V5), and MST (medial supe- ment that is up to 8.7D away, providing support to Parks’ rior temporal visual area), and play a major role in the clinical observations. appreciation of stereopsis and in generating disparity vergence (motor fusion). In the presence of strabismus, inputs from the same 4.2.1 Binocular Correspondence: point in space will stimulate nonadjacent ocular domi- Anomalous, Normal, or Both? nance columns, cells that would ordinarily not communicate with each other horizontally, or synapse with the Interestingly, one of the questions raised by Parks and same binocular cell further downstream in visual pro- debated for decades is whether the binocular vision that is cessing. Unrepaired, large angle infantile-onset strabis- the prominent feature of MFS should be called ARC, nor- mus has been shown to have devastating eff ects on the mal correspondence (NRC) with an expansion of Panum’s population of binocular cells. Th e supply of binocular fusional space in the peripheral fi eld (Parks’ conclusion), cells throughout area V1 is decimated [3]. Yet objective or even a combination of the two. Some authors have evidence of binocular cortical processing has been found found NRC in the central visual fi eld, with ARC in the in human subjects with small angle strabismus and MFS periphery [8, 12]; others have found ARC centrally, and [4, 5]. Th e question then arises, how is it that these NRC peripherally [13]. Certainly, the angle of strabismus patients can achieve fusion and stereopsis? is small enough and the peripheral receptive fi elds large One theory is that the cortical adaptation that occurs enough that it is conceivable peripheral fusion might be in response to a small angle ocular deviation is limited to achieved without requiring a rewiring of the visual cortex suppression of the foveal ocular dominance columns in (see Sect. 4.2). On the other hand, it seems unlikely that area V1. Th is would preserve the parafoveal columns and stereoacuity as fi ne as 70 seconds of arc, which has been allow for normal, though limited binocular communica- found in MFS, could be consistent with a foveal suppres- tion with gross stereopsis [3]. Th is theory also implies sion scotoma of up to 5° with NRC. Perhaps stereoacuity at that the anomalous motor fusion present in MFS is also this level is the result of an expansion of Panum’s area sur- driven by the disparity-sensitive neurons that are located rounding the fi xation point. However, such an adaptation, at this earliest stage of binocular processing [6]. In this should it be found, would surely be termed anomalous. paradigm, retinal correspondence would be considered What do we mean when we say a patient has ARC? normal, as no cortical rewiring would be needed to main- Th e state of retinal correspondence has historically been tain fusion in the presence of a small deviation. defi ned as characteristic responses to specifi c clinical sen- Other researchers have found evidence of an adapta- sory tests; responses which can be manipulated by many tion that results in binocular vision in MFS; one that diff erent external factors [14]. Test results are also infl u- occurs further downstream from area V1, in areas V2, V3, enced by both the patient’s ability to communicate and and beyond. Th is adaptation does involve a rewiring that the examiner’s interpretation of the response. It is not could be considered the anatomic basis of anomalous uncommon for the same subject to demonstrate charac- retinal correspondence (ARC) [7, 8]. For example, it has teristic ARC responses on some tests and NRC responses been demonstrated in esotropic cats that if the angle of on others. It has been assumed that ARC is the result of a strabismus is small (<10°), the binocular neurons in the shift in the perceptual mapping of the deviated eye under lateral suprasylvian cortex (area LS) may be spared, binocular conditions, and these tests are designed to though their receptive fi elds are shift ed so that normally determine the subjective visual direction of at least one noncorresponding retinal elements may communicate retinal element. However, in human subjects with ARC, [9, 10]. Area LS of the cat is functionally analogous to area no cortical shift in topography was found with pattern MT in the primate. VEP, though this does not rule out a shift occurring in Regardless of where the adaptation takes place, it cortical areas further downstream [7]. appears that the visual cortex may be most successful in It is important to remember that the concepts of the achieving fusion in the presence of a tropia when it can horopter, Panum’s fusional space, and binocular corre- combine information from cell populations that are no spondence are simply geometric and psychophysical con- more than two cortical neurons distant [11]. At approxi- structs used to describe binocular vision. Until we know mately 7 mm in length, the typical cortical neuron is how this binocular vision is achieved in the visual cortex, 4.3 MFS with Manifest Strabismus 35 perhaps it is more important to recognize that patients alignment and fusional vergence is immature in neonates, with MFS indeed have binocular correspondence, rather but more oft en results in transient over-convergence as than how we label that correspondence. Either way, as dis- opposed to over-divergence [20]. Pathways for nasally cussed earlier, animal studies are beginning to reveal a directed pursuit are more developed at birth compared possible anatomical basis for the clinical observations with those for temporally directed pursuit. Interruption of described in MFS. Until these anomalous neural connec- maturation due to an insult such as early-onset, unrepaired tions can be shown in a human subject with the clinical strabismus, leads to permanent monocular naso-temporal features of MFS, the debate remains unresolved. pursuit asymmetry [21]. It may also lead to latent nystagmus, which typically features a pathologic nasally directed pursuit movement of the fi xating eye, followed by a physiologic temporal-ward refi xation saccade [18]. Th ese motor 4.3 MFS with Manifest Strabismus fi ndings associated with infantile esotropia seem to sug- Th e majority of patients with MFS have a manifest strabis- gest that the infant visual system is biased to convergent mus, and esotropia is the most prevalent form by a wide alignment when normal development is interrupted. margin. Th e prevalence of micro-esotropia in several large series of primary and secondary MFS has been reported from 61 to 90% [1, 15]. MFS with small angle exotropia is 4.3.2 Esotropia Allows for Better less common, occurring in 8–21% [1, 15, 16]. Th e preva- Binocular Vision lence of MFS associated with small angle vertical strabismus is extremely low at 0–3% in large series [1, 15, 16]. Fusion and stereopsis may be more likely to develop if the Choi and Isenberg described 40 cases of MFS with a ver- ocular deviation is less than 9D though presumably, the tical tropia; however, the prevalence of this variety of greater the number of cortical neurons necessary to link MFS cannot be determined from their report [17]. nonadjacent ocular dominance columns, the poorer the quality of the resulting binocular vision. Deviations up to 20D have been shown to support peripheral sensory fusion [14], if not stereopsis, so it is no surprise that 4.3.1 Esotropia is the Most Common peripheral fusion is a feature of MFS. However, in a recent Form of MFS study, the maximum angle of horizontal strabismus con- Apparently, monofi xation can be achieved and main- sistent with true stereopsis was found to be only 4D [16], tained with any type of strabismus. However, the esotro- which happens to correspond with the approximate pic variety of MFS is so prevalent it is unlikely that this length of one cortical neuron. occurs by chance. New evidence suggests that a conver- Th e maximum angle of strabismus that still allows for gent deviation may be the default position if orthotropia fusional vergence is not yet known, though the most robust with bifi xation is not possible [6]. convergence response to binocular image disparity in As discussed in Sect. 4.2, studies comparing normal and monkeys with MFS occurs at 4.0–4.5D of crossed disparity strabismic monkeys have found that an early onset unre- [22], once again corresponding with the length of the aver- paired strabismus will deplete the supply of binocular con- age cortical neuron. Th e motor fusion amplitudes of nections in area V1, as well as cause low metabolic activity human subjects with MFS have been found to be within (suppression) in ocular dominance columns correspond- the normal range by some [1, 13, 23], and present but sub- ing to the deviating eye [3, 6, 18]. Binocular processing normal by others [24]. Th ough patients with MFS oft en begins in the layers above and below input layer have fusional vergence suffi cient to overcome small angles 4 of area V1 in the striate cortex, but continues in several of strabismus, most patients with MFS maintain a manifest diff erent populations of binocular cells within and beyond strabismus. Th e logical conclusion is that, in patients with area V1 that are sensitive to either relative or absolute reti- MFS, there is a greater functional benefi t to keeping the nal image disparity. Th ese cell groups give rise to stereopsis eyes slightly misaligned, particularly on the esotropic side. or fusional vergence, respectively [19]. Vergence neurons MFS with esotropia diff ers slightly from MFS with sensitive to crossed disparity (convergence) appear to be exo- or hypertropia. Not only is it more common, but it is naturally more numerous than those coding for uncrossed the form that allows for the best binocular vision. In a disparity (divergence) in normal monkeys [6]. It is possible large series, the micro-ET group out-performed the other that more convergence neurons survive the early insult sim- two alignment categories by a wide margin in each of the ply because there is a preponderance of them to begin with. three sensory categories: sensory fusion, motor fusion, Th e timing of the insult is probably also contributory and stereopsis [15]. Th e most striking diff erence in the to the prevalence of small angle esotropia in MFS. Eye sensory exam was found in the motor fusion category. 36 4 The Monofi xation Syndrome: New Considerations on Pathophysiology

Both primary and secondary micro-esotropes were sig- even in the presence of high-quality binocular vision [15, nifi cantly more likely to have disparity vergence than the 36, 37]. Twenty-four to 26% of MFS cases deteriorate over exotropes or hypertropes. a period of 5.5–17.5 years [15, 36, 37]. In these studies, Why might binocular vision be better in MFS with deterioration was not the result of loss of sensory status. 4 esotropia? In esotropia, the fovea of the fi xating eye must Following treatment, 48–80% of subjects were able to communicate with a nonfoveal point on the nasal retina regain monofi xation status. of the deviating eye to achieve fusion. In exotropia, the Stability of MFS with exo- or hypertropia appears to fi xating fovea must link with a point on the temporal be more vulnerable to insults to the visual system such as retina of the deviating eye. However, not all areas of the dense amblyopia or a signifi cant change in the refractive retina are created equal. Temporal retina is at a competi- error over time [15]. Dense amblyopia appears to be dis- tive disadvantage, even in the normal, nonstrabismic ruptive to an already fragile binocular connection in visual system. Cones and ganglion cells are 1.5-fold less exotropia, and may contribute to instability in the major- numerous in the temporal retina [25–28]. LGN layers ity of exotropic patients. Drastic changes in refractive receiving input from the ipsilateral temporal retina have error in MFS with exotropia appear to have a similar fewer cells and less volume [29]. And in the visual cortex, destabilizing eff ect. Neither of these factors appears to temporal ocular dominance columns occupy less terri- have an eff ect on long-term stability in micro-esotropia, tory than nasal columns, with the diff erence increasing however. dramatically with retinal eccentricity [30]. Temporal Instability of alignment in MFS is also associated with retina matures slower than nasal retina in normal human the presence of vertically incomitant horizontal strabis- infants [31]. Spatial resolution and vernier acuity are mus, oblique dysfunction, and a history of large-angle poorer in the temporal retina of normal eyes [32–34]. infantile esotropia. Micro-esotropes were statistically less Th e critical period for the development of the temporal likely to have a history of any of these associated motility retina and its connections in the visual cortex begins disorders in one study [15]. later and takes longer to complete than that for nasal retina [31]. And fi nally, the neural mechanisms underlying disparity detection from uncrossed disparity (as would occur in exotropia) are naturally more sensitive to 4.4 Repairing and Producing MFS image decorrelation than those from crossed disparity Any mechanic will tell you that one of the best ways to [35]. If the critical period is interrupted by strabismus, understand something is to take it apart and reassemble the temporal retina should be selectively penalized, it. Can MFS be “taken apart” or cured? Curing MFS means potentially magnifying the anatomic and physiological elimination of the foveal suppression scotoma, which is asymmetry. relatively simple to accomplish, and restoring bifi xation Th is presents a particular problem for exotropia. If with fusion and high grade stereopsis, which is consider- inputs from the temporal retina are less numerous, ably more diffi cult. Most researchers (including Parks) delayed in development, relatively suppressed, and more believe that a patient with MFS cannot be restored to vulnerable to the deleterious eff ects of image decorrela- bifi xation [1, 13, 38, 39]. Th ere is also very little in the tion, the foveal cortical neurons of the dominant eye current literature to suggest that this is possible. A single would have comparatively few neurons from the deviated study claims to have cured MFS in nine patients [40], and eye with which to work. Th e larger the angle of exotropia, another reports a spontaneous resolution of MFS and the fewer are the temporal cortical neurons available to amblyopia in a small group of older children and teenag- link with the columns of the dominant eye because of the ers [41]. In the former study, of 30 patients with amblyo- increase in the ratio of dominance with retinal eccentric- pia and eccentric fi xation, nine improved stereoacuity ity. Th e relative suppression of these temporal neurons below the threshold for MFS (60 s of arc or better) that may result in poor quality communication, even if a link coincided with improvement in visual acuity. However, could be established. since stereoacuity is dependent on spatial resolution as well as alignment, and at least seven of these patients had no manifest strabismus prior to occlusion therapy, it may be that the treatment simply cured amblyopia, rather 4.3.3 Esotropia is the Most Stable Form than MFS. Good binocular vision is associated with stability, but does To the contrary, there seems to be opinion backed by not guarantee lasting alignment. Studies have found that evidence to suggest that MFS cannot be cured, but more stability of alignment in microtropia is not permanent, importantly, a cure should not be attempted [42]. 4.4 Repairing and Producing MFS 37

Antisuppression and treatment of ARC typically lead to Case 4.1 insuperable diplopia (see Case 4.1) [39, 43]. For those with an associated small tropia, nothing is gained by A 5-year old female was diagnosed with monofi x- attempted correction of the deviation with surgery or ation syndrome following a failed pre-school vision prism, because the monofi xation persists and the devia- screening. Th e patient completed a course of optom- tion recurs. Normal or near-normal fusional vergence in etric vision training designed to eliminate the foveal these patients assures that alignment will be maintained suppression scotoma in the left eye. Once constant, at the visual system’s preferred angle, regardless of intractable diplopia was present, and the patient was attempts at intervention. In addition, patients with MFS referred to an orthoptist and pediatric ophthalmolo- are typically asymptomatic and already enjoy high quality gist for the management of diplopia. Th e patient was binocular vision. If bifi xation could be restored, it may 6-years old when presented to the ophthalmologist. not result in a signifi cant improvement in quality of life. Vsc: 20/20 Can MFS be restored once deconstructed? If it is pos- 20/25 sible to lose the suppression ability due to trauma, occlusion or loss of vision in the preferred eye, or therapeutic Motility: intervention, might it be possible to restore it? Very little Dsc has been published in this area. Th e cases in the literature LET 5Δ with simultaneous prism and cover test suggest that suppression cannot be relearned once Builds to E 20Δ with prism and alternate cover unlearned [43]. However, the prognosis may depend on Nsc what caused the loss of suppression, as Case 4.2 shows. LET 5Δ with simultaneous prism and cover Because MFS cannot or perhaps should not be cured Builds to E(T) 20Δ with prism and alternate once established, a better question might be “Can MFS be cover prevented in favor of bifi xation?” Parks hypothesized Sensory: that, for those cases of secondary MFS, correction of the Constant, uncrossed diplopia at distance and near, underlying pathology, whether strabismus or anisometro- unrelieved with any combination of prism. pia, before 6 months of age may be the answer. Th is is a Amblyoscope examination: challenging hypothesis to study in human subjects; such Objective angle (Grade I target) = +20 early intervention is oft en logistically diffi cult. An animal Subjective angle (Grade I target) = +5 model is better suited to answer this question. Grade II: constant, variable diplopia, no sensory fusion, no suppression; as image approaches +5 on amblyoscope, diplopia converts from uncrossed to crossed. 4.4.1 Animal Models for the Study of MFS Management: Th ere are several valid methods for creating the clinical Bilateral medial rectus recessions were done for the conditions associated with the development of MFS. decompensating near deviation. At the 1-day post- Large angle esotropia has been surgically induced in operative visit, the diplopia was unchanged. Exam- infant monkeys [23], or simulated with the use of prism ination results at that visit are as follows. glasses [18, 21]. One can create large angle sensory stra- Post-op Motility: bismus through monocular or binocular occlusion early Dsc: in life [44–46]. One can also create an animal model for LET 5Δ with simultaneous prism and cover test anisometropia by using optical defocus with minus lenses Builds to E 20Δ with prism and alternate cover test [47, 48]. With each of these methods, the timing of the Nsc: repair of the induced strabismus or anisometropia deter- LET 5Δ with simultaneous prism and cover test mines the sensory outcome. If the image decorrelation is Builds to E 20Δ with prism and alternate cover test repaired in the infant monkey by 3 weeks of age (corre- Post-op Exam 2: lates to 3 months in human infants), bifi xation can result At 1 month following surgery, the motility and sen- in some animals [3, 6]. If delayed for up to 24 weeks, sory examinations were unchanged from presenta- bifi xation is not possible, but MFS can result. If delayed tion. Th e patient was off ered an occlusion foil to longer than 24 months, not only do the monkeys show a alleviate the diplopia. Th e mother declined as she lack of sensory fusion, motor fusion and stereopsis, but viewed this as “a step backwards” aft er all the vision they tend to develop latent nystagmus, asymmetry of pur- therapy that was done. suit and OKN, A- and V- pattern incomitance, and 38 4 The Monofi xation Syndrome: New Considerations on Pathophysiology

Case 4.2 4.5 Primary MFS (Sensory Signs of Infantile-Onset Image Decorrelation) A 16-year old female with a history of monofi xation syndrome presents with a 3-month history of constant hori- Th e problem of primary MFS is one that has perplexed 4 zontal diplopia. Th e onset of the diplopia was abrupt, Parks and others. Primary MFS accounts for 16–19% of following closed head trauma without loss of conscious- all cases of monofi xation [1, 15]. Th is subpopulation is ness, secondary to a motor vehicle accident. Th e patient interesting, as it may represent the visual cortex’s active reports that the diplopic image is always present, but is choice when bifoveal fi xation is not possible for some rea- not always in the same location relative to fi xation and son. But what is that reason? One theory that has been appears to be constantly moving. Previous records doc- debated for decades is the possibility that some individu- ument a stable RET 6Δ, with a superimposed phoria of als have an inherent inability to bi-fi xate. However, no up to 18Δ in addition to the sensory features of monofi x- evidence of a genetic absence of disparity detectors has ation syndrome. been uncovered thus far. In a recent study, there was no data found to support the hypothesis that MFS is a motor Vcc: 20/20 OD Rx: +0.50 +1.00 ´ 090 adaptation to an inherent ARC [13]. 20/20 OS Plano To the contrary, since animal studies have demon- Motility: strated that even a brief interval of image decorrelation Dcc: early in the critical period of development can lead to RET variable from 8Δ to 25Δ MFS, one answer may be that the patient had strabismus Ncc: or anisometropia that spontaneously resolved in early RET variable from 10Δ to 25Δ infancy. In a recent study, the presence of image decorre- Sensory: lation for only 3 days, if occurring at the height of the Constant uncrossed horizontal diplopia of variable critical period, was found to cause dramatic changes in magnitude, unrelieved with prism. Th e addition of cortical processing of binocular input in monkeys [49]. base-out prism in free space appears to cause an Th e fi rst change caused by early image decorrelation is increase in the esodeviation, with diplopia. suppression in area V1, beyond the input level in layer Amblyoscope examination: four. Apparently, once begun, this process of low meta- Objective angle (Grade I targets) = +25 bolic activity spreads quickly. Th e longer the period of Th ere was no subjective angle at which the patient image decorrelation, the more prevalent the suppression could appreciate sensory fusion with either Grade I becomes in all layers of V1. Once suppression is estab- or II targets. lished, the developing cortex may have no choice but to work around it to achieve the best binocular vision pos- Management: sible under the circumstances. Th e patient was prescribed a dense occlusion foil (Bangerter Light Perception foil) for the right lens of the glasses. At her 2-week follow-up, the fi lter strength was reduced to Bangerter 0.2. Two weeks 4.5.1 Motor Signs of Infantile-Onset later, the strength was reduced again to Bangerter Image Decorrelation 0.4. Th e fi lter was discontinued 1 month later, with Secondary abnormalities of ocular motility associated complete resolution of the diplopia. Her sensory with early-onset image decorrelation are well documented. and motor examination returned to baseline level, Patients with uncorrected infantile-onset strabismus oft en and has been stable for over 2 years. develop latent nystagmus, dissociated vertical deviation, and A- or V-pattern incomitance, as well as demonstrate persistent naso-temporal pursuit and OKN asymmetry dissociated vertical deviation [18, 21, 45, 46]. Research (see Sect. 4.5). Th e age of onset of binocular decorrelation shows that in the animal model, shorter durations of appears to determine whether these signs will be present, image decorrelation result in better sensory outcomes. and the duration of binocular decorrelation determines Th ese results imply that excellent outcomes may be pos- the severity [18, 21, 44–46]. Occasionally these motor sible in the human if alignment is restored within 90 days signs may be observed in cases of secondary MFS (see of the onset of strabismus. Th is suggests that, if monofi x- Sect. 4.3.3) following strabismus repair, but they are par- ation is to be prevented in favor of bifi xation, very early ticularly rare in primary MFS. Th e only secondary abnor- detection and intervention is necessary. mality that has been found consistently thus far is References 39 asymmetry of the motion VEP response, which appears to 4. Fawcett SL, Birch EE (2000) Motion VEPs, stereopsis, and be associated with foveal suppression [4]. bifoveal fusion in children with strabismus. Invest One possible explanation for this lack of motor evi- Ophthalmol Vis Sci 41:411–416 dence of the long-term image decorrelation in MFS is 5. Struck MC, VerHoeve JN, France TD (1996) Binocular cor- that the motor signs such as pursuit asymmetry are pres- tical interactions in the monofi xation syndrome. J Pediatr ent, but subclinical. Another possibility is that the angle of Ophthalmol Strabismus 33:291–297 strabismus is so small in primary MFS that the cortex 6. Tychsen L (2007) Causing and curing infantile esotropia in does not recognize the decorrelation and the motor path- primates: the role of decorelated binocular input. Trans ways develop normally. A third possibility is that it is the Am Ophthalmol Soc 105:564–593 high quality of the binocular vision that is present in MFS 7. McCormack G (1990) Normal retinotopic mapping in that somehow prevents the development of these motor human strabismus with anomalous retinal correspondence. sequelae. Th is is yet another query to be added to Parks’ Invest Ophthalmol Vis Sci 31:559–568 long list of questions about Th e Monofi xation Syndrome. 8. Sireteanu R, Fronius M (1989) Diff erent patterns of retinal correspondence in the central and peripheral visual fi eld of strabismics. Invest Ophthalmol Vis Sci 30:2023–2033 Summary for the Clinician 9. Grant S, Berman NE (1991) Mechanism of anomalous reti- ■ Th e MFS has much to teach us about both nor- nal correspondence: maintenance of binocularity with mal and abnormal binocular vision. Even as alteration of receptive-fi eld position in the lateral suprasyl- some answers begin to reveal themselves, more vian (LS) visual area of strabismic cats. Vis Neurosci questions arise. 7:259–281 ■ Monofi xation may be preventable if the cause of 10. Sireteanu R, Best J (1992) Squint-induced modifi cation of the image decorrelation is detected and repaired visual receptive fi elds in the lateral syprasylvian cortex of promptly, probably within 60–90 days of onset. the cat: binocular interaction, vertical eff ect, and anoma- ■ Once monofi xation is present, attempting a cure is lous correspondence. Eur J Neurophysiol 4:235–242 unwise. MFS, particularly with small angle esotro- 11. Wong AMF, Lueder GT, Burkhalter A, Tychsen L (2000) pia, is relatively stable and allows for good bin ocular Anomalous retinal correspondence: neuro-anatomic vision so there is little to be gained. Antisuppression mechanism in strabismic monkeys and clinical fi ndings in and anti-ARC therapies designed to restore bi- strabismic children. J AAPOS 4:168–174 foveal fi xation typically result in intractable diplo- 12. Fronius M, Sireteanu R (1989) Monocular geometry is pia. Attempted repair of the associated strabismus selectively distorted in the central visual fi eld of strabismic with surgery or prism will not create bifi xation amblyopes. Invest Ophthalmol Vis Sci 30:2034–2044 once monofi xation is established. 13. Harwerth RS, Fredenburg PM (2003) Binocular vision ■ MFS can decompensate with time, even in the with primary microstrabismus. Invest Ophthalmol Vis Sci presence of good binocular vision. Patients with 44:4293–4306 this condition should be followed periodically, 14. Arnoldi K (2004) Th e VII Burian memorial lecture: factors and any changes in acuity or refractive error contributing to the outcome of sensory testing in patients addressed promptly to minimize the risk of dete- with anomalous binocular correspondence. In: Verlohr D, rioration with loss of binocular vision. Georgievski Z, Rydberg A (eds) Global perspectives converge downunder, the transactions of the Xth international orthoptic congress. International Orthoptic Association, Melbourne, Australia, pp 73–80 References 15. Arnoldi K (2001) Monofi xation with eso-, exo-, or hypertropia: is there a diff erence? Am Orthopt J 51:55–66 1. Parks MM (1969) Th e monofi xation syndrome. Tr Am 16. Leske DA, Holmes JM (2004) Maximum angle of horizon- Ophthalm Soc 67:609–657 tal strabismus consistent with true stereopsis. J AAPOS 2. Hubel DH, Wiesel TN (1977) Functional architecture of 8:28–34 macaque monkey visual cortex. Philos Trans Roy Soc Lond. 17. Choi DG, Isenberg SJ (2001) Vertical strabismus in 198:1–59 monofi xation syndrome. J AAPOS 5:5–8 3. Tychsen L (2005) Can ophthalmologists repair the brain in 18. Richards M, Wong A, Foeller P, Bradley D, Tychsen L (2008) infantile esotropia? Early surgery, stereopsis, monofi xation Duration of binocular decorrelation predicts the severity of syndrome, and the legacy of Marshall Parks. J AAPOS latent (fusion maldevelopment) nystagmus in strabismus 9:510–521 macaque monkeys. Invest Ophthalmol Vis Sci 49:1872–1878 40 4 The Monofi xation Syndrome: New Considerations on Pathophysiology

19. Neri P, Bridge H, Heeger DJ (2004) Stereoscopic processing treated for cataract. Invest Ophthalmol Vis Sci 34: of absolute and relative disparity in human visual cortex. J 3501–3509 Neurophysiol 92:1880–1991 34. Merigan WH, Katz LM (1990) Spatial resolution across the 20. Horwood A (2003) Neonatal ocular misalignments refl ect macaque retina. Vis Research 30:985–991 4 vergence development but rarely become esotropia. Br 35. Cisarik PM, Harwerth RS (2008) Th e eff ects of interocular J Ophthalol 87:1146–1150 correlation and contrast on stereoscopic depth magnitude 21. Hasany A, Wong A, Foeller P, Bradley D, Tychsen L (2008) estimation. Optom Vis Sci 85:164–173 Duration of binocular decorrelation in infancy predicts the 36. Arthur BW, Smith JT, Scott WE (1989) Long-term stability severity of nasotemporal pursuit asymmetries in strabis- of alignment in the monofi xation syndrome. J Pediatr mic macaque monkeys. Neuroscience 156:403–411 Ophthalmol Strabismus 26:224–231 22. Tychsen L, Scott C (2003) Maldevelopment of convergence 37. Hunt MG, Keech RV (2005) Characteristics and course of eye movements in macaque monkeys with small- and patients with deteriorated monofi xation syndrome. large-angle infantile esotropia. Invest Ophthalmol Vis Sci J AAPOS 9:533–536 44:3358–3368 38. Pratt-Johnson JA, Tillson G (2001) Management of strabis- 23. Harwerth RS, Smith EL, Crawford ML, von Noorden GK mus and amblyopia, 2nd edn. Th ieme, New York, Stuttgart, (1997) Stereopsis and disparity vergence in monkeys with pp 113 subnormal binocular vision. Vis Res 37:483–493 39. vonNoorden GK, Campos EC (2002) Binocular vision and 24. Borman DK, Kertesz AE (1985) Fusional responses of stra- ocular motility, 6th edn. Mosby, St. Louis, pp 544 bismics to foveal and extrafoveal stimulation. Invest 40. Houston CA, Cleary M, Dutton GN, McFadsean RM (1998) Ophthalmol Vis Sci 26:1731–1739 Clinical characteristics of microtropia – is microtropia a 25. Curcio CA, Allen KA (1990) Topography of ganglion cells fi xed phenomenon? Br J Ophthalmol 82:219–224 in human retina. J Comp Neurol 300:5–25 41. Keiner EC (1978) Spontaneous recovery in microstrabis- 26. Curcio CA, Sloan KR, Kalina RE, Hendrickson AE (1990) mus. Ophthalmologica 177:280–283 Human photoreceptor topography. J Comp Neurol 42. vonNoorden GK, Campos EC (2002) Binocular vision and 292:497–523 ocular motility, 6th edn. Mosby, St. Louis, pp 344–345 27. Perry VH, Silveira LC, Cowey A (1990) Pathways mediat- 43. Quéré MA, Lavenant G, Péchereau A (1993) Les diplopies ing resolution in the primate retina. Cib Found Symp incoercibles post-thérapeutiques. J Fr Orthopt 25:191 155:5–14 44. Das VE, Fu LN, Mustari MJ, Tusa RJ (2005) Incomitance in 28. Wässle H, Grünert U, Röhrenbeck J, Boycott BB (1990) monkeys with strabismus. Strabismus 13:33–41 Retinal ganglion cell density and cortical magnifi cation 45. Fu LN, Tusa RJ, Mustari MJ, Das VE (2007) Horizontal sac- factor in the primate. Vis Research 30:1897–1911 cade disconjugacy in strabismic monkeys. Invest 29. Tychsen L, Kim D, Burkhalter A (1994) Naso-temporal Ophthalmol Vis Sci 48:3107–3114 asymmetries in geniculo-striate pathway of normal adult 46. Tusa RJ, Mustari MJ, Das VE, Boothe RG (2002) Animal macaque. Invest Ophthalmol Vis Sci Suppl 35:1773 models for visual deprivation-induced strabismus and nys- 30. Tychsen L, Burkhalter A (1997) Nasotemporal asymmetries tagmus. Ann NY Acad Sci 956:346–360 in V1: ocular dominance columns of infants, adult, and 47. Wensveen JM, Harwerth RS, Smith EL (2003) Binocular strabismic macaque monkeys. J Comp Neurol 388:32–46 defi cits associated with early alternating monocular defocus. 31. Lewis TL, Maurer D (1992) Th e development of the tem- I. Behavioral observations. J Neurophysiol 90: 3001–3011 poral and nasal visual fi elds during infancy. Vis Research 48. Zhang B, Matsura K, Mori T, Wensveen JM, Harwerth RS, 32:903–911 Smith EL Chino Y (2003) Binocular defi cits associated 32. Beirne RO, Zlatkova MB, Anderson RS (2005) Changes in with early alternating monocular defocus, neurophysiolog- human short-wavelenth-sensitive and achromatic resolu- ical observations. J Neurophysiol 90:3012–3023 tion acuity with retinal eccentricity and meridian. Vis 49. Zhang B, Bi H, Sakai E, Maruko I, Zheng J, Smith EL, Chino Neurosci 22:79–86 YM (2005) Rapid plasticity of binocular connections in 33. Bowering ER, Maurer D, Lewis TL, Brent HP (1993) developing monkey visual cortex (V1). Pro Natl Acad Sci Sensitivity in the nasal and temporal hemifi elds in children USA 102:9026–9031 Chapter 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment 5 Lawrence Tychsen

Core Messages ■ Proper alignment of the eyes requires informa- reliably in normal primates by impeding the mat- tion sharing (fusion) between monocular visual uration of fusional/binocular connections in V1. input channels in the CNS; the fi rst locus for ■ Infantile esotropia occurs predominantly in fusion in the CNS of primates is the striate cere- human infants who have perinatal insults that bral cortex (area V1). would impair correlated visual input to V1. ■ Fusion behaviors and V1 binocular connections ■ Surgical realignment of the eyes during the criti- are immature at birth, maturing during a critical cal period of normal binocular maturation may period in the fi rst months of life; maturation of achieve functional sensory and motor cures. fusion and V1 binocular connections requires ■ If surgery fails to restore bifoveal fusion, subnor- correlated (synchronized) input from each eye. mal fusion (micro-esotropia/monofi xation) may ■ Nasalward biases are present innately in the neu- be achieved within boundaries set by the proper- ral pathways of normal primates before matura- ties of neurons in V1 and extrastriate cortex. tion of binocularity. ■ Late-onset (e.g., accommodative) esotropia is ■ Esotropia and the associated nasalward gaze easier to treat because the fusional connections in biases of infantile strabismus can be produced V1 matured substantially before the emergence of eye misalignment.

early-onset esotropia are predominantly emmetropic [1], 5.1 Esotropia as the Major Type of Developmental Strabismus whereas late-onset esotropia is associated commonly with a substantial hypermetropic refractive error (accommo- Esotropia is the leading form of developmental strabismus. dative esotropia). Th e most prevalent form of develop- Th erefore, unraveling the causal mechanism and response mental strabismus in humans is concomitant, constant, to treatment is an important public health issue. Th e pur- nonaccommodative, early-onset esotropia. Most of these pose of this chapter is to review knowledge gained over the cases have onset in the ﬁ rst 12 months of life, i.e., infan- last two decades that: (a) implicates cerebral cortex malde- tile-onset. Infantile esotropia may be considered the para- velopment as the cause, and (b) explains how repair of cor- digmatic form of strabismus in all primates, as it is also tical circuits may be the key to functional cures. the most frequent type of natural strabismus observed in monkeys [2].

5.1.1 Early-Onset (Infantile) Esotropia 5.1.2 Early Cerebral Damage as the Esotropia has a bimodal, age-of-onset distribution. Th e Major Risk Factor largest peak (comprising ~40% of all strabismus) occurs at or before age 12–18 months, with a second, smaller If infantile esotropia is a paradigmatic form of strabis- “late onset” esotropia peak at age 3–4 years. Children with mus, investigations designed to reveal pathophysiologic 42 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment

mechanisms should begin by asking what factors con- 5.1.4 Genetic Infl uences on Formation tribute to its causation. At highest risk are infants who of Cerebral Connections suff er cerebral maldevelopment from a variety of causes (Table 5.1), especially insults to the parieto-occipital cor- Genetic factors also play a causal role. Large-scale studies 5 tex and underlying white matter (geniculostriate projec- have documented that ~30% of children born to a strabis- tions or optic radiations) [3, 5–7]. Periventricular and mic parent will themselves develop strabismus [18]. Twin intraventricular hemorrhage in the neonatal period studies reveal a concordance rate for monozygous twins of increases the prevalence of infantile strabismus 50–100- 73% [19]. Less than 100% concordance implies that intra- fold. Less specifi c cerebral insults, e.g., from very low uterine or perinatal (“environmental”) factors alter the birth weight (with or without retinopathy of prematu- expression of the strabismic genotype. Maumenee and rity) or Down syndrome, increase the risk above that of associates analyzed the pedigrees of 173 families containing otherwise healthy infants by factors of 20–30-fold probands with infantile esotropia [20]. Th e results sug- [4, 7–10]. gested a multifactorial or Mendelian codominant inheritance pattern. Codominant means that both alleles of a single gene contribute to the phenotype but with diff erent thresholds for expression of each allele. Th ese genes could 5.1.3 Cytotoxic Insults conceivably encode cortical neurotrophins, or axon guid- to Cerebral Fibers ance and maturation. Any of these genetically modulated Th e occipital lobes in newborns are vulnerable to dam- factors could increase the susceptibility to disruption of age [6, 12–14]. Premature infants frequently suff er visual cortical connections in otherwise healthy infants. injury to the optic radiations near the occipital trigone. Balanced binocular input requires equally strong pro- 5.1.5 Development of Binocular Visuomotor jections from each eye through this periventricular Behavior in Normal Infants zone. Th e fi bers connect the lateral geniculate laminae to the ocular dominance columns (ODCs) of the striate Esotropia is rarely present at birth. For this reason alone, cortex. Th e projections are immature at birth and the “infantile esotropia” is a more appropriate descriptor than quality of signal fl ow would be critically dependent “congenital esotropia.” Constant misalignment of the visual upon the function of oligodendrocytes, which insulate axes appears typically aft er a latency of several months, the visual fi bers. Neonatal oligodendrocytes are espe- becoming conspicuous on average between the ages of 2 cially vulnerable to cytotoxic insult [15]. Th e striate and 5 months [11, 21, 22]. To understand visuomotor cortex is also susceptible to hypoxic injury because it maldevelopment in strabismic infants during this period, it has the highest neuron-to-glia ratio in the entire cere- is helpful to understand the development of binocular brum [16] and the highest regional cerebral glucose fusion and vergence in normal infants (Table 5.2) during consumption [17]. the same 2–5-month postnatal interval.

Table 5.1. Cerebral damage risk factors for infantile-onset strabismus

Type Prevalence strabismus (%) Author(s) Intraventricular hemorrhage with hydrocephalus 100 [3] Cerebral visual pathway white matter injury 76 [4] Occipitoparietal hemorrhage or leukomalacia 54–57 [5, 6] Very low birth weight infants (<1,500 g) 33a [7] Very low birth weight (<1,251 g) and prethreshold 30 [8] retinopathy of prematurity Very low birth weight (<1,251 g) and normal 17 [4] neuroimaging Down syndrome 21–41 [9, 10] Healthy full-term infants 0.5–1.0 [11] aAdditional 17% of infants had persistent asymmetric OKN 5.1 Esotropia as the Major Type of Developmental Strabismus 43

Table 5.2. Binocular development and visuomotor behaviors in infant primate

Immature behavior Chief ﬁ ndings before onset Investigator(s) of mature behavior Binocular disparity Stereo-blindness [23] sensitivity absent Convergent disparity sensitivity [24, 25] before ~3–5 mos emerges earlier than divergent [26] Binocular sensorial Equal attraction to rivalrous vs. [27, 25] fusion absent before fusible stimuli [28] ~3–5 mos Fusional (binocular) Binocular alignment errors common [29, 30] vergence unstable despite accommodative capacity [27] before ~3–5 mos [31] [32, 33] Nasalward bias of vergence Transient convergence errors 4X [34] pronounced divergence errors before ~3–5 mos Convergent disparity sensitivity present earlier than divergent Convergence fusion range exceeds [32, 33] divergence by 2:1 Nasalward bias of cortically mediated Motion VEP nasotemporal asymmetry [35, 36] motion sensitivity before ~6 mos Stronger preferential sensitivity [37] to nasalward motion [38] [39] Nasalward bias of pursuit/OKN Nasalward motion evokes stronger [40] before ~6 mos OKN/pursuit [41] Nasotemporal asymmetry resolves [42] aft er onset binocularity [43] [44] [45] Nasalward bias of gaze-holding Nasalward slow phase drift [42] before ~6 mos of eye position [46] Persists as latent ﬁ xation [47] nystagmus with binocular maldevelopment

life, achieving adult-like levels of sensitivity. Sensitivity to 5.1.6 Development of Sensorial Fusion and Stereopsis crossed (near) disparity appears on average several weeks before that to uncrossed (far) disparity [24]. During this Binocular disparity sensitivity and binocular fusion are same interval infants begin to display an aversion to stimuli absent in infants less than several months of age, as demon- that cause binocular rivalry (i.e., nonfusable stimuli). strated by several methods, most notably studies that have Visually evoked potentials in normal infants, recorded using used forced preferential looking (FPL) techniques [23–25, dichoptic viewing and dichoptic stimuli, show comparable 27, 28]. Th e FPL studies show that stereopsis emerges results [43, 48, 49]. Onset of binocular signal summation abruptly in humans during the ﬁ rst 3–5 months of postnatal occurs aft er, but not before, ~3 months of age. 44 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment

5.1.7 Development of Fusional 5.1.9 Development and Maldevelopment Vergence and an Innate of Cortical Binocular Connections Convergence Bias Knowledge of visual cortex development (Table 5.3) is 5 Fusional vergence eye movements mature during an important for understanding the neural mechanisms that equivalent period in early infancy. In the fi rst 2 months of could cause strabismus, for several reasons. First, the life, alignment is unstable and the responses to step or visual cortex is the initial locus in the CNS at which visual ramp changes in disparity are oft en markedly inaccurate signals from the two eyes are combined and a combina- [32, 33]. Th e inaccuracy cannot be ascribed to errors of tion of visual signals is necessary to generate the vergence accommodation. Accommodative precision during this error commands that guide eye alignment. Second, the period consistently exceeds that of fusional (disparity) most common form of strabismus (esotropia) appears vergence [29, 30, 33]. coincident with maturation of cortically mediated, bin- Studies of fusional vergence development in normal ocular, sensorimotor behaviors in normal infants. Th ird, infants reveal an innate bias for convergence [32, 33]. perinatal insults to the immature visual cortex are linked Transient convergence errors of large degree exceed strongly to subsequent onset of strabismus. And fi nally, divergence errors by a ratio of 4:1. Th e fusional vergence the constellation of sensory and motor defi cits in infantile response to crossed (convergent) disparity is also intact strabismus can be explained by known cortical pathway earlier and substantially more robust than that to diver- mechanisms. gent disparity. Th e innate bias favoring fusional convergence in primates persists aft er full maturation of normal binocular disparity sensitivity. Fusional convergence capacity exceeds the range of divergence capacity by a 5.1.10 Binocular Connections Join Monocular mean ratio of 2:1 [50, 51]. Compartments Within Area V1 (Striate Cortex) Aff erents from each eye are segregated in monocular lamina of the lateral geniculate nucleus (LGN) and at the input layer (4C) of ODCs of the striate cortex, or visual 5.1.8 Development of Motion Sensitivity area V1 (Fig. 5.1) [52, 53]. Th e fi rst stage of binocular and Conjugate Eye Tracking processing in the primate CNS is made possible by hori- (Pursuit/OKN) zontal connections between ODCs of opposite ocularity, Th e innate nasalward bias of the vergence pathway has above and below layer 4C [52, 68, 70]. Physiological analogs in the visual processing of horizontal motion, recordings in normal neonatal and adult monkeys show both for perception and conjugate eye tracking. In the monocular responses in layer 4C and binocular responses fi rst months of life, VEPs elicited by oscillating grating from the majority of neurons in V1 layers 4B and 2–6 stimuli (motion VEPs) show a pronounced nasotempo- [52, 54, 63]. Th e binocular responses in the neonate are ral asymmetry under conditions of monocular viewing cruder and weaker than those recorded in normal adult [35–38]. Th e direction of the asymmetry is inverted [58, 59, 77]. Binocular disparity sensitive neurons are when viewing with the right vs. left eye. Monocular FPL present in the neonatal cortex, but the spatial tuning is testing reveals greater sensitivity to nasalward motion poor and they are characterized by a high binocular sup- [39]. Monocular pursuit and optokinetic tracking show pression (inhibition) index. Th e immature neuronal strong biases favoring nasalward target motion when response properties are attributed to unrefi ned, weak viewing with either eye [40, 41, 43–45]. Optokinetic excitatory horizontal binocular connections between aft er-nystagmus (slow phase eye movement in the dark ODCs. Th ese axonal connections help defi ne the segrega- aft er extinction of stimulus motion) is characterized by a tion of ODCs [62, 77]. ODC borders are immature (fuzzy) consistent nasalward drift of eye position [42]. Th ese at birth but adult-like (sharply defi ned) by 3–6 weeks nasalward motion biases are most pronounced before postnatally [60, 78] (the equivalent of 3–6 months in the onset of sensorial fusion and stereopsis, but system- humans, 1 week of monkey visual development is compa- atically diminish thereaft er. rable with 1 month in humans [79]). 5.1 Esotropia as the Major Type of Developmental Strabismus 45

Table 5.3. Development of neural pathways in normal and strabismic primate

Neurobiological principle Physiology/anatomy Investigator(s)

Striate cortex (area V1) is the ﬁ rst Right and left eye inputs remain [52, 53] CNS locus for binocular processing segregated in LGN and input layer (4C) in V1 Binocular responses recorded from [54] neurons in V1 lamina beyond layer 4C Neurons in V1 layers 2–6 are sensitive [55] to binocular disparity Binocular structure + function in Segregation of RE/LE ODCs immature at birth [56] V1is immature at birth Binocular (disparity sensitive) neurons [57] present at birth but tuning poor Immature binocular neurons have weak [58, 59] excitatory horizontal connections [60, 61] between ODCs and high suppression index [62] Maturation of binocular connectivity Absence of correlation causes lack of disparity [63, 64, 65] in V1 requires correlated RE/LE input sensitivity and loss of horizontal [66] connections in V1 [67, 68, 69, 70] V1 feeds forward to extrastriate visual areas Extrastriate areas MT/MST mediate [71, 72] MT/MST which control ipsiversive eye pursuit/OKN and recieve feedforward [73, 74] tracking and gaze holding (binocular)projections from V1 lamina [75] 4B Lesions of MST impair ipsiversive pursuit/OKN and gaze holding V1 feed forward connections to MT/MST Before maturation of binocularity, a nasalward [76] at birth are monocular from ODCs movement bias is apparent when viewing with either driven by the contralateral eye eye (RE viewing evokes left ward pursuit/OKN/gaze drift ; LE viewing evokes rightward pursuit/OKN/gaze drift ) Nasalward + temporalward neurons are [77] present in = numbers within V1/MT but [13] nasalward have innate connectivity advantage MST inputs from the ipsilateral eye require If binocularity matures, monocular viewing [76] maturation of binocular V1/MT evokes equal nasalward/temporalward eye movement + [13, 47] connections stable gaze MST neurons encode both vergence Disparity sensitive neurons in MST also [81] and pursuit/OKN mediate vergence [80] If binocularity fails to mature, monocular viewing evokes [105] nasalward pursuit/OKN and inappropriate convergence [82, 47] Convergence motoneurons are Convergence neurons outnumber divergence neurons 3:2 in [122, 123] more numerous the midbrain of normal primates 46 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment

a 2/3

Fusion/stereopsis 4B Alignment and 4C 5 Balanced Gaze

Ocular Dominance Columns RRL L of V1 (Striate Cortex) Correlated LGN 2/3 Activity 4B 4C b

Stereo-blindness

RRLL Esotropia and Gaze Asymmetries Periventricular White Matter RRL L Projections

De-Correlated Fig. 5.1 Neuroanatomic basis for binocular vision. Monocular Activity retinogeniculate projections from left eye (temporal retina-nasal visual hemiﬁ led) and right eye (nasal retina-temporal hemiﬁ eld) remain segregated up to and within the input layer of ocular Fig. 5.2 Horizontal connections for binocular vision in V1 of dominance columns (ODCs) in V1, layer 4C (striate visual cor- normal (correlated activity) vs. strabismic (decorrelated) pri- tex). Binocular vision is made possible by horizontal connec- mate, layer 2–4B. (a) V1 of normal primates is characterized by tions between ODCs of opposite ocularity in upper layers 4B equal numbers of monocular and binocular connections. (b) In and 2/3 (as well as lower layers 5/6, not shown). RE inputs red; strabismic primates, the connections are predominantly mon- LE inputs blue ocular (i.e., a paucity of binocular connections). RE inputs red; LE blue; binocular violet

5.1.12 Projections from Striate Cortex (Area V1) 5.1.11 Too Few Cortical Binocular to Extrastriate Cortex (Areas MT/MST) Connections in Strabismic Primate Projections from V1 layer 4B feed forward to regions of extrastriate visual cortex, in particular the middle tempo- Maturation of binocular connections in V1 requires ral and middle superior temporal area (MT/MST) [75]. correlated (synchronous) activity between right and MT and MST mediate pursuit/OKN and a closely related left eye inputs (Fig. 5.2a) [66]. Decorrelation of inputs, type of tracking movement, ocular following [73, 74]. by natural strabismus [68, 70], or as a consequence of MT/MST neurons are directionally selective and sensi- experimental manipulations that produce retinal image tive to binocular disparity, guiding both conjugate and noncorrespondence [66, 67], causes loss of binocular disconjugate (near-far) tracking [80–82]. In normal pri- horizontal connections (Fig. 5.2b). Monocular connec- mates, greater than 90% of MT/MST neurons exhibit bal- tions between ODCs of the same ocularity are anced, binocular responses. In strabismic primates, the maintained. Th e loss is due to excessive pruning of responses are predominantly monocular, indicating that connections, beyond the normal process of axon retrac- the loss of binocularity found in V1 is passed on in the tion and refi nement that takes place within and between projections to MT/MST. ODCs in the fi rst weeks of life. (Captured in the neuroscience dictum: “Cells that fi re together, wire together. Cells that fi re apart, depart.”) Th e paucity of binocular 5.1.13 Inter-Ocular Suppression Rather than connections is accompanied by loss of binocular Cooperation in Strabismic Cortex responsiveness and disparity sensitivity, measured electrophysiologically, in V1 neurons [55, 63, 64]. Th e When the eyes are misaligned, suppression is necessary companion behavioral defi cits are stereoblindness and to avoid diplopia or visual confusion. Suppression is a absence of fusional vergence [47, 65]. major sensorial abnormality in humans and monkeys 5.1 Esotropia as the Major Type of Developmental Strabismus 47 with infantile strabismus. Visual inputs may be suppressed strabismus and amblyopia, as compared with strabismus from one eye continuously (causing unilateral amblyo- alone (that is, alternating fi xation). Th e metabolic abnor- pia), or commonly in infantile strabismus, from each eye malities are found throughout V1 when suppression is alternately ~50% of the time (alternate fi xation) [83, 84]. widespread; alternatively, suppression is confi ned to In normal animals, horizontal connections between zones of V1 that match retinotopically the location of a ODCs can mediate suppression when confl icting stimuli suppression scotoma. Th e metabolic suppression is not activate neurons in neighboring ODCs [85, 86]. found in the LGN, which is composed of neurons driven Th e mitochondrial enzyme cytochrome oxidase (CO) monocularly from each eye without binocular interac- is used to reveal neuronal activity within ODCs [87–89]. tion. Th ese fi ndings imply that abnormal binocular inter- In normal primates, the input layer of area V1, layer 4C, action in V1 leads to heightened competition between shows a uniform pattern of CO activity in right eye and left ODCs of opposite ocularity, with suppression of meta- eye columns (Fig. 5.3a), refl ecting equal activity (absence bolic activity in opposite-eye ODCs. Th e abnormalitis of inter-ocular suppression). Unequal CO activity is a gen- add to our knowledge of the brain damage caused by eral fi nding in area V1 of primates who have strabismus unrepaired strabismus. As noted in the preceding sec- [78, 90], amblyopia [91], or both [92]. Th e unequal activity tions, the eff ects include an ~50% reduction in long- is seen as reduced CO activity (metabolic suppression) in range, excitatory binocular horizontal connections the ODCs driven by one eye in each cerebral hemisphere joining ODCs of opposite ocularity [70, 93]. In the pres- (Fig. 5.3b). When strabismus is combined with amblyopia, ence of strabismus, the remaining 50% of binocular con- metabolic suppression is more pronounced. nections (long-range, short-range or a combination) may Th e CO abnormality in monkey cortex correlates with be predominantly inhibitory. clinical observations in strabismic humans. Binocularity is impaired to a greater degree, and suppression tends to be more pronounced, in patients who have combined 5.1.14 Naso-Temporal Inequalities of Cortical Suppression a 2/3 Psychophysical studies of the development of the visual hemifi elds in normal human infants indicate that tempo- 4B Equal Neuronal ral retina sensitivity matures slower than nasal retina sen- Metabolic Activity 4C sitivity [94, 95]. Th e nasotemporal asymmetry in sensitivity diminishes if the infant develops normal vision, but lower RRL L temporal sensitivity remains permanently if early binocular development is disrupted by strabismus or amblyopia Normal [96–98] (for review, see [78]). In strabismic animals, metabolic suppression tends to b be most apparent in ODCs driven by the ipsilateral eye in V1 of both the right and left hemispheres. Ipsilateral inputs Inter-ocular originate from the temporal hemi-retinae of each eye, Metabolic implying that inputs to V1 from the temporal hemiretinae Suppression are at a developmental disadvantage [78, 92, 99]. Th e human psychophysical fi ndings, together with the monkey RRL L anatomic fi ndings, reinforce the conclusion that abnormal Strabismic binocular experience in early infancy unfairly punishes visual neurons that are slow to develop and fewer in num- Fig. 5.3 Metabolic activity in neighboring ODCs within V1 of ber, that is, those driven by the temporal hemiretina [78]. normal vs. strabismic primate. (a) In normal, Layer 4C stains uniformly for the metabolic enzyme cytochrome oxidase (CO) (shown as brown), indicating equal activity in right-eye vs. left -eye columns. (b) In strabismic, a narrow monocular zone within the 5.1.15 Persistent Nasalward Visuomotor dominant ODCs (shown here as left -eye) shows normal meta- Biases in Strabismic Primate bolic activity (brown), but ODCs belonging to the suppressed eye (shown as right-eye) and binocular border zones between ODCs If normal maturation of binocularity is impeded by eye are pale, connoting abnormally low – i.e., suppressed – activity misalignment, the innate nasalward biases of eye tracking 48 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment

do not resolve – they persist and become pronounced [46, phase) gaze drift . In newborns, the outputs from V1 to 100–102]. Normally, area MST in each cerebral hemi- each area MST appear to favor innately the contralateral sphere encodes ipsiversive eye tracking and gaze holding eye (i.e., inputs from the right eye make stronger connec- (Fig. 5.4). Ablations within MST impair ipsiversive pur- tion – through area V1 of both hemispheres – to area 5 suit/OKN, and excitation of MST evokes ipsiversive (slow MST of the left hemisphere) [13, 76]. Th e contralateral-

Strabismic Normal

chi

RE LE RE LE RE LE RE LE

call

nasalward gaze stable gaze instability

Fig. 5.4 Neural network diagrams showing visual signal fl ow for pursuit and gaze holding in strabismic vs. normal primates. Paucity of mature binocular connections explains behavioral asymmetries evident as asymmetric pursuit/OKN and latent fi xation nystagmus. Note that in all primates, pursuit area neurons in each hemisphere encode ipsilaterally directed pursuit. Signal fl ow is initiated by a moving stimulus in the monocular visual fi eld, which evokes a response in visual area neurons (i.e., V1/MT). Each eye at birth has access – through innate, monocular connections – to the pursuit area neurons (e.g., MSTd) of the contralateral hemisphere. Access to pursuit neurons of the ipsilateral hemisphere requires mature, binocular connections. Strabismic/nasalward gaze instability: moving from top to bottom, starting with target motion in monocular visual fi eld of right eye. Retinal ganglion cell fi bers from the nasal and temporal hemiretinae (eye) decussate at the optic chiasm (chi), synapse at the LGN, and project to alternating rows of ODCs in V1 (visual area rectangles). In each V1, ODCs representing the nasal hemiretinae (temporal visual hemi-fi eld) occupy slightly more cortical territory than those representing the temporal hemiretinae (nasal hemifi eld), but each ODC contains neurons sensitive to nasally directed vs. temporally directed motion (half circles shaped like the matching hemifi eld, arrows indicate directional preference). Visual area neurons (including those beyond V1 in area MT) are sensitive to both nasally directed and temporally directed motion, but only those encoding nasally directed motion are wired innately – through monocular connections – to the pursuit area. Normal/stable gaze: binocular connections are present, linking neurons with similar orientation/directional preferences within ODCs of opposite ocularity (diagonal lines between columns). Viewing with the right eye, visual neurons preferring nasally directed motion project to the left hemisphere pursuit area; visual neurons preferring temporally directed motion project to the right hemisphere pursuit area. Temporally directed visual area neurons gain access to pursuit area neurons only through binocular connections. Call corpus callosum, through which visual area neurons in each hemisphere project to opposite pursuit area. Bold lines active neurons and neuronal projections 5.1 Esotropia as the Major Type of Developmental Strabismus 49 eye-to-MST connectivity advantage is consistent with an eye ODCs gain equal access to neurons within areas MST innate, contralateral-eye-to-V1 connectivity advantage. of the right and left hemisphere, and the nasalward bias (Captured in twin dictums: “fi rst come, fi rst served ”and disappears. (Captured in the dictum: “Tracking from ear “majority rules.”) V1 neurons in each hemisphere, driven to nose will balance as binocularity grows.”) If binocular by the nasal hemiretinae (contralateral eye), develop ear- connections are lost, the nasalward bias persists and is lier and outnumber (by a ratio of ~53:47 in primate) neu- exaggerated. Th e bias is evident clinically (Fig. 5.5) as a rons from the temporal hemiretinae (ipsilateral eye). Area pathologic naso-temporal asymmetry of pursuit/OKN MST on the side ipsilateral to the viewing eye can only be and a nasalward (slow phase) drift of gaze-holding (latent accessed through binocular V1/MT connections. nystagmus) [103, 104]. Th e contralateral eye-to-MST connectivity bias pro- Area MST neurons are sensitive to binocular disparity vides a mechanism for the nasalward tracking bias, evi- and also drive fusional vergence eye movements [80, 82]. dent before onset of binocularity (Fig. 5.4). Right eye Eye movement recordings in a primate with infantile viewing activates right eye ODCs in each area V1. Right esotropia showed inappropriate activation of conver- eye ODCs connect preferentially to the left area MST. Th e gence whenever nasalward monocular OKN was evoked left area MST mediates ipsiversive/left ward tracking, [105]. Neuroanatomic analysis of V1 in this monkey which is nasalward tracking with respect to the viewing showed a paucity of binocular connections and metabolic (right) eye. When binocular connections mature, right evidence of heightened interocular suppression. Th e

Fusional Vergence (esotropia) Fig. 5.5 Nasalward vergence and gaze asymmetries in strabismic humans and monkeys. Fusional vergence: esodeviation of the nonfi xating eye, evident as alternating esotropia. Tracking pursuit/OKN: horizontal smooth pursuit is asymmetric during monocular viewing. Pursuit is smooth (normal) when target motion is nasalward in Tracking (pursuit/OKN) the visual fi eld. Pursuit is cogwheel (low gain-abnormal) when the target moves temporalward. Th e movements of the two eyes are conjugate, and the direction of the asymmetry reverses instantaneously with a change of fi xating eye, so that the direction of robust pursuit is always for nasalward motion in the visual fi eld. Gaze holding- latent nystagmus: viewing Gaze Holding (latent nystagmus) with the right-eye, both eyes have a nasalward slow-phase drift , followed by temporalward refoveating fast-phase microsaccades. Th e direction of the nystagmus reverses instantaneously when the left eye is fi xating, so that the slow phase is nasalward with respect to the fi xating eye 50 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment

conclusion drawn from these observations was that MST showed that if stable, binocular alignment was not neurons promote esotropia (i.e., a bias for nasalward ver- achieved until age 24 months, the chances of repairing gence) when binocularity fails to develop in V1. Th e stereopsis were nil. If stable alignment was achieved by mechanism is attractive, because it ties together the age 6 months, the chances of repairing stereopsis were 5 nasalward biases of vergence, pursuit/OKN and gaze good, and a substantial percentage of the infants regained holding (latent nystagmus) in cortical regions vulnerable robust stereopsis, i.e., random dot stereopsis with thresh- to perinatal damage. olds on the order of 60–400 arcsec. Outputs from the cortical areas noted earlier (V1, MT/ Scrutiny of early alignment data in infantile esotropia MST) and related cortical areas descend to brainstem has produced more refi ned and forceful conclusions. visual relay and premotor neuron pools immediately Figure 5.6a is replotted data on stereopsis outcomes in adjacent to the motor nuclei (Fig. 5.5) [106]. Even in the over 100 consecutive infantile esotropes [112]. Th e Y-axis absence of cortical maldevelopments, the vergence sys- is prevalence of stereopsis aft er surgical alignment, and tem is unbalanced, favoring convergence. Midbrain pre- the X-axis is age of onset or duration of misalignment motor neurons driving convergence outnumber those before surgery. Th e dashed line at 40% represents the driving divergence, by a ratio of 3:2. average prevalence of stereopsis when all infants operated upon by 2 years of age are grouped together, without regard to age at correction or duration before correction. Th e noise in the data – relating age at alignment to stere- 5.1.16 Repair of Strabismic Human Infants: opsis outcome – is related to the fact that onset of strabis- The Historical Controversy mus is idiosyncratic, varying considerably from infant to Is repair of binocular V1 connections possible, restoring infant, and distributed randomly in the interval 2–6 normal fusion and stereopsis, while preventing or revers- months of age. Th ere is no systematic relationship between ing the constellation of ocular motor maldevelopments? age of onset of esotropia and subsequent attainment of Th e answer to this question is rooted in a debate between stereopsis. However, when the data is reanalyzed with two competing twentieth century schools of treatment strict attention to duration of misalignment, a strong cor- philosophy, derived from the eminent British strabismol- relation is evident between shorter durations of misalign- ogists, Claude Worth and Bernard Chavasse. Worth pos- ment and restoration of stereopsis (Fig. 5.6b). Excellent tulated in 1903 that esotropic infants suff ered “an outcomes are achievable in infants operated upon within irreparable defect of the fusion faculty” [107]. Th eir brain 60 days of onset of strabismus (“early surgery”) [112]. Th e was congenitally incapable of achieving substantial bin- clinical dictum that follows is that age at surgery should ocular vision. Early surgical treatment was therefore be tailored to age of onset and not chronological age. unfounded because it was futile. Chavasse on the other Esotropic infants who regain high grade stereopsis hand – attracted by the Pavlovian physiology of the 1920 also regain robust fusional vergence [112–114]. Clinical and 1930s – believed that the brain machinery for fusion observation also suggests that they have a lower preva- was present in esotropic infants, but the development of lence of recurrent esotropia (or exotropia), pursuit/OKN “conditioned refl exes” for binocular fusion were impeded asymmetry, motion VEP asymmetry, latent nystagmus, by factors such as weakness of the motor limb [108]. He and dissociated vertical deviation (DVD). However, ocu- postulated (in his text published in 1939) that if the eyes lar motor recording is diffi cult to perform in children and could be realigned during what he believed to be a period detailed, quantitative information is lacking. of refl ex learning, binocular fusion could be restored.

5.1.18 Timely Restoraion of Correlated 5.1.17 Repair of High-grade Fusion is Possible Binocular Input: The Key to Repair New knowledge of stereopsis development in the 1980s Eye movement studies of strabismic infant monkeys have bolstered the rationale in favor of early surgery, as articu- helped fi ll gaps in clinical knowledge. Th e studies have lated by disciples of Chavasse in the U.S., most notably shown that normal motor and sensory pathway develop- August Costenbader, Marshall Parks, and a series of ment can be restored when the timeliness of therapy con- Parks’ trainees [109, 110]. Th e new knowledge prompted forms to that of early surgery in humans [47, 115]. If a gradual reexamination of old data and inspired impor- binocular image correlation is restored in strabismic tant case studies – in the 1980 and 1990s – on the effi cacy monkeys within 3 weeks of onset of strabismus (the of early strabismus surgery [111–114]. Th ese reports equivalent of 3 months in humans), fusional vergence, 5.2 Visual Cortex Mechanisms in Micro-Esotropia (Monofi xation Syndrome) 51

a 100 pursuit/OKN and gaze holding return to normal (Fig. 5.6c). Th e repair of ocular motor behavior occurs 80 with repair of stereopsis and restoration of normal motion responses (motion VEPs). If decorrelation per- 60 sists in strabismic monkeys until the equivalent of 12 months’ duration in humans, esotropia and stereoblind- 40 ness persist. Prolonged-decorrelation animals exhibit 20 latent nystagmus, pursuit/OKN asymmetry, motion VEP

% Children with Stereopsis with Children % asymmetry, and DVD. Th e quality of behavioral repair 0 correlates with the quality of neuroanatomic repair in V1 123456 (Fig. 5.6c). “Early repair” monkeys (i.e., those who have Age on Onset (months) shorter durations of decorrelation) have a normal complement of binocular horizontal excitatory connections b 100 between ODCs of opposite ocularity, and “delayed repair” (longer durations of decorrelation) monkeys a 80 paucity. Th e restoration of binocular connections in V1 60 of “early repair” monkeys appears to have equally benefi - cal eff ects on downstream areas of extrastriate cortex 40 (MT/MST) driving the ocular motor neurons of the brainstem. Th e benefi t is evident as symmetric naso-

20 temporal eye tracking, stable gaze holding, and more % Children with Stereopsis with Children % 0 normal fusional vergence. 0-2 3-5 6-8 9-11 12-18 19-24 Duration of Misalignment (months)

5.2 Visual Cortex Mechanisms in Micro- c 40 Esotropia (Monoﬁ xation Syndrome) Pur Asym As outlined earlier, recent data on early correction of 30 Nyst infantile strabismus suggests that it is a curable disorder. But early surgery is the exception rather than the rule of 20 Stereo Eso current clinical practice in the U.S. and Europe. Th e DVD 10 majority of infants who have esotropia are corrected 6 or

(SD multiples) (SD V1 binoc more months aft er onset of misalignment. Th e chances of Magnitude of Deficit of Magnitude 0 rescuing bifoveal fusion aft er this interval are slim. Most infants are aligned to within 8 PD of orthotropia (microe- 0 36912 24 sotropia) and regain a degree of subnormal stereopsis and Duration of Decorrelation (weeks) motor fusion, i.e., monofi xation syndrome. Monofi xation syndrome occurs as a primary disor- Fig. 5.6 Repair of random-dot stereopsis aft er surgical realignment of the eyes in children with infantile esotropia, der (prevalence 1%) or, more commonly, as a secondary and analogous fi ndings in strabismic monkeys. (a) Prevalence phenomenon, aft er delayed treatment of large magni- of stereopsis as a function of age-of-onset of strabismus. No tude strabismus [116, 117]. Th e syndrome also occurs systematic relationship is evident. (b) High prevalence (~80%) in monkeys [118]. Th e major sensory and motor fea- of stereopsis in infants who were aligned within 2 months of tures of monofi xation syndrome are listed in Table 5.4. onset of strabismus. Probability of stereopsis was negligible in infants who had durations of strabismus exceeding ~12 Neural mechanisms for the fi rst two features listed in months. Redrawn from data of Birch et al. [112]. (c) Magnitude Table 5.4 are not diffi cult to explain. Receptive fi elds in of behavioral defi cits increases systematically as a function of V1 – representing the fovea – are tiny and have narrow decorrelation-duration in monkeys. One week of monkey tolerances. Any defocusing or other decorrelation of visual development is equivalent of 1 month in humans. Pur one eye’s inputs would produce a confl ict in neighboring Asymm horizontal pursuit asymmetry; Nyst velocity of latent nystagmus; Stereo random dot stereopsis defi cit; Eso angle of V1 columns and promote suppression of ODCs corre- esotropia; DVD magnitude of dissociated vertical deviation; sponding to the weaker eye. Th e fovea subtends ~5° of V1 binoc reduction in binocular connections between RE and the retinotopic map of V1, thus a suppression scotoma LE ODCs in V1 (striate cortex) of ≤5° makes sense. Feature two, subnormal stereopsis, 52 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment

Table 5.4. Monoﬁ xation (Microstrabismus) Syndrome

Clinical Feature Possible Neural Mechanism

1. Foveal suppression scotoma of 3-5 deg in the Inhibitory-connection-mediated metabolic suppression of 5 non-preferred eyea when viewing binocularly decorrelated activity in V1 foveal ODCs of non-preferred eye 2. Subnormal stereopsis (threshold 60-3000 arc sec) Broader disparity tuning of parafoveal neurons in V1/MT (foveal neurons suppressed) 3. Stable microesotropiab less than ~ 4-8 PD (~2.5-5 deg) Small angle ≈ average horizontal neuron length in V1, eso by default to convergent disparity coding of major MST population 4. Fusional vergence amplitudes intact for disparities V1 excitatory horizontal binocular connections (and V1/MT/ >2.5-5 deg (>4-8 PD) MST disparity neurons) intact beyond region of foveal suppression asubnormal acuity (amblyopia) in the non-preferred eye in 34% of corrected infantile esotropes and 100% of anisometropes. bmicroexotropia in ≤10%

could be explained along similar lines. Stereoscopic In a primate with microesotropia and a right eye fi xa- thresholds increase exponentially from the fovea to more tion preference (Fig. 5.7), a neuron within a foveolar (0°) eccentric positions along the retinotopic map of the visual column of the fi xating, right eye must link up with a nonfi eld. If foveal ODCs are suppressed and parafoveal ODCs adjacent column representing the pseudo-foveola of the are left to mediate stereopsis; stereopsis is degraded but deviated, left eye. Based on retinotopic maps of V1 in not obliterated. But it is features three and four of the macaque monkey, a horizontal axon ~7 mm in length monofi xation syndrome, the visuomotor signs, that are could join ODCs (and receptive fi elds) that were up to but most intriguing. If binocular development is perturbed so not further than 2.5° apart, or converting deg to PD, not that right and left eye foveal ODCs (receptive fi elds) do more than 4.4 PD. Shown here is a 2-dimensional map not enjoy perfectly correlated activity, why should the fall representing V1 from the right cerebral hemisphere (left back position of visual cortex be set so predictably ~2–4° visual hemi-fi eld) of a microesotropic macaque. Th e sulci (~4–8 prism diopters or PD) of micro-esotropia (Fig. 5.7)? and gyri have been unfolded and the visual fi eld represen- And if the heterotropia exceeds that range, why is fusional tation superimposed using standard retinotopic land- vergence typically absent? marks. One horizontal axon, originating within the foveal representation at 0–1° eccentricity, could link to a receptive fi eld shift ed 2.5° or 4.4 PD distant (Fig. 5.7). Two neu- 5.2.1 Neuroanatomic Findings in Area V1 rons strung together could join receptive fi elds 5° or 8.7 of Micro-Esotropic Primates PD apart. Th e conclusion that emerges is that the 4–8 PD “rule” of the monofi xation syndrome is explicable as a Studies of ODCs and neuronal axons in area V1 have combination of innate V1 neuron size and V1 topography. revealed a possible mechanism. Th e overall pattern and Th e visuomotor system of the strabismic primate appears width of ODCs in V1 (~400 mm [0.40 mm]) is the same in to achieve subnormal, but stable binocular fusion so long normal and strabismic monkeys [70, 78]. Horizontal axon as the angle of deviation is confi ned to a distance corre- length was measured for neurons within the V1 region sponding to not more than one to two V1 neurons [119]. corresponding to visual fi eld eccentricities of 0–10° (i.e., the representation of the fovea, parafovea and macula). Th e length is similar in both normal and strabismic mon- 5.2.2 Extrastriate Cortex in Micro-Esotropa keys, on average ~7 mm [70, 119]. In a primate with normal eye alignment, the ODC representing the foveola (or Neuronal response properties of the vergence-related 0° eccentricity) of the left eye is immediately adjacent to region of extrastriate visual cortex, MST, may also the column representing the foveola of the right eye. Th e explain the 2.5°-microesotropia rule in monofi xation side-by-side arrangement of the “foveolar” columns in syndrome. MST receives downstream projections from normal V1 is well within the range of horizontal axonal disparity-sensitive cells, both in V1 and in MT. Th e connections needed to allow those ODCs to communi- majority of binocular neurons in V1, MT and MST cate for high-grade binocular fusion. encode absolute disparity [82, 120]. Absolute disparity 5.2 Visual Cortex Mechanisms in Micro-Esotropia (Monofi xation Syndrome) 53

Fig. 5.7 (a) Monofi xator/ a microesotrope exhibits a deviation of the visiual axes on cover testing of approxi- 2.5°° (4.4 PD) mately 4 PD (~2.5°), which in Left Esotropia this case is shown as a left eye microesotropia (dark arrowhead pseudofovea position in deviated eye). When fusional vergence or prism adaptation is tested in 0°° 2.5°° 0°° such a patient, the angle of deviation tends to persis- tently return to that 2.3° angle. (b) Two-dimensional map representing V1 from the right cerebral hemisphere (left visual hemi-fi eld) of a microesotropic primate. Th e sulci and gyri have been unfolded and the visual fi eld representation superimposed b using standard retinotopic Right V-1 Left Visual Field landmarks. One horizontal axon (average length ~7 mm), D 180°° originating within the foveal Monocular 135°° 10°° representation at 0–1° Region 5°° 2.5°° 80°° 20°° 0°° eccentricity, could link to a 40°° 40°° receptive fi eld shift ed 2.5° or 80°° 20°° 4.4 PD distant. Two neurons 4.5°° strung together could join 10°° M FOVEA H.M. receptive fi elds 5° or 8.7 PD L apart. Th e conclusion that H.M. emerges is that the 4–8 PD “rule” of monofi xation/ 135°° microesotropia syndrome is explicable as a combination 180°° 45°° V 0°° of innate V1 neuron size (one to two axon lengths) 4.4 PD ≈ 1axon and V1 topography 7 mm 8.7 PD ≈ 2axon 14 mm

sensitivity (the location of an image on each retina with binocular MST neurons. Th e probability of surviving an respect to the foveola, or 0° eccentricity) guides ver- insult would be the greatest for the most populous neu- gence, as opposed to relative disparity sensitivity (the rons: those encoding ~2.5° (~4.4 PD) of convergence. In location of an image in depth with respect to other the presence of a generally weakened pool of disparity- images), which is necessary for stereopsis. Th e largest sensitive neurons, the vergence system may default to the population of vergence-related neurons in MST of nor- vergence commanded by the surviving population. A 2.5° mal monkeys drives the eyes to ~2.5° of convergent convergence angle could be kept stable (preventing dete- (crossed) disparity [82]. (Th e next largest population rioration to large angle strabismus) by the next most pop- encodes ~2.5° of divergence.) Normal primates have the ulous remaining neurons, those encoding 2.5° of strongest short-latency vergence responses to conver- divergence. Th ese mechanism are attractive because they gent disparities of ~2.5° [121]. can account for the direction, approximate magnitude, Insults that impair the development of binocular con- and stability of microesotropia, with retention of a capac- nections in immature V1 would be expected to impair the ity for fusional (e.g., prism) vergence responses evoked by (downstream) development of the entire population of disparities >2.5°. 54 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment

17. Phelps M, Mazziotta J, Kuhl D, et al (1981) Tomographic References mapping of human cerebral metabolism: visual stimula- 1. Graham PA (1974) Epidemiology of strabismus. Br tion and deprivation. Neurology 31:517–529 J Ophthalmol 58:224–231 18. Lorenz B (2002) Genetics of isolated and syndromic stra- 5 2. Kiorpes L, Boothe RG, Carlson MR, et al (1985) Frequency bismus: facts and perspectives. Strabismus 10:147–156 of naturally occurring strabismus in monkeys. J Pedriatr 19. Paul TO, Hardage LK (1994) Th e heritability of strabismus. Ophthalmol Strabismus 22:60–64 Ophthalmic Genet 15:1–18 3. Tamura EE, Hoyt CS (1987) Oculomotor consequences of 20. Maumenee IH, Alston A, Mets MB, et al (1986) Inheritance intraventricular hemorrhages in premature infants. Arch of congenital esotropia. Tr Am Ophthalmol Soc 84: Ophthalmol 105:533–535 85–93 4. Khanna S, Sharma A, Ghasia F, et al (2009) Prevalence of 21. Archer SM, Sondhi N, Helveston EM (1989) Strabismus in the ocular motor signs of the infantile strabismus complex infancy. Ophthalmology 96:133–137 in children with and without cerebral visual pathway white 22. Birch E, Stager D, Wright K, et al (1998) Th e natural his- matter injury. Invest Ophthalmol Vis Sci Abstracts 09-A- tory of infantile esotropia during the fi rst six months of life. 5876 J AAPOS 2:325–328 5. Pike MG, Holmstrom G, de Vries LS, et al (1994) Patterns 23. Fox R, Aslin RN, Shea SL, et al (1980) Stereopsis in human of visual impairment associated with lesions of the preterm infants. Science 207:323–324 infant brain. Dev Med Child Neurol 36:849–862 24. Birch EE, Gwiazda J, Held R (1982) Stereoacuity develop- 6. Hoyt CS (2003) Visual function in the brain-damaged ment for crossed and uncrossed disparities in human child. Eye 17:369–384 infants. Vision Res 22:507–513 7. van Hof-van Duin J, Evenhuis-van Leunen A, Mohn G, et 25. Birch EE, Shimojo S, Held R (1985) Preferential-looking al (1989) Eff ects of very low birth weight (VLBW) on visual assessment of fusion and stereopsis in infants aged 1 to 6 development during the fi rst year aft er term. Early Hum months. Invest Ophthalmol Vis Sci 26:366–370 Dev 20:255–266 26. O’Dell C, Boothe RG (1997) Th e development of 8. VanderVeen DK, Coats DK, Dobson V, et al (2006) stereoacuity in infant rhesus monkeys. Vision Res 37: Prevalence and course of strabismus in the fi rst year of life 2675–2684 for infants with prethreshold retinopathy of prematurity. 27. Birch EE, Gwiazda J, Held R (1983) Th e development of Arch Ophthalmol 124:766–773 vergence does not account for the onset of stereopsis. 9. Hiles DA, Hoyme SH, McFarlane F (1974) Down’s syn- Perception 12:331–336 drome and strabismus. Am Orthopt J 24:63–68 28. Gwiazda J, Bauer JAJ, Held R (1989) Binocular function 10. Shapiro MB, France TD (1985) Th e ocular features of in human infants: correlation of stereoptic and fusion- Down’s syndrome. Am J Ophthalmol 99:659–663 rivalry discriminations. J Pediatr Ophthalmol Strabismus 11. Pediatric eye disease investigator group (2002) Th e clinical 26:128–132 spectrum of early-onset esotropia: experience of the con- 29. Aslin RN (1977) Development of binocular fi xation in genital esotropia observational study. Am J Ophthalmol human infants. J Exp Child Psychol 23:133-150. 133:102–108 30. Aslin RN, Jackson RW (1979) Accommodative-convergence 12. Bailey P, von Bonin G (1951) Th e isocortex of man. in young infants: development of a synergistic sensory- University of Illinois, Urbana motor system. Can J Psychol 33:222–231 13. Tychsen L (1999) Infantile esotropia: current neurophysio- 31. Hainline L, Riddell PM (1995) Binocular alignment and logic concepts. In: Rosenbaum AL, Santiago AP (eds) Clinical vergence in early infancy. Vision Res. 35:3229–3236 strabismus management. WB Saunders, Philadephia 32. Horwood AM (1993) Maternal observations of ocular 14. Volpe JJ (1987) Hypoxic-ischemic encephalopathy: neuro- alignment in infants. J Pediatr Ophthalmol Strabismus pathology and pathogenesis. In: Volpe JJ, Markowitz M 30:100–105 (eds) Neurology of the newborn. W.B.Saunders Company, 33. Horwood AM, Riddell PM (2004) Can misalignments in Philadelphia typical infants be used as a model for infantile esotropia? 15. Noetzel MJ, Brunstrom JE (2001) Th e vulnerable oligoden- Invest Ophthalmol Vis Sci 45:714–720 drocyte. Infl ammatory observations on a cause of cerebral 34. Riddell PM, Horwood AM, Houston SM, et al (1999) Th e palsy. Neurology 56:1254–1255 response to prism deviations in human infants. Curr Biol 16. Huttenlocher P, de Courten C, Garey L, et al (1982) 9:1050–1052 Synaptogenesis in human visual cortex: evidence for syn- 35. Norcia AM, Garcia H, Humphry R, et al (1991) Anomalous apse elimination during normal development. Neurosci motion VEPs in infants and in infantile esotropia. Invest Lett 33:247–252 Ophthalmol Vis Sci 32:436–439 References 55

36. Norcia AM (1996) Abnormal motion processing and bin- 52. Hubel DH, Wiesel TN (1977) Ferrier lecture. Functional ocularity: infantile esotropia as a model system for eff ects architecture of macaque monkey visual cortex. Proc R Soc of early interruptions of binocularity. Eye 10:259–265 Lond Biol Sci 198:1–59 37. Brown RJ, Norcia AM (1997) A method for investigating 53. Hubel DH (1982) Exploration of the primary visual cortex, binocular rivalry in real-time with the steady-state VEP. 1955–78. Nature 299:515–524 Vision Res 37:2401–2408 54. Poggio GF, Fischer B (1977) Binocular interaction and 38. Birch EE, Fawcett S, Stager D (2000) Co-development of depth sensitivity in striate and prestriate cortex of behav- VEP motion response and binocular vision in normal ing rhesus monkey. J Neurophysiol 40:1392–1405 infants and infantile esotropes. Invest Ophthalmol Vis Sci 55. Wiesel TN (1982) Postnatal development of the visual 41:1719–1723 cortex and the infl uence of environment. Nature 299: 39. Bosworth RG, Birch EE (2007) Direction-of-motion detec- 583–591 tion and motion VEP asymmetries in normal children and 56. Hubel DH, Wiesel TN, LeVay S (1977) Plasticity of ocular children with infantile esotropia. Invest Ophthalmol Vis dominance columns in monkey striate cortex. Philos Trans Sci 48:5523–5531 R Soc Lond B Biol Sci 278:377–409 40. Atkinson J (1979) Development of optokinetic nystagmus 57. LeVay S, Wiesel TN, Hubel DH (1980) Th e development of in the human infant and monkey infant: an analogue to ocular dominance columns in normal and visually deprived development in kittens. In: Freeman RD (ed) Developmental monkeys. J Comp Neurol 191:1–51 neurobiology of vision. Plenum, New York 58. Chino Y, Smith EL, Hatta S, et al (1996) Suppressive bin- 41. Naegele JR, Held R (1982) Th e postnatal development of ocular interactions in the primary visual cortex (V1) of monocular optokinetic nystagmus in infants. Vision Res infant rhesus monkeys, Society for Neuroscience. 22:341–346 Washington, D.C 42. Schor CM, Narayan V, Westall C (1983) Postnatal develop- 59. Chino YM, Smith IIIEL, Hatta S, et al (1997) Postnatal ment of optokinetic aft er nystagmus in human infants. development of binocular disparity sensitivity in neurons Vision Res. 23:1643–1647 of the primate visual cortex. J Neurosci 17:296–307 43. Wattam-Bell J, Braddick O, Atkinson J, et al (1987) 60. Horton JC, Hocking DR (1996) An adult-like pattern of Measures of infant binocularity in a group at risk for stra- ocular dominance columns in striate cortex of newborn bismus. Clin Vis Sci 4:327–336 monkeys prior to visual experience. J Neurosci 16: 44. Jacobs M, Harris C, Taylor D (1994) Th e Development of 1791–1807 eye movements in infancy. In: Lennerstrand G (ed) Update 61. Horton JC, Hocking DR (1997) Timing of the critical on strabismus and pediatric ophthalmology. Proceedings period for plasticity of ocular dominance columns in of the Joint ISA and AAPO&S Meeting. Vancouver, macaque striate cortex. J Neurosci 17:3684–3709 Canada. June 19 to 23, 1994. CRC, Boca Raton 62. Endo M, Kaas JH, Jain N, et al (2000) Binocular cross-ori- 45. Tychsen L (2001) Critical periods for development of entation suppression in the primary visual cortex (V1) of visual acuity, depth perception and eye tracking. In: Bailey infant rhesus monkeys. Invest Ophthalmol Vis Sci DB Jr, et al (ed) Critical thinking about critical periods. 41:4022–4031 Paul H. Brookes, Baltimore 63. Crawford MLJ, von Noorden GK (1979) Th e eff ects of 46. Tychsen L, Hurtig RR, Scott WE (1985) Pursuit is impaired short-term experimental strabismus on the visual system but the vestibulo-ocular refl ex is normal in infantile strain Macaca mulatta. Invest Ophthalmol Vis Sci 18:496–505 bismus. Arch Ophthalmol 103:536–539 64. Crawford MLJ, Smith IIIEL, Harwerth RS, et al (1984) 47. Wong AMF, Foeller P, Bradley D, et al (2003) Early versus Stereoblind monkeys have few binocular neurons. Invest delayed repair of infantile stabismus in macaque monkeys: Ophthalmol Vis Sci 25:779–781 I. Ocular motor eff ects. J AAPOS 7:200–209 65. Crawford ML, Harwerth RS, Smith EL, et al (1996) Loss of 48. Birch E, Petrig B (1996) FPL and VEP measures of fusion, stereopsis in monkeys following prismatic binocular dis- stereopsis and stereoacuity in normal infants. Vision Res. sociation during infancy. Behav Brain Res 79:207–218 36:1321–1327 66. Lowel S, Singer W (1992) Selection of intrinsic horizontal 49. Skarf B, Eizenman M, Katz LM, et al (1993) A new VEP connections in the visual cortex by correlated neuronal system for studying binocular single vision in human activity. Science 255:209–212 infants. J Pediatr Ophthalmol Strabismus 30:237–242 67. Trachtenberg JT, Stryker MP (2001) Rapid anatomical 50. Mellick A (1949) Convergence. An investigation into the nor- plasticity of horizontal connections in the developing mal standards of age groups. Br J Ophthalmol 33: 755–763 visual cortex. J Neurosci 21:3476–3482 51. Tait EF (1949) Fusional vergence. Am J Ophthalmol 68. Tychsen L, Burkhalter A (1995) Neuroanatomic abnor- 32:1223–1230 malities of primary visual cortex in macaque monkeys 56 5 Visual Cortex Mechanisms of Strabismus: Development and Maldevelopment

with infantile esotropia: preliminary results. J Pediatr 84. Pratt-Johnson JA, Tillson G (1984) Suppression in strabis- Ophthalmol Strabismus 32:323–328 mus – an update. Br J Ophthalmol 68:174–178 69. Tychsen L, Yildirim C, Anteby I, et al (2000) Macaque 85. Hirsch JA, Gilbert CD (1991) Synaptic physiology of hori- monkey as an ocular motor and neuroanatomic model of zontal connections in primary visual cortex. J Neurosci 5 human infantile strabismus. In: Lennerstrand G, Ygge J 11:1800–1809 (eds) Advances in strabismus research: basic and clinical 86. Weliky M, Kandler K, Fitzpatrick D, et al (1995) Patterns of aspects. Wenner-Gren International Series. Portland, excitation and inhibition evoked by horizontal connec- London, U.K tions in visual cortex share a common relationship to ori- 70. Tychsen L, Wong AMF, Burkhalter A (2004) Paucity of entation columns. Neuron 15:541–552 horizontal connections for binocular vision in V1 of natu- 87. Deyoe EA, Trusk TC, Wong-Riley MT (1995) Activity cor- rally-strabismic macaques: cytochrome-oxidase compart- relates of cytochrome oxidase-defi ned compartments in ment specifi city. J Comp Neurol 474:261–275 granular and supragranular layers of primary visual cortex 71. Pasik T, Pasik P (1964) Optokinetic nystagmus: an of the macaque monkey. Vis Neurosci 12:629–639 unlearned response altered by section of chiasma and cor- 88. Horton JC (1984) Cytochrome oxidase patches: a new pus callosum in monkeys. Nature 203:609–611 cytoarchitectonic feature of monkey visual cortex. Philos 72. Pasik P, Pasik T (1977) Ocular movements in split-brain Trans R Soc Lond B Biol Sci 304:199–253 monkeys. Adv Neurol 18:125–135 89. Wong-Riley MTT (1994) Primate visual cortex: dynamic 73. Dürsteler MR, Wurtz RH, Newsome WT (1987) Directional metabolic organization and plasticity revealed by cyto- pursuit defi cits following lesions of the foveal representa- chrome oxidase. In: Peters A, Rockland KS (eds) Cerebral tion within the superior temporal sulcus of the macaque cortex. Plenum, New York monkey. J Neurophysiol 57:1262–1287 90. Fenstemaker SB, Kiorpes L, Movshon JA (2001) Eff ects of 74. Dürsteler MR, Wurtz RH (1988) Pursuit and optokinetic experimental strabismus on the architecture of macaque defi cits following chemical lesions of cortical areas MT and monkey striate cortex. J Comp Neurol 438:300–317 MST. J Neurophysiol 60:940–965 91. Horton JC, Hocking DR, Kiorpes L (1997) Pattern of ocu- 75. Ungerleider LG, Desimone R (1986) Cortical connections lar dominance columns and cytochrome oxidase activity of visual area MT in the macaque. J Comp Neurol 248: in a macaque monkey with naturally occurring anisome- 190–222 tropic amblyopia. Vis Neurosci 14:681–689 76. Kiorpes L, Walton PJ, O’Keefe LP, et al (1996) Eff ects of 92. Wong AMF, Burkhalter A, Tychsen L (2005) Suppression artifi cial early-onset strabismus on pursuit eye movements of metabolic activity caused by infantile strabismus and and on neuronal responses in area MT of macaque mon- strabismic amblyopia in striate visual cortex of macaque keys. J Neurosci 16:6537–6553 monkeys. J AAPOS 9:37–47 77. Hatta S, Kumagami T, Qian J, et al (1998) Nasotemporal 93. Tychsen L (2007) Causing and curing infantile eostropia in directional bias of V1 neurons in young infant monkeys. primates: the role of de-correlated binocular input (an Invest Ophthalmol Vis Sci 39:2259–2267 American ophthalmological society thesis). Trans Am 78. Tychsen L, Burkhalter A (1997) Nasotemporal asymme- Ophthalmol Soc 105:564–593 tries in V1: ocular dominance columns of infant, adult, and 94. Lewis TL, Maurer D (1992) Th e development of the tem- strabismic macaque monkeys. J Comp Neurol 388:32–46 poral and nasal visual fi elds during infancy. Vision Res 79. Boothe RG, Dobson V, Teller DY (1985) Postnatal develop- 32:903–911 ment of vision in human and nonhuman primates. Ann 95. Lewis TL, Maurer D, Blackburn K (1985) Th e development Rev Neurosci 8:495–546 of young infants’ ability to detect stimuli in the nasal visual 80. Kawano K (1999) Ocular tracking: behavior and neuro- fi eld. Vision Res. 25:943–950 physiology. Curr Opin Neurobiol 9:467–473 96. Bowering ER, Maurer D, Lewis TL, et al (1993) Sensitivity 81. Maunsell JHR, Van Essen DC (1983) Functional properties in the nasal and temporal hemifi elds in children treated for of neurons in middle temporal visual area of the macaque cataract. Invest Ophthalmol Vis Sci 34:3501–3509 monkey. II. Binocular interactions and sensitivity to bin- 97. Sireteanu R, Fronius M (1982) Naso-temporal asymme- ocular disparity. J Neurophysiol 49:1148–1167 tries in human amblyopia: consequence of long-term 82. Takemura A, Inoue Y, Kawano K, et al (2001) Single-unit interocular suppression. Vision Res 21:1055–1063 activity in cortical area MST associated with disparity- 98. Sireteanu R, Fronius M (1989) Visual fi eld losses in strabis- vergence eye movements: evidence for population coding. mic amblyopes. Klin Monatsbl Augenheilkd 194:261–269 J Neurophysiol 85:2245–2266 99. Horton JC, Hocking DR, Adams DL (1999) Metabolic map- 83. Jampolsky A (1955) Characteristics of suppression in stra- ping of suppression scotomas in striate cortex of macaques bismus. Arch Ophthalmol 54:683 with experimental strabismus. J Neurosci 19:7111–7129 References 57

100. Schor CM, Levi DM (1980) Disturbances of small-fi eld 112. Birch EE, Fawcett S, Stager DR (2000) Why does early sur- horizontal and vertical optokinetic nystagmus in amblyo- gical alignment improve stereopsis outcomes in infantile pia. Invest Ophthalmol Vis Sci 19:668–683 esotropia? J AAPOS 4:10–14 101. Tychsen L, Lisberger SG (1986) Maldevelopment of visual 113. Ing MR (1995) Surgical alignment prior to six months of motion processing in humans who had strabismus with age for congenital esotropia. Trans Am Ophthalmol Soc onset in infancy. J Neurosci 6:2495–2508 93:135–146 102. Tychsen L, Rastelli A, Steinman S, et al (1996) Biases of 114. Wright KW, Edelman PM, McVey JH, et al (1994) High- motion perception revealed by reversing gratings in grade stereo acuity aft er early surgery for congenital humans who had infantile-onset strabismus. Dev Med esotropia. Arch Ophthalmol 112:913–919 Child Neurol 38:408–422 115. Tychsen L, Wong AMF, Foeller P, et al (2004) Early versus 103. Hasany A, Wong A, Foeller P, et al (2008) Duration of bin- delayed repair of infantile strabismus in macaque monocular decorrelation in infancy predicts the severity of keys: II. Eff ects on motion visually evoked responses. Invest nasotemporal pursuit asymmetries in strabismic macaque Ophthalmol Vis Sci 45:821–827 monkeys. Neuroscience 156:403–411 116. Lang J (1968) Evaluation in small angle strabismus or 104. Richards M, Wong A, Foeller P, et al (2008) Duration of microtopia, Strabismus symposium Gieben. Karger, Basel binocular decorrelation predicts the severity of latent 117. Parks MM (1969) Th e monofi xation syndrome. Tr Am (fusion maldevelopment) nystagmus in strabismic Ophthalmol Soc 67:609–657 macaque monkeys. Invest Ophthalmol Vis Sci 49: 118. Tychsen L, Scott C (2003) Maldevelopment of convergence 1872–1878 eye movements in macaque monkeys with small and large- 105. Yildirim C, Tychsen L (2000) Disjunctive optokinetic nys- angle infantile esotropia. Invest Ophthalmol Vis Sci tagmus in a naturally esotropic macaque monkey: interac- 44:3358–3368 tions between nasotemporal asymmetries of versional eye 119. Wong AMF, Lueder GT, Burkhalter A, et al (2000) movement and convergence. Ophthalmic Res 32:172–180 Anomalous retinal correspondence: neuroanatomic mech- 106. Leigh RJ, Zee DS (1999) Th e neurology of eye movements. anism in strabismic monkeys and clinical fi ndings in stra- Oxford University, New York bismic children. J AAPOS 4:168–174 107. Worth C (1903) Squint. Its causes, pathology, and treat- 120. Cumming BG, Parker AJ (1999) Binocular neurons in V1 ment. Blakiston, Philadelphia of awake monkeys are selective for absolute, not relative, 108. Chavasse F (1939) Worth’s squint or the binocular refl exes disparity. J Neurosci 19:5602–6218 and the treatment of strabismus. 7th Bailliere Tindall and 121. Masson GS, Busettini C, Miles FA (1997) Vergence eye Cox, London movements in response to binocular disparity without 109. Costenbader FD (1961) Infantile esotropia. Trans Am depth perception. Nature 389:283–286 Ophthalmol Soc 59:397–429 122. Mays LE (1983) Neurophysiological correlates of vergence 110. Ing M, Costenbader FD, Parks MM, et al (1966) Early sur- eye movements. In: Schor CM, Ciuff reda KJ (eds) Vergence gery for congenital esotropia. Am J Ophthalmol eye movements: basic and clinical aspects. Butterworths, 61:1419–1427 Boston 111. Birch EE, Stager DR, Everett ME (1995) Random dot stere- 123. Mays LE (1984) Neural control of vergence eye move- oacuity following surgical correction of infantile esotropia. ments: convergence and divergence neurons in midbrain. J J Pediatr Ophthalmol Strabismus 32:231–235 Neurophysiol 51:1091–1108 Chapter 6

Neuroanatomical Strabismus 6 Joseph L. Demer

Core Messages ■ Strabismus may arise from identifi able structural ■ Strabismus may also arise from abnormalities of abnormalities of the extraocular muscles (EOMs) peripheral innervation of the EOMs. Congenital or their innervation. Congenital or acquired cranial dysinnervation disorders (CCDDs) typi- myopathies aff ect EOM function or structure to cally produce hypoplasia and loss of function of impair normal relaxation and force generation. insuffi ciently innervated EOMs, with contracture Abnormalities of EOM paths may produce stra- of their more normally innervated antagonists. bismus by altering EOM pulling directions. Path High resolution imaging can directly demonstrate abnormalities arise from abnormalities of the hypoplastic and misdirected motor nerves to the location and stability of the connective tissue pul- EOMs in the CCDDs, sometimes with additional leys that infl uence EOM paths. Pulley disorders abnormalities of the optic or other cranial nerves. may be congenital or acquired, and produce pat- Some forms of strabismus may be associated with tern strabismus, divergence paralysis esotropia, abnormalities of the brainstem or cerebellum that and horizontal or vertical incomitant strabismus. are demonstrable by clinical imaging. However, Structural abnormalities of EOMs or their associ- typical forms of developmental strabismus such ated connective tissues may be demonstrated by as concomitant esotropia and exotropia are not clinical orbital imaging. associated with EOM abnormalities.

6.1 General Etiologies of Strabismus 6.2 Extraocular Myopathy Strabismus, defi ned as misalignment of the visual direc- 6.2.1 Primary EOM Myopathy tions of the two eyes, may arise from several general causes. Th ese include primary myopathies of extraocu- Primary EOM myopathy may be due to congenital meta- lar muscles (EOMs), disorders of the connective tissues bolic disorder, acquired infl ammation, or mechanical that comprise the globe’s gimbal system, peripheral dis- trauma. Chronic progressive external ophthalmoplegia orders of nerves controlling the EOMs, and central dis- (CPEO) features insidious onset of slowly progressive, orders of fusional vergence commands (Table 6.1). Th is typically symmetric, external ophthalmoplegia [1]. chapter emphasizes causes of strabismus that can be Manifestations of CPEO range from involvement limited characterized as mechanistically specifi c pathologies of to the eyelids and EOMs to systemic and encephalopathic the subcortical nervous system, EOMs, and associated features. Tissues with high oxidative metabolism such as connective tissues. Such pathologies are termed neuro- muscle, brain, and heart are most aff ected [2]. Th e asso- anatomical because their causes can, at least in principle, ciation between CPEO and heart block is called Kearns– be demonstrated anatomically using appropriate clinical Sayre syndrome [3]. Ragged red fi bers, as demonstrated methods, and are distinct from developmental forms of on modifi ed trichrome stain, can be seen in limb and strabismus that arise from complex abnormalities in EOMs in nearly all cases of Kearns–Sayre syndrome and cerebral cortex. occasionally in isolated CPEO [3]. Molecular diagnosis of 60 6 Neuroanatomical Strabismus

Table 6.1. Etiologies of strabismus

Category Examples

Primary myopathies Mitochondrial myopathy, 6 endocrine myopathy, traumatic myopathy Orbital connective Pulley heterotopy, pulley tissue disorders instability, pulley hindrance Peripheral motor Congenital cranial dysinnervation neuropathies disorders (CCDDs), acquired peripheral ocular motor neuropathy Subcortical vergence Horizontal gaze palsy and disorders progressive scoliosis, cerebellar disease Cortical disorders Infantile strabismus, intermittent of vergence exotropia, accommodative esotropia Fig. 6.1 Coronal T1-weighted magnetic resonance imaging (MRI) of a right orbit of a patient with chronic progressive external ophthalmoplegia (CPEO) demonstrating abnormal CPEO is problematic, since most cases are caused by spo- bright signal within extraocular muscles that are of generally radic mitochondrial DNA deletions. More clinically use- normal size. IR inferior rectus muscle; LR lateral rectus muscle; ful may be T1-weighted magnetic resonance imaging ON optic nerve; SO superior oblique muscle; SR superior rectus muscle (MRI), which in CPEO demonstrates abnormal bright signal within clinically weak EOMs having generally normal size [1] (Fig. 6.1). Other cases of chronic, ﬁ xed EOM weakness are associated with obvious EOM atrophy (Table 6.2).

6.2.2 Immune Myopathy Immune EOM myopathy, also known as endocrine myopathy or thyroid eye disease (TED), is typically associated with immune dysthyroidism but may follow an independent temporal course [4]. TED begins with infl ammation and infi ltration of EOMs, orbital connective tissues, or both. A classical presentation of TED involves infl ammatory enlargement of EOMs producing upper eyelid retraction, proptosis, and restrictive ophthalmoplegia. Chronic EOM enlargement and fi brosis persists following resolution of infl ammation. Orbital imaging by MRI or computed X-ray tomography (CT) typically demonstrates enlargement of EOM bellies, sparing the terminal tendons. MRI demonstrates abnormal internal signal in involved EOMs (Fig. 6.2). Rectus EOMs, particularly the inferior and medial rectus (MR) muscles, demonstrate Fig. 6.2 the most common clinical involvement, although all Coronal T1-weighted MRI of both orbits of a patient with thyroid eye disease (TED) demonstrating enlargement and EOMs, including the obliques (Fig. 6.2), may be involved. in all rectus and the SO muscles. IR inferior rectus muscle; LR Restrictive strabismus is typical in TED, most commonly lateral rectus muscle; MR medial rectus muscle; ON optic nerve; involving limitation of supraduction. SO superior oblique muscle; SR superior rectus muscle 6.2 Extraocular Myopathy 61

Table 6.2. Types of extraocular myopathy

Cause Main clinical features Imaging ﬁ ndings Laboratory diagnostic tests

Metabolic Progressive weakness Normal EOM size, bright T1 MRI Muscle biopsy for ragged red signal fi bers, electrocardiogram Immune Restriction, EOM belly enlargement Th yroid function tests myopathy infl ammatory signs and/or orbital fat enlargement Infl ammatory Restriction and/or weakness, EOM belly and tendon Tests for vasculitis, myositis infl ammatory signs enlargement infl ammation, sarcoidosis Neoplastic Restriction and/or weakness, Nodular EOM enlargement, Metastatic evaluation, EOM myopathy and/or infl ammatory signs or orbital mass biopsy Mechanical Weakness or restriction EOM discontinuity or displacement, possible orbital fracture

Another presentation in TED is infl ammatory enlarge- transect or avulse EOM bellies, or avulse motor nerves to ment of the nonmuscular orbital connective tissues, par- EOMs. Clinically unrecognized penetration of the orbit ticularly orbital fat. Proptosis is the main feature, but by thin, sharp objections may occur in the setting of strabismus may arise due to forward displacement of the more widespread facial trauma, since entry wounds globe relative to fi xed structures such as the fi xed anchors of through the eyelid crease or conjunctival fornix are con- the trochlea and the soft pulley system of the other EOMs. cealed by edema and heal very quickly. High-resolution orbital imaging by CT or MRI may be valuable in the evaluation of strabismus associated with facial trauma, 6.2.3 Infl ammatory Myositis to detect direct EOM trauma and distinguish this from Myositis of EOMs not due to thyroid ophthalmopathy weakness of structurally intact EOMs due to traumatic typically involves both the EOM belly and tendon. cranial neuropathy [7]. Immunologic mechanisms with a host of triggers are Blunt orbital trauma may produce blow-out fractures believed to be the cause [5]. of the orbital walls, most commonly the thinner medial and inferior walls [8, 9]. In larger orbital fractures, EOMs and orbital connective tissues herniate into the adjacent 6.2.4 Neoplastic Myositis sinuses via relatively large bony defects. Large blow-out fractures are associated with enophthalmos, but not oft en Primary or metastatic neoplasms may cause strabismus with strabismus unless there is direct EOM trauma. Smaller by inducing EOM weakness or restriction. In such cases, orbital fractures may exhibit a trap-door mechanism, with orbital imaging may demonstrate nodular EOM enlarge- a displaced bone fragment trapping an EOM or part of the ment or a contiguous orbital mass [6]. Biopsy of the connective tissue pulley system. Especially in children in involved EOM may be helpful for diagnosis if likely meta- whom infl ammatory signs may not be clinically evident, static source is not already known. an EOM may become entrapped and strangulated in a trap-door orbital fracture. Entrapment and strangulation Summary for the Clinician of an EOM constitutes a situation demanding emergent ■ Old orbital fractures may complicate the presen- surgical release, while immediate repair is not typically tation of strabismus of recent origin. critical for most blow-out fractures. An entrapped EOM is ■ Patients may not recall old orbital fractures. very likely to exhibit clinical weakness on force generation testing, as well as producing a mechanical restriction to forced duction testing. Old, forgotten blow-out fractures may complicate the presentation of acquired strabismus 6.2.5 Traumatic Myopathy due to other causes [10]. Direct trauma to EOMs may compromise their function Even in the absence of EOM entrapment in an orbital and produce strabismus. Sharp objects penetrating the fracture, connective tissues of the orbital pulley system orbit may disinsert EOM tendons from the globe, may become entrapped in the fracture. Such a situation 62 6 Neuroanatomical Strabismus

may be associated with the clinical fi ndings of limitation ses. Malpositioning of the entheses, or malpositioning of of active duction in the EOM’s fi eld of action due to pulley the orbital bones to which the entheses join, can therefore hindrance (discussed below), as well as mechanical restric- cause signifi cant alterations in rectus EOM pulling direction to the opposite direction of passive rotation using for- tions. More signifi cant still, the pulling directions of the 6 ceps. In the usual clinical setting of generalized orbital and four horizontal rectus EOMs can be purely horizontal eyelid edema, these clinical fi ndings can be indistinguish- only if their respective pulleys all lie on a horizontal line able from those of EOM entrapment in an orbital fracture. exactly transverse to the mid-sagittal plane of the skull. It is therefore desirable to promptly obtain an adequate Any other orientation of the horizontal rectus pulleys in imaging study, such as a CT or MRI scan, that can identify the two orbits will impart vertically imbalanced actions to any possible tissue entrapped in an orbital fracture. the binocularly yoked agonist pairs: the MR in one orbit Expeditious, if not emergent, release of entrapped EOMs and the LR in the opposite orbit. Th is eff ect is not related or pulley tissue should be performed within several days to the activity of the oblique EOMs, and probably cannot before scarring makes repositioning impossible [11]. be counteracted by them. Symmetric heterotopy of the rectus pulley arrays in the orbits produces two clinical fi ndings: imbalanced ver- Summary for the Clinician sions in oblique gaze directions (formerly but incorrectly ■ Orbital pulley disorders can cause strabismus. attributed to oblique EOM dysfunction) and vertically ■ Strabismus due to pulley disorders can clinically incomitant horizontal strabismus [16, 17]. MRI has dem- mimic restrictive or paralytic strabismus. onstrated the coronal plane locations of rectus EOM pulleys to be stereotypic in normal [17, 18] and most strabismic subjects [18]. Th e 95% confi dence intervals of coronal plane pulley coordinates are less than ±1 mm 6.3 Congenital Pulley Heterotopy [18]. A computer model of binocular alignment incorpo- Th e direction of ocular rotation imparted by any EOM is rates passive elastic pulleys [19] and is now available as defi ned by the relative locations of its scleral insertion the application Orbit. Th e expected eff ect of coronal plane and pulley; EOM path direction posterior to the pulley is heterotopy (malpositioning) of pulleys can be computed not directionally important [12–14]. Every EOM can using Orbit [20]. Many cases of incomitant cyclovertical produce horizontal, vertical, and torsional actions, in strabismus are associated with heterotopy of one or more relative proportions depending on pulley and insertion rectus EOM pulleys exceeding two standard deviations locations. Th us, alterations in positions of the horizontal from normal. Patterns of incomitance in individual rectus pulleys can impart substantial vertical and tor- patients consistently match those predicted by Orbit sim- sional actions to the medial and lateral rectus (LR) EOMs, ulation based on measured pulley locations, suggesting while alterations in positions of the vertical rectus pulleys that pulley heterotopy caused the strabismus [21, 22]. can impart substantial horizontal actions to the vertical When the LR pulley is located superiorly to the MR rectus EOMs (Table 6.3). pulley in both orbits (Fig. 6.3a), the MR exerts an infra- Th e MR and LR pulleys are directly suspended by ducting action in adduction relative to that of the LR, fi broelastic connective tissues from anteriorly located causing excessive infraduction in extreme adduction, entheses, or anchors, on the orbital bones [15]. Th e medial since only the abducting eye can fi xate a target in this enthesis is at the posterior lacrimal crest, while the lateral position. Th is heterotopic pulley confi guration is typi- enthesis is at Whitnall’s tubercle. Th e inferior (IR) and cally associated with a nasal placement of the SR pulley superior rectus (SR) pulleys are somewhat indirectly relative to the IR pulley, such that the array of the four supported by, in both cases, the medial and lateral enthe- rectus pulleys has been incyclo rotated about the orbital

Table 6.3. Pattern strabismus associated with pulley heterotopy and eyelid conﬁ guration

Incomitance Horizontal Vertical pulleys Lateral canthal pulleys inclination

LR MR IR SR A pattern Superior Inferior Temporal Nasal Superior V pattern Inferior Superior Nasal Temporal Inferior 6.4 Acquired Pulley Heterotopy 63

Fig. 6.3 Coronal T2 fast spin echo MRI showing typical pulley conﬁ gurations of both orbits for (a) A and V (b) pattern strabismus

center. In supraversion, the SR exerts an adducting action, need not be bilaterally symmetrical; when asymmetrical, while in infraversion, the IR exerts an abducting action. the resulting strabismus may be horizontally as well as Binocular alignment is consequently more divergent in vertically incomitant, resembling dysfunction of a single infraversion than in supraversion, constituting an A pat- oblique EOM. tern strabismus. Osseous deformity with pulley heterotopy may be When the LR pulley is located inferiorly to the MR suspected when external facial features are asymmetri- pulley in both orbits (Fig. 6.3b), the MR exerts a supra- cal, or when there is a signifi cant inclination to one or ducting action in adduction relative to that of the LR, both the palpebral apertures [12, 23]. Th e medial and causing excessive supraduction in extreme adduction, lateral canthal tendons normally insert on the orbital since only the abducting eye can fi xate a target in this bones near the medial and lateral entheses of the pulley position. Th is heterotopic pulley confi guration is typi- system, respectively. A superior (“mongoloid”) inclina- cally associated with a temporal placement of the SR pul- tion of the lateral palpebral canthus is associated with A ley relative to the IR pulley, such that the array of the four pattern incomitance, while an inferior inclination of the rectus pulleys has been excyclo rotated about the orbital lateral palpebral canthus is associated with V pattern center [16]. In supraversion, the SR exerts an abducting incomitance. action, while in infraversion, the IR exerts an adducting action. Binocular alignment is consequently more convergent in infraversion than in supraversion, constituting 6.4 Acquired Pulley Heterotopy a V pattern strabismus. Bony deformity of the orbits, such as that associated Th e inferior oblique’s (IO’s) orbital layer inserts partly on with craniosynostosis, is a common cause of congenital the conjoined IO–IR pulleys, partly on the IO sheath pulley heterotopy. Such a deformity and pulley heterotopy temporally and partly on the LR pulley’s inferior aspect 64 6 Neuroanatomical Strabismus

[15, 24]. Consequently, the IO exerts a tonic nasalward When horizontal pulley sag occurs symmetrically, there force on the IR pulley, and a tonic inferior force on the LR is no eff ect on horizontal binocular alignment, since the pulley [24]. In youth, these active muscular forces are bal- MR and LR muscles experience balanced force reduc- anced by the elastic stiff ness of the pulley connective tis- tions [25]. Th e additional infraducting force contributed 6 sue suspensions, particularly by the elasticity of a ligament by the horizontal rectus EOMs is most likely to be the connecting the LR with the SR pulleys that is termed the cause of the predictably reduced supraducting ability of LR–SR band [15, 25]. Th e suspensory tissues of the orbital older people [28]. pulleys become gradually attenuated during normal aging More severe LR–SR band degeneration may permit the [15, 25], causing predictable inferior shift s in horizontal LR to shift farther inferiorly than does the MR pulley rectus pulley positions [26], and making the pulleys of (Fig. 6.4). In this case, more of LR abducting force is con- order people more susceptible to the eff ects of trauma verted to infraducting force than is the corresponding and surgery. situation for MR adducting force. Th e imbalance leads to a convergent shift in alignment most evident during distance viewing when the visual axes of the eyes should be Summary for the Clinician parallel, while there may be little or no esodeviation dur- ■ Pulley connective tissue degeneration in older ing near viewing where physiologic convergence is people can cause horizontal or vertical stra- required. Th is situation has been described as “diver- bismus. gence paralysis esotropia,” a clinical entity in which ■ Involutional eyelid changes and blepharopto- there is esotropia predominantly or exclusively present sis suggest that pulley tissues may also be during distance but not near viewing, and in which degenerating. there is no evidence of LR paresis, e.g., abducting saccadic velocities and abduction range are normal [27]. When bilaterally symmetrical, the vertical eff ect in the two eyes is matched, avoiding vertical strabismus. 6.5 “Divergence Paralysis” Esotropia “Divergence paralysis esotropia” due to LR pulley sag While the locations of the vertical rectus pulleys remain typically occurs in older people with retracted upper constant during the lifespan of a normal person, the hori- eyelid creases and blepharoptosis due to dehiscence of zontal rectus pulleys gradually “sag” inferiorly by 2–3 mm the levator tendon from the tarsal plate [25]. Both the by the seventh decade of life [26]. Th is converts some of blepharoptosis and strabismus presumably result from the horizontal force of the horizontal rectus EOMs to orbital connective tissue degeneration in the absence of infraducting force, without any abducens neuropathy or EOM neuropathy or myopathy. Patients typically retain defi ciency of the magnitude of LR force generation. excellent fusional convergence and binocular fusional Abducting saccades maintain normal peak velocities [27]. potential. While divergence paralysis esotropia can be

Fig. 6.4 Coronal histological sections of human left orbits of ages ranging from childhood to the ninth decade of life, showing attenuation and ultimate rupture of the LR–SR band with inferior sag of the LR pulley relative to the center of the medial rectus pulley (denoted by the yellow horizontal line). Masson’s trichrome stains collagen blue and muscle dark red. (Copyright nonexclu- sively assigned to American Academy of Ophthalmology, 2008.) 6.5 “Divergence Paralysis” Esotropia 65

Table 6.4. Alignment eff ect of LR–SR band degeneration diagnosis of sagging eye syndrome (Fig. 6.5). While it may sometimes be possible to surgically repair the rup- Symmetry Resulting strabismus tured or stretched LR–SR band to normalize LR pulley Bilaterally symmetric Divergence paralysis position, severe degeneration may render this ligament esotropia irreparable. In that event, posterior surgical ligature Asymmetric Hypotropia ± Esotropia between the lateral margin of the SR muscle and the superior margin of the LR muscle may be required to normalize LR path [25]. very successfully treated by multiple conventional strabismus surgical approaches that counteract esodeviation (e.g., MR recession or LR resection), it is the author’s 6.5.2 Postsurgical and Traumatic experience that the required surgical dosage must be Pulley Heterotopy about double that required for other forms of esotropia. Rectus pulley suspensions may be damaged by surgical Surgical repair of LR pulley sag is not typically required dissections. Again, the LR pulley is most susceptible to this in divergence paralysis esotropia (Table 6.4). eff ect of aggressive anterior dissection at strabismus, retinal, or orbital surgery. For instance, damage to the LR–SR band during endoscopic orbital decompression surgery 6.5.1 Vertical Strabismus Due may present as restrictive hypotropia in adduction. to Sagging Eye Syndrome Asymmetric stretching or catastrophic rupture of the LR–SR band may suddenly impart a marked infraduct- 6.5.3 Axial High Myopia ing action to the involved LR muscle, even creating restriction to passive supraduction [25] (Fig. 6.5). Th e Inferior displacement of the LR muscle is also a well-rec- clinical presentation may be acute onset of hypotropia ognized cause of strabismus in high myopes [29]. Known with defi ciency of supraduction that might be mistaken as “heavy eye” syndrome or myopic strabismus fi xus, this for SR paralysis or IR restriction in the absence of ade- syndrome is characterized by esotropia and hypotropia quate orbital imaging. Orbital imaging secures the due to conversion of LR muscle action from abduction to infraduction [29, 30]. Patients with “heavy eye” syndrome have impaired abduction and supraduction due to degeneration of the LR–SR band, allowing inferior LR pulley displacement causing inferior shift in LR muscle path that may become so extreme as to approach that of the LR. Abducting LR force is converted into infraducting force, resulting in large-angle esotropia and hypotropia. Since axial length in this condition is typically 30 mm or more, strabismus associated with axial high myopia was formerly (but misleadingly) termed the “heavy eye syndrome” under the assumption that an enlarged globe would sink inferiorly in the orbit [31]. Clinical orbital imaging is of great value in diagnosis of this condition, since it confi rms the diagnosis of LR displacement, and excludes alternative or coexisting conditions that may require diff erent surgical treatment, or preclude treatment altogether. For example, with or without inferior displacement of the LR pulley, a severely staphyomatous globe may fi ll the bony orbit so completely that duction is limited [32], or the LR muscle may have suff ered neuropathic paralysis and have become atrophic. If the cause of Fig. 6.5 Coronal MRI of left orbit of older patient demonstrating marked inferior displacement of LR pulley in sagging eye the esotropia is simply inferior displacement of the LR syndrome associated with acute onset hypotropia. LR lateral pulley due to LR–SR band degeneration, an eff ective rectus muscle; MR medial rectus muscle treatment may be identical to that used in the sagging eye 66 6 Neuroanatomical Strabismus

Summary for the Clinician 6.6 Congenital Peripheral Neuropathy: The Congenital Cranial Dysinnervation ■ Numerous structural abnormalities of extraocu- Disorders (CCDDs) lar muscles and associated connective tissues may cause strabismus. 6 Certain congenital forms of strabismus occur despite ■ Structural causes of strabismus may mimic neu- normal orbital connective tissues and pulleys, as the result rological causes of strabismus. of defi ciency or misdirection of motor nerves to the ■ High-quality orbital imaging is generally neces- EOMs. Genetic causes of many of the CCDDs are described sary to diagnose structural abnormalities of in chapter 7 in this volume by Antje Neugebauer and Julia Fricke, extraocular muscles and associated connective and will not be discussed here in this chapter that emphasizes tissues that cause strabismus. the patho physiology of strabismus. It is useful to understand two general principles in the functional anatomy of these CCDDs. First, EOMs with insuffi cient motor innerva- syndrome in the absence of high myopia: posterior surgi- tion are hypoplastic and hypofunctional. Second, eff ectively cal ligature between the lateral margin of the SR muscle innervated antagonists of congenitally noninnervated EOMs and the superior margin of the LR muscle. exhibit contracture and increased stiff ness (Table 6.5).

Table 6.5. Main imaging ﬁ ndings in CCDDs

Disorder Orbital ﬁ ndings Skull base ﬁ ndings

Congenital oculomotor Variable hypoplasia of inferior oblique Profound hypoplasia of oculomotor palsy (IO), IR, medial rectus (MR), SR, and LPS; nerves hypoplasia of intraorbital oculomotor nerve branches Congenital fi brosis Profound hypoplasia of SR and LPS; Profound hypoplasia of oculomotor of extraocular muscles nerves ± moderate MR, IO, SO hypoplasia; ± LR dysplasia; hypoplasia of intraorbital motor nerves; mild ON hypoplasia Congenital trochlear palsy Aff ected SO hypoplasia None (normal trochlear nerve usually too small to image) Duane syndrome Hypoplasia or aplasia of superior LR; ± Ipsilateral abducens nerve hypoplasia dysplasia of inferior LR; ± longitudinal LR splitting; ± abducens nerve aplasia; oculomotor nerve innervates inferior LR Moebius syndrome Hypoplasia of deep portions Normal subarachnoid cranial nerves of all myopathies of extraocular innervating orbit muscles (EOMs); curvature of anterior rectus EOMs; narrowing of deep bony orbits; ON straightening; Intraorbital motor nerve hypoplasia Horizontal gaze palsy with Normal Hypoplastic and fi ssured medulla and progressive scoliosis pons 6.6 Congenital Peripheral Neuropathy: The Congenital Cranial Dysinnervation Disorders (CCDDs) 67

typifi ed by bilateral congenital blepharoptosis and oph- 6.6.1 Congenital Oculomotor (CN3) Palsy thalmoplegia, with the eyes restricted to infraduction Congenital oculomotor (CN3) palsy is typically partial. below the horizontal midline [34]. Horizontal strabismus It may appear clinically bilateral or unilateral, although may coexist (Tables 6.5, 6.6). on careful evaluation apparently unilateral cases may be Forced duction testing in CFEOM1 demonstrates discovered to be bilateral albeit highly asymmetrical restriction to passive supraduction, consistent with [33]. Patients may present with variable defi ciencies of surgical observations of increased extraocular muscle adduction, supraduction, and infraduction, along with (EOM) stiffness. Older pathologic reports of speci- variable mydriasis and blepharoptosis. Aff ected EOMs mens of resected EOMs in CFEOM suggested replace- are hypoplastic, corresponding to their functional defi - ment by fibrous tissue [35–37]. The classic concept of ciencies. Intraorbital motor nerves to EOMs innervated CFEOM as a primary myopathy, however, was chal- by CN3 are hypoplastic, as is the subarachnoid CN3 lenged by autopsy findings in a subject from a pedigree (Fig. 6.6). with the KIF21A mutation [34]. Engle et al. alternatively suggested that CFEOM1 is a primary disorder of EOM motor neuron development, leading to hypoplasia or atrophy of the EOMs they innervate, and sec- 6.6.2 Congenital Fibrosis ondary contracture of their antagonists [34]. Older of the Extraocular Muscles (CFEOM) reports of “fibrosis” in EOM tendons are likely to have In many fundamental respects similar to congenital CN3 been artifacts of inadvertent biopsy of distal EOM ten- palsy, CFEOM is a heritable congenital CN3 hypoplasia dons [34]. with frequent misdirection of remaining fi bers, more Orbital MRI in CFEOM1 demonstrates hypoplasia profoundly aff ecting the superior than inferior division of the motor nerves normally innervated by CN3, most of CN3. Th ree distinct phenotypes, CFEOM1–3, are rec- profound for the SR and levator palpebrae superioris ognized. Th e classic form, CFEOM1 (MIM 135700), is corresponding to the clinically prominent hypotropia

Fig. 6.6 FIESTA MRI demonstrating hypoplasia of the subarachnoid oculomotor nerve (CN3). (a) Normal subject. (b) Dominant Duane retraction syndrome (DRS) linked to chromosome 2 (DURS2). (c) Congenital oculomotor palsy. (d) Congenital ﬁ brosis of the extraocular muscles type 1 (CFEOM1) 68 6 Neuroanatomical Strabismus

Table 6.6. Imaging features in acquired neuropathic extraocular muscle palsy

Muscle Size Contractility Path

Inferior oblique Reduced 40% Reduced Normal 6 IR Small posteriorly Reduced Centrifugal inﬂ ection Lateral rectus Reduced 50–90% posteriorly Reduced Centrifugal inﬂ ection Levator palpebrae superioris Small Cannot evaluate Normal Medial rectus Small posteriorly Reduced Normal SO Reduced 40–50% Reduced Normal SR Small posteriorly Reduced Normal

and blepharoptosis (Fig. 6.7a, b) [38]. Intraorbital motor More direct evidence of this misrouting is provided by branches of CN3 are also hypoplastic (Fig. 6.7c). high-resolution MRI showing innervation of the inferior MRI in CFEOM1 demonstrates marked hypoplasia of zone of the LR by a branch of CN3 that would normally the subarachnoid CN3. Signiﬁ cant but usually subclinical be fated to innervate the IR. In most cases, when a patient optic nerve (ON) hypoplasia occurs in CFEOM1, as may with CFEOM1 attempts deorsumversion, the eyes abduct superior oblique (SO) muscle hypoplasia presumably due dye to LR contraction, increasing the exotropia present in to trochlear nerve (CN4) hypoplasia. Th e posterior parts central gaze. In CFEOM1, CN6 innervates the superior of multiple EOMs may be dysplastic in CFEOM, although zone of the LR muscle. their anterior portions generally appear normal both by Patients with CFEOM2 (OMIM 602078) have congeni- MRI and at EOM surgery. tally bilateral exotropic ophthalmoplegia and blepharop- Th e frequent occurrence of synergistic eye movements tosis. Th is rare recessive disorder occurs in consan and the Marcus Gunn jaw winking phenomenon in guineous pedigrees. Th e orbital and cranial nerve pheno- CFEOM1 [39, 40] suggests motor axonal misrouting. type of CFEOM2 have not been studied in detail.

Fig. 6.7 Typical orbital MRI ﬁ ndings in CFEOM1. (a) Sagittal view showing profound hypoplasia of the SR and levator palpebrae superioris. (b) Coronal view in mid-orbit showing profound hypoplasia of the SR. (c) Deep orbital view demonstrating proximity and presumed innervation of the inferior zone of the LR by an aberrant of the inferior division of the oculomotor nerve (CN3) 6.6 Congenital Peripheral Neuropathy: The Congenital Cranial Dysinnervation Disorders (CCDDs) 69

Th e third CFEOM variant, CFEOM3, encompasses 6.6.4 Duane’s Retraction Syndrome (DRS) patients with CFEOM not classifi able as either CFEOM1 or CFEOM2. Th is “atypical” group includes unilateral Pure congenital abducens (CN6) palsy is exceptionally cases who have orthotropic central gaze, or whose central rare except as secondary to an obvious intrauterine or gaze is hypotropic but who can supraduct above the cen- neonatal pathology such as tumor or hydrocephalus. tral position. Subjects with CFEOM3 have asymmetrical Rather, in congenital developmental CN6 palsy, the LR is blepharoptosis, limited supraduction, variable ophthal- innervated or coinnervated by a branch of CN3, usually a moplegia, and are usually exotropic. MRI demonstrates motor branch ordinarily fated to innervate the MR. In asymmetrical levator palpebrae superioris and SR atro- this respect, the situation is similar to CFEOM. DRS is phy correlating with blepharoptosis and defi cient supra- characterized by congenital abduction defi cit, narrowing duction, and small orbital motor nerves [41]. While at of the palpebral fi ssure on adduction, and globe retrac- least one subarachnoid CN is hypoplastic, ophthalmople- tion with occasional upshoot or downshoot in adduction gia occurs only when subarachoid CN3 width is less than [43]. Early electrophysiological studies suggested absence the 2.5th percentile of normal. Multiple EOMs exhibit of normal abducens (CN6) innervation to the LR muscle variable hypoplasia, correlating with duction in individ- as the cause of DRS, with paradoxical LR innervation in ual orbits. A-pattern exotropia is frequent in CFEOM3, adduction [44, 45]. Absence of the CN6 nerve and motor correlating with LR misinnervation by CN3. ON cross- neurons has been confi rmed in one sporadic unilateral sections are slightly subnormal, but rectus pulley loca- [46] and another bilateral autopsy case of DRS [47]. Parsa tions are normal [42]. Some cases of CFEOM3 are et al. fi rst used MRI to demonstrate absence of the suba- associated with brain abnormalities including corpus cal- rachnoid portion of CN6 in DRS [48], a fi nding that has losum hypoplasia. been confi rmed in 6 of 11 additional cases [49], and later correlated with the presence of residual abduction in multiple cases [50, 51]. Innervation of the LR by CN6 is defi cient in both DRS Summary for the Clinician and CN6 palsy, although unlike CN6 palsy, the eyes in ■ CFEOM is not a primary muscle disorder, but central gaze are frequently aligned in DRS [52]. While rather a cranial nerve disorder. most DRS cases are sporadic, a dominant form DURS2 is linked to chromosome 2. MRI demonstrated that DRS linked to the DURS2 locus is associated with bilateral abnormalities of many orbital motor nerves, and structural abnormalities of all EOMs except those innervated 6.6.3 Congenital Trochlear (CN4) Palsy by the inferior division of CN3 [53]. Orbital motor nerves While SO hypoplasia may coexist with other CCDDs are typically small, with CN6 oft en nondetectable. Lateral such as CFEOM, SO dysfunction may not be clinically rectus (LR) muscles are oft en structurally abnormal, oft en evident in the setting of diff use external ophthalmople- with MRI and motility evidence of oculomotor nerve gia or anomalous innervation of other EOMs. Isolated (CN3) innervation from vertical rectus EOMs leading to congenital CN4 palsy is oft en suspected in the presence A or V patterns of strabismus. Cases may include SO, SR, of clinical evidence of ipsilateral hypertropia increasing and LPS hypoplasia, sparing only the MR, IR, and IO on contralateral gaze, and with head tilt toward the ipsi- EOMs. Th e subarachnoid CN3 may be small. Th erefore, lateral shoulder. While the congenital nature of the dis- DURS2-linked DRS is a diff use CCDD involving but not order appears clear when there is a history of lifelong limited to CN6. spontaneous head tilt to the contralateral shoulder, in many cases present aft er many years of compensation for what the history suggests has been a progressive Summary for the Clinician condition without identifi able cause. Whether lifelong or insidious, orbital imaging in presumably congenital ■ CCDDs are nonprogressive developmental SO palsy demonstrates reduction in SO muscle size, and disorders featuring reduced and aberrant reduction in the normal contractile increase in SO innervation. cross-section due to infraduction (Fig. 6.8). Since even ■ Subnormal innervation of some EOMs in the normal subarachnoid CN4 cannot be reliably CCDDs leads to secondary EOM hypoplasia, imaged by MRI, correlations with CN4 size have not dysplasia, and weakness. been made in congenital CN4 palsy. 70 6 Neuroanatomical Strabismus

■ Antagonists of hypoplastic EOMs become sec- impairment. Moebius syndrome is a heterogeneous ondarily stiﬀ . clinical disorder whose clinical deﬁ nition has evolved in ■ Neuromuscular features may vary between orbits the recent literature. Minimum criteria include congeni- of the same patient, and among patients with tal facial palsy with impairment of ocular abduction 6 identical genetic CCDDs. [54–56]. Th e wide clinical spectrum and multiple areas ■ High-resolution imaging of EOMs and their of brainstem involvement in patients with Moebius syn- peripheral innervation can be clinically valuable drome have led to its early conceptualization as a devel- for strabismus management in CCDDs. opmental disorder of the brainstem, rather than an isolated cranial nerve developmental disorder [56]. However, Moebius syndrome may present with total facial paralysis and complete external ophthalmoplegia, where MRI demonstrates a normal brainstem and suba- 6.6.5 Moebius Syndrome rachnoid portions of motor cranial nerves innervating Moebius syndrome typically presents as a sporadic trait the orbit, but marked hypoplasia of the deep portions of with congenital facial (CN7) palsy and abduction the EOMs.

Fig. 6.8 Coronal T2-weighted MRI of both orbits in left SO palsy demonstrating marked reduction in SO cross-section, as well as reduction in normal contractile increase in cross-section from up to down gaze 6.7 Acquired Motor Neuropathy 71

6.7 Acquired Motor Neuropathy Summary for the Clinician On orbital imaging, the hallmarks of EOM denervation ■ Th e three-step test is not specifi c for trochlear are atrophy of the EOM belly, and loss of normal contrac- palsy. tile increase in EOM cross-section in the EOM’s fi eld of ■ Orbital imaging confi rms neurogenic atrophy of action. the SO muscle.

High-resolution MRI has quantifi ed normal changes in SO 6.7.1 Oculomotor Palsy cross-section with vertical gaze, and SO atrophy and loss of Chronic oculomotor palsy is associated with neurogenic gaze-related contractility typical of SO palsy [23, 65–67]. atrophy of the associated EOMs, but the degree of atrophy Following experimental intracranial trochlear neurectomy appears to be related to the presence of any residual inner- in monkey, the SO atrophies within 5 weeks to a stable vation or reinnervation, either normal or aberrant [7]. overall size 60% of normal; this atrophy occurs entirely Little or no EOM atrophy may be present when there is within the global layer, where fi ber size is reduced by 80%, aberrant innervation, even if this innervation would nor- sparing the orbital layer [68]. A striking and consistent mally have been directed to another EOM. High-resolution MRI fi nding has been nonspecifi city of the three-step test imaging in chronic oculomotor palsy also demonstrates for structural abnormalities of the SO belly, tendon, and atrophy of the intraorbital branches of the oculomotor trochlea, found in only in ~50% of patients [69]. Even in nerve [7], similar to that observed in CFEOM. patients selected because MRI demonstrated profound SO atrophy, there was no correlation between clinical motility and IO size or contractility [67]. Multiple conditions can simulate the “SO palsy” pat- 6.7.2 Trochlear Palsy tern of incomitant hypertropia [70]. Vestibular lesions Th eoretical, experimental, and much clinical evidence produce head-tilt-dependent hypertropia, also known as support the idea that acute, unilateral SO palsy produces skew deviation [71] that can mimic SO palsy by the three- a small ipsilateral hypertropia that increases with contral- step test [72]. Pulley heterotopy can simulate SO palsy ateral gaze, and with head tilt to the ipsilateral shoulder [16, 73], and is probably not its result, since SO atrophy is [57, 58]. Th e basis of this “three-step test” is traditionally not associated with signifi cant alterations in pulley posi- believed to be related to Ocular Counter Rolling (OCR), tion in central gaze [21]. so that the eye ipsilateral to head tilt is normally intorted by the SO and SR muscles whose vertical actions cancel [59]. However, ipsilateral to a palsied SO, unopposed SR 6.7.3 Abducens Palsy elevating action is supposed to create hypertropia. Th e three-step test has been the cornerstone of diagnosis and Denervation of the LR is associated with muscle belly classifi cation cyclovertical strabismus for generations of atrophy [74, 75], loss of contractile thickening during clinicians [60]. When the three-step test is positive, clini- attempted abduction, and a centrifugal bowing of the LR cians infer SO weakness and attribute the large amount of path away from the orbital center with accentuation of interindividual alignment variability to secondary changes the transverse infl ection in LR path near the posterior [61] such as “IO overaction” and “SR contracture.” Much mouth of the LR pulley sleeve (Fig. 6.9) [76]. Such evidence, however, indicate that the three-step test’s changes in atrophic LR path elongate its length, a factor mechanism is misunderstood. Kushner has pointed out that tends to increase passive elastic tension of the para- that if traditional teaching were true, then IO weakening, lyzed LR [77]. the most common surgery for SO palsy, should increase the head-tilt-dependent change in hypertropia; however, the opposite is observed [62]. Among numerous inconsisten- 6.7.4 Inferior Oblique (IO) Palsy cies with common clinical observations [62], bilateral should cause greater head-tilt-dependent change in Since the inferior division of CN3 innervates the IO, IO hypertropia than unilateral SO palsy; however, the oppo- palsy commonly accompanies weakness of multiple site is found [63]. Simulation of putative eff ects head tilt EOMs produced by a proximal lesion to this large motor in SO palsy suggests that SO weakness alone cannot nerve. However, the IO’s motor nerve follows a relatively account for typical three-step test fi ndings [64]. lengthy isolated course along the lateral margin of the IR 72 6 Neuroanatomical Strabismus

Fig. 6.9 Orbital T2-weighted MRI in chronic left abducens palsy. Axial view above shows thinning and lateral inﬂ ection of palsied LR 6 muscle, which in coronal view below is seen to have reduced cross-section

muscle, entering the IO in the EOM’s posterior surface 6.8.2 Cerebellar Disease relatively superfi cially in the orbit when compared with innervation to the other EOMs. Isolated acquired neuro- Th e cerebellum contributed toward binocular alignment pathic IO palsy is thus anatomically possible. When it [81]. Hereditary cerebellar degeneration is oft en associ- occurs, IO palsy is associated with denervation atrophy of ated with convergence insuffi ciency, and in advanced the IO belly [78]. cases oft en produces cerebellar atrophy [82]. Cerebellar or brainstem tumors may be associated with acute onset of concomitant esotropia in children [83]. Acquired cer- 6.8 Central Abnormalities of Vergence ebellar damage, such as by infarction, may produce skew and Gaze deviation other strabismus. Several common causes of strabismus are not associated with abnormalities of the EOMs, motor nerves, or orbital connective tissues. Th e forms of strabismus arise from 6.8.3 Horizontal Gaze Palsy and Progressive Scoliosis abnormalities in the central nervous system, some of which are structural lesions that may be imaged. Horizontal gaze palsy and progressive scoliosis is a recessive disorder of axon path fi nding in the central nervous system. Patients have essentially complete horizontal Summary for the Clinician ophthalmoplegia despite intact EOMs and peripheral ■ Developmental esotropia and exotropia are not motor innervation to them, but MRI demonstrates dys- associated with structural abnormalities in the plasia of the hindbrain suggestive of a sagittal fi ssure orbit. interrupting decussating white matter tracts [84].

6.8.1 Developmental Esotropia and Exotropia References

Human infantile and accommodative esotropia are not 1. Ortube MC, Bhola R, Demer JL (2006) Orbital magnetic known to be associated with anatomic abnormalities in resonance imaging of extraocular muscles in chronic pro- the orbit. Naturally and artifi cially esotropic and exo- gressive external ophthalmoplegia: Specifi c diagnostic tropic monkeys also exhibit latent nystagmus and other fi ndings. J AAPOS 10:414–418 features typical of human childhood strabismus, but 2. Biousse V, Newman NJ (2001) Neuro-ophthalmology of these monkeys have normal horizontal EOM structure, mitochondrial diseases. Semin Neurol 3:275–291 normal EOM locations, and normal intraorbital EOM 3. Vilarinho L, Santorelli FM, Cardoso ML, et al (1998) innervation [79]. However, strabismic monkeys have Mitochondrial DNA analysis in ocular myopathy. Observations microscopic evidence of abnormalities in visual cortical in 29 portuguese patients. Euro Neurol 39: 148–153 area V1 [80], and most likely also other visual cortical 4. Kuriyan AE, Phipps RP, Feldon SE (2008) Th e eye and thy- areas. roid disease. Cur Opin Ophthalmol 19:499–506 References 73

5. Wolf AB, Yang MB, Archer SM (2007) Postoperative myosi- 24. Demer JL, Oh SY, Clark RA, et al (2003) Evidence for a tis in reoperated extraocular muscles. J AAPOS 11: 373–376 pulley of the inferior oblique muscle. Invest Ophthalmol 6. Capone AJ, Slamovits TL (1990) Discrete metastasis of Vis Sci 44:3856–3865 solid tumors to extraocular muscles. Arch Ophthalmol 25. Rutar T, Demer JL (2009) “Heavy eye syndrome” in the 108:237–243 absence of high myopia: A connective tissue degeneration 7. Demer JL (2003) A 12 year, prospective study of extraocu- in elderly strabismic patients. J AAPOS 13(1):36–44 lar muscle imaging in complex strabismus. J AAPOS 26. Clark RA, Demer JL (2002) Eff ect of aging on human rec- 6:337–347 tus extraocular muscle paths demonstrated by magnetic 8. Koornneef L (1982) Current concepts on the management resonance imaging. Am J Ophthalmol 134:872–878 of orbital blow-out fractures. Ann Plast Surg 9:185–200 27. Lim L, Rosenbaum AL, Demer JL (1995) Saccadic velocity 9. Wojno TH (1987) Th e incidence of extraoular muscle and analysis in patients with digergence paralysis. J Pediatr cranial nerve palsy in orbital fl oor blow-out fractures. Ophthalmol Strabismus 32:76–81 Ophthalmol 94:682–687 28. Clark RA, Isenberg SJ (2001) Th e range of ocular move- 10. Ortube MC, Rosenbaum AL, Goldberg RA, et al (2004) ments decreases with aging. J AAPOS 5:26–30 Orbital imaging demonstrates occult blow out fracture in 29. Krzizok TH, Schroeder BU (1999) Measurement of recti complex strabismus. J AAPOS 8:264–273 eye muscle paths by magnetic resonance imaging in highly 11. Tse R, Allen L, Matic D (2007) Th e white-eyed medial myopic and normal subjects. Invest Ophthalmol Vis Sci blowout fracture. Plast Reconstr Surg 119:277–286 40:2554–2560 12. Demer JL (2004) Pivotal role of orbital connective tissues 30. Kowal L, Troski M, Gilford E (1994) MRI in the heavy eye in binocular alignment and strabismus. Th e Friedenwald phenomenon. Aust N Zeal J Ophthalmol 22:125–126 lecture. Invest Ophthalmol Vis Sci 45:729–738 31. Ward DM (1956) Th e heavy eye phenomenon. Trans 13. Demer JL (2006) Current concepts of mechanical and neu- Ophthalmol Soc U K 87:717–726 ral factors in ocular motility. Cur Opin Neurol 19:4–13 32. Demer JL, von Noorden GK (1989) High myopia as an 14. Demer JL (2006) Ocular motility in a time of revolutionary unusual cause of restrictive motility disturbance. Surv paradigm shift . Invest Ophthalmol Vis Sci Ophthalmol 33:281–284 15. Kono R, Poukens V, Demer JL (2002) Quantitative analysis 33. Wu J, Isenberg S, DJ L (2006) Magnetic resonance imaging of the structure of the human extraocular muscle pulley demonstrates neuropathology in congenital inferior divi- system. Invest Ophthalmol Vis Sci 43:2923–2932 sion oculomotor palsy. J AAPOS 10:473–475 16. Clark RA, Miller JM, Rosenbaum AL, et al (1998) 34. Engle EC, Goumnerov BC, McKeown CA, et al (1997) Heterotopic muscle pulleys or oblique muscle dysfunc- Oculomotor nerve and muscle abnormalities in congeni- tion? J AAPOS 2:17–25 tal fi brosis of the extraocular muscles. Ann Neurol 17. Miller JM, Demer JL Biomechanical modeling in strabis- 41:314–325 mus surgery. In: Rosenbaum AL, Santiago P (eds) (1999) 35. Apt L, Axelrod N (1978) Generalized fi brosis of the Clinical strabismus management: principles and tech- extraocular muscles. Am J Ophthalmol 85:822–829 niques. Mosby, St. Louis 36. Brodsky MC, Pollock SC, Buckley EG (1989) Neural misdi- 18. Clark RA, Miller JM, Demer JL (1997) Location and stabil- rection in congenital ocular fi brosis syndrome: Implications ity of rectus muscle pulleys inferred from muscle paths. and pathogenesis. J Pediatr Ophthalmol Strabismus Invest Ophthalmol Vis Sci 38:227–240 26:159–161 19. Miller JM (2007) Understanding and misunderstanding 37. Harley RD, Rodrigues MM, Crawford JS (1978) Congenital extraocular muscle pulleys. J Vision 7:1–15 fi brosis of the extraocular muscles. Trans Am Ophtlalmol 20. Miller JM, Pavlovski DS, Shaemeva I (1999). Orbit 1.8 gaze Soc 76:197–226 mechanics simulation. Eidactics, San Francisco 38. Demer JL, Clark RA, Engle EC (2005) Magnetic resonance 21. Clark RA, Miller JM, Demer JL (1998) Displacement of the imaging evidence for widespread orbital dysinnervation in medial rectus pulley in superior oblique palsy. Invest congenital fi brosis of extraocular muscles due to mutations Ophthalmol Vis Sci 39:207–212 in KIF21A. Invest Ophthalmol Vis Sci 46:530–539 22. Demer JL, Clark RA, Miller JM (1999) Heterotopy of 39. Brodsky MC (1998) Hereditary external ophthalmoplegia, extraocular muscle pulleys causes incomitant strabismus. synergistic divergence, jaw winking, and oculocutaneous In: Lennerstrand G (ed) Advances in strabismology. hypopigmentation. Ophthalmology 105:717–725 Aeolus, Buren, Th e Netherlands 40. Yamada K, Andrews C, Chan W-M, et al (2003) 23. Velez FG, Clark RA, Demer JL (2000) Facial asymmetry in Heterozygous mutations of the kinesin KIF21A in congen- superior oblique palsy and pulley heterotopy. J AAPOS ital fi brosis of the extraocular muscles type 1 (CFEOM1). 4:233–239 Nat Genet 35:318–321 74 6 Neuroanatomical Strabismus

41. Demer JL, Clark RA, Engle EC (2009) Magnetic resonance 56. Verzijl HT, van der Zwaag B, Cruysberg JR, et al (2003) imaging evidence of an asymmetrical endophenotype in Mobius syndrome redefi ned: a syndrome of rhomben- congenital fi brosis of extraocular muscles type 3. Invest cephalic maldevelopment. Neurology 61:327–333 Ophthalmol Vis Sci (in preparation) 57. Bielschowsky A (1939) Lectures on motor anomalies. XI. 6 42. Clark RA, Engle EC, Demer JL (2009) Magnetic resonance Etiology, prognosis, and treatment of ocular paralyses. Am imaging (MRI) of the endophenotype of congenital fi brosis J Ophthalmol 22:723–734 of the extraocular muscles type 3 (CFEOM3). Abstracts of 58. von Noorden GK, Murray E, Wong SY (1986) Superior 35th Annual Meeting of the American Association for oblique paralysis. A review of 270 cases. Arch Ophthalmol Pediatric Ophthalmology and Strabismus. J Am Assoc 104:1771–1776 Pediatr Ophthmol Strabismus 13(1):e13 59. Adler FE (1946) Physiologic factors in diff erential diagno- 43. Duane A (1905) Congenital defi ciency of abduction asso- sis of paralysis of superior rectus and superior oblique ciated with impairment of adduction, retraction move- muscles. Arch Ophthalmol 36:661–673 ments, contraction of the palpebral fi ssure and oblique 60. Scott WE, Kraft SP (1986) Classifi cation and treatment of movements of the eye. Arch Ophthalmol 34:133–159 superior oblique palsies: II. Bilateral superior oblique pal- 44. Huber A (1974) Electrophysiology of the retraction syn- sies. In: Caldwell D (ed) Pediaric ophthalmology and stra- dromes. Br J Ophthalmol 58:293–300 bismus: transactions of the New Orleans academy of 45. Strachan IM, Brown BH (1972) Electromyography of ophthalmology. Raven, New York extraocular muscles in Duane’s syndrome. Br J Ophthalmol 61. Straumann D, Steff en H, Landau K, et al (2003) Primary 56:594–599 position and Listing’s law in acquired and congenital tro- 46. Miller NR, Kiel SM, Green WR, et al (1982) Unilateral chlear nerve palsy. Invest Ophthalmol Vis Sci 44:4282–4292 Duane’s retraction syndrome (type 1). Arch Ophthalmol 62. Kushner BJ (2004) Ocular torsion: Rotations around the 100:1468–1472 “WHY” axis. J AAPOS 8:1–12 47. Hotchkiss MG, Miller NR, Clark AW, et al (1980) Bilateral 63. Kushner BJ (1988) Th e diagnosis and treatment of bilateral Duane’s retraction syndrome. A clinical-pathologic case masked superior oblique palsy. Am J Ophthalmol report. Archives of Ophthalmology 98:870–874 105:186–194 48. Parsa CF, Grant E, Dillon WP, et al (1998) Absence of the 64. Robinson DA (1985) Bielschowsky head-tilt test–II. abducens nerve in Duane syndrome verifi ed by magnetic Quantitative mechanics of the Bielschowsky head-tilt test. resonance imaging. Am J Ophthalmol 125:399–401 Vision Res 25:1983–1988 49. Ozkurt H, Basak M, Oral Y, et al (2003) Magnetic reso- 65. Chan TK, Demer JL (1999) Clinical features of congenital nance imaging in Duane’s retraction syndrome. J Pediatr absence of the superior oblique muscle as demonstrated by Ophthalmol Strabismus 40:19–22 orbital imaging. J AAPOS 3:143–150 50. Kim JH, Hwang J-M (2005) Hypoplastic oculomotor nerve 66. Demer JL, Miller JM (1995) Magnetic resonance imaging and absent abducens nerve in congenital fi brosis syndrome of the functional anatomy of the superior oblique muscle. and synergistic divergence with magnetic resonance imag- Invest Ophthalmol Vis Sci 36:906–913 ing. Ophthalmology 112:728–732 67. Kono R, Demer JL (2003) Magnetic resonance imaging of 51. Kim JH, Hwang JM (2005) Presence of abducens nerve the functional anatomy of the inferior oblique muscle in according to the type of Duane’s retraction syndrome. superior oblique palsy. Ophthalmology 110:1219–1229 Ophthalmology 112:109–113 68. Demer JL, Poukens V, Ying H, et al (2008) Eff ects of acute 52. DeRespinis PA, Caputo AR, Wagner RS, et al (1993) Duane’s trochlear denervation on primate superior oblique (SO) Retraction Syndrome. Surv Ophthalmol 38:257–288 muscle: diff erential sparing of orbital layer. ARVO 53. Demer JL, Clark RA, Lim KH, et al (2007) Magnetic reso- Abstracts: #4495 nance imaging evidence for widespread orbital dysinner- 69. Demer JL, Miller MJ, Koo EY, et al (1995) True versus mas- vation in dominant Duane’s retraction syndrome linked to querading superior oblique palsies: muscle mechanisms the DURS2 locus. Invest Ophthalmol Vis Sci 48:194–202 revealed by magnetic resonance imaging. In: Lennerstrand 54. Kumar D (1990) Moebius syndrome. J Med Genet 27: G (ed) Update on strabismus and pediatric ophthalmology. 122–126 CRC, Boca Raton (FL) 55. MacDermot KD, Winter RM, Taylor D, et al (1991) 70. Kushner BJ (1987) Errors in the three-step test in the diag- Oculofacialbulbar palsy in mother and son: review of 26 nosis of vertical strabismus. Ophthalmology 96:127–132 reports of familial transmission within the ‘Mobius spec- 71. Brodsky ME (2003) Th ree dimensions of skew deviation. trum of defects’. J Med Genet 28:18–26 Br J Ophthalmol 87:1440–1441 References 75

72. Donahue SP, Lavin PJ, Hamed LM (1999) Tonic ocular tilt size in isolated inferior oblique palsy. J AAPOS 12(6): reaction simulating a superior oblique palsy: diagnostic con- 602–607 fusion with the 3-step test. Arch Ophthalmol 117:347–352 79. Narasimhan A, Tychsen LT, Poukens V, et al (2007) 73. Kono R, Okanobu H, Ohtsuki H, et al (2008) Displacement Horizontal rectus muscle anatomy in naturally and artifi - of the rectus muscle pulleys simulating superior oblique cially strabismic monkeys. Invest Ophthalmol Vis Sci palsy. Jpn J Ophthalmol 52:36–43 48:2576–2588 74. Clark RA, Rosenbaum AL, Demer JL (1999) Magnetic 80. Tychsen L, Wong AM, Burkhalter A (2004) Paucity of hori- resonance imaging aft er surgical transposition defi nes the zontal connections for binocular vision in V1 of naturally anteroposterior location of the rectus muscle pulleys. strabismic macaques: Cytochrome oxidase compartment J AAPOS 3:9–14 specifi city. J Comp Neurol 474:261–275 75. Clark RA, Demer JL (2002) Rectus extraocular muscle pul- 81. Takagi M, Tamargo R, Zee DS (2003) Effects of lesions ley displacement aft er surgical transposition and posterior of the cerebellar oculomotor vermis on eye move- fi xation for treatment of paralytic strabismus. Am ments in primate: binocular control. Prog Brain Res J Ophthalmol 133:119–128 142: 19–33 76. Demer JL (2008) Infl ection in inactive lateral rectus mus- 82. Durig JS, Jen JC, Demer JL (2002) Ocular motility in cle: Evidence suggesting focal mechanical eff ects of con- genetically defi ned autosomal dominant cerebellar ataxia. nective tissues. Invest Ophthalmol Vis Sci 49:4858–4864 Am J Ophthalmol 133:718–721 77. Clark RA, Demer JL (2008) Posterior infl ection of weak- 83. Williams AS, Hoyt CS (1989) Acute comitant esotropia in ened lateral rectus path: Connective tissue factors reduce children with brain tumors. Arch Ophthalmol 107: response to lateral rectus recession. Am J Ophthalmol 376–378 147(1):127–133.e2 84. Jen JC, Chan WM, Bosley TM, et al (2004) Mutations in a 78. Ela-Dalman N, Velez FG, Demer JL, et al (2008) High reso- human ROBO gene disrupt hindbrain axon pathway cross- lution MRI demonstrates reduced inferior oblique muscle ing and morphogenesis. Science 304:1509–1513 Chapter 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives to Understand Ocular Motility Disorders 7 Antje Neugebauer and Julia Fricke

Core Messages ■ Congenital cranial dysinnervation disorders syndromic forms of Duane syndrome and horizon- (CCDDs) are a group of neurodevelopmental distal gaze palsy with progressive scoliosis (HGPPS) eases of the brainstem and the cranial nerves. to be related to mutations in genes that play a role ■ Endogenic or exogenic disturbances lead to a pri- in brainstem and cranial nerve development. mary dysinnervation of structures supplied by ■ By clinical features and theoretic considerations cranial nerves. Motility disturbances and poten- some forms of congenital ptosis, congenital fourth tially structural changes occur. nerve palsy, Möbius syndrome and Marcus Gunn ■ Secondary dysinnervation occurs if fi bers of other jaw winking phenomenon are understood as cranial nerves innervate the primarily misinner- CCDDs. vated structures. Synkinetic movements or ■ Other congenital disturbances of ocular motility cocontractions of antagonists result and may lead with fi brotic features such as congenital Brown to structural changes in the muscles involved. syndrome, congenital monocular elevation palsy ■ Neurogenetic studies proved congenital fi brosis of and vertical retraction syndrome may be discussed the extraocular muscles (CFEOM), isolated and as CCDDs.

cranial nerve development. Th e typical motility patterns 7.1 Congenital Cranial Dysinnervation in these diseases and the muscular anomalies can now be Disorders: Facts About Ocular Motility Disorders explained as changes secondary to incomplete, absent or paradoxical innervation of the eye muscles. Electromyographic, clinicopathologic, neuroradiologic and genetic studies changed the view upon some congenital ocular motor disorders dramatically during the 7.1.1 The Concept of CCDDs: Ocular Motility last decades [1–8]. Disorders as Neurodevelopmental Defects Many of them that were formerly understood as congenital structural anomalies of the extraocular muscles With the term congenital cranial dysinnervation disor- [9] can now be explained as consequent to disorders in ders (CCDDs) coined in 2002 [16] a new entity was estab- brainstem or cranial nerve development. lished that convincingly encompasses diff erent congenital, Neurogenetic studies and amongst them particularly nonprogressive diseases sharing etiopathologic features. those of the workgroup of E. Engle improved our Th e underlying concept postulates a defect in the pre- understanding of classic representatives of congenital eye natal development of the neuronal structures supplying motility disorders such as congenital fi brosis of the innervation of the cranial region. extraocular muscles (CFEOM) and Duane retraction As to the nature of this defect, primary genetic disor- syndrome [2, 3, 6, 10–15, 21, 22]. In familial cases, ders in the neurodevelopmental plan or exogenic infl u- mutations were found in genes that play crucial roles in ences are a possibility. 78 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

So it has to be stressed that although by genetic inves- prominent example for this. Th e hox homeobox cluster tigations in familial cases of congenital cranial dysinner- encoding sequential processes of diff erentiation both in vation single gene defects could be found to be responsible time and space has been studied in the genome of for hereditary forms of CCDDs the mechanism by which Drosophila melanogaster. In mammals related sequences 7 congenital cranial dysinnervations may occur is not nec- that encode diff erent steps in hindbrain diff erentiation essarily genetic. Nevertheless, the proof that mutations in are identifi ed on four chromosomes thus multiplying the genes playing a role in brainstem development are caus- information for single developmental steps [17–19]. ative for the phenotypes of CCDDs was important to Genes for axonal guidance are preserved through the elicit the neurodevelopmental nature of the disorders. species as well and that is why basic research in this fi eld Whether the cause of a single disorder in cranial nerve is helpful to understand disease mechanisms in CCDDs. development is genetic or exogenic, the consequences of A good example is the interaction between slits and lack of innervation of the target muscles are common fea- netrin as proteins expressed in the midline of the nervous tures: the underaction of the non- or underdeveloped system and growing neurons that express receptors that cranial nerve is referred to as primary dysinnervation, interact with them. Generally proteins of the slit group which may lead to secondary fi brotic changes in the tar- act as repellents from the midline and netrin acts as an get muscles. Substitutional innervation of the target mus- attractant. In the hindbrain an intricate interplay between cles by cranial nerve fi bers originally destined for other slits and the receptors of the robo-group and dcc that is a muscles is referred to as secondary dysinnervation, in netrin receptor guides growing axons either away from or these cases paradoxical and sometimes synkinetic and across the midline. Further guidance molecules are the cocontractive motility patterns result. semaphorins and ephrins, which interact with various As CCDDs of ocular motility namely the development receptor complexes [17–20]. of the third, fourth and sixth cranial nerves and the for- By now we have only narrow insight into some of the mation of brainstem structures involved in ocular motor genetically determined interactions in normal cranial control are of interest. development. Future investigations with linkage analysis A brief summary of the steps involved in proper devel- in familial disorders and investigations targeting on can- opment of the brainstem structures supplying ocular didate genes are likely to elucidate the role of further motility may indicate diff erent stages at which hazardous genes in these processes. infl uences can induce specifi c lesions. Hitherto mutations in six genes are identifi ed as causative in CCDDs, more gene loci are mapped. Two genes are involved in the pathologic process in CFEOM [21, 22], 7.1.1.1 Brainstem and Cranial Nerve Development most probably interacting in axon function and nuclear From the fi rst induction of neural tissue in the developing formation, three genes up to now are found mutated in dif- organism to the proper innervation of an extraocular eye ferent subgroups of Duane retraction syndrome [6, 10, 23, muscle by a cranial nerve a lot of consecutive steps have 24]. Th e example of the diff erent mutated genes causing to be taken that depend on the inborn genetic plan for Duane retraction syndrome shows that the interference development and on the conditions in the surroundings with diff erent steps of development may lead to similar of the organism. phenotypes: one gene is a homeobox gene controlling the Major steps are anterior–posterior patterning of the development of one hindbrain segment: one gene is a pre- neural system as well as dorsal–ventral patterning, segmen- sumed transcription factor and one gene seems to regulate tation with formation of brainstem nuclei, axon sprouting axonal outgrowth in cranial nerves. One gene is found and axon guidance requiring neuronal interaction with mutated in a complex disorder of horizontal gaze, termed chemoattractants and chemorepellents that interact with horizontal gaze palsy with progressive scoliosis (HGPPS), axonal receptors and guide the axonal growth cone away this gene encodes for one of the transmembrane receptors from or toward the midline and toward the target muscle. in the slit-robo interaction [15]. Some genes involved in these developmental processes are highly conserved during the development of species. Th at is why insight into the developmental plans of inver- 7.1.1.2 Single Disorders Representing CCDDs tebrates helps us to understand the developmental steps in mammals. Congenital Fibrosis of the Extraocular Th e role of so called homeobox genes that form a Muscles (CFEOM) genomic sequence that is encoding developmental steps CFEOM was described already in 1879 by Heuck [25]. in anterior–posterior patterning and segmentation is a Th is disorder drew the attention of Elizabeth Engle to the 7.1 Congenital Cranial Dysinnervation Disorders: Facts About Ocular Motility Disorders 79 entity of ocular motility disorders [2] and in 2001 it was CFEOM2 is inherited in an autosomal recessive mode; the fi rst congenital eye motility disorder in which a gene features are bilateral ptosis and an exotropia with adduc- relevant in cranial nerve development was identifi ed to tion defi ciency and varying disorders in vertical alignment be mutated in familial cases [21]. and motility. In this entity a lack of innervation both of the Clinically, CFEOM is characterized by gross motility third and the fourth cranial nerves is presumed [2, 29]. disorders and sometimes paradoxical motility [26–28] in Mutations in the gene ARIX/PHOX2A have been eye muscles and in the lid muscle that are supplied by the found in several pedigrees. From animal experiments it third cranial nerve and in some forms by the third and can be derived that ARIX is necessary for proper third fourth cranial nerves (Fig. 7.1). According to clinical and fourth nerve development [21, 29, 30]. traits, three subgroups have been described, and a recent CFEOM3 is an autosomal dominant disorder with review [29] covers these disorders. varying penetrance and varying symptoms including CFEOM1 is an autosomal dominant anomaly charac- unilateral or bilateral ptosis and motility defi ciencies of terized by bilateral ptosis and bilateral elevation defi - the muscles usually supplied by the third nerve. KIF21A ciency of the eyes, both leading to a compensatory has been found mutated in this phenotype but there chin-up head posture. Intraoperatively passive motility is seems to be a heterogeneous genetic background because found to be restricted, and especially the elevation of the linkage analyses in diff erent families also indicate other globe is hindered. Clinicopathologic studies showed genetic loci. Clinical overlap with congenital motility disfi brous changes in the eye muscles that formerly led to the orders classifi ed as vertical retraction syndrome is possi- assumption that the disorder was primarily myogenic. ble [31, 32]. More recent neuropathologic studies revealed abnormalities in the inferior part of the oculomotor nucleus and Duane Retraction Syndrome absence of the superior part of the nerve and hypoplasia Duane retraction syndrome represents the most frequent of the target muscles of this nerve, which are the superior and the most prominent congenital cranial dysinnerva- rectus and the levator palpebrae [14]. With mutations tion disorder (CCDD). In 1905 Alexander Duane pub- found in the gene KIF21A [22] in families with this disor- lished a paper titled “Congenital defi ciency of abduction, der, it could be shown that alterations in a kinesin pro- associated with impairment of adduction, retraction moting axonal transport processes in neurons play an movements, contraction of the palpebral fi ssure and etiopathologic role in CFEOM1. Th us clinic, pathologic oblique movements of the eye” [33]. Th is title still gives and genetic fi ndings are consistent in this disorder with the full description of the main features of the syndrome the notion of a primary defective innervation in the mus- known today as Duane or retraction syndrome (Fig. 7.2). cles usually supplied by the superior part of the third In primary gaze, esotropia is the most common fi nd- nerve, stemming from neurons located in the inferior ing but a considerable number of patients are orthotropic part of the third nerve nucleus. Th e fi brous changes in the and about 20% are exotropic [34]. Many patients adopt a noninnervated muscles can be understood as secondary head posture to maintain binocular single vision. changes due to noninnervation of the muscle fi bers. Although this constellation of ocular motility disorders had been described earlier by others, it was the merit of Alexander Duane to set up a large series of own and published cases, thus accumulating the data of 54 patients. Th e early etiopathologic theories put forward mainly focused on mechanical changes in the horizontal rectus muscles. In 1959, Breinin performed electromyographic examinations in Duane retraction syndrome and found no potential in the lateral rectus muscle on abduction but a response in the lateral rectus on intended adduction [1]. Th us a paradoxical innervation of the lateral rectus was realized. A further milestone were clinicopathologic studies by Hotchkiss and Miller who found absent sixth nerves in Duane retraction syndrome and confi rmed pathologic fi ndings by Mantucci dating from 1946 where a hypoplas- Fig. 7.1 Patient with bilateral congenital fi brosis of the extraocular muscles (CFEOM). Aft er bilateral inferior rectus recession, tic sixth nerve nucleus and absence of the sixth nerve the patient still adopts a 10° chin-up head posture to fi xate due were described. Miller showed that lateral rectus innerva- to ptosis and residual elevation defi ciency tion was taken over by fi bers of the third nerve [4, 7, 8]. 80 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

Fig. 7.2 Patient with Duane ad syndrome in the left eye. Near alignment in primary gaze (b), adduction deﬁ ciency and downward 7 movement on right gaze (a), abduction deﬁ ciency on left b gaze (c). Lateral view of the globe on left gaze (d), retraction of the globe on e right gaze (e) c

Neuroradiologic studies later on also diagnosed hypopla- in HOXA1 were found to be causative [2, 38, 40, 41]. sia of the sixth nerve in Duane syndrome [35–37]. HOXA1 encodes one homeobox gene that is important for In a thorough review De Respinis [34] gives data on hindbrain segmentation. Individuals suff ering from the demographic and epidemiologic features of the disease. Athabascan brainstem dysgenesis syndrome (ABDS), a Duane syndrome is estimated to account for 1–4% of sporadic disorder that beyond the traits of BSAS causes strabismus cases. Pooled data of major studies showed a central hypoventilation, mental retardation and varying predilection of left eyes with 59%; 23% occurred in the accompanying signs including cardiac anomalies and facial right eye and 18% were bilateral cases. Sixty percent of weakness were found to have homozygous HOXA1 the patients were female. mutations. Th e spectrum of associated nonocular fi ndings In patients with isolated Duane anomaly, no abnor- encompasses miswiring syndromes as Marcus Gunn phe- malities in the HOXA1 gene were found [38, 42]. nomenon and crocodile tears, vertebral anomalies as the Th e third gene involved in the genesis of Duane syn- Klippel-Feil anomaly and hearing problems. Syndromes drome is CHN1. It has been found mutated in several encompassing Duane syndrome are Wildervanck or cer- pedigrees with familial Duane syndrome inherited as a vico-oculo-acoustic syndrome with Duane syndrome, dominant trait [23]. Clinically these patients displayed sensorineural deafness and the Klippel-Feil anomaly as not only reduced abduction and the pattern of oft en bilat- traits and Okihiro syndrome that combines Duane syn- eral Duane syndrome but also some abnormalities in the drome with radial ray anomalies. vertically acting eye muscles innervated by the third An induction of Duane syndrome by teratogens is nerve. Th e gene CHN1 encodes a2-Chimaerin, a protein possible; some patients with thalidomide embryopathy that plays a role in the information fl ow induced by eph- suff er from uni- or bilateral Duane syndrome [34, 86]. rin and ephrin-receptor interaction that leads to growth Th e fi rst mutation to be identifi ed as causative for cone changes infl uencing the guidance of a growing axon Duane retraction syndrome was found in patients with [44]. In a chick in ovo model, it could be shown that familial Okihiro syndrome or Duane radial ray syndrome changes comparable with those induced by the gain of (DRRS) [6, 10] in SALL4, a gene that encodes a transcrip- function mutations found in CHN1 lead to incomplete tion factor. Th e molecular mechanisms by which Duane outgrowth of ocular motoneurons [23]. syndrome and radial anomalies are induced are not yet Th e current pathophysiologic concept for Duane syn- clear. In sporadic cases of Duane syndrome up to now no drome putting together clinical, electrophysiologic, clini- mutations in SALL4 were found [39]. copathologic, neuroradiologic and genetic fi ndings looks In the recently described Bosley-Salih-Alorainy syn- upon the disorder as a CCDD in which innervation of the drome (BSAS), bilateral Duane syndrome combines lateral rectus by sixth nerve fi bers is not full or absent and variably with sensorineural deafness, carotid artery third nerve fi bers, mainly those primarily intended for malformations, delayed motor development and some- the medial rectus take over some innervation of the lat- times autistic disorders. Th e syndrome is inherited in an eral rectus. Th us, in primary position the underlying autosomal recessive mode. In diff erent pedigrees, mutations paresis is partly or fully compensated for the lateral rectus 7.1 Congenital Cranial Dysinnervation Disorders: Facts About Ocular Motility Disorders 81 receives nerve impulses of the third nerve, thus keeping a the angle of squint in primary gaze relatively small with regard to the motility defi ciency in abduction. Sometimes even overcompensation with a divergent angle in primary position or synergistic divergence on adduction occurs. Th e most common pattern of motility in Duane syndrome is an abduction defi ciency, accompanied by a slighter adduction defi ciency that results from the lateral rectus cocontracting on intended adduction. Th is cocon- b traction results in retraction of the globe and narrowing of the palpebral fi ssure on adduction.

Horizontal Gaze Palsy with Progressive Scoliosis (HGPPS) A disturbance in the SLIT/ROBO signaling pathway has been found out to be the cause of a complex CCDD that c leads to a horizontal gaze palsy with unaff ected vertical eye movements. In the entity of so-called HGPPS hindbrain anomalies and ocular motor anomalies can be explained by disorders of the pathfi nding of fi bers that normally cross the midline in the hindbrain. Mutations in the ROBO3 gene that encodes a transmembrane receptor molecule that normally seems to promote midline crossing of some hindbrain axons were identifi ed in d patients with HGPPS [15, 45]. Th e neuroanatomic correlates for the typical eye motility pattern (Fig. 7.3) are not fully understood, special neuroradiologic techniques could show that the typical hypoplastic appearance of the hindbrain on conventional NMR goes along with non- crossing fi bers of ascending and descending tracts [46–48]. Neurologic examinations confi rm that atypical e lack of crossing fi bers exists [49]. Th e nature of the progressive scoliosis which means a signifi cant impairment to the patients may be to be neurogenic.

7.1.1.3 Disorders Understood as CCDDs Fig. 7.3 Patient with familial horizontal gaze palsy and progres- Th e pathophysiological concept of CCDDs also helps to sive scoliosis (HGPPS). Patient aft er bilateral medial rectus reces- understand other congenital, nonprogressive disorders sion for esotropia. Fixation in primary gaze with binocular and syndromes in which the proper motor innervation of functions (c). Only slight adduction movements on intended right (b) and left (d) gaze. Unimpaired elevation (a) and depression (e) cranial muscles is lacking, defi cient or substituted. Such syndromes in which the causative mechanism is not yet fully understood encompass disturbances in the third, fourth, sixth and seventh cranial nerves. Congenital synkinetic movements of the lid on jaw Congenital ptosis is a part of the features of CFEOM movements oft en with congenital ptosis are referred to as and in this context proven to be a CCDD. As an isolated Marcus Gunn phenomenon and hint to a paradoxical trait it is in some forms also presumed to represent a innervation in the levator palpebrae by fi bers of the motor minor variant of dysinnervation in the target area of the portion of the fi ft h nerve (Fig. 7.4). Misrouting of sixth third nerve. Familial cases hint to genetic causes and gene and seventh nerve fi bers into the levator palpebrae also loci already have been identifi ed. has been described [28, 47, 50, 51]. One of our patients 82 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

a b

Fig. 7.4 Patient with Marcus Gunn lid synkinesis (a, b). Opening of the right lid on sucking on the paciﬁ er (b)

displays lid opening on intended downgaze on adduc- CCDDs by Traboulsi [52, 53]. Familial cases are described tion, hinting to a possible miswiring of fourth nerve neu- [54, 55] but an associated gene locus is not yet identiﬁ ed. rons in this case (Fig. 7.5). In a study targeting on ARIX as a candidate gene in con- Congenital fourth nerve palsy may represent a CCDD genital trochlear palsy, no mutation was identiﬁ ed yet the with only primary dysinnervation resulting in elevation of authors hint to a high rate of polymorphisms [55]. the eye on adduction and reduced depression on adduc- Synkinetic movements of the superior oblique on mouth tion (Fig. 7.6). Th e disorder was put into the context with opening and swallowing have been described [47].

a b c

d e f

g h i

Fig. 7.5 Patient presumed to have aberrant innervation of the right lid by fourth nerve ﬁ bers. Lid opening on left downgaze (i), slightly widened right palpebral ﬁ ssure in primary gaze position (e), slightly ptotic lid on abduction of the right eye (d, g)

a b

Fig. 7.6 Patient with congenital fourth nerve palsy in the right eye. Normal right gaze (a), elevation on adduction on left gaze (b) 7.2 Congenital Cranial Dysinnervation Disorders: Perspectives to Understand Ocular Motility Disorders 83

Some more descriptions of single synkinetic disorders In 1949, H.W. Brown (1898–1978) at the First Strabismus concerning the sixth nerve and its target muscle such as Symposium in Iowa City gave a lecture on congenital struc- abduction of the globe on mouth opening, upgaze and tural muscle anomalies. In this talk and in the subsequent drinking exist [47]. publication, he discussed congenital motility disorders with A typical combination of mostly bilateral sixth nerve fi brotic features such as retraction syndrome, strabismus and seventh nerve underaction can be observed in Möbius fi xus, vertical retraction syndrome and general fi brosis syn- syndrome. Recent publications hint to the total spectrum drome. Furthermore, under the name of superior oblique of Möbius syndrome that is broader and encompasses tendon sheath syndrome, he introduced a special form of also combinations of horizontal gaze palsies or bilateral congenital elevation defi ciency in this context that since Duane syndrome and facial weakness and presumably then is known as congenital Brown syndrome [9, 58]. lower brainstem disorders such as pharyngeal and tongue We investigate whether there is evidence that more anomalies. But also third nerve anomalies reminding of congenital eye motility disorders than currently listed, CFEOM are described. Furthermore limb anomalies and namely Brown syndrome, Double elevator palsy and ver- problems of motor coordination occur. Th us Möbius syn- tical retraction syndrome represent congenitial cranial drome covers features of a more generalized developmen- dysinnervation disorders. tal brainstem syndrome [56, 57]. Isolated uni- or bilateral facial palsy is described as a familial disorder; gene loci are mapped [16, 53]. 7.2.1 Congenital Ocular Elevation Defi ciencies: A Neurodevelopmental View Summary for the Clinician 7.2.1.1 Brown Syndrome ■ A group of congenital ocular motility disorders are caused by developmental disturbances. Th ese Motility Findings are nonprogressive, incomitant forms of strabis- Brown syndrome is an oculomotor disturbance charac- mus with certain typical motility patterns and terized by an elevation defi ciency on adduction, normal clinical features such as synkinetic movements or near normal elevation on abduction, mild elevation that help to establish the diagnosis. defi ciency in straight upgaze, positive forced duction test ■ Because of the developmental origin some of and no or only slight superior oblique hyper function as these motility disorders occur in syndromatic cardinal features. Sometimes a head posture is adopted, constellations. A thorough general examination hypotropia of the aff ected eye in primary position may is necessary. occur, a relative divergence of the eyes in upgaze may exist and sometimes widening of the lid fi ssure on adduction can be observed [59, 60] (Fig. 7.7). In acquired cases, Brown syndrome results from dam- 7.2 Congenital Cranial Dysinnervation age that hinders the passage of the superior oblique ten- Disorders: Perspectives to Understand don through the trochlea. Th e pathogenesis in congenital Ocular Motility Disorders cases is not completely understood [60–63]. While some of the congenital ocular motility disorders Brown’s initial assumption that a congenital palsy with restrictive features are explained, others are not yet of the inferior oblique leads to secondary changes in the understood. superior oblique tendon sheath was disproven by a b c

d e f

Fig. 7.7 Patient with right-sided Brown syndrome. Minimal hypotropia in primary gaze (e). Slight elevation deﬁ ciency in right upgaze (a), marked elevation deﬁ ciency in left upgaze (c). Slight depression on adduction in left gaze (f) 84 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

electromyography, which showed normal innervation in Up to now CCDDs with secondary dysinnervation of the inferior oblique. Brown subsequently regarded the ocular target muscles by nerve fi bers intended for other disorder to be caused by a structural anomaly in a supe- eye muscles are described for defects in the sixth nerve, rior oblique tendon sheath [59, 64]. Many studies report for the third nerve and for combined defects of the third 7 structural anomalies in the tendon and its surrounding and fourth nerve but not for isolated defects in the fourth tissue. Current textbooks explain Brown syndrome as a nerve. form of restrictive strabismus and suggest varying Misinnervation by fi bers normally intended for the anomalies in the superior oblique muscle or its tendon antagonists of the primary dysinnervated muscles occurs and the trochlea complex including the surrounding tis- in Duane syndrome and oft en keeps the deviation of the sues [60–63]. eyes in primary position remarkably small. Th e notion of Brown syndrome as a misinnervation A misinnervation of a non- or underinnervated supe- syndrome was put forward already in 1969 by Papst and rior oblique muscle by fi bers intended for the inferior Stein who in an electromyographic study demonstrated oblique or the medial rectus would eliminate the eleva- paradoxical innervation of the superior oblique muscle on tion on adduction found in congenital fourth nerve palsy. intended elevation in adduction of the globe. Th e authors Furthermore, the vertical and torsional angles of devia- interpreted this fi nding in analogy to the paradoxical tion in primary position would be kept small by a coin- coinnervation found in Duane retraction syndrome and nervation by fi bers normally running to the inferior postulated a neurodevelopmental origin of the syndrome. oblique muscle. First, because the antagonist of the pri- Other authors confi rmed the results by electromyography, marily paretic superior oblique muscle might receive less so that a total of fi ve cases with electromyographic record- nerve fi bers and second, because its tone now simultane- ing of paradoxical innervation to the superior oblique are ously is antagonized by a tone in the superior oblique. reported by three diff erent investigators [43, 65, 66, De An aberrant innervation in the superior oblique by Decker, personal communication, 2004]. Nevertheless, fi bers intended for the inferior oblique would result in this explanation currently is not widely accepted. One blockage of elevation in adduction by cocontraction of argument put forward against the hypothesis of a para- the two muscles. Th is could be the explanation for the doxical innervation refers to an electromyographic study elevation defi ciency on adduction. Primary dysinnerva- by Catford and Hart [67] who could not fi nd paradoxical tion in some muscular regions and cocontraction of the innervation in patients with Brown syndrome. But the muscle against the action of the inferior oblique could patients examined by Catford and Hart mostly displayed lead to structural changes in the superior oblique and late onset of Brown syndrome and may represent acquired thus explain restriction against elevation in adduction in cases. A second counter-argument points to the common the forced duction test. fi nding of a positive forced duction test under anesthesia A cocontraction of the superior and inferior oblique in congenital Brown syndrome that hints to a mechanical that both have their functional origin anterior to their component rather than to a mere innervational one [62]. insertion could also be claimed to explain widening of the Discussing the question whether a passive restriction of lid fi ssure on adduction. Th is would be an eff ect reverse to the globe under anesthesia on forced duction to elevation the narrowing of the lid fi ssure on adduction by retraction in adduction contradicts the hypothesis of a primary mis- of the globe in Duane syndrome. A paradoxical coinnerva- innervation, one has to consider that a misinnervation tion in the lid due to compensation of a hypoplasia in the could lead to secondary changes in the muscle, tendon, subnucleus of the levator palpebrae could also be possible. trochlea and surrounding connective tissues. In the publi- As well passive forces by a secondarily tight superior cation by Gutowski that defi nes CCDDs it is summarized oblique as active forces by a potential coinnervation of that “dysinnervation may be associated with secondary the superior oblique by fi bers originally destined for the muscle pathology and/or other orbital and bony struc- medial rectus would explain depression of the globe on tural abnormalities” [16]. adduction. Moreover, an overcompensation of the pri- In the light of the understanding of CCDDs, we think mary defect by misrouting of axons intended for the it worthwhile to reconsider the question whether Brown antagonist of the underinnervated muscle could occur syndrome represents a misinnervation disorder. as it is the case in the subset of Duane syndrome with Th e hypothesis is that a primary developmental dysin- exotropia. nervation of the superior oblique muscle as it occurs in Clarke described three cases with a depression on congenital fourth nerve palsy is accompanied by a sec- adduction of the globe that was primarily diagnosed as ondary dysinnervation of the superior oblique by fi bers Brown syndrome but was in this publication presented as of the third nerve. an own entity. In these cases, an innervation of the 7.2 Congenital Cranial Dysinnervation Disorders: Perspectives to Understand Ocular Motility Disorders 85 superior oblique by fi bers primarily destined to the a medial rectus would be possible [68]. At last even a miswiring of fi bers prone to the superior rectus could be discussed. Th is would explain why many patients show also a minor elevation defi ciency in abduction. Th us, the motility fi ndings of Brown syndrome could b be explained by an aberrant innervation of a primarily dysinnervated superior oblique muscle. We analyzed the literature and our own data of 87 patients examined for congenital Brown syndrome in our clinic in the years 1995–2007 for information supporting c or contradicting the hypothesis that the typical features of congenital Brown syndrome result from primary and secondary misinnervation.

Saccadic Eye Movements Barton [69] in a study on vertical saccades described the d eye tracking of vertical saccades in a patient with Brown syndrome. Reproducibly, there occurred a marked and punctuated lateral shift , described as a “horizontal fl ip,” of the globe in the upward saccades and a medial shift in the downward saccades. Under the proposed hypothesis, this would be explained by an additional abductor acting by cocontraction of the superior oblique when the eye comes Fig. 7.8 Patient with right-sided Duane and left -sided Brown syn- into the fi eld of action of the inferior oblique. drome. Right upgaze (a) shows abduction defi ciency in the right eye and elevation defi ciency on adduction on the left side. Right Th e authors compare the fl ip movement of the eye to gaze (b) shows abduction defi ciency in the right eye and widening that in horizontal saccades in Duane syndrome. With the of the palpebral fi ssure in the left eye. Eyes shown in primary gaze onset of cocontraction, a fl ip could occur by the sudden (c). Left gaze (d) shows narrowing of the right lid fi ssure action of the antagonist. misinnervation disorder in Brown syndrome. In these Comorbidity cases a bilateral disturbance of trochlear nerve develop- In the majority of cases Brown syndrome represents an iso- ment could be postulated that in the side with Brown syn- lated disease. Among the diseases reported to accompany drome is answered by a misinnervation or restrictive Brown syndrome interestingly CCDDs such as Duane syn- alteration in the superior oblique and in the other side drome, congenital Ptosis, crocodile tears and Marcus Gunn leads to the symptoms of fourth nerve palsy. phenomenon [70] are prevailing. Contralateral congenital fourth nerve palsy is frequent as well [71–73]. Moreover, Epidemiologic Features colobomata and cardiac malformations are named. Under the hypothesis of a similar etiology, we compared In our 87 patients, three demonstrated additional epidemiologic data for Brown and Duane syndrome Duane syndrome, two ptosis, one incomplete lid clo- because both the fourth and sixth cranial nerves have sure and one Marcus Gunn yaw winking phenomenon. developmentally an origin of rhombomeres which is dif- In 13/87 patients, (14.9%) contralateral fourth nerve ferent from the third nerve [52]. palsy with superior oblique underaction in downgaze was documented. Figure 7.8 shows a patient with Laterality right-sided Duane syndrome and left -sided Brown De Respinis [34] reviewed publications on Duane syn- syndrome. drome and fi gured out side distribution from pooled data Th e coincidence with CCDDs could be caused by of diff erent studies. We pooled the data of ten studies on common pathogenetic mechanisms interfering with Brown syndrome [60, 65, 74–81] and of our own series. brainstem and cranial nerve development. In a total of 11 studies, including 246 patients with con- Th e high incidence of contralateral fourth nerve genital Brown syndrome the right side was aff ected in palsies also is of interest with regard to a potential 53%, the left side in 38% and both sides in 9%. 86 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

In the series on Duane syndrome [34], a total of 835 to have been congenital in this series. Wright summarizes cases were analyzed. the incidence of inheritance by 2% in Brown syndrome and In 59% the left eye was aff ected, in 23% the right eye found 1 of 38 cases, thus 3%, with inheritance in his own and in 18% bilaterality was found. series. Wright hints to eight reports of inheritance in the lit- 7 Assuming that in Duane syndrome the pathophysio- erature with a total of 23 involved patients, he himself add- logic mechanism has a tendency to aff ect rather the left eye, ing another one [60]. Lobefalo [82] reported a family with these data seem contradictory to a common pathogenesis. autosomal distal arthrogryposis multiplex congenita and Th is contradiction resolves because the fi bers of the Brown syndrome; thus, we overlook a total of ten descrip- fourth nerve are crossing and the nucleus of the fourth tions of familial Brown syndrome. nerve lies contralaterally. Th e hypothesis stating a primary Th ree of the reports of familial Brown syndrome brainstem related pathophysiologic mechanism of Brown involve monocygotic twins with mirror images. In Duane syndrome, the data concerning laterality show an interest- syndrome, mirror images in twins are also described. ing parallel between Duane and Brown syndrome. But, although there are as in Brown syndrome far Nevertheless a higher bilateral incidence in Brown more sporadic than familial cases, the amount of heredi- syndrome has to be noticed. tary cases in Duane syndrome with about 10% is greater But if according to the hypothesis congenital Brown than in Brown syndrome. As well in Brown syndrome as syndrome would represent a subgroup of congenital in Duane syndrome, the familial cases are presumed to be fourth nerve palsy in which paradoxical coinnervation mostly inherited by an autosomal dominant transmission occurs, cases with a contralateral fourth nerve palsy [34, 83, 84]. should be understood as bilateral with regard to the A genetic study performed under the assumption that underlying pathology, thus the percentage of bilateral Brown syndrome might be looked upon in the context of cases would increase signifi cantly. the other congenital strabismus syndromes already has been done in a family with familial aff ection [85]. ARIX Sex Distribution was not found to be mutated. But the case reports of the Pooled data of ten and our own studies [60, 65, 74–81] patients should be read carefully for the late onset of symp- encompassing 246 patients showed the aff ection of 55% toms in the teenage years should also let an acquired pathol- females and 45% males. For Duane syndrome de Respinis ogy maybe on the basis of a familial rheumatic disposition [34] found in pooled data of 835 patients, 58% were being taken into consideration. Th us, this paper in our women and 42% were men. Again, an analogy between opinion does not contradict the hypothesis in question. the entities of Brown and Duane syndrome under the Of our 87 patients, 21 patients had a positive family hypothesis of a similar pathophysiologic mechanism history in regard to strabismus or amblyopia (24.1%). could be drawn. Th ree patients (3.4%) had relatives with Brown syndrome: two pairs of brothers, amongst them one pair Incidence of twins with “mirror images” and one parent child Incidence of Brown syndrome is estimated to be 1 per constellation. 430–450 strabismus cases, i.e., 0.22% [60]. Duane syn- One patient’s grandfather was reported to us to be drome occurs in at least 1% of strabismus cases [34]. “unable to move the eyes to the right or left .” We had no Both syndromes are rare but a 4 times greater inci- opportunity to examine the patient but a video of him dence of Duane syndrome remains to be explained. showed a condition that might represent bilateral Duane Stating a failed innervation of the superior oblique mus- syndrome or horizontal gaze palsy. cle by fi bers of the fourth nerve and paradoxical innervation of the superior oblique in Brown syndrome one Potential Induction of the Syndrome would have to add the cases of uni- or bilateral congenital Among the developmental defects caused by thalidomide fourth nerve palsy to fi gure out the incidence of the there are also cranial miswiring syndromes. We investi- underlying pathophysiologic entity of a developmental gated whether in thalidomide embryopathy also Brown fourth nerve disorder. syndrome is described. In 21 patients with thalidomide embryopathy and ocular motility disorders, Miller [86] Heredity describes nine patients with Duane syndrome and two In Brown syndrome, most cases seem to occur spontane- patients with decreased function of the right-sided infe- ously. Of the 126 cases in the 1973 report of Brown [59] 2 are rior oblique; furthermore, patients suff ered from gaze familial, although it cannot be confi rmed whether all 126 paresis, isolated abduction weakness, aberrant lacrimation cases were congenital ones, but at least 100 can be estimated and facial nerve palsy. 7.2 Congenital Cranial Dysinnervation Disorders: Perspectives to Understand Ocular Motility Disorders 87

It could be discussed whether the patients described Intra-and Postoperative Findings with inferior oblique underaction were patients with a Structural changes in the superior oblique tendon in paradoxical coinnervation in fourth nerve hypo- or Brown syndrome have been described by many surgeons aplasia. [79, 92, 103]. In our series, 28 patients underwent oper- Saito in a neurological work-up of the data of 137 ation, in 20 cases the surgical protocol mentions tightness patients with thalidomide embryopathy described three of the tendon, in one case in which a tucking procedure patients with disturbances of the fourth nerve [87]. was performed on the inferior oblique also fi brotic changes in this muscle were reported. Radiologic Findings Surgical results are oft en disappointing as indicated by Imaging studies in Brown syndrome displayed diff erent the multitude of approaches suggested. Surgeons oft en pathologies. Enlargement and irregularities in the tro- recognize a disappointing discrepancy between intraop- chlear complex were shown by Sener et al. Bhola et al. erative fi ndings aft er interventions on the superior oblique examined three patients with Brown syndrome, two of in that passive motility is improved aft er the procedure whom showed hypoplasia on NMR tomography in the but active motility in the postoperative course is still not muscular portion of the superior oblique – a remarkable improved signifi cantly. fi nding with regard to the hypothesis of a primary devel- Papst and Stein in their thorough early discussion of a opmental disorder in the fourth nerve underlying Brown potential misinnervation already hinted to this fi nding as syndrome. an argument for an innervational abnormality in Brown To test the hypothesis of a fourth nerve dysinnerva- syndrome [43, 66, 93]. tion in Brown syndrome, Kolling and coworkers exam- We summarize from our studies that the hypothesis of ined the trochlear nerve with nuclear magnetic resonance Brown syndrome as a neurodevelopmental disorder imaging and presented their results at the 12th meeting should still be pursued to be verifi ed or falsifi ed. of the Bielschowsky society in 2007 [unpublished data]. In two of four patients, the trochlear nerve was found absent on the side of the motility defi ciency a fi nding in favor of the hypothesis. Muscular anomalies were not 7.2.1.2 Congenital Monocular Elevation Defi ciency and Vertical Retraction found in these patients [80, 88]. Syndrome Natural Course in Brown Syndrome While in congenital Brown syndrome an elevation defi - As to the natural course of the disease, reports are incon- ciency of the eye exists if the globe is adducted, in consistent. Whereas Wright states that congenital Brown genital monocular elevation defi ciency or in “double syndrome yields rather stable fi ndings, many authors elevator palsy,” elevation of the globe is hindered in report spontaneous improvement or even resolution adduction as well as in abduction. [89–91]. In most of our patients fi ndings were quite stable An early description of the disorder is given by White but in single cases – for example, at the age of 2 years in in 1942 [94]. one of the twins with mirror image – we saw signifi cant Acquired and congenital cases are reported. spontaneous improvement. Congenital cases are characterized by orthotropia or Th e fi nding of spontaneous resolutions challenges hypotropia in primary position, true ptosis or pseu- the hypothesis of a dysinnervation. But one has to con- doptosis in the majority of cases. In a considerable sider that the hypothesis states secondary fi brotic number of cases restriction of the globe to forced duc- changes. Also under the assumption of a mere mechan- tion into elevation is found. Oft en the lid shows para- ical cause of Brown syndrome, spontaneous improve- doxical movements on yaw movements, i.e., the Marcus ments remain to be explained. Any explanation such as Gunn phenomenon. Furthermore dissociated vertical growth changes of the orbital anatomy or changes in deviation (DVD) is present, sometimes it occurs aft er fi brotic tissues would serve under both assumptions. In operation. Oft en Bell’s phenomenon is preserved the setting of cocontraction, changes in fi brous strands although elevation on following movements, saccades even may be more probable. Furthermore, the postna- and in compensatory eye movements cannot be elicited tal plasticity of the neuromuscular connections with [62] (Fig. 7.9). potential processes of initial polyneuronal innervation Olson and Scott report a series of 31 patients with con- and gradual synapse elimination in the eye muscles is genital monocular elevation defi ciency in which they reg- not well examined especially under the condition of istered pseudptosis in 90%, true ptosis in 64%, chin-up coinnervation [18]. head position in 77%, hypotropia in primary gaze in 97% 88 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

a Further, the fi nding that elevation is hindered in abduction, which means in the fi eld of action of the superior rectus, and in adduction, which means in the fi eld of action of the inferior oblique, led many observers to 7 exclude a nuclear disorder: for the third nerve, the subnucleus for the innervation of the superior rectus lies contralaterally, and for the inferior oblique, it lies ipsilaterally in the mesencephalon. b Remarkably, as in Brown syndrome, which was initially understood as a paresis of the inferior oblique in a case of so-called double elevator palsy, innervation of the inferior oblique was found normal in an electromyographic examination [98]. It was speculated that a longstanding palsy of the superior rectus alone also would impede elevation on c adduction and that an inferior oblique palsy not necessarily is required to produce the typical motility pattern, [62, 94] thus a nuclear origin confi ned to the subnucleus of the superior rectus was not out of discussion. In cases with resistance to forced duction, impairment of Bell’s phenomenon also exists, where sometimes a primarily fi brotic origin is presumed. Th us supranuclear, nuclear, fascicular and muscular d etiologies are discussed for the rare disorder of congenital monocular elevation defi ciency. With the Marcus Gunn phenomenon, ptosis and restriction as accompanying signs some features exist that could be compatible with a neurodevelopmental origin of double elevator palsy. A case with the combination of Duane syndrome and double elevator palsy has been reported [99]. In our series of 23 patients, two showed contralateral fourth nerve palsy. Th ree of our patients showed retraction of the globe on vertical eye movements. Fig. 7.9 Patient with congenital monocular elevation defi ciency in the right eye. Elevation of the right eye hindered in right Th is leads to similarities with vertical retraction syn- upgaze (a), straight upgaze (b) and left upgaze (c). Higher eleva- drome that also had been included by Brown into the tion of the right eye on lid closure, Bell’s phenomenon, (d) than structural anomalies [9]. on elevation (b) Descriptions of vertical retraction syndrome are inconsistent in that some authors describe only anomalies in vertical eye movements with retraction of the globe with a mean of 20 PD, Marcus Gunn jaw winking in 28%, with narrowing of the lid fi ssure; others describe vertical reduced or absent Bell’s phenomenon in 75% and restriction motility disorders with retraction combined with hori- to elevation on forced duction in 42% of those tested [95]. zontal abnormalities that resemble Duane syndrome. In our own series of 23 patients with double elevator Vertical retraction syndrome seems to be even rarer palsy in eight cases Bell’s phenomenon was positive. than congenital monocular elevation defi ciency. The fact that in some cases elevation of the globe is A secondary misinnervation as cause for the retrac- preserved under the conditions of Bell’s phenomenon, tion of the globe on vertical movements would be a pos- DVD or under anesthesia [96] led several authors to sible explanation. conclude that double elevator palsy represents a supra- Th e view upon congenital double elevator palsy and nuclear disorder and seemed to exclude an infranuclear vertical retraction syndrome as neurodevelopmental dis- disorder. Some authors discuss a fascicular lesion orders would require a model that solves the question [62, 94, 97]. why Bell’s phenomenon remains intact in some cases. 7.2 Congenital Cranial Dysinnervation Disorders: Perspectives to Understand Ocular Motility Disorders 89

sequential activation of genes and the building up of 7.2.2 A Model of some Congenital Elevation gradients of mediators for developmental steps. Defi ciencies as Neurodevelopmental Diseases Th e crossing of fi bers in certain segments of the brainstem depends on the integrity of the cascade of interac- Our inquiries into the fi eld of congenital elevation defi - tions between substances mediating attraction to and ciencies lead us to hypothesize that these disorders might repulsion from the midline and their receptors. Th e muta- represent rather a continuum of developmental disorders tions in the ROBO3 gene leading to HGPPS are an exam- than distinct diseases. ple of a locally defi ned failure of midline crossing of Clinically, it is sometimes hard to diff erentiate certain neurons. between a Brown syndrome and a congenital monocular If such a failure occurred in the lower mesencephalic elevation defi ciency. Wright in his review hinted to 70% region, an isolated uni- or bilateral fourth nerve palsy of patients that had been operated and that demon- could result. If fi bers of the third nerve e.g., fi bers intended strated signifi cant elevation defi ciency in abduction for the superior rectus or the inferior oblique would enter [60]. In 76 own examinations of patients with congenital the superior oblique paradoxical innervation could result Brown syndrome, we found 66% to have remarkable in the motility pattern of Brown syndrome (Fig. 7.10). If hindrance of elevation in abduction. We remarked that the defect extended higher to the region of the crossing some patients with typical Brown syndrome display a fi bers of the third nerve, the subnucleus sending fi bers slight ptosis on the aff ected side. Patients with congeni- across the midline that lies next to the fourth nerve and tal monocular elevation defi ciency may display retrac- innervates the levator palpebrae muscle would be aff ected tion on up- or downgaze so that clear diff erentiation and fourth nerve palsy or Brown syndrome accompanied from vertical retraction syndrome may be diffi cult. At by ptosis would result. A substitutional innervation, e.g., last even diff erentiation between a unilateral congenital by fi bers of the motor portion of the fi ft h nerve or of the fi brosis syndrome and these disturbances may be diffi - third nerve would compensate the primary dysinnerva- cult. Th us one might ask for an explanation taking into tion partially but lead to synkinetic movements of the lid account that borders are not clear cut. on jaw movements as Marcus Gunn phenomenon or on In prenatal development segmentation, anterior– downgaze producing a lid lag or on adduction producing posterior and dorso–ventral patterning is achieved by widening of the lid fi ssure.

superior rectus, MIF

superior rectus, SIF Fig. 7.10 Model of N. III- congenital Brown syndrome nucleus as a neurodevelopmental disorder. A schematic levator palpebrae, SIF drawing shows the third and N. IV-nucleus fourth nerve nuclei in the brainstem. A unilateral superior oblique, SIF gradual disturbance exists that mostly aﬀ ects the fourth x nerve nucleus or its crossing neurons. An x indicates N.IV disruption of normal fourth nerve innervation. Dashed lines indicate secondary misinnervation of the superior oblique by third nerve ﬁ bers. Note that this misinnervation does not run topographically in the way N.III- shown. Th e lines just indicate fibers which muscles might share innervation 90 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

Further extension would encompass the subnucleus homogeneous innervation. In consequence of the idea of for the superior rectus. Brown syndrome and ptosis would a dual innervation of the eye muscles, concepts of supra- be accompanied by an elevation defi ciency in abduction, nuclear disorders in general have to be reconsidered. thus completing the image of congenital monocular ele- Th e motoneuron group innervating the MIF of the 7 vation defi ciency. If the superior rectus is innervated by superior rectus is found in the so-called S-group, which in fi bers of its main antagonist, retraction movements as man lies in the cranial part of the nucleus. Th e functional well as depression defi ciency result. role of the MIF fi bers is not yet elucidated but they are Interestingly, recent studies on the functional neuro- presumed to play a role in tonic muscle activity [100, 101]. anatomy of the third nerve nucleus state a dual innerva- One could speculate that MIF neurons play a role in the tion of the eye muscles. So called single innervated muscle mediation of Bell’s phenomenon and further that these fi bers (SIF) and multiple innervated muscle fi bers (MIF) neurons either by their special cytologic features or just by receive input each from a special subset of motoneurons their cranial position are not reached by the pathologic that diff er in their histologic appearance from neurons process hindering midline crossing. Th is would explain innervating SIF fi bers. Th ese are located in distinct why Bell’s phenomenon remains intact in some cases of regions of the third nerve nucleus [100, 101]. monocular elevation defi ciency. Th us the concept of a Such a dual innervation would make it necessary to supranuclear disorder would not be necessary. reconsider the presumption of a fi nal common path in eye Th is model would explain Brown syndrome, congeni- muscle innervation. Th e principle itself as introduced by tal monocular elevation defi ciency and vertical retraction Sherrington referred to the motoneuron as the fi nal path syndrome as disorders of mesencephalic disturbance of [102] and is not in question but it has been adopted in a midline crossing of fourth and third nerve fi bers with way that looked upon the eye muscle as a structure with dysinnervation (Fig. 7.11).

superior rectus, MIF

N.III

superior rectus, SIF N. III- nucleus

N. IV- levator palpebrae, SIF nucleus

N.V- superior oblique, SIF fibers x x x N.IV

N.III- fibers

Fig. 7.11 Model of congenital monocular elevation defi ciency as a neurodevelopmental disorder. A schematic drawing shows the third and fourth nerve nuclei in the brainstem. A unilateral gradual disturbance exists that mostly aff ects the fourth and third nerve nuclei or their crossing neurons. An x indicates disruption of normal fourth nerve innervation and disruption of the crossing fi bers of the third nerve, resulting in primary misinnervation of the superior oblique, superior rectus and levator palpebrae. Dashed lines indicate secondary misinnervation of these muscles by third nerve fi bers originally intended and leading impulses for the medial rectus, inferior oblique and inferior rectus. Note that this misinnervation does not run topographically in the way shown. Th e lines just indicate which muscles might share innervation. Green line indicates multiple innervated muscle fi bers (MIF) for tonic innervation of the superior rectus not aff ected by the lesion References 91

Th e clinical fi ndings seem consistent, future studies 8. Miller NR, Kiel SM, Green WR, et al (1982) Unilateral namely genetic studies in familial cases or on candidate Duane’s retraction syndrome (TypI). Arch Ophthalmol genes will help to test this model. 100(9):1468–1472 9. Brown HW (1950) Congenital structural muscle anomalies. In: Allen JH (ed) Strabismus ophthalmic symposium I. Summary for the Clinician VC Mosby, St. Louis ■ Congenital Brown syndrome, congenital mon- 10. Al-Baradie R, Yamada K, St Hilaire C, et al (2002) Duane ocular elevation defi ciency and vertical retrac- radial ray syndrome (Okihiro syndrome) maps to 20q13 tion syndrome as nonprogressive forms of and results from mutations in SALL4, a new member of the strabismus with fi brotic changes share features SALL family. Am J Hum Genet 71(5):1195–1199 Epub with CCDDs and are found to be accompanied 2002 Oct 22 intraindividually or familial by other CCDDs. 11. Engle EC (2002) Applications of molecular genetics to the ■ Electromyographic, neuroradiologic and surgi- understanding of congenital ocular motility disorders. cal fi ndings support the hypothesis that Brown Ann N Y Acad Sci 956:55–63 syndrome represents a CCDD. 12. Engle EC (2006) Th e genetic basis of complex strabismus. ■ Genetic linkage analysis or examination of candi- Pediatr Res 59(3):343–348 date genes might prove or disprove a model that 13. Engle EC (2007) Genetic basis of congenital strabismus. hypothesizes a continuous spectrum of congenital Arch Ophthalmol 125(2):189–195 neurodevelopmental elevation disorders. 14. Engle EC, Goumnerov BC, McKeown CA, et al (1997) Oculomotor nerve and muscle abnormalities in congenital fi brosis of the extraocular muscles. Ann Neurol 41(3): 314–325 Acknowledgment The data of our own Brown syn- 15. Jen JC, Chan WM, Bosley TM, et al (2004) Mutations in a drome [103] series and the literature on this topic as dis- human ROBO gene disrupt hindbrain axon pathway cross- cussed in chapter 7.2.1.1 were evaluated in cooperation ing and morphogenesis. Science 304(5676):1509–1513. with Gregor Schaaf. Epub 2004 Apr 22 16. Gutowski NJ, Bosley TM, Engle EC (2003) Th e congenital cranial dysinnervation disorders. Neuromuscul Disord 13:573–578 References 17. Hans ten Donkelaar J, Lammens M, Hori A (2006) Clinical 1. Breinin GM (1957) Electromyography: a tool in ocular and neuroembryology. Springer, Berlin neurologic diagnosis. II. Muscle palsy. Arch Ophthalmol 18. Purves D (ed) (2008) Neuroscience, 4th edn. Sinauer 57:165–175 Associates, Massachusetts 2. Engle EC (2007) Oculomotility disorders arising from dis- 19. Sanes DH (ed) (2006) Development of the nervous system, ruptions in brainstem motor neuron development. Arch 2nd edn. Elsevier, Amsterdam Neurol 64(5):633–637 20. Woods CG (2004) Crossing the Midline. Science 304: 3. Engle EC, Leigh RJ (2002) Genes, brainstem development, 1455–1456 and eye movements. Neurology 59(3):304–305 21. Nakano M, Yamada K, Fain J, et al (2001) Homozygous 4. Hotchkiss MG, Miller NR, Clark AW, et al (1980) Bilateral mutations in ARIX(PHOX2A) result in congenital fi brosis of Duane’s retraction syndrome. A clinical-pathologic case the extraocular muscles type 2. Nat Genet 29(3): 315–320 report. Arch Ophtalmol 98:870–874 22. Yamada K, Andrews C, Chan WM, et al (2003) 5. Huber A, Esslen E (1969) Duane’s syndrome; observa- Heterozygous mutations of the kinesin KIF21A in congen- tions on the pathogenesis and etiology of diff erent formes ital fi brosis of the extraocular muscles type 1 (CFEOM1). of the Stilling-Duane-Turk-retraction syndrome. Doc Nat Genet 35(4):318–321. Epub 2003 Nov 2 Ophthalmol 26:619–628 23. Miyake N, Chilton J, Psatha M, et al (2008) Human CHN1 6. Kohlhase J, Heinrich M, Schubert L, Liebers M, et al (2002) mutations hyperactivate alpha2-chimaerin and cause Okihiro syndrome is caused by SALL4 mutations. Hum Duane’s retraction syndrome. Science 321(5890):839–843. Mol Genet 11(23):2979–2987 Epub 2008 Jul 24 7. Miller NR (2005) Strabismus syndromes: the congenital 24. Tischfi eld MA, Bosley TM, Salih MA (2005) Homozygous cranial dysinnervation disorders (CCDDs). In: Taylor D, HOXA1 mutations disrupt human brainstem, inner ear, Hoyt CS (eds) Pediatric ophthalmology and strabismus. cardiovascular and cognitive development. Nat Genet 37 Elsevier, Saunders, Edinburgh, London (10):1035–1037. Epub 2005 Sep 11 92 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

25. Heuck G (1879) Über angeborenen vererbten Beweglich- 41. Bosley TM, Alorainy IA, Salih MA, et al (2008) Th e clinical keitsdefect der Augen. Klin Monatsbl Augenheilkd 17: spectrum of homozygous HOXA1 mutations. Am J Med 253–278 Genet A 146A(10):1235–1240 26. Brodsky MC, Pollock SC, Buckley EG (1989) Neural 42. Holve S, Friedman B, Hoyme HE, et al (2003) Athabascan 7 misdirection in congenital ocular fi brosis syndrome: brainstem dysgenesis syndrome. Am J Med Genet A implications and pathogenesis. Pediatr Ophthalmol 120A(2):169–173 Strabismus 26(4):159–161 43. Stein HJ, Papst W (1969) Elektromyographische Unter- 27. Cibis GW (1984) Congenital familial external ophthal- suchungen zur Pathogenese und Th erapie des Musculus moplegia with co-contraction. Ophthalmic Paediatr Genet obliquus superior-Sehnenscheidensyndroms (Brown- 4:163–167 Syndrom). Ber Zusammenkunft Dtsch Ophthalmol Ges 28. Yamada K, Hunter DG, Andrews C, et al (2005) A novel 69:618–624 KIF21A mutation in a patient with congenital fi brosis of 44. Wegmeyer H, Egea J, Rabe N, et al (2007) EphA4-dependent the extraocular muscles and Marcus Gunn jaw-winking axon guidance is mediated by the RacGAP a2-Chimaerin. phenomenon. Arch Ophthalmol 123(9):1254–1259 Neuron 55:756–767 29. Heidary G, Engle EC, Hunter DG (2008) Congenital fi brosis 45. Chan WM, Traboulsi EI, Arthur B, et al (2006) Horizontal of the extraocular muscles. Semin Ophthalmol 23(1):3–8 gaze palsy with progressive scoliosis can result from com- 30. Yazdani A, Chung DC, Abbaszadegan, et al (2005) A novel pound heterozygous mutations in ROBO3. J Med Genet PHOX2A/ARIX mutation in an Iranian family with con- 43(3):e11 genital fi brosis of extraocular muscles type 2 (CFEOM2). 46. Haller S, Wetzel SG, Lütschg J (2008) Functional MRI, DTI Am J Ophthalmol 136(5):861–865 and neurophysiology in horizontal gaze palsy with pro- 31. Hanisch F, Bau V, Zierz S (2005) Congenital fi brosis of gressive scoliosis. Neuroradiology 50:453–459 extraocular muscles (CFEOM) and other phenotypes of 47. Pieh C, Lagrèze WA (2007) Angeborene Fehlinnerva- congenital cranial dysinnervation syndromes (CCDD). tionssyndrome. Ophthalmologe 104:1083–1096 Nervenarzt 76(4):395–402 48. Sicotte NL, Salamon G, Shattuck DW, et al (2006) Diff usion 32. Yamada K, Chan WM, Andrews C, et al (2004) Identifi cation tensor MRI shows abnormal brainstem crossing fi bers of KIF21A mutations as a rare cause of congenital fi brosis associated with ROBO3 mutations. Neurology 67(3): of the extraocular muscles type 3 (CFEOM3). Invest 519–521 Ophthalmol Vis Sci 45(7):2218–2223 49. Amoiridis G, Tzagournissakis M, Christodoulou P, et al 33. Duane A (1905) Congenital defi ciency of abduction, asso- (2006) Patients with horizontal gaze palsy and progressive ciated with impairment of adduction, retraction move- scoliosis due to ROBO3 E319K mutation have both ments, contraction of the palpebral fi ssure and oblique uncrossed and crossed central nervous system pathways movements of the eye. Arch Ophthalmol 34:139–159 and perform normally on neuropsychological testing. 34. De Respinis PA, Caputo AR, Wagner RS, et al (1993) Duane’s Neurol Neurosurg Psychiatry 77:1047–1053; originally retraction syndrome. Surv Ophthalmol 38: 257–288 published online 13 Jun 2006 35. Kim HJ, Hwang JM (2005) Presence of the abducens nerve 50. Brodsky MC (1991) Platysma-levator synkinesis in con- according to the type of Duane’s retraction syndrome. genital third nerve palsy. Arch Ophthalmol 109:620 Ophthalmology 112(1):109–113 51. Brodsky MC (1998) Hereditary external ophthalmoplegia, 36. Ozkurt H, Basak M, Oral Y, et al (2003) Magnetic reso- synergistic divergence, jaw winking and oculocutaneous nance imaging in Duane’s retraction syndrome. J Pediatr hypopigmentation. Ophthalmology 105:717–725 Ophthalmol Strabismus 40(1):19–22 52. Traboulsi EI (2004) Congenital abnormalities of cranial 37. Parsa CF, Grant E, Dillon WP Jr (1998) Absence of the nerve development: overview, molecular mechanisms, and abducens nerve in Duane syndrome verifi ed by magnetic further evidence of heterogeneity and complexity of syn- resonance imaging. Am J Ophthalmol 125(3):399–401 dromes with congenital limitation of eye movements. 38. Tischfi eld MA, Chan WM, Grunert JF, et al (2006) HOXA1 Trans Am Ophthalmol Soc 102:373–389 mutations are not a common cause of Duane anomaly. Am 53. Traboulsi EI (2007) Congenital cranial dysinnervation dis- J Med Genet A 140(8):900–902 orders and more. J AAPOS 11(3):215–217 39. Wabbels BK, Lorenz B, Kohlhase J (2004) No evidence of 54. Astle WF, Rosenbaum AL (1985) Familial congenital SALL4-mutations in isolated sporadic duane retraction fourth cranial nerve palsy. Arch Ophthalmol 103:532–535 “syndrome” (DURS). Am J Med Genet A 131(2):216–218 55. Jiang Y, Matsuo T, Fuliwara H, et al (2005) ARIX and 40. Bosley TM, Salih MA, Alorainy IA, et al (2007) Clinical PHOX2B Polymorphisms in patients with congenital supe- characterization of the HOXA1 syndrome BSAS variant. rior oblique muscle palsy. Acta Med Okayama 59(2): Neurology 69(12):1245–1253 55–62 References 93

56. De Souza-Dias CR, Goldchmitt M (2007) Further consider- 75. Crawford JS, Orton RB, Labow-Daily L (1980) Late results ations about the ophthalmic features of the Möbius sequence, of superior oblique muscle tenotomy in true Brown’s syn- with data of 28 cases. Arq Bras Oft almol 70(3): 451–457 drome. Am J Ophthalmol 89:824–829 57. Verzijl HT, van der Zwaag B, Cruysberg JR, et al (2003) 76. Eustis HS, O’Reilly C, Crawford JS (1987) Management of Möbius syndrome redeﬁ ned: a syndrome of rhomben- superior oblique palsy aft er surgery for true Brown’s syn- cephalic maldevelopment. Neurology 61(3):327–333 drome. J Pediatr Ophthalmol Strabismus 24:10–16 58. Von Noorden GK (2002) Th e history of strabismology. 77. Hadjadj E, Conrath J, Ridings B, et al (1998) Syndrome de Wayenborgh, Ostende Brown: actualités. J Fr Optalmol 21:276–282 59. Brown HW (1973) True and simulated superior oblique 78. Maggi R, Maggi C (2002) Tendon surgery in Brown’s syn- tendon sheath syndromes. Doc Ophthalmol 34:123–136 drome. J Pediatr Opthalmol Strabismus 39:33–38 60. Wright KW (1999) Brown’s syndrome: diagnosis and man- 79. Parks MM, Eustis HS (1987) Simultaneous superior agement. Trans Am Ophthalmol Soc 47:1023–1107 oblique tenotomy and inferior oblique recession in Brown’s 61. Esser J, Mühlendyck H (2004) Jaensch-Brown-Syndrom. syndrome. Ophthalmology 94:1043–1048 In: Kaufmann H (ed) Strabismus. Georg Th ieme, Stuttgart, 80. Sener EC, Özkan SB, Aribal ME, et al (1996) Evaluation of New York congenital Brown’s syndrome with magnetic resonance 62. Von Noorden GK, Campos EC (2002) Binocular vision imaging. Eye 10:492–496 and ocular motility, 6th edn. Mosby, St. Louis 81. Von Noorden GK, Olivier P (1982) Superior oblique tenec- 63. Wright KW (ed) (2003) Pediatric ophthalmology and stra- tomy in Brown’s syndrome. Ophthalmology 89:303–309 bismus, 2nd edn. Springer, New York 82. Lobefalo L, Mancini AT, Petitti MT, et al. (1999) A family with 64. D’Esposito M, Cotticelli L, Caccia-Perugini G, et al (1974) autosomal dominant distal arthrogryposis multiplex congen- La Pseudoparalisis dell’oblique inferiore. Acta Neurol ita and Brown syndrome. Ophthalmic Genet 20(4): 233–241 (Napoli) 29(6):625–658 83. Mc Kusick VA (1990). Mendelian inheritance in Man, 9th 65. Feric-Seiwerth F, Celic M (1972) Contribution to the edn. John Hopkins Univ, Baltimore knowledge of the superior oblique tendon sheath syn- 84. Paul OT, Hardage LK (1994) Th e heritability of strabismus. drome. In: Mein J et al (ed) Orthoptics – proceedings of the Ophthalmic Genet 15(1):1–18 second international orthoptic congress, Amsterdam 1971. 85. Iannaccone A, McIntosh N, Ciccarelli ML (2002) Familial Excerpta Medica, Amsterdam, pp 354–359 unilateral Brown syndrome. Ophthalmic Genet 23(3): 66. Papst W, Stein HJ (1969) Zur Ätiologie des Musculus- 175–184 obliquus-superior-Sehnenscheidensyndroms. Klin Monatsbl 86. Miller MT (1991) Th alidomide embryopathy: a model for Augenheilkd 154:506–518 the study of congenital incomitant horizontal strabismus. 67. Catford GV, Hart JCD (1971) Superior oblique tendon Trans Am Ophthalmol Soc 89:623–674 sheath syndrome. An electromyographical study. Brit J 87. Kida M (ed) (1987) Th alidomide embryopathy in Japan. Ophthalmol 55:155–160 Kodansha, Tokyo 68. Clarke WN, Noël LP (1985) Depression in adduction syn- 88. Bhola R, Rosenbaum AL, Ortube MC, et al (2005) High- drome. Can J Ophthalmol 20:23–28 resolution magnetic imaging demonstrates varied anatomic 69. Barton JJ, Intriligator JM (2001) Vertical saccades in supe- abnormalities in Brown syndrome. J AAPOS 9(5): 438–448 rior oblique palsy and Brown’s syndrome. J Neuroophthalmol 89. Capasso L, Torre A, Gagliardi V (2001) Spontaneous 21:250–255 resolution of congenital bilateral Brown’s Syndrome. 70. Wilson ME, Eustis HS, Parks MM (1989) Brown’s syn- Ophthalmologica 215:372–375 drome. Surv Ophthalmol 34:153–172 90. Gregersen E, Rindziunski E (1993) Brown’s syndrome. 71. Bhola R, Sharma P, Saxena R, et al (2004) Magnetic reso- Acta Ophthalmol 71:371–376 nance imaging of an unusual case of Brown’s Syndrome with 91. Kaban TJ, Smith K, Orton RB, et al (1993) Natural history contralteral superior oblique palsy. J AAPO 8(2): 196–197 of presumed congenital Brown syndrome. Arch 72. Castanera de Molina A, Munoz GL (1991) Brown syndrome Ophthalmol 111:943–946 associated with contralateral superior oblique palsy: a case 92. Mühlendyck H (1996) Jaensch-Brown-Syndrom – Ursache report. J Pediatr Ophthalmol Strabismus 28: 310–313 und operatives Vorgehen. Klin Monatsbl Augenheilkd 73. Clarke WN, Noël LP (1993) Brown’s syndrome with con- 208:37–47 tralateral inferior oblique overaction: a possible mecha- 93. Crawford JS (1976) Surgical treatment of true Brown’s syn- nism. Can J Ophthalmol 28(5):213–216 drome. Am J Ophthalmol 81:289–296 74. Berk AT, Erkan D, Sener C, et al (1994) Congenital Brown’s 94. White JW (1942) Paralysis of the superior rectus and infe- syndrome: clinical and surgical approach. Eur J Ophthalmol rior oblique muscles in the same eye. Arch Ophthalmol 4:138–143 27:366–371 94 7 Congenital Cranial Dysinnervation Disorders: Facts and Perspectives

95. Olson RJ, Scott WE (1998) Dissociative phenomena in con- 100. Büttner-Ennever JA (2006) Th e extraocular motor nuclei: genital monocular elevation deﬁ ciency. J AAPOS 2:72–8 organization and functional neuroanatomy. In: Büttner- 96. Mims JL 3rd (2005) “Double elevator palsy” eye supraducts Ennever JA (ed) Neuroanatomy of the oculomotor system. during stage II general anesthesia supporting hypothesis of Elsevier, Amsterdam 7 (supra)nuclear etiology. Binocul Vis Strabismus Q 20(4): 101. Horn AK, Eberhorn A, Härtig W, et al (2008) 199–204 Perioculomotor cell groups in monkey and man deﬁ ned by 97. Leigh RJ, Zee DS (2006) Th e neurology of eye movements, their histochemical and functional properties: reappraisal 4th edn. Oxford University, Oxford, New York of the Edinger-Westphal nucleus. J Comp Neurol 507(3): 98. Bell J A, Fielder A, Viney S (1990) Congenital double ele- 1317–1335 vator palsy in identical twins. J Clin Neuro-ophthalmol 102. Sherrington CS (1979) Selected writings of Sir Charles 10(1):32–34 Sherrington. In: Denny-Brown D (ed) Oxford University, 99. Verma MJ, Faridi MM (1992) Ocular motility disturbances Oxford (Duane retraction syndrome and double elevator palsy) with 103. Parks MM, Brown M (1975) Superior oblique tendon congenital heart disease, a rare association with Goldenhar sheath syndrome of Brown. Am J Ophthalmol 79(1): syndrome–a case report. Indian J Ophthalmol 40(2):61–62 82–86 Chapter 8 The Value of Screening for Amblyopia Revisited 8 Jill Carlton and Carolyn Czoski-Murray

Core Messages ■ Vision screening for children may be considered occurs in weeks 4–12. In some cases, further in terms of detection of amblyopia, strabismus, amblyopia therapy may not be required. and/or refractive error. Variations exist within ■ Children who undergo amblyopia therapy at an and between countries regarding vision screening early age have been found to respond more for children in terms of program content, referral quickly to occlusion than older children, and criteria, and personnel. Recommendations state require less occlusion in total. Th ere is evidence pre-school vision screening programs be con- to suggest that successful treatment of children ducted by orthoptists or by professionals trained aged over 7 years can be achieved in cases of and supported by orthoptists. anisometropic, strabismic, and mixed etiology ■ Th e justifi cations of vision screening for children amblyopia. include an increased risk of blindness to the ■ Atropine has been found to be as eff ective as healthy eye as a result of injury or disease in adults patching in the treatment of both moderate and with amblyopia. An increased risk of blindness is severe amblyopia. present as the non-amblyopic eye of an amblyope ■ Recurrence of amblyopia may occur following may become diseased or injured. treatment, with reported rates of 7–27%. Factors ■ A recent report found that screening for amblyo- infl uencing recurrence include age of the child at pia could not be considered as cost-eff ective, but cessation of treatment, VA at the time of cessation acknowledged that much uncertainty exists sur- of treatment, and the type of amblyopia that is rounding the short- and long-term implications present. of the condition(s). Further research is needed to ■ Reported health-related quality of life (HRQoL) provide such evidence. implications of amblyopia include the impact of ■ Treatment of amblyopia associated with refrac- the condition upon stereoacuity; fi ne motor skills; tive error should incorporate a period of observa- reading speed; and interpersonal relationships. tion with glasses-wear alone to allow for ■ Th e reported HRQoL implications of strabismus “refractive adaptation” (also known as “optical are related to physical appearance, particularly treatment of amblyopia”). Improvements in visual upon self-image and interpersonal relationships. acuity (VA) can occur up to and beyond 20 weeks Surgical correction of strabismus has been aft er glasses are prescribed. Most improvement reported to improve HRQoL.

whether the study sample was taken from a clinical cohort 8.1 Amblyopia (where a greater prevalence would be expected), or a pop- Amblyopia is a sensory anomaly defi ned as defective uni- ulation-based study. However, the most important factor lateral or bilateral visual acuity (VA). Th ere are a number that can account for the diff erences in the reported preva- of classifi cations of amblyopia based on the etiological lence rates is that of amblyopia defi nition. Over the recent cause(s). Th e reported prevalence of amblyopia varies years, a defi nition of amblyopia based upon a diff erence widely, from 1–5%. Diff erences in prevalence can be in VA of two or more Snellen or logMAR lines between attributed to the population studied (e.g. ethnicity), and eyes has been adopted. However, there is no universally 96 8 The Value of Screening for Amblyopia Revisited

accepted defi nition of amblyopia in terms of VA defi cit. 8.2.1 Screening for Amblyopia, Studies that report on amblyopia prevalence, diagnosis, Strabismus, and/or Refractive Errors and/or treatment must be interpreted carefully, and oft en cannot be directly compared. Nonetheless, amblyopia is Screening for amblyopia, strabismus, and/or refractive 8 considered to be a common condition which occurs in errors has long been an emotive and contentious issue. childhood, and if left untreated, will remain present Diff erences in health care provision from one country to throughout adult life. Th is chapter will explore what is another can make it diffi cult to draw inferences on the meant by screening; detection of amblyopia and strabis- possible benefi ts and risks associated with the implemen- mus through screening programs; amblyopia treatment; tation or withdrawal of such programs. For example, dif- and consequences of amblyopia and its treatment (both ferences exist between the UK and the United States of in the long and short term). America (USA). Within the UK, vision screening of children was developed as part of the child health surveillance programs established during the 1960s and 1970s. Th e appropriateness of such programs was called into 8.2 What Is Screening? question following a systematic review of their eff ective- Th e purpose of screening is to identify persons as being ness [2]. In 2003, the Health For All Children Report at greater or lesser risk of developing, or having, a par- (also known as Hall 4) recommended changes in the way ticular condition. Th e United Kingdom (UK) National children are monitored and referred for suspected ambly- Screening Committee (NSC) defi ned screening as “a opia and strabismus [3], and the Child Health Promotion public health service in which members of a defi ned Program (CHPP) recommended all children to be population, who do not necessarily perceive that they screened for visual impairment between 4 and 5 years of are at risk of, or are already aff ected by, a disease or its age by an orthoptist-led service [4]. Th is recommenda- complications, are asked a question or off ered a test to tion has been adopted regionally in the UK, although not identify those individuals who are more likely to be universally. helped than harmed by further tests or treatment to Within the USA, there are also widespread diff er- reduce the risk of a disease or its complications” [1]. ences regarding pre-school vision screening guidelines, Th ere are recognized criteria for screening relating to policies, and procedures. Recommendations from the the condition itself, diagnosis, treatment, and cost. Th ese American Academy of Ophthalmology (AAO), American are summarized in Table 8.1. Association for Pediatric Ophthalmology and Strabismus

Table 8.1. Summary of criteria for screening [72]

Category Criteria

Condition Th e condition should be an important health problem, whose epidemiology and natural history are understood. Th ere should be a recognizable risk factor or early symptomatic stage Diagnosis Th ere should be a simple, safe, precise, and validated screening test which is acceptable to the population. Th ere should be an agreed policy on further investigation of individuals with a positive test result Treatment Th ere should be an eff ective treatment or intervention for those identifi ed as having the disease or condition, with evidence of early treatment leading to better outcome than late treatment. Th ere should be agreed evidence-based policies regarding which individuals should be off ered treatment Program Th ere should be evidence from high-quality randomized controlled trials (RCTs) that the screening program is eff ective in reducing mortality or morbidity. Th ere should be evidence that the complete screening program (including the test, diagnostic procedures, and treatment) is clinically, socially, and ethically acceptable. Th e benefi t of the program should outweigh the physical and psychological harm. Th e cost of the program should be economically balanced in relation to expenditure on medical care as a whole (i.e. value for money) 8.2 What Is Screening? 97

(AAPOS), and the American Academy of Pediatrics of the strabismus would be suggestive that amblyopia is (AAP) are that vision screening should be performed on likely to develop within the critical period of vision children between the ages of 3 and 3 ½ years [5]. Despite development. the existence of such recommendations, current practice within the USA is totally non-standardized, with much variability by state and locality. Th is was highlighted by 8.2.1.3 Screening for Refractive Error Ciner et al. [6], who recommended that specifi c components of a pre-school vision screening program ought to Screening for refractive error alone is not commonplace. be considered, including the tests to be conducted, Th e justifi cation would be that the presence of signifi - parental education on the condition, and recording and cant refractive error may impact upon educational prog- referral criteria. ress and daily living. Th e existence of unequal refractive Over recent years, there has been a call to make any error (anisometropia) could be deemed an amblyogenic recommendations for vision screening for children more risk factor. Indeed, the correction of any clinically sig- evidenced-based, and advances in the literature regarding nifi cant refractive error during the critical period of screening test accuracy and treatment of amblyopia will vision development supports the notion of pre-school only serve to facilitate this. However, the implementation vision screening. of any recommendations is oft en driven by political rather than clinical factors. 8.2.1.4 Screening for Other Ocular Conditions Any form of pre-school vision screening is likely to result 8.2.1.1 Screening for Amblyopia in detection of other ocular conditions. Th ese may include Th e purpose of pre-school vision screening for amblyopia ocular pathologies such as cataract or retinoblastoma; or is to detect children with unilateral or bilateral amblyo- may be related to motility, such as Duane’s or Brown’s pia. Accurate detection of amblyopia is primarily achieved syndrome. Whilst such conditions are of great clinical through VA testing. Th e value of conducting other tests importance, not least because of their association with for the purpose of screening for amblyopia alone is mini- systemic health problems, the justifi cation of screening mal; some would argue additional tests could be included for detection of these conditions alone cannot be justi- in the screening program to detect amblyogenic factors fi ed. To screen for such conditions in isolation is neither (e.g. strabismus or refractive error). practical nor appropriate. Th e economic benefi t of adding such conditions to a screening program for amblyopia and/or strabismus is negligible. 8.2.1.2 Screening for Strabismus Th e purpose or value for pre-school vision screening for 8.2.2 Diff erence Between a Screening strabismus alone could be questioned. It may be argued and Diagnostic Test that large, cosmetically apparent strabismus would be observed by parents or guardians and/or health care Th ere is diff erence between a screening test and a diag- practitioners. Once noted, appropriate referral to an nostic test. As the name implies, a screening test is used ophthalmologist would be initiated. Th erefore, the jus- to identify and eliminate those with a given problem(s); tifi cation of pre-school vision screening for large-angled there is no requirement for it to quantify the extent of strabismus may not be valid. Th e detection of small- any defi cit or problem, or indeed for it to provide any angle strabismus, however, is not as easy and requires information for diagnosis. A diagnostic test provides expert testing from orthoptists and ophthalmologists. information that can be used to help make a clinical Th e value of such detection remains under debate. If the diagnosis, and/or infl uence the management plan of the strabismus is so small that it is not cosmetically obvious, condition. A diagnostic test oft en quantifi es the extent or then it is unlikely that surgical treatment for the condi- severity of the condition. For example, photoscreening is tion would be undertaken. To that end, the value of used to detect refractive error (screening test); however, screening may be questioned. An argument for screen- the results would not be used to diagnose the extent of ing could be that the presence of a small-angle strabis- the refractive error present or indeed for the prescription mus is an amblyogenic factor: amblyopia may not be of glasses. Th is would be achieved through refraction present at the time of screening; however, the existence (diagnostic test). 98 8 The Value of Screening for Amblyopia Revisited

amblyopia had almost three times the risk of visual 8.2.3 Justifi cation for Screening for Amblyopia and/or Strabismus impairment in their better-seeing eye compared with people without amblyopia. Th e justifi cation of pre-school vision screening for ambly- More recently, Van Leeuwen et al. [10] examined 8 opia and/or strabismus remains a controversial issue. the excess risk of bilateral visual impairment among Referring to the NSC criteria of screening, the condition individuals with amblyopia as part of the Rotterdam to be screened should be an important clinical condition. study (a population-based prospective cohort study of Th e evidence relating to the condition’s importance and the frequency and determinants of common cardiovas- impact relate primarily to the consequence of amblyopia cular, locomotor, neurological, and ophthalmological and/or strabismus in the short or long term. It has been diseases). They found that the estimated lifetime risk recognized that there is a detrimental eff ect of having of bilateral visual impairment is almost doubled in reduced vision in one eye (as is the case with unilateral those who also have a diagnosis of amblyopia. The amblyopia). Brown et al. [7] stated that in the presence of authors reported that the number of individuals needed ocular disease, yet good VA in both eyes, subjects reported to treat to prevent one case of binocular visual impair- to have a higher HRQoL than those with good VA in only ment is 12.5. one eye. When vision loss in the non-amblyopic eye in the One of the arguments regarding the consequence of presence of amblyopia does occur (through injury or dis- amblyopia refers to the risk of blindness to the healthy ease), the eff ect on the individual is oft en devastating. eye as a result of injury or disease. Rahi et al. [8] reported Th ere have been reported cases of plasticity in the visual on the fi ndings of the British Ophthalmological system, even in adulthood, whereby improvements in VA Surveillance Unit (BOSU), a national surveillance in the amblyopic eye have been observed [11]. scheme for the study of rare ophthalmological disorders Another argument for the notion of pre-school or events. Over a 2-year period, the number of indi- vision screening for amblyopia and/or strabismus is the viduals with unilateral amblyopia with a newly acquired impact of having either condition on quality of life. Th is loss of vision in the non-amblyopic eye was recorded. will be examined in more detail towards the end of the Th e authors were able to report on the total population chapter. lifetime risk and annual rate of permanent visual impairment or blindness attributable to loss of vision in the non-amblyopic eye. In addition, the projected lifetime risk and annual rate of permanent visual impair- 8.2.4 Recent Reports Examining Pre-School Vision Screening ment or blindness attributable to loss of vision in the non-amblyopic eye in individuals with amblyopia were Th e scarcity of evidence that would allow decision makers reported. It was found that the lifetime risk of visual in the UK NHS to fund screening programs with confi - impairment increased substantially from the age of 15 dence that it is an effi cient use of limited health care to 64 years and by 95 years of age (incidence per 100,000 resources has made screening for amblyopia problematic. total UK population, 5.67 [4.33–7.01 CI] compared To be cost-eff ective, a program has to demonstrate that it with 32.98, [29.06–36.89 CI]). Th is can be attributed to is fi rst clinically eff ective. Issues of how disinvestment in the increased prevalence of other ocular disorders that existing technologies or health care programs is carried occur with increasing age (such as cataract and age- out is becoming increasingly important in the UK health related macular degeneration). Th e authors stated that care setting, as new evidence-based technologies are man- every year as a result of disease aff ecting the non- dated by the National Institute for Health and Clinical amblyopic eye, at least 185 people in the UK with uni- Excellence (NICE). Decisions concerning which programs lateral amblyopia have vision loss to a level that is can continue to be funded from the health care budgets associated with detriment to quality of life. It is possible that are under increasing pressure due to the mandated that the incidence rates are greater than this, with only programs from NICE are being made in local areas. Th e the minimum estimates of the risk of visual impairment problems associated with older established programs aft er disease in the non-amblyopic eye being reported. relate mainly to the reality that oft en these were imple- Th e authors stated that the lifetime risk of serious vision mented many years ago when evidence was limited, or loss for an individual with amblyopia was substantial they were never subject to the level of scrutiny that is cur- and in the region of 1.2–3.3%. Th is was supported by rently expected for any new technology or program. Th e Chua and Mitchell [9], who found that people with recent review of screening for amblyopia is one such area. 8.2 What Is Screening? 99

In 2008, the Health Technology Assessment report on outcomes for removing the amblyogenic risk were con- pre-school vision screening was updated, examining both sidered to be between 0 and 30%. the clinical and cost eff ectiveness of screening programs Carlton et al. [12] reported that the available evidence for amblyopia and strabismus in children up to the ages of did not support the screening program for amblyopia and 4–5 years [12]. amblyogenic factors. Economic evaluation showed that A systematic review of the literature examining the screening for amblyopia and strabismus in children could clinical and cost eff ectiveness of screening children for not be considered as a cost-eff ective use of resources. amblyopia and strabismus before the age of 5 years was Analysis of cost eff ectiveness using the available research undertaken. Cost eff ectiveness and expected value of per- data found that screening was not cost-eff ective at cur- fect information (EVPI) modeling was reported. EVPI rently accepted quality adjusted life years (QALY) values. modeling is used in cost-eff ectiveness analysis to attempt (QALYs are used in cost-utility studies, and consider both to establish the benefi ts of undertaking research that the duration of health states and their impact on HRQoL would reduce the costs of uncertainty. Th e cost of uncer- [13]). However, the lack of evidence highlighted a need tainty in this case is that the wrong disinvestment deci- for further research on the impact of amblyopia and sion could be made. amblyogenic factors in the long-term. Th e lack of evi- Following a review of the literature, a natural history dence surrounding the long-term impact of amblyopia model was constructed which described the incidence increased the level of uncertainty in the model. By mak- and progression of amblyopia up to the age of 7 years. As ing a number of assumptions on utility loss (i.e. the is customary, a separate model which extrapolated the impact on quality of life), the model demonstrated that costs and eff ects of amblyopia over an individual’s remain- screening could become highly cost-eff ective. EVPI mod- ing lifetime was also constructed. Th ese models were eling showed that the value of eliminating uncertainty incorporated into a separate screening model that repre- ranges between £17,000 to over £100,000 per QALY. In sented the potential impact of treatment. Th e expected other words, the impact of amblyopia upon a person’s health outcome for the individual was defi ned as the quality of life (in the short or long term) is still unknown, expected number of cases remaining in a population of and guesstimates of such impact lead only to more 7-year-olds, that is, those children for whom treatment uncertainty. was either unsuccessful or who had failed to be detected. Th ese fi ndings may not provide the ideal result for A post-screening model was constructed to estimate decision makers, as the answers are not clear cut. Cost the long-term eff ects of childhood amblyopia on a cohort eff ectiveness alone should not be the deciding factor in of individuals who would have bilateral or unilateral the provision of pre-school vision screening. For exam- vision loss over a 93-year time horizon. Th e costs associ- ple, the issue of equity may also need to be considered. ated with the screening program and the benefi ts Th is is particularly relevant in communities where there (expressed as utility weights) were applied to both vision may be a greater prevalence of amblyopia or strabismus loss across the model’s time horizon, which allowed us to which could not be detected or acted upon by parental give the estimated costs, and to the consequences of observation alone. Th e fi gures reported earlier, linking amblyopia. the cost per QALY, are those which are applied to new Th e model population was informed by the literature technologies. Th e QALY threshold for disinvestment is reviews. It was identifi ed during the data extraction pro- undefi ned at present. cess that there was a signifi cant lack of quantitative data Th e German Institute for Quality and Effi ciency in available which could be used in the model. Th is prob- Healthcare (IQWIG) is an independent scientifi c institute lem was addressed by having a pragmatic approach to that investigates the benefi ts and harms of medical inter- estimate the transitions in the model for which amblyo- ventions. In producing reports on the assessment of an genic factors translated into a number of VA states. A intervention (such as screening), IQWIG adheres to strict number of experts, who were able to confi rm or reject inclusion and exclusion criteria in the reviewing of exist- the plausibility of the assumptions that were made, were ing literature surrounding the given subject. In 2008, consulted. It was not possible to use any empirical data IQWIG assessed the benefi ts of screening for visual which could have informed the eff ectiveness of treat- impairment in children up to the age of 6 years [14]. Th ey ment for amblyogenic factors. It was assumed that by concluded that “no robust conclusions” could be directly removing the risk factor for refractive error, the out- inferred from the studies identifi ed in their review. To come would be 100% eff ective. Strabismus treatment that end, the notion of pre-school vision screening could is acknowledged to be less successful; therefore, the neither be supported nor rejected. 100 8 The Value of Screening for Amblyopia Revisited

condition in a population who are correctly identifi ed by Summary for the Clinician a screening test. Specifi city is the proportion of individu- ■ Th e purpose of screening is to identify persons als free of the target condition in a population who are as being at greater or lesser risk of developing, or correctly identifi ed by a screening test. Positive predictive 8 having a particular condition. Screening should values describe the proportion of individuals with a posi- be considered in terms of the condition, diagno- tive result who have a target condition; and negative pre- sis, treatment, and the screening program itself. dictive value is the proportion of individuals who test ■ Vision screening for children may be considered negative and who do not have a target condition. in terms of detection of amblyopia, strabismus, and/or refractive error. Variations exist within and between countries regarding vision screen- 8.3.1 Vision Tests ing for children in terms of program content, referral criteria, and personnel. Th e use of crowded logMAR acuity is the gold-standard ■ Th e justifi cations of vision screening for children VA measure in adults both within clinical and research include an increased risk of blindness to the settings. Th is is also becoming the case with VA mea- healthy eye as a result of injury or disease in surement in children. Steps have been made to identify adults with amblyopia. normative values of pediatric VA using diff erent vision ■ An increased risk of blindness is present, as the tests, protocols of testing, and repeatability of testing non-amblyopic eye of an amblyope may become [15–19]. Th e preference as to which vision test that is to diseased or injured. be included in a screening program is not always clear. ■ Recent reports indicate that further evidence is Oft en a number of vision tests may be included within required to support the notion of pre-school the one screening program to incorporate factors such vision screening despite seminal research exam- as a child’s comprehension and ability to perform a test. ining diagnosis, treatment, and consequence of It is outside the scope of this chapter to report upon the amblyopia, strabismus, and/or refractive error. relative sensitivity and specifi city of each vision test. However, it should be noted that the cut-off points used for referral within a screening program should be directly related to the specifi c vision tests used within that screening program. In other words, it should not be 8.3 Screening Tests for Amblyopia, generic, with an arbitrary referral point (such as 0.2 log- Strabismus, and/or Refractive Error MAR or worse). A VA level that is achieved using one Th e accurate detection of amblyopia, strabismus, and/or vision test may be diff erent from that achieved using an refractive error undoubtedly forms a critical factor in the alternative vision test. Th e referral criteria should be reported success of any pre-school vision screening pro- stipulated for each vision test that could be used within gram. However, much variation exists both within and the screening program. between countries as to the content of vision screening programs. Th is includes the age at which the child is screened, referral criteria of the screening program, and 8.3.2 Cover-Uncover Test indeed, the personnel administering the tests that form the screening program. Owing to such diff erences, it is Th e cover-uncover test is used to detect the presence of oft en diffi cult to make direct comparisons between stud- strabismus, and is deemed to be the gold standard for ies that report on vision screening success. Much has detecting strabismus. However, there are few studies that been contributed to the literature over recent years, largely report on the sensitivity and specifi city of the test itself. through the work of the Vision in Preschoolers Study Williams et al. [20] were able to report on the sensitivity (VIP). VIP is a multi-centre study, conducted in the USA, and specifi city of the cover-uncover test on children who whose purpose is to evaluate whether there are tests, or had been screened at the ages of 8, 12, 18, 25, 31 and 37 combinations of tests, that can be used eff ectively in pre- months. At 37 months, the sensitivity of the test was cal- school vision testing. culated to be 75% (95% CI, 0.577–0.899%), with a speci- Th e eff ectiveness of a screening test in detecting a confi city of 100%. dition is considered in terms of sensitivity, specifi city, and Th e VIP study also assessed the eff ectiveness of the positive and negative predictive values. Sensitivity is cover-uncover test in detecting strabismus, amblyopia, defi ned as the proportion of individuals with the target reduced VA, and refractive error [21]. Th e results are 8.3 Screening Tests for Amblyopia, Strabismus, and/or Refractive Error 101

Table 8.2. Sensitivity of cover-uncover test when speciﬁ city was set to 0.94 [21] Test Amblyopia n = 75 Strabismus n = 48 Refractive error Reduced VA n = 132 (95% CI) (95% CI) n = 240 (95% CI) (95% CI) Cover-uncover 0.27 (0.17–0.37) 0.60 (0.46–0.74) 0.16 (0.11–0.21) 0.06 (0.02–0.10) n = number of children

summarized in Table 8.2. Th e results of this study indicated Th e VIP has reported on the testability of two diff erent that the cover-uncover test is more sensitive at detecting stereotests used to screen for vision disorders, the Random the presence of strabismus compared with detecting the Dot E and the Stereo Smile test [21, 23]. Th e results presence of amblyopia, refractive error, or reduced VA. reported by condition type are summarized in Table 8.3. Th e results indicated that both the stereotests are more accurate at detecting the presence of amblyopia and stra- 8.3.3 Stereoacuity bismus compared with that for reduced VA or refractive Th e inclusion of stereoacuity tests within pre-school error. vision screening programs could be considered as a con- In a further study, VIP examined the sensitivity of the tentious issue. VIP [22] stated that most guidelines rec- same stereotests when the specifi city was set at 0.94. Th e ommend a test of stereopsis. However, if a child was results are summarized in Table 8.4, and show that found to have normal VA, no strabismus, and no clini- the Stereo Smile test was more accurate than the Random cally signifi cant refractive error, yet failed to demonstrate Dot E in detecting most target conditions of screening. adequate evidence of stereoacuity, should they be referred for further investigation? A number of stereotests are available for use as part of a pre-school vision screening 8.3.4 Photoscreening and/or Autorefraction program; however, normative pediatric values of stereopsis have not been identifi ed for some of these tests. In the Th e use of photoscreeners and/or autorefractors in absence of such data, the appropriateness of inclusion of pre-school vision screening is extremely varied. Within such tests could be questioned. Stereotests that involve a the USA, they are commonplace, and the variety of dif- pass/fail response could be deemed as more appropriate ferent makes and models make summarizing literature for the purpose of screening for vision problems. extremely diffi cult. Th e use of such instruments within

Table 8.3. Sensitivity of Random Dot E and stereo smile by condition typea [23]

Stereotest Amblyopia Reduced VA Strabismus Refractive error Speciﬁ city

Year 1 n = 796 n = 75 n = 132 n = 48 n = 240 Random Dot E 0.63 0.38 0.60 0.47 0.90 Year 2 n = 1037 n = 88 n = 114 n = 62 n = 299 Stereo smile 0.77 0.30 0.68 0.51 0.91 n = number of children; amay have more than one condition

Table 8.4. Sensitivity of Random Dot E and stereo smile when speciﬁ city was set to 0.94a [21]

Test Amblyopia Strabismus Refractive error Reduced VA (95% CI) (95% CI) (95% CI) (95% CI)

Random Dot E 0.28 (0.18–0.38) 0.29 (0.16–0.42) 0.23 (0.18–0.23) 0.24 (0.17–0.31) Stereo smile 0.61 (0.51–0.71) 0.58 (0.46–0.70) 0.37 (0.32–0.42) 0.20 (0.13–0.27) aMay have more than one condition 102 8 The Value of Screening for Amblyopia Revisited

UK pre-school vision screening programs is much less recommend that these children ought to be referred or frequent. When considering the appropriateness of pho- retested at a later date possibly with a diff erent test. Th e toscreeners and/or autorefractors in pre-school vision impact of recall and re-testing, or automatic referral will screening, it is important to recognize their accuracy undoubtedly aff ect the overall clinical and cost eff ective- 8 when compared with a gold standard (usually a refrac- ness of any pre-school vision program. tion performed under full cycloplegia). Th ere are notable advantages and disadvantages of photoscreening when compared with autorefraction. One of the main diff er- 8.3.6 Who Should Administer ences is that of cost. Aft er the initial expense of purchase, the Screening Program? there is minimal additional cost to autorefraction. Photoscreening, however, requires printing of the image, Within the UK, it is recommended that pre-school vision and depending upon who is administering the test, inter- screening programs be conducted by orthoptists or by pretation of the results. Th e implications of both these professionals trained and supported by orthoptists [3, 4]. factors lead to a higher overall expense when incorpo- In the USA, pre-school vision screening is usually con- rated into a vision screening program. ducted by nurses and lay people. Th e use of lay people to It should also be noted that the primary aim of the use administer screening tests does have advantages, particu- of a photoscreener or autorefractor is the detection of larly when considering the economic burden of a screen- refractive error. Th at is, it may detect an amblyogenic fac- ing program. Lay screeners are a cheaper alternative to tor, but not amblyopia itself. Similarly, the presence of eye care professionals, such as orthoptists, optometrists, strabismus may also be detected, although understand- or ophthalmologists. ably, the sensitivity and specifi city rates of these are con- Concerns regarding training and assessment of lay siderably lower than those of detecting refractive error. screeners have been raised; are lay screeners as accurate It is beyond the scope of this chapter to review and as eye care professionals in detecting amblyopia, strabis- appraise literature describing specifi c photorefractors mus, and/or refractive error? Th is question was addressed and/or autorefractors. Important points to note when by VIP, who assessed the performance of lay screeners in considering such articles include the study population administering pre-school vision screening tests compared (including age, ethnicity, and whether general or clinical); to nurse screeners [25]. In this study, the screening tests test setting (e.g. environment); sensitivity and specifi city conducted included assessment of refractive error, VA, of the test; the personnel conducting the test; and whether and stereoacuity. Two hand-held autorefractors were used any comparison is made to the gold standard (in this case, to detect the presence of refractive error. VA was assessed full refraction under cycloplegia). at two diff erent testing distances; a linear test was performed at 10 feet, and a single, crowded test administered at 5 feet. Th e results of the study demonstrated that although nurse screeners appeared to have slightly higher 8.3.5 What to Do with Those Who sensitivities in the assessment of refractive error and pres- Are Unable to Perform Screening Tests? ence of stereoacuity compared with lay screeners, the dif- Successful testing of children is largely dependent on the ferences were not statistically signifi cant. child’s cooperation and compliance. Th e decision about However, when examining the results of VA testing, whether to refer those children who are unable to per- the authors reported that nurse screeners achieved sig- form screening tests is diffi cult. Some would argue that nifi cantly higher sensitivity than lay screeners with the such children ought to be referred for further investiga- linear VA test. Whilst the authors made no recommen- tion, for the reason that they are unable to perform the dations for future screening protocol strategies, their screening tests due to the presence of an ocular condition. results could be interpreted in two ways. Th e lack of sta- Others would say that this may not be the case, and that tistically signifi cant diff erences in detection of refractive cooperation may be the true issue. Th e prevalence of ocu- error or stereoacuity with tests administered by lay lar conditions amongst children who were unable to per- screeners could support the use of such personnel in form pre-school screening tests has been investigated and vision screening programs. However, the diff erences it was found that pre-school children who were unable to observed in VA testing between lay screeners and nurse perform the screening test were at a higher risk of higher screeners could suggest that nurse screeners would be amblyopia, strabismus, signifi cant refractive error, or more eff ective in detecting vision anomalies. Diff erences unexplained low VA compared with those who had in screening programs between countries will undoubt- passed the screening test [24]. Th is led the authors to edly continue to exist; however, recommendations as to 8.4 Treatment of Amblyopia 103 who should conduct screening based upon personnel 8.4.1 Type of Treatment costs alone may not be appropriate. Amblyopia is treated by obscuring the image from the good eye to promote the use of the amblyopic eye. Th is Summary for the Clinician can be achieved through occlusion treatment (patching ■ Content of vision screening programs vary widely. or pharmacological occlusion, in the form of atropine), or Most involve assessment of VA for which a large through optical penalization. Th ere are notable advan- number of tests are available. Th e gold standard is tages and disadvantages to diff erent treatment modalities a crowded logMAR-based test. Referral criteria in terms of compliance, ease of administration, and VA should be specifi c for the test used. outcome. Comparison of studies investigating the eff ec- ■ Th e use of photoscreeners and/or autorefractors tiveness of treatment of amblyopia is hindered, due to dif- in vision screening programs is not universal. fering defi nitions of both “amblyopia” and “treatment Th e use of photoscreeners and/or autorefractors success”. In addition, clinicians have long recognized that will have an impact upon the cost eff ectiveness the amount of treatment prescribed and the amount of of screening. treatment actually undertaken may diff er. Objective mea- ■ Th e inclusion of stereotests in pre-school vision surement of the amount of occlusion worn has been made screening programs could be questioned. possible with the introduction of occlusion dose moni- ■ Recommendations state that pre-school vision tors (ODM). ODMs were developed and validated by the screening programs be conducted by orthoptists Monitored Occlusion Treatment for Amblyopia Study or by professionals trained and supported by (MOTAS) Cooperative (UK), and since then, have been orthoptists. used to examine whether there is a dose response to occlusion therapy.

8.4 Treatment of Amblyopia 8.4.2 Refractive Adaptation Th e clinical management of amblyopia is determined One of the main concepts that have arisen over the recent following careful consideration on a case-per-case basis, years in amblyopia treatment is that of refractive adapta- taking into account a number of factors including the tion (or “optical treatment of amblyopia” as it is some- type of amblyopia present, the patient’s age, and the times known [26]. Th ere has been increasing evidence to level of VA in the amblyopic eye. Nonetheless, advances suggest that the treatment of amblyopia in the presence of in evidence-based medicine have led to a number of refractive error should incorporate observation of VA fol- recognized studies that have reinforced or altered clini- lowing the prescription of glasses alone [26–29]. Th ese cal practice in the management of this condition. Th e studies report increases in VA in subjects such that some Pediatric Eye Disease Investigator Group (PEDIG), did not require any additional treatment for their ambly- based in the USA, is a multi-centre group dedicated to opia. Prior to such studies, it was uncertain whether clinical research in strabismus, amblyopia and other eye observed improvements in VA achieved were the result of disorders aﬀ ecting children. Funded by the National amblyopia therapy (i.e. occlusion) or due to glasses-wear Eye Institute (NEI), this group has investigated many alone. aspects of the clinical course of amblyopia and its treat- It is becoming increasingly clear that refractive adap- ment. Th e Monitored Occlusion Treatment of Amblyopia tation is a recognized period in amblyopia therapy. Th e Study Cooperative (MOTAS Cooperative) is a multi- time taken to reach this period, however, remains under disciplinary group of ophthalmologists, orthoptists, debate. Th e MOTAS studies utilized a period of 18-week basic scientists, and statisticians dedicated to investigat- observation [27–29]; however, the PEDIG reported that ing amblyopia treatment. Based in London (UK), it is 83% of their study group demonstrated stability of funded by the charities Guide Dogs for the Blind improvement in VA before 15 weeks, but one patient Association, and Fight for Sight. Th ey have conducted improved in 30 weeks [26]. Improvements in VA have two clinical trials to identify the response of amblyopia been described to occur aft er 20 weeks, but not consider- to occlusion therapy. Data from both the studies con- ably, with the majority of improvement having occurred ducted by PEDIG and the MOTAS Cooperative have in weeks 4–12 [30]. contributed to our understanding of the management of One of the arguments supporting the notion of vision amblyopia. screening is the detection of bilateral refractive error. 104 8 The Value of Screening for Amblyopia Revisited

Wallace et al. [31], as part of the PEDIG study, examined One disadvantage of pharmacological occlusion is that the improvements in VA in children with bilateral refrac- the eff ects are not readily reversible; it can take several tive amblyopia aged between 3 and 10 years. Th ey reported weeks for the eff ects of atropine to wear off . Concerns that correction of refractive error improved VA, with only also exist regarding its effi cacy as a treatment modality, 8 12% of the cohort requiring additional amblyopia therapy with some clinicians believing it to be a less eff ective in the form of occlusion or atropine. treatment when compared with conventional occlusion. Studies conducted by PEDIG examined the eff ectiveness of conventional occlusion vs. pharmacological occlusion 8.4.3 Conventional Occlusion in the treatment of moderate amblyopia (20/40–20/80) Patching treatment is oft en initiated as the fi rst-line [34] and severe amblyopia (20/100–20/400) [35]. Either approach in amblyopia therapy. One advantage of patching treatment modality was found to be appropriate with treatment is that the eff ects are reversible; that is, once the similar improvements in VA in either group. Th e decision patch is removed, the non-amblyopic eye is favored, which towards which therapy should be adopted may now be is not the case with pharmacological occlusion. Since the based on other factors. One such factor may be the instil- acknowledgement of refractive adaptation, it has been nec- lation of the atropine itself. Th e eff ect of diff erent atropine essary to confi rm that occlusion therapy is also eff ective in regimens in the treatment of moderate amblyopia (20/40– the management of amblyopia. PEDIG compared the eff ect 20/80) was investigated. Comparisons were made between of daily patching vs. a control group of amblyopes in chil- the observed eff ects of daily atropine instillation and dren aged 3–7 years, following a period of refractive adap- those of weekend-only atropine instillation [36]. Both tation. An improvement in VA was observed in both the groups were observed to show improvements in VA of groups aft er 5 weeks, and as expected, a greater improve- similar magnitudes. It could be argued that the need for ment was reported in the patched group [32]. daily atropine instillation is redundant, thereby improv- Th e MOTAS Cooperative investigated the amount of ing the therapeutic experience for the child. Th is in itself occlusion required to improve VA and explored the dose- may encourage parents and/or clinicians to adopt this response relationship in amblyopia therapy [28]. Th ey treatment modality. found that most children required between 150 and 250 h of occlusion, irrespective of the type of amblyopia present. Specifi c characteristics were observed to aff ect the response, 8.4.5 Optical Penalization such as the age of the patient; where older children required a greater amount of occlusion to achieve similar gains in Another treatment option in the management of amblyo- VA compared with their younger counterparts. Younger pia is that of optical penalization. Th is is where lenses are children have been observed to respond more quickly and used to induce a defocused image of the non-amblyopic with less occlusion than older children; however, the fi nal eye. Tejedor and Ogallar [37] directly compared the level of VA achieved has been similar for all ages [29]. eff ects of atropine vs. optical penalization in the treat- Traditionally, clinicians have recommended near-visual ment of mild to moderate amblyopia (VA of at least activities whilst occlusion therapy is undertaken; however, 20/60). Th is small study found greater improvements in there has been little research to justify such advice. Th e VA in the atropine group aft er 6 months of therapy, which PEDIG investigated whether performing such activities may be attributed to the child peeking over or around the infl uenced the improvement in VA outcome when treating glasses and thereby not achieving the desired eff ect of amblyopia in conjunction with occlusion therapy [33]. No optical penalization. Although optical penalization statistical evidence to support the notion that near visual remains a useful treatment option in specifi c clinical situ- activities improved VA outcome in their study group was ations, it is oft en not considered as an appropriate fi rst- found. It should be noted that the study group were pre- line choice of therapy in the management of amblyopia. scribed only 2 h of patching per day, and that the authors made no inference as to whether the results would be similar in subjects patched for a greater or lesser time. 8.4.6 Eff ective Treatment of Amblyopia in Older Children (Over the Age of 7 Years)

8.4.4 Pharmacological Occlusion Th ere has been strong evidence that treatment for amblyopia is more eff ective prior to the age of 7 years. Pharmacological occlusion (i.e. atropine) has notable Despite this, amblyopia therapy has been reported to be benefi ts; it could be argued that it carries with it less of a successful in older children with either anisometropic social stigma compared with the wearing of an eye patch. [38–42] and/or strabismic amblyopia [40–42]. Treatment 8.4 Treatment of Amblyopia 105 of strabismic amblyopia in the older child should be 8.4.8 Other Treatment Options for Amblyopia pursued with caution, as there is a notable risk of reducing the density of suppression, and thereby inducing Th e use of photorefractive keratectomy (PRK) for the intractable diplopia in these patients. A number of stud- treatment of anisometropia in children has not been fully ies that reported on improvements in VA in older chil- investigated and concerns exist surrounding the long- dren with strabismic or mixed etiology amblyopia term response to refractive surgery in terms of VA and following treatment have not reported on whether the corneal status. However, it could be postulated that if the density of suppression had been measured, or if any amblyopic risk factor of high anisometropia is removed other side-eff ects had been observed [40–42]. Despite early, then the possibility of development of dense ambly- some evidence to suggest that successful treatment of opia would be reduced. Paysse et al. [45] reported the amblyopia in the older child is possible, earlier inter- results of a small study of children with high anisometro- vention is more advantageous, and to that end supports pia, and found improvements in both VA and stereopsis the notion of pre-school vision screening. following treatment. However, compliance with amblyopia therapy remained unaff ected in this study group following treatment. Th e use of refractive surgery in children 8.4.7 Treatment Compliance is not commonplace and there remains a need for a large randomized clinical trial to fully investigate the possible Th e successful management of amblyopia is intrinsically benefi ts of this form of treatment. linked to treatment compliance and adherence to therapy. Th is in itself is multi-factorial in nature. Th e development and application of ODMs has meant that reasons 8.4.9 Recurrence of Amblyopia for non-compliance can be more thoroughly investi- Following Therapy gated. In particular, ODMs have highlighted the discrepancy between the amount of occlusion prescribed Recurrence of amblyopia has been observed in patients and the amount administered. Clinicians have long rec- following the cessation of treatment, with rates varying ognized that the amount of occlusion carried out oft en widely. Some recent studies have sought to identify fac- falls short of their recommended treatment plan. Stewart tors that may infl uence whether recurrence is likely to et al. [29] reported on the eff ect of 6 h a day occlusion occur [46–49]. Th ese include age of termination of treat- compared with 12 h a day occlusion in the treatment of ment, VA at the time of cessation of treatment, and the strabismic and/or anisometropia amblyopia. Th ey found type of amblyopia present. Recurrence in amblyopia was that the amount of occlusion received was 66 and 50% of noted in 7–27%, with a low reported recurrence in chil- their prescribed 6 and 12 h a day, respectively. Such dren who underwent treatment aft er the age of 7 years information ought to be taken into account when pre- [49]. Age of the child at the cessation of treatment does scribing occlusion therapy. appear to be a factor, with recurrence inversely correlated Loudon et al. [43] examined some of the limiting fac- with patient age [46]. tors of occlusion therapy for amblyopia and reported that parental fl uency in the national language and level of education were both predictors of low compliance. Summary for the Clinician Parental understanding of the condition and treatment ■ Treatment of amblyopia associated with refrac- has also been reported as being an important factor in the tive error should incorporate a period of obser- successful management of amblyopia. vation with glasses-wear alone to allow for Adherence to treatment must be considered not only “refractive adaptation” or “optical treatment of in terms of the child complying with therapy, but in the amblyopia”. Improvements in VA can occur up parent/guardian administering the treatment as advo- to and beyond 20 weeks aft er glasses are pre- cated by the ophthalmologist and/or orthoptist. Searle scribed, but most improvement occurs in weeks et al. [44] found two variables that were signifi cant pre- 4–12. In some cases, further amblyopia therapy dictors of compliance with occlusion therapy. Th ey may not be required. reported that self-effi cacy (the belief in the ability to patch ■ Th ere is evidence to suggest that children who their child) was positively associated with treatment com- undergo amblyopia therapy at an early age pliance. Th e parental belief that occlusion therapy inhib- respond more quickly to occlusion than older its the child’s activities was negatively associated with children, and require less occlusion in total. treatment compliance. 106 8 The Value of Screening for Amblyopia Revisited

■ Pharmacological occlusion, in the form of atro- 8.5.2 Stereoacuity and Motor pine, has been found to be as eff ective as conven- Skills in Children with Amblyopia tional occlusion (patching) in the treatment of both moderate and severe amblyopia. Weekend- Stereoacuity and motor skills have been reported to be 8 only atropine instillation has been shown to pro- impaired in children with amblyopia. Webber et al. [50] duce similar improvements in VA as daily investigated the functional impact of amblyopia in chil- atropine instillation in the treatment of moder- dren by assessing the fi ne motor skills of those with ate amblyopia. amblyopia compared with age-matched control subjects. ■ Th ere is evidence to suggest that successful treat- It was noted that the subjects with amblyopia performed ment of children aged over 7 years can be signifi cantly worse in most of the fi ne motor skills tests achieved in cases of anisometropic, strabismic, conducted as part of the study, particularly in the tasks and mixed etiology amblyopia. related to time. Th e results were even more noticeable in ■ Th e development of ODM has informed not only those children with a diagnosis of amblyopia and strabis- the occlusion-dose response of amblyopia treat- mus. Hrisos et al. [51] investigated the infl uence of VA ment, but also reasons for poor treatment com- and stereoacuity on the performance of pre-school chil- pliance. Parental understanding of the condition dren undertaking tasks that required visuomotor skills and belief in therapy may infl uence treatment and visuospatial ability. Th e authors reported that reduced outcome. monocular VA itself did not relate to any ability of task ■ Recurrence of amblyopia may occur following performance, but stereoacuity was found to aff ect task treatment, with reported rates of 7–27%. Factors performance, with subjects with reduced steroacuity infl uencing recurrence include age of the child at noted to have poorer responses to neurodevelopment cessation of treatment, VA at the time of cessa- tasks. Such studies support the notion that amblyopia is tion of treatment, and the type of amblyopia associated with negative implications to HRQoL. present.

8.5.3 Reading Speed and Reading Ability in Children with Amblyopia

8.5 Quality of Life Reading speed and reading ability has been assessed in children with amblyopia. Stift er et al. [52] reported that When considering the application of any screening pro- maximum reading speed was signiﬁ cantly reduced in gram, thought should be made regarding the impact of those with the condition. Th erefore, they could be deemed testing for the target condition, the impact that the target to have a functional reading impairment when compared condition has upon a person, and the impact that subse- with normal-sighted controls. It is recognized that read- quent treatment of that target condition may have upon a ing ability is multi-factorial in nature, and is inﬂ uenced person. One of the ways in which the health impact of a by comprehension. Th e study does not imply that chil- disease or condition can be assessed is through measures dren with unilateral amblyopia are poor readers under of quality of life, or HRQoL. Over recent years, there has binocular conditions, for the binocular VA and reading been a growing body of evidence which has examined the acuity of the two groups were comparable. impact of amblyopia and/or strabismus upon a person’s physical and emotional well-being.

8.5.4 Impact of Amblyopia Upon Education Chua and Mitchell [9], as part of the Blue Mountains Eye 8.5.1 The Impact of Amblyopia Study in Australia (a population-based survey of people Upon HRQoL aged 49 years or older), examined the consequences of Th ere have been a number of studies that have investi- amblyopia on education, occupation, and long-term gated the impact of amblyopia upon HRQoL. Th ese have vision loss. In their study population, the presence of examined the eff ect of amblyopia upon stereoacuity and amblyopia was not found to be signifi cantly associated motor skills [50, 51], reading speed ability [52], educa- with lifetime occupational class. However, fewer people tional attainment [9], and emotional well-being [44, with amblyopia were found to have completed higher 53–58]. university degrees. Th is fi nding was supported by Rahi 8.5 Quality of Life 107 et al. [59], who reported on fi ndings of the 1958 British adverse reactions from their peers. Th ey compared two birth cohort with respect to any association of amblyopia groups that had been off ered pre-school vision screening with diverse educational, health, and social outcomes. at the age of 3 years with those who had not; and asked Th e authors could fi nd no statistical evidence between the children at age 8 years whether they had been bullied the presence of amblyopia and educational attainment or through a standard structured interview. Th e authors paid employment. reported an almost 50% reduction in children who reported having been bullied in the group that had been off ered pre-school screening, compared with the group who had not. 8.5.5 Emotional Well-Being Not all children undertaking amblyopia therapy fi nd and Amblyopia the treatment a negative experience. Indeed, in a study by Th e psychosocial impact of amblyopia and its treatment Choong et al. [53], the authors found no signifi cant has been explored from both the parental and child per- changes in parental (carer’s) stress or the child’s psycho- spective [56]. Children have reported feelings of shame social well-being between an occluded and non-occluded and negativity associated with amblyopia, particularly group. One factor that did result in changes in parental following the start of treatment. Th e initiation of therapy attitude towards the child was the issuing of glasses. A can draw adverse attention from others, and children statistically signifi cant diff erence was found, where carers have reported that they felt interrogated by others about felt more negative towards their child once glasses were their treatment (particularly if their treatment involved prescribed. As glasses form an integral part of amblyopia the wearing of glasses and a patch). therapy, it could be deemed that the results do in fact It is important to recognize that the impact of ambly- demonstrate psychosocial implications of amblyopia opia therapy may be experienced not only by the child, treatment, particularly from the carer’s perspective. but also by family members [54]. Th is could result in Confl icting evidence exists in the adult population. impaired relationships between the child and parent/ Rahi et al. [59] reported that adults with amblyopia were guardian, but also between siblings. Parents oft en state no more likely to be bullied (either at the age of 7 or 11 that their child may be more clingy or demanding when years), and could fi nd no evidence for an association occlusion is worn; that the child’s compliance with occlu- between the presence of amblyopia and participation in sion can lead to negative behavioral changes or that their social activities in either childhood or adult life. Th e child appears to be less confi dent when wearing their authors also stated that those with amblyopia were no patch or glasses [56]. more likely to report depression or psychological distress Th e issue of peer victimization and bullying associated in adult life. with amblyopia has been recognized [55, 56, 58]. Th is Th is fi nding was not supported by Packwood et al. [57], may be in response to the wearing of glasses and/or occlu- who explored the psychosocial implications of growing up sion therapy. Horwood et al. [58], as part of the Avon and living with amblyopia in a group of adult subjects. Th e Longitudinal Study of Parents and Children (ALSPAC) authors reported that those with amblyopia experienced conducted in the UK, investigated whether wearing more distress in several areas of psychological well-being, glasses, having manifest strabismus, or having a history of including somatization, obsession-compulsion, interper- wearing an eye patch pre-disposed pre-adolescent chil- sonal sensitivity, anxiety, and depression. dren to being victimized more frequently at school. In Taken in isolation, the impact of any one of the afore- this study, the outcome measure used to assess whether mentioned problems may be minimally associated with bullying had occurred was through a structured face-to- detriment to HRQoL. However, the cumulative eff ect of face interview, conducted with the child at the age of 8.5 impaired reading, motor skills, and psychosocial impact years. Children were asked if they had experienced or of amblyopia, for example, might infl uence HRQoL to a used any forms of overt or relational bullying. Th e authors greater degree. reported that those children who wore glasses or had a history of wearing an eye patch were 35–37% more likely to be victims of physical or verbal bullying (aft er adjust- 8.5.6 The Impact of Strabismus Upon HRQoL ment for social class and maternal education). Williams et al. [55] argued the case for pre-school Th e psychosocial implications of strabismus are more vision screening in that those who had undertaken accepted and recognized, particularly in cases of cos- screening were likely to have concluded amblyopia ther- metically obvious strabismus. Detrimental implications apy early (i.e. before school starts), and thus would avoid of strabismus include a negative self-image, reduced 108 8 The Value of Screening for Amblyopia Revisited

self-confi dence, low self-esteem, and poor interpersonal fi ndings of each study are equally valid; however, it must relationships [60]. Th e presence of a cosmetically notice- be recognized that there may be levels of bias exerted able strabismus has also been reported to impact upon a depending upon which methodology is applied. For person’s ability to gain employment [61, 62], and in a example, studies that report from the parental perspec- 8 person’s ability to attract a partner [63]. Furthermore, tive [53, 54, 56, 70] may in fact be capturing parental the presence of strabismus does not only aff ect those in opinion regarding the condition and/or its treatment, adulthood. Uretman et al. [64] determined that children rather than a true measure of HRQoL changes. Studies with strabismus were perceived in a negative light by that involve adults with a history of amblyopia and/or adults. Th e age at which the emergence of negative atti- strabismus [57] are asking subjects to recall childhood tudes towards those with strabismus develops has been experiences. It is possible that adult experiences have studied. Paysse et al. [65] reported that at approximately since “tainted” the recall of such events, either exaggerat- 6 years of age, children begin to express a negative atti- ing or diminishing the true changes in HRQoL experi- tude towards strabismus. enced as a child. Perhaps, studies that report from the In adults, it has been documented that those with child perspective [55, 56, 58] could be considered the strabismus experience more social anxiety and use most valid. Th ey deliver insight into what is experienced social avoidance strategies compared with the general at the time. However, they are not without their weak- population [66, 67]. It could be argued, therefore, that nesses. What they fail to do is inform as to whether the surgical correction of strabismus serves to provide psy- impact of amblyopia and/or strabismus (as a condition, chosocial benefi ts, and thus improves HRQoL. or its treatment) is appreciated in the longer term, that is, Improvements in quality of life following strabis- into adulthood. mus surgery are well documented in adults [66–69]; however, its eff ect on children is not as extensively researched. Archer et al. [70] reported on a group of 98 8.5.8 The Impact of the Condition children who underwent strabismus surgery (although or the Impact of Treatment? it is unclear whether the purpose of surgery was purely cosmetic or functional in nature). Th e authors stated It can be diffi cult to fully distinguish whether any that following surgery, there were signifi cant improve- reported detriment to HRQoL in amblyopia is due to ments in a number of quality of life dimensions, includ- the condition itself or its treatment. Th is is not a factor ing those of anxiety, social relations, and developmental when considering strabismus. Strabismus (particularly satisfaction (parental response). Th e results concur that of large angle strabismus) is cosmetically notice- with those found in an adult population, and it can able and it is the impact that that has upon the person therefore be deemed that the psychosocial benefi ts which can aff ect HRQoL. Th erefore, it can be said that reported in adults following strabismus surgery are any study that reports on HRQoL and strabismus is also applicable to children. reporting on the eff ect that the condition has upon a person’s well-being. With amblyopia, this is not the case. Th e condition itself cannot be identifi ed by peers. What is noted is the eff ect of treatment upon HRQoL, 8.5.7 Critique of HRQoL Issues with the instigation of glasses or occlusion therapy. in Amblyopia Studies that report on changes in HRQoL in amblyopia, Methods of determining the impact of amblyopia and/or frequently report on the impact of the treatment upon strabismus upon HRQoL diff er greatly from one study to quality of life rather than the condition itself [44, 53–55, another. Some report changes in psychosocial behavior 55–57]. Alternative studies do report on the impact of and well-being using a purpose-designed questionnaire amblyopia; however, the measures of these studies are [60, 62, 63, 67, 71]. Whilst their fi ndings are of great clini- of adult-related issues (such as employment, educa- cal importance, it can be diffi cult to compare one study tional attainment, and risk of losing vision in the non- with another due to diff erences in methodologies. amblyopic eye) [9, 59]. It is not possible to determine One key component that must be considered when whether the same HRQoL changes that occur in child- addressing the issue of HRQoL and amblyopia and/or hood are appreciated in adulthood, because the mea- strabismus is that of the perspective from which the sures used in the identifi ed studies are so diff erent. results are taken. Th at is, are the results taken from Nonetheless, it can be concluded that there is evidence responses from the parent, the child, or from an adult to suggest that there are HRQoL issues related to ambly- with a history of amblyopia and/or strabismus? Th e opia and/or strabismus and its treatment. References 109

6. Ciner E, Dobson V, Schmidt P, Allen D, Cyert L, Maguire M, Summary for the Clinician et al (1999) A survey of vision screening policy of pre- ■ Th ere have been a number of studies investigat- school children in the United States. Surv Ophthalmol ing HRQoL implications of amblyopia and/or 43(5):445–457 strabismus over recent years. Th ese have involved 7. Brown MM, Grown GC, Sharma S, Busbee B, Brown H studies with children who have the condition, or (2001) Quality of life associated with unilateral and bilat- adults who had previously undergone treatment. eral good vision. Ophthalmology 108:643–648 ■ Studies have reported amblyopia to impact upon 8. Rahi JS, Logan S, Timms C, Russell-Eggitt I, Taylor D stereoacuity, fi ne motor skills, and reading speed. (2002) Risk, causes, and outcomes of visual impairment ■ Th e presence of amblyopia does not appear to aft er loss of vision in the non-amblyopic eye: a population- have any impact on educational attainment or based study. Lancet 360:597–602 paid employment in adult life. 9. Chua B, Mitchell P (2004) Consequences of amblyopia on ■ Amblyopia (more specifi cally amblyopia treat- education, occupation, and long term vision loss. Br ment) has been shown to impact negatively upon J Ophthalmol 88:1119–1121 a child’s emotional well-being; and may also 10. Van Leeuwen RE (2007) Risk of bilateral visual impair- aff ect relationships between the child and par- ment in individuals with amblyopia: the rotterdam study. ent/guardian. Br J Ophthalmol 91(11):1450–1451 ■ Th e issue of bullying and amblyopia treatment 11. Rahi JS, Logan S, Borja MC, Timms C, Russell-Eggitt I, requires further investigation. Some studies Taylor D (2002) Prediction of improved vision in the reported that children who had glasses or had a amblyopic eye aft er visual loss in the non-amblyopic eye. history of occlusion therapy were more likely to Lancet 360:621–622 be victims of bullying. However, other studies 12. Carlton J, Karnon J, Czoski-Murray C, Smith KJ, Marr J refuted this. (2008) Th e clinical eff ectiveness and cost-eff ectiveness of ■ Taken in isolation, the impact of any one of the screening programmes for amblyopia and strabismus in aforementioned problems may be minimally children up to the age of 4–5 years: a systematic review and associated with detriment to HRQoL. However, economic evaluation. Health Technol Assess 12(25):iii-194 the cumulative eff ect of impaired reading, motor 13. Gold M, Siegel J, Russell L, Weinstein M (1996) Cost- skills, and psychosocial impact of amblyopia, for eff ectiveness in health and medicine. Oxford University, example, might infl uence HRQoL to a greater New York degree. 14. IQWiG (2008) Screening for visual impairment in chil- ■ Th e reported HRQoL implications of strabismus dren: executive summary (translation of the executive are related to physical appearance and the impact summary of the fi nal report). IQWiG Reports – of strabismus upon self-image and interpersonal Commission No S05–02 relationships. Surgical correction of strabismus 15. Drover JR, Felius J, Cheng CS, Morale SE, Wyatt L, Birch has been reported to improve HRQoL. EE (2008) Normative pediatric visual acuity using single surrounded HOTV optotypes on the electronic visual acuity tester following the amblyopia treatment study protocol. J AAPOS 12(2):145–149 16. Sonksen PMW (2008) Th e Sonksen logMAR test of visual References acuity: II. Age norms from 2 years 9 months to 8 years. 1. Health departments of the United Kingdom (2000) Second J AAPOS 12(1):18–22 report of the National Screening Committee 17. Shea SJ, Gaccon L (2006) In the absence of strabismus what 2. Snowdon SK, Stewart-Brown SL (1997) Pre-school vision constitutes a visual defi cit in children? Br J Ophthalmol screening. Health Technol Assess 1:1–98 90(1):40–43 3. Hall DMB, Elliman D (2003) Health for all children, 4th 18. Birch EE, Strauber SF, Beck RW, Holmes JM, Pediatric eye edn. Oxford University, Oxford disease investigator group (2008) Comparison of the 4. Department of health (2008) Updated child health promo- amblyopia treatment study HOTV and the electronic-early tion programme treatment of diabetic retinopathy study visual acuity pro- 5. A Joint Statement of the American Association for Pediatric tocols in amblyopic children aged 5 to 11 years. J AAPOS Ophthalmology and Strabismus and the American 13(1):75–78 Academy of Ophthalmology. Vision screening for infants 19. Chen SI, Chandna A, Norcia AM, Pettet M, Stone D (2006) and children (2007) Th e repeatability of best corrected acuity in normal and 110 8 The Value of Screening for Amblyopia Revisited

amblyopic children 4 to 12 years of age. Invest Ophthalmol A randomized trial to evaluate 2 hours of daily patching for Vis Sci 47(2):614–619 strabismic and anisometropic amblyopia in children. 20. Williams C, Harrad RA, Harvey I, Sparrow JM, ALSPAC Ophthalmology 113(6):904–912 study group (2001) Screening for amblyopia in preschool 33. Pediatric eye disease investigator group (2008) A random- 8 children: results of a population-based randomised con- ized trial of near versus distance activities while patching trolled trial. Ophthalmic Epidemiol 8:279–295 for amblyopia in children aged 3 to less than 7 years. 21. Th e vision in preschoolers study group (2005) Sensitivity Ophthalmology 115(11):2071–2078 of screening tests for detecting vision in preschoolers- 34. Pediatric eye disease investigator group (2002) A random- targeted vision disorders when specifi city is 94%. Optom ized trial of atropine vs. patching for treatment of moder- Vis Sci 82:432–438 ate amblyopia in children. Arch Ophthalmol 120:268–278 22. Vision in preschoolers study group (2006) Random Dot E 35. Holmes JM, Kraker RT, Beck RW, Birch EE, Cotter SA, stereotest: testability and reliability in 3- to 5-year-old Everett DF, et al (2003) A randomized trial of prescribed children. J AAPOS 10(6):507–514 patching regimens for treatment of severe amblyopia in 23. Th e vision in preschoolers study group (2004) Comparison children. Ophthalmology 110:2075–2087 of preschool vision screening tests as administered by 36. Repka MX, Cotter SA, Beck RW, Kraker RT, Birch EE, licensed eye care professionals in the vision in preschoolers Everett DF, et al (2004) A randomized trial of atropine study. Ophthalmology 111:637–650 regimens for treatment of moderate amblyopia in children. 24. Maguire MG, Vision in preschoolers study group (2007) Ophthalmology 111(11):2076–2085 Children unable to perform screening tests in vision in 37. Tejedor J, Ogallar C (2008) Comparative effi cacy of penal- preschoolers study: proportion with ocular conditions and ization methods in moderate to mild amblyopia. Am impact on measures of test accuracy. Invest Ophthalmol J Ophthalmol 145(3):562–569 Vis Sci 48(1):83–87 38. Menon V, Shailesh G, Sharma P, Saxena R (2008) Clinical 25. Vision in preschoolers study group (2005) Preschool vision trial of patching versus atropine penalization for the treat- screening tests administered by nurse screeners compared ment of anisometropic amblyopia in older children. with lay screeners in the vision in preschoolers study. J AAPOS 12:493–497 Invest Ophthalmol Vis Sci 46:2639–2648 39. Patwardhan NA (2007) Is age relevant for the success 26. Cotter SA, Pediatric eye disease investigator group, of treatment of anisometropic amblyopia? Indian Edwards AR, Wallace DK, Beck RW, Arnold RW, et al J Ophthalmol 55(6):469–470 (2006) Treatment of anisometropic amblyopia in children 40. Brar GS, Bandyopadhyay S, Kaushik S, Raj S (2006) with refractive correction. Ophthalmology 113(6): Effi ciency of occlusion therapy for management of ambly- 895–903 opia in older children. Indian J Ophthalmol 54:257–260 27. Stewart CE, Moseley MJ, Stephens DA, Fielder AR (2005) 41. Park KH, Hwang J-M, Ahn JK (2004) Effi cacy of amblyo- On behalf of the MOTAS Cooperative. Refractive adapta- pia therapy initiated aft er 9 years of age. Eye 18:571–574 tion in amblyopia: quantifi cation of eff ect and implications 42. Pediatric eye disease investigator group (2004) A prospec- for practice. Br J Ophthalmol 88:1552–1556 tive, pilot study of treatment of amblyopia in children 10 28. Stewart CE, Stephens DA, Fielder AR, Moseley MJ (2007) to <18 years old. Am J Ophthalmol 137(3):581–583 Modeling dose-response in amblyopia: toward a child- 43. Loudon SE, Fronius M, Looman CW, Awan M, Simonsz B, specifi c treatment plan. Invest Ophthalmol Vis Sci van der Maas PJ, et al (2006) Predictors and a remedy for 48(6):2589–2594 noncompliance with amblyopia therapy in children mea- 29. Stewart CE, Stephens DA, Fielder AR, Moseley MJ, sured with the occlusion dose monitor. Invest Ophthalmol ROTAS C (2007) Objectively monitored patching regi- Vis Sci 47(10):4393–4400 mens for treatment of amblyopia: randomised trial. BMJ 44. Searle A, Norman P, Harrad R, Vedhara K (2002) 335(7622): 707 Psychosocial and clinical determinants of compliance with 30. Chen PL, Chen JT, Tai MC, Fu JJ, Chang CC, Lu DW (2007) occlusion therapy for amblyopic children. Eye 16:150–155 Anisometropic amblyopia treated with spectacle correc- 45. Paysse EA, Coats DK, Hussein MA, Hamill MB, Koch DD tion alone: possible factors predicting success and time to (2006) Long-term outcomes of photorefractive keratectomy start patching. Am J Ophthalmol 143(1):54–60 for anisometropic amblyopia in children. Ophthalmology 31. Wallace DK, Chandler DL, Beck RW, Arnold RW, Bacal DA, 113(2):169–176 Birch EE, et al (2007) Treatment of bilateral refractive 46. Bhola R, Keech RV, Kutschke P, Pfeifer W, Scott WE (2006) amblyopia in children three to less than 10 years of age. Recurrence of amblyopia aft er occlusion therapy. Am J Ophthalmol 144(4):487–496 Ophthalmology 113:2097–2100 32. Wallace DK, Pediatric eye disease investigator group, 47. Holmes JM, Melia M, Bradfi eld YS, Cruz OA, Forbes B, Edwards AR, Cotter SA, Beck RW, Arnold RW, et al (2006) Pediatric eye disease investigator group (2007) Factors References 111

associated with recurrence of amblyopia on cessation of Findings from 1958 British birth cohort. BMJ 332(7545): patching. Ophthalmology 114(8):1427–1432 820–825 48. Tacagni DJ, Stewart CE, Moseley MJ, Fielder AR (2007) 60. Satterfi eld D, Keltner JL, Morrison TL (1993) Psychosocial Factors aff ecting the stability of visual function following aspects of strabismus study. Arch Ophthalmol 111: cessation of occlusion therapy for amblyopia. Graefes Arch 1100–1105 Clin Exp Ophthalmol 245(6):811–816 61. Coats DK, Paysse EA, Towler AJ, Dipboy RL (2000) Impact 49. Hertle RW, Scheiman MM, Beck RW, Chandler DL, Bacal of large angle horizontal strabismus on ability to obtain DA, Birch E, et al (2007) Stability of visual acuity improve- employment. Ophthalmology 107:402–405 ment following discontinuation of amblyopia treatment 62. Mojon-Azzi SM, Mojon DS (2008) Strabismus and employ- in children aged 7 to 12 years. Arch Ophthalmol 125(5): ment: the opinion of headhunters. Acta Ophthalmol epub 655–659 ahead of print 50. Webber AL, Wood JM, Gole GA, Brown B (2008) Th e 63. Mojon-Azzi SM, Potnik W, Mojon DS (2008) Opinions of eff ect of amblyopia on fi ne motor skills in children. Invest dating agents about strabismic subjects’ ability to fi nd a Ophthalmol Vis Sci 49(2):594–603 partner. Br J Ophthalmol 92:765–769 51. Hrisos Clarke S (2006) Unilateral visual impairment and 64. Uretmen O, Egrilmez S, Kose S, Pamukcu K, Akkin C, neurodevelopmental performance in preschool children. Palamar M (2003) Negative social bias against children Br J Ophthalmol 90(7):836–838 with strabismus. Acta Ophthalmol Scand 81:138–142 52. Stift er E, Burggasser G, Hirmann E, Th aler A, Radner W 65. Paysse EA, Steele EA, Brady McCreery KM, Wilhelmus (2005) Monocular and binocular reading performance in KR, Coats DK (2001) Age of the emergence of negative children with microstrabismic amblyopia. Br J Ophthalmol attitudes toward strabismus. J AAPOS 5:361–366 89:1324–1329 66. Jackson S, Harrad R, Morris M, Rumsey N (2006) Th e psy- 53. Choong YF, Lukman H, Martin S, Laws DE (2004) chosocial benefi ts of corrective surgery for adults with Childhood amblyopia treatment: psychosocial implica- strabismus. Br J Ophthalmol 90:883–888 tions for patients and primary carers. Eye 18:369–375 67. Nelson BA, Gunton KB, Lasker JN, Nelson LB, Drohan LA 54. Parkes LC (2001) An investigation of the impact of occlusion (2008) Th e psychosocial aspects of strabismus in teenagers therapy on children with amblyopia, its eff ect on their fami- and the impact of surgical correction. J AAPOS 12:72–76 lies, and compliance with treatment. Br Orthopt J 58:30–37 68. Burke J, Leech C, Davis H (1997) Psychosocial implica- 55. Williams C, Horwood J, Northstone K, Herrick D, Waylen tions of strabismus surgery in adults. J Pediatr Ophthalmol A, Wolke D, et al (2006) Th e timing of patching treatment Strabismus 34:159–164 and a child’s wellbeing.[see comment]. Br J Ophthalmol 69. Menon V, Saha J, Tandon R, Mehta M, Kokhar S (2002) 90(6):670–671 Study of psychosocial aspects of strabismus. J Pediatr 56. Koklanis K, Abel LA, Aroni R (2006) Psychosocial impact Ophthalmol Strabismus 39:203–208 of amblyopia and its treatment: a multidisciplinary study. 70. Archer SM, Musch DC, Wren PA, Guire KE, Del Monte MA Clin Exp Ophthalmol 34:743–750 (2005) Social and emotional impact of strabismus surgery 57. Packwood EA, Cruz OA, Rychwalski P, Keech RV (1999) Th e on quality of life in children. J AAPOS 9:148–151 psychosocial eff ects of amblyopia study. J AAPOS 3:15–17 71. van de Graaf ES, van der Sterre GW, Polling JR, van 58. Horwood J, Waylen A, Herrick D, Williams C, Wolke D, Kempen H, Simonsz B, Simonsz HJ (2004) Amblyopia and Avon longitudinal study of parents and children study team Strabismus Questionnaire: design and initial validation. (2005) Common visual defects and peer victimization in Strabismus 12:181–193 children. Invest Ophthalmol Vis Sci 46(4):1177–1181 72. UK National screening committee (2003) Criteria for 59. Rahi JS, Cumberland PM, Peckham CS (2006) Does appraising the viability, eff ectiveness and appropriateness amblyopia aff ect educational, health, and social outcomes? of a screening programme Chapter 9

The Brückner Test Revisited 9 Michael Gräf

Core Messages ■ Th e Brückner test is useful to detect various Any optically relevant opacity will be apparent by amblyogenic disorders. Aft er a short training, a shadow in the red refl ex. every physician can perform the test. ■ Detection of refractive error can be improved by ■ Th e test as originally described consists of four extending the test distance up to 4 m and observ- elements to observe: (1) the position of the fi rst ing the brightness of the red refl ex in both eyes Purkinje images (corneal light refl exes), (2) the simultaneously. While usually at a distance of 1 m, fundus red refl ex in the pupil, (3) pupillary light the red refl ex is brighter in the more ametropic refl exes, and (4) any movement of the eyes when eye, the refl ex in this eye becomes increasingly illumination alters from one eye to the other. darker with increasing test distance. With increas- ■ Asymmetry in corneal light refl exes on both eyes ing test distance, myopia and hypermetropia, may indicate strabismus. However, small devia- which are not compensated by accommodation, tions are not reliably detected, and asymmetry cause signifi cant dimming, and anisometropia can also be caused by diff erent angle kappa in causes increasing asymmetry. both eyes. ■ Th e test sensitivity to detect microstrabismus by ■ Performance of the red refl ex test requires a direct asymmetric fundus red refl ex is low. ophthalmoscope. Substitution by an otoscope, ■ Testing pupillary light refl exes is recommendable indirect ophthalmoscope, or any other light to assess visual aff erence, pupillomotor eff erence source causes loss of test validity. and pupil responsiveness. It is hardly suitable to ■ Th e red refl ex test allows for detection of refrac- diagnose or exclude amblyopia and amblyogenic tive error, strabismus and organic disorders such disorders. as opacities of the optic media and distinct ■ Testing for fi xation movements caused by switch- pathologies of the fundus. ing illumination from one eye to the other is sim- ■ Media opacity is easily detected at a test distance ilar to the cover test. Data on diagnostic validity of 0.3 m and less, examining each eye separately. of this procedure are lacking.

9.1 Amblyopia and Amblyogenic Disorders 9.1.1 Early Detection of Amblyopia Amblyopia is estimated to aff ect approximately 2–5% of Early detection of amblyopia and amblyogenic factors the population in Western countries and is a signifi cant requires objective methods that are independent of any preventable cause of vision loss in children and adults verbal response of the child. Refractive error and strabis- [1–8]. Amblyogenic risk factors include ptosis, media mus are the most frequent causes of amblyopia. So, meth- opacity, fundus pathologies, strabismus and refractive ods are necessary that indicate ametropia and strabismus error [9–11]. When these risk factors are detected at an with a high sensitivity and specifi city. Refractometry or early age, amblyopia can be prevented or minimized more retinoscopy in cycloplegia is the most reliable way to eff ectively [3, 12–14]. One signifi cant limiting factor of detect and measure ametropia in childhood. However, most amblyopia screening programs is the reliance on the this requires experience of the examiner and the possibil- subjective responses of the child being tested. ity to perform both cycloplegia and measurement. Th ese 114 9 The Brückner Test Revisited

conditions as well as parental readiness are oft en lacking. the lids and the root of the nose. In infants and toddlers, Non-cycloplegic photorefractive screening is not a tanta- as well as in Asians, epicanthus which is nasally covering mount substitute of refractometry in cycloplegia [15, 16]. the lid fi ssure can be suggestive of esotropia. Besides, the technical equipment is relatively expensive, 9 and therefore hardly any paediatrician or general practi- tioner performs photorefractometry. Even the Brückner 9.2.1 Physiology test is not routinely used by paediatricians, although pre- conditions for performance are ideal and the test is rec- Purkinje described that when the eye is being illumi- ommended for paediatric screening examinations in nated by an examination light, refl exes appear from the Germany [17]. Th e Brückner test is a readily available corneal surface, the corneal endothelium, and both the screening tool that can be used with newborns, infants anterior and posterior surface of the lens. Th e fi rst and preverbal children by non-ophthalmologists [18, 19]. Purkinje image coming from the corneal tear fi lm is Th e test requires not more than a direct ophthalmoscope brightest. Usually it appears slightly nasally of the centre and only few seconds for performance. of the cornea and the pupil, when the eye is fi xating a light source which is held directly below the pupil of the observer. Slight eccentricity of the corneal light refl ex is caused by the diff erence between the visual line and the 9.1.2 Brückner’s Original Description pupillary axis, the angle k, which is similar to the angle g In 1962, Roland Brückner (1912–1996), an ophthalmolo- [21]. When the eye turns in a distinct direction, the gist in Basel, Switzerland, reported on ‘Exact strabismus position of the corneal light refl ex relative to the pupil diagnostic in ½- to 3-year-old children by a simple proce- will shift to the opposite direction. Conjugate gaze dure, the “transillumination test” ’ [18]. Brückner illumi- movements induce parallel shift of the images in both nated both pupils from a distance of 1 m and assessed the eyes. Th is causes asymmetry in the two images, if their following criteria: positions were symmetric at fi rst. For instance, right gaze induces nasal shift of the image in the right eye and ■ Position of fi rst Purkinje images relative to the pupil temporal shift of the image in the left eye. Th e same will ■ Colour of the fundus red refl ex in the pupil happen, when the light source is moved to the right- ■ Size and constriction of the pupils hand side from the observer’s point of view or when the ■ Eye movements with and illumination of the pupils observer assesses the image position from left -hand side beside the light source. Non-conjugate eye movements Assessment of the fi rst two criteria requires simultaneous or manifest strabismus cause a non-parallel shift or illumination of both eyes, whereas assessment of the fol- position, respectively, of the images on both eyes. For lowing two criteria requires alternate illumination. Th ree instance, when the left eye fi xates the light and the right years later, Brückner added an article on ‘Practical exer- eye is esotropic, then the fi rst Purkinje image on the cises with the transillumination test for early diagnosis of right eye will be temporally dislocated. So, this method strabismus’, emphasizing the essential component of the in principle allows for detection of strabismus. test, which is the assessment of the red refl ex of the fun- Th e idea to measure squint angles by using corneal dus when the pupil is lighted and viewed with a direct light refl exes arose at the end of the nineteenth century ophthalmoscope [19]. Th is particular component was [22, 23]. Hirschberg assumed that 1-mm shift of the cor- new concerning strabismus diagnostic and in the aft er- neal light refl ex corresponded to an angle of 7° by which math called Brückner test in the closer sense. It has also the eye is turned [22]. At the end of the twentieth century, been called the Brückner refl ex [3, 20]. empiric studies proved that within the range of small and moderate deviation the correct ratio is 12°/mm [10, 24, 25]. Nevertheless, up to the twenty-fi rst century, the wrong ratio of 7°/mm is still wide spread. Recognition of asymmetry in the Purkinje images can be improved by evalu- 9.2 Corneal Light Refl exes ating photographs [26]. In laboratory trials, photographic (First Purkinje Images) Hirschberg testing was eff ective in approximately 80% of Assessment of the fi rst Purkinje images in the two eyes cases in detecting a deviating eye in strabismus of about 5 allows for more exact strabismus diagnostic than mere prism dioptres [27]. Regarding more accurate diagnostic, assessment of the position of the cornea within the palpe- the alteration of relative position of the fi rst and the fourth bral fi ssure. Th e latter depends on the confi guration of Purkinje images due to deviation of the visual axis have 9.3 Fundus Red Refl ex (Brückner Refl ex) 115 been studied [11, 28–32]. However, the fourth Purkinje image is not visible clearly enough by performing the Brückner test.

9.2.2 Performance Fig. 9.1 Corneal light refl exes in a 12-month-old girl. In this Assessment of the corneal light refl ex for symmetry on case, asymmetry of the corneal light refl ex between both eyes is both eyes requires a small light source, which must be caused by fl ashlight position beside the objective of the camera. fi xated by the patient. To avoid glaring the patient, the So, the image of the fl ashlight on the right eye is more and the image on the left eye is less nasally decentred. At 9 o’clock in light should not be too bright. Th e observer compares front of both pupils, images of a window the position of the corneal refl ex images in the two eyes in relation to the pupils. Physiologically, the images appear approximately 0.5 mm nasal to the centre of the Summary for the Clinician pupil. Th e eccentricity depends on the individual angle k. ■ Th e images may be better visible when the observer Evaluating the corneal light refl exes in both eyes looks above the ophthalmoscope. Th en the pupils for symmetry allows to detect manifest strabis- appear black and there is more luminance contrast of mus and to estimate its size. Exclusion of strabis- the images. If the iris is dark brown with low contrast to mus is impossible because slight asymmetry the black pupil, looking through the ophthalmoscope is corresponding to small squint angle can hardly advantageous. Favourite test distances are around be recognized and asymmetry in the angles k in 0.5 m. Closer test distance may cause defence in chil- both eyes can both, simulate or mask strabismus. dren and also adequate convergence might not be war- Bias occurs when the patient fi xates a point ranted. Larger distance makes it diffi cult to detect small beside the examination light or when the light is asymmetry. not on the examiner’s visual line.

9.2.3 Shortcomings and Pitfalls 9.3 Fundus Red Refl ex (Brückner Refl ex) False-negative fi ndings are likely in case of small squint Performing the ‘transillumination’ test requires a direct angle. Since misalignment of 6° corresponds to not more ophthalmoscope. Looking through the ophthalmoscope, than 0.5 mm asymmetry in the position of the corneal the examiner can see the patient’s pupil shining red, light refl exes, it is evident that small angle strabismus can caused by the light refl ected by the choroid and the retinal hardly be identifi ed by this method. Asymmetry in the surface of the eye. Th e fundus refl ex was also called angle k between both eyes can veil strabismus. Brückner refl ex [3, 20]. Colour and brightness of the fun- Ectopia and anomalies of the pupil have to be consid- dus refl ex depend on brightness of the examination light, ered. False-positive fi nding of strabismus can be caused consistence and refractive quality of the optical media, by parallel shift of the refl ex images in the two eyes when pigmentation of the fundus and refractive state of the eye. the light is horizontally displaced. Th e light source must Any opacity of the optic media causes an abnormally dark be exactly beneath (not beside!) the visual axis of the or lacking red refl ex in the region of the opacity. Slight observer’s fi xating eye. Severe bias occurs when the light nuclear cataract may be visible by a darker ring, which is is hold under one eye while the other eye is fi xating: Taken caused by the equator of the nucleus (Fig. 9.2). Posterior the angle k were equal in both eyes, the interpupillary pole cataract causes a black shadow in the centre of the distance were 60 mm, and the examination distance were pupil. Frequently, a very small shadow is visible nasally 0.5 m, then the resulting asymmetry would correspond to below the centre of the pupil as the correlate of Mittendorf’s 12°. A similar mistake occurs by evaluating fl ashlight spot. With eye movement these shadows move to the photographs, which were recorded with the fl ashlight opposite direction within the pupil while shadow caused beside the objective (Fig. 9.1). With the fl ashlight coaxi- by corneal opacity or anterior cataract will move to the ally or above the objective, this bias can be avoided, but it same direction. An examiner who is familiar with the cannot be assured that the child was really fi xating the Brückner test will probably detect every optically relevant camera [33]. cataract, albeit we are not aware of any scientifi c study on 116 9 The Brückner Test Revisited

Fig. 9.2 Visualization of organic pathologies in the fundus refl ex test. Top (better left ), nuclear cataract OS>OD. OD, beginning cataract visible by a dark ring corresponding to the equator of the lens. OS, advanced cataract causing signifi cant central shadow. Bottom (better right), large peripheral retinoblastoma OS already visible by partial leukocoria when looking above the ophthalmoscope. Both examples show that organic fi ndings are better visible with magnifi cation by shorter distance compared to “armlength” distance

the sensitivity of the Brückner test to detect media opac- 9.3.1 Physiology ity. Visualizing media opacity and pathologies of the fundus the Brückner refl ex is extremely important for Examination of the fundus red refl ex can roughly be paediatricians, general practitioners and others who are compared with direct ophthalmoscopy performed at a not equipped to perform slitlamp biomicroscopy and large distance so that only very small part of the fundus is indirect ophthalmoscopy. Abnormally, bright, white or visible. Provided central fi xation of the patient, the fun- dark fundus refl ex can also be caused by the optic nerve dus red refl ex represents the patient’s fovea. To explain head and by pathologies of the fundus, such as coloboma, the dimming of the red refl ex when the patient takes up retinoblastoma, toxoplasmosis scars and medullated fi xation, Brückner discussed various factors [18]. nerve fi bres. Pupillary constriction, diff erent refl ectivity of the central When the patient takes up central fi xation of the oph- and peripheral retinal surface and accuracy of accommo- thalmoscope light, there is normally a constriction of the dation were assumed to be the major causes of dimming pupils and dimming of both fundus refl exes [18]. By and change in colour [18, 35–37]. Backscattering of the interfering with this dimming phenomenon, manifest light by the retinal nerve fi bre layer proportional to the strabismus and anisometropia can produce asymmetry in thickness of the layer and changes arising from variation the brightness and colour of the fundus refl exes in both in retinal pigment epithelium density, with the retina dis- eyes. Brückner stressed the point that strabismus could playing the characteristics of a diff use refl ector, were fur- be reliably detected by this asymmetry. Traditionally, the ther discussed but not as primary factors of dimming deviated or more ametropic eye was described to have the [35]. Mere pupillary constriction does not explain asym- brighter refl ex [9, 18]. Regarding ametropia, however, metric dimming due to strabismus, but it may amplify examination distance is a decisive factor. At larger dis- eff ects of defocus and retinal refl ectivity. Brückner’s idea tance, the more ametropic eye yields the darker fundus that diff erence in refl ectivity between the central and refl ex [34]. para-central or peripheral retinal surface contribute to 9.3 Fundus Red Refl ex (Brückner Refl ex) 117 the dimming phenomenon was refreshed by Roe and If an eye is deviated, off -axis optical aberrations will Guyton who described specular refl ection of the retina decrease the conjugacy of the ophthalmoscope light and from the internal limiting membrane that changes slope the retina. If the fovea is not exactly conjugate to the light with ocular rotation [35, 36]. Th e fundus refl ex is not source, the light from the retina spills passed the light solely caused by refl ection from the choroid and the reti- source into the examiner’s eye, increasing the brightness nal pigment epithelium but, to a minor part, also by of the red refl ex [35, 36]. Th is hypothesis might fi t with refl ection from the retinal surface. If signifi cant light were the observation that – at the traditional examination dis- refl ected from the internal limiting membrane of the ret- tance of 1 m – the fundus red refl ex in the (more) ame- ina, the slope of the foveal pit would refl ect enough light tropic eye is usually brighter compared to an emmetropic away from the pupil. Because this part of light would not eye. Th e hypothesis corresponds to the assumption that be refl ected back to the observer, the red refl ex would accuracy of accommodation is one reason of dimming. appear darkened [35, 36]. Misalignment of one eye with Foveal dimming of the red refl ex allows for sensitive light being refl ected from the para-foveal retinal surface, discrimination between subsequent central and eccentric which is rather perpendicular to the direction of the illumination of the same eye. Dimming occurred in 97.2% incoming light, increases coaxial refl ection and thus the of trials with fi xation of the light compared with fi xation of brightness of the fundus refl ex (Fig. 9.3). a target between 2.5 and 10° beside the light, regardless of Th is might also explain the lack of dimming in new- the angle of eccentricity. Th is rate did not decrease when borns and young infants as a consequence of develop- the pupil was dilated by mydriatic eye drops (Gräf et al., ment of the foveal pit. While most infants 8 months of age MS in preparation). However, the static inter-ocular dif- and older show dimming of the fundus refl exes in both ference in the refl exes due to strabismus was less apparent. eyes occurring with central fi xation, neonates and most In young adults, simulated esotropia with squint angles up infants younger than 2 months of age do not show dim- to 5° was detected in not more than 62%. Th e deviated eye ming of the fundus refl ex with fi xation and between 2 and was identifi ed by the brighter red refl ex in 48%. Esotropia 8 months of age up to 28% of infants have asymmetric of 7.5 and 10° was detected in 85 and 97% with identifi ca- dimming of the fundus refl exes in the two eyes [9]. So, in tion of the deviated eye in 75 and 86% (Table 9.1). To newborns and young infants, asymmetry may represent a achieve these rates, very discreet red refl ex asymmetry normal stage of development and symmetry does not was considered. Th e rate of false-positive fi ndings was exclude strabismus. 36% (Gräf et al., MS in preparation). Th ese results con- Another mechanism might be off -axis aberration fi rm prior fi ndings [38]. When esotropia of, for example, resulting in poor image formation on the retina. Roe and 8 prism dioptres was simulated by fi xating a near target, Guyton believed the fundus refl ex would appear darker in not more than two thirds of strabismus conditions were an eye that is fi xating and focusing on the ophthalmo- detected [27]. One might argue that these were only labo- scope light because the light source in the ophthalmo- ratory studies, but an increase in sensitivity and specifi c- scope and its retinal image are conjugate to one another. ity in young children compared with highly cooperative

200 µm

Fig. 9.3 Optic coherence tomography (spectralis OCT) of the normal central fundus. Part of the light is already refl ected from the surface of the retina. Due to the slope of the foveal pit part of the light is refl ected away from the pupil. Th is might in part explain that the red refl ex darkens when the patient takes up central fi xation of the ophthalmoscope light 118 9 The Brückner Test Revisited

Table 9.1. Results of red reﬂ ex test in simulated esotropia and orthotropia (control condition)

Simulated esotropia Number of trials Test negative (%) Test positive (%) Correct localization (%)

Esotropia 2–5° 100 38 62 48 9 Esotropia 7.5° 100 15 85 76 Esotropia 10° 100 3 97 86 Orthotropia 300 64 36 – Test negative symmetric red refl ex; test positive inter-ocular asymmetry in red refl ex; correct localization brighter red refl ex in the deviated eye Gräf et al., (in preparation)

adults is rather unlikely. Strabismus detection will hardly distance between the observer and the patient, the por- improve by extending the test distance, except indirectly, tion of the refl ected light bundle reaching the observer’s by detection of anisometropia which frequently accom- pupil decreases. So, when the observer moves back- panies esotropia [35]. Th ere might be a chance to improve wards, the brighter refl ex, which at a distance of 1 m, test sensitivity and specifi city by using a short-pass fi lter usually corresponds to the (more) ametropic eye, that blocks the refl exes coming from the retinal pigment becomes darker (Fig. 9.4) [34]. Th e test sensitivity to epithelium and the choroid and thus augments asymme- detect unilateral refractive error by the weak refl ex in try caused by asymmetric light refl ection from the inter- the ametropic eye at a test distance of 4 m is better com- nal limiting membrane. pared with the traditional test at a distance of 1 m or less Considering optical basics, examination distance [30]. Using a direct ophthalmoscope, unilateral myopia must be an essential factor infl uencing the red refl ex in of 1–4 diopters was detected in 60–82% of trials at 1 m case of refractive error. Uncorrected ametropia causes but in 100% of trials at 4 m (Table 9.2). Unilateral hyper- defocus of the retinal image of the light source. On the metropia of 1–4 diopters was detected in 34–80% of tri- way back to the observer, this image is projected through als at 1 m but in 52–98% of trials at 4 m. Compared with the pupil. A myopic eye focuses the light beams at the experts, results of students were weaker at 1 m but far point of the eye. Beyond the far point, the light bun- equivalent at 4 m [34]. Th e low rate of false-positive dle is divergent. In case of hypermetropia, which is not fi ndings shows that rather discreet asymmetry was not compensated by accommodation, the light beams depart considered pathologic in that study, in contrast to the the eye as a primarily divergent bundle. With increasing study on simulated strabismus, (Fig. 9.5).

Fig. 9.4 Anisometropia of 5 dioptres (emmetropia OD, hypermetropia OS). Fundus red refl ex recorded at distances of 1 m (top) and 4 m (bottom). Th is amount of anisometropia causes red refl ex asymmetry already at the traditional distance with the refl ex from the more ametropic eye being somewhat brighter. At the extended distance the red refl ex of the (more) ametropic eye is much darker 9.3 Fundus Red Refl ex (Brückner Refl ex) 119

Table 9.2. Sensitivity (50 trials for each condition) and false-positive ﬁ ndings (in 225 trials) of the Brückner reﬂ ex to detect unilateral spherical ametropia [34]

Simulated unilateral Experts 1 m (%) Experts 4 m (%) Students 1 m (%) Students 4 m (%) ametropia

Hypermetropia 1 diopters 34 52 8 60 Hypermetropia 2 diopters 58 94 40 100 Hypermetropia 3 diopters 76 96 56 100 Hypermetropia 4 diopters 80 98 64 100 Myopia 1 diopters 60 100 32 68 Myopia 2 diopters 80 100 28 100 Myopia 3 diopters 74 98 40 100 Myopia 4 diopters 82 100 36 100 False-positive tests 3.1 4.0 1.5 3.0

Results for unilateral astigmatism showed also the [34]. Th ese rates that depend on patient selection and higher detection rates at 4 m distance (Table 9.3). observer experience are not representative for a real On the basis of these results, it is recommendable to per- screening situation in early infancy. form the test also at a distance of 4 m to detect refractive error more sensitively [34]. Paysse et al. compared the ability of paediatric resi- 9.3.2 Performance dents to diff erentiate asymmetric from symmetric red refl ex in ten patients and six control subjects. Four It is commonly recommended to perform the test at a dis- patients were anisometropic by 2.25–5.5 dioptres without tance of about 1 m or less (‘arm’s length distance’) by strabismus. In the entire group, paediatric residents simultaneously illuminating both eyes of a patient, and to achieved a test sensitivity of 61% and a specifi city of 71% compare colour and brightness of the pupillary red [3]. Gole and Douglas reported a test sensitivity of 86% refl exes for symmetry [3, 18, 19, 35, 37, 38, 40, 41]. Th e and a specifi city of not more than 65%. Th e Brückner test room light should be dimmed but the room should not be was performed by a medical student [20]. In these two completely dark [18]. studies, the test distance was 1 m. In a group of anisome- Using a direct ophthalmoscope is mandatory. Otoscope tropic patients, we achieved a sensitivity of 32.5% at that or fl ashlight illumination will not yield the same optical distance, and a specifi city of 93.3%. At a distance of 4 m, phenomena because the characteristic of the emitted light sensitivity increased to 77.5% and specifi city was 80% is diff erent. Th e light beam must be directed simultaneous

Table 9.3. Sensitivity (50 trials for each condition) and false-positive ﬁ ndings (in 400 trials) of the Brückner reﬂ ex to detect unilateral astigmatismus simplex [30]

Simulated astigmatism With the rule Against rule With the rule Against rule 1 m (%) 1 m (%) 4 m (%) 4 m (%)

Hypermetropic 1 diopters 44 44 62 46 Hypermetropic 2 diopters 58 60 88 72 Hypermetropic 3 diopters 76 66 100 82 Hypermetropic 4 diopters 88 72 100 100 Myopic 1 diopters 50 22 44 74 Myopic 2 diopters 60 48 74 98 Myopic 3 diopters 60 70 86 100 Myopic 4 diopters 70 80 92 100 False-positive tests 5.5 5.25 120 9 The Brückner Test Revisited

into both eyes to enable accurate comparison of the red reﬂ exes. Light intensity can be varied during the examination. It should not be too high to avoid glare. Detection of small media opacity is easier at 0.5–0.1 m, examining each a 9 eye separately and using a convex lens in the ophthalmoscope, if necessary. To improve detection of refractive error, the examiner should then go 4 m backwards continuously observing the pupils simultaneously for luminance of the red reﬂ exes. By the same way, it is possible to check for correct spectacle correction. b

9.3.3 Possibilities and Limitations Severe media opacity is visible by lacking or dark fundus red refl ex, regardless of distance. Small media opacity and pathologies of the fundus are best visible at a c short distance. Any asymmetry in brightness and colour of the refl exes is predictive of amblyogenic risk factors [18, 19, 37, 38]. While the test is very sensitive to detect media opacity, detection of small-angle strabismus is limited. Detection of ametropia (particularly myopia) and anisometropia can be improved by extending the test distance, but nev- d ertheless, isometropic hypermetropia cannot be detected reliably. Inter-ocular asymmetry in the red refl ex can be caused by anisocoria and is also frequent in the age range below 8 months.

Summary for the Clinician e ■ Th e fundus red refl ex test is an excellent complement and a possibility for ophthalmologists, orthoptists, paediatricians, and general practitioners to recognize various eye disorders very early. It is also a valuable tool for use in developmental countries. At a short distance relevant media opacity can be reliably detected by darkening of f the red refl ex. Testing for refractive error is better performed at an extended distance. Uni- or bilateral partially or completely weak or lacking red refl ex is always pathological. Fig. 9.5 Brückner refl ex at 4 m distance in case of emmetropia OU (a) and simulated hypermetropia OS (anisometropia) of 1 diopter (b), 2 diopters (c), 3 diopters (d), and 4 diopters (e). For comparison, simulated myopia OS of 1 diopter (f) [34] 9.4 Pupillary Light Refl exes Pupillary light refl exes are being tested to detect patholo- 1. Dimming of the red refl ex when the child is centrally gies in the iris, in the eff erent branch of the pupillary light fi xating the ophthalmoscope light. refl ex loop, in the midbrain, or in the aff erent branch of 2. Reduced direct pupillary light refl ex in the amblyopic the refl ex loop. Regarding strabismus diagnostic, Brückner eye compared with the direct light refl ex in the non- described two criteria: amblyopic eye. 9.4 Pupillary Light Refl exes 121

Th is step requires monocular illumination of the pupils. has a stronger pupillomotor eff ect compared with para- Brückner reported pupillary constriction in the deviated central illumination. Pupillary constriction is also induced eye when the light beam was changed from the fi xating by the increased light sensitivity of the ‘dark-adapted’ eye. eye onto the strabismic eye, as soon as this eye took up In the clinical situation, it is hardly possible to discriminate fi xation. Permanent fi xation with the previously illumi- between these two mechanisms. If the strabismic eye fails nated dominant eye yields an eccentric retinal image of to take up central fi xation, an aff erent pupillomotor defect the ophthalmoscope light in the deviated eye. Despite component may be simulated when this eye is being illumi- dark adaptation of the deviated eye, the pupillomotor nated or there is in fact a relative aff erent pupillary defect eff ect of the eccentric illumination can be weaker than (RAPD) due to amblyopia [47–50]. Figure 9.6 shows that that of the central illumination in the fellow eye. So, the already minimal eccentricity of illumination reduces pupil- response to alternating illumination may either look like lary constriction compared with a central illumination. relative aff erent pupillomotor defi cit or dimming of the red refl ex in the amblyopic eye occurs aft er some latency when the amblyopic eye takes up fi xation. 9.4.2 Performance Th e examiner directs the light cone on the patient’s right 9.4.1 Physiology eye and observes constriction of each pupil. Th e procedure is repeated illuminating the patient’s left eye. If both Illumination of one eye causes symmetric constriction of pupils are normally reactive, which is mostly the case, both pupils [42–46]. In unilateral amaurosis, pupillary con- comparison of the direct light refl exes of both eyes will be striction is lacking in both eyes when only the blind eye is suffi cient [51–52]. If only one pupil is reactive, this pupil being illuminated. Illumination of the other eye causes can be used to compare the constriction with subsequent normal constriction of the pupils in both eyes. Less severe illumination of the right and the left eye. Th e pupillary aff erent disorders show a similar pattern except residual constriction has to be equal in latency, speed and ampli- reaction to illumination of the (more severely) concerned tude, regardless of the eye illuminated. eye. Discreet aff erent disorders can be found by the swinging fl ash light test [42–44]. Aiming at strabismus diagnostic, the observer has to watch any eye movement occurring 9.4.3 Possibilities and Limitations aft er the change of the illumination to the other eye. If the previously deviated eye which is now being illuminated RAPD is typical of severe asymmetric retinal lesion or takes up fi xation, the movement of this eye may be visible, asymmetric lesion of the optic nerve including the optic and the pupils will constrict because foveal illumination chiasm. Amblyogenic disorders, such as refractive error,

gaze direction

10º

5º

0º

–5º Fig. 9.6 Video-oculographic registration of the change in –10º pupil diameter with alternating ﬁ xation of the pupil diameter ophthalmoscope light and low illuminated visual 6 mm targets 2.5, 5, 7.5, and 10° right (positive values) and 4 mm left (negative values) of the ophthalmoscope light. 2 mm Fixation of the ophthalmoscope light induced more 0 mm pupillary constriction than ﬁ xation of a target as few as 0 5 10 15 20 25 30 35 40 45 50 55 60 2.5° beside time / seconds 122 9 The Brückner Test Revisited

media opacity or any other pre-retinal disorder, gener- mination test allows for detection of refractive ally do not cause an apparent RAPD. Th ompson reported error, particularly at an extended test distance. that a careful look revealed small RAPD in less than half Nevertheless, reliable detection of amblyogenic of amblyopic eyes. Th is defect was generally less than 0.5 ametropia requires refractometry or retinoscopy 9 log units [46], and the size of possible RAPD did not cor- in cycloplegia. relate well with the visual acuity of the amblyopic eye [47–50]. Regarding strabismus diagnostic it may be an advantage that children usually look directly to the light. Manifest strabismus may be detected by the eye move- References ment when illumination changes from one eye to the other and the child changes fi xation. However, it is hardly 1. Ehrlich MI, Reinecke RD, Simons K (1983) Preschool possible to detect strabismus by RAPD. vision screening for amblyopia and strabismus. Programs, methods, guidelines. Surv Ophthalmol 28:145–163 Summary for the Clinician 2. Flynn JT (1991) Amblyopia revisited. J Pediatr Ophthalmol ■ Severe unilateral amblyopia might be detected Strabismus 28:183–201 by RAPD in the amblyopic eye, but usually, the 3. Paysse EA, Williams GC, Coats DK, Williams EA (2001) pupillary light refl exes are hardly suitable to Detection of red refl ex asymmetry by pediatric residents detect strabismus or amblyopia. using the Brückner refl ex versus the MTI photoscreener. Pediatrics 108:E74 4. Tychsen L (1992) Binocular vision. In: Hart WM Jr (ed) 9.5 Eye Movements with Alternating Adler’s physiology of the eye: clinical application. Mosby, Illumination of the Pupils St Louis, pp 837–838 Provided central fi xation and absence of strabismus, 5. Rahi J, Logan S, Timms C (2002) Risk, causes and out- alternation of illumination to the other eye should not comes of visual impairment aft er loss of vision in the non- elicit any gaze movement. In case of manifest strabismus, amblyopic eye. Lancet 360:597–602 there may be a movement of the illuminated eye from its 6. Rahi J S, Logan S, Borja MC, Timms C, Russell-Eggitt, previous strabismic position towards the light, together Taylor D (2002) Prediction of improved vision in the with a conjugate movement of the other eye. However, in amblyopic eye aft er visual loss in the non-amblyopic eye. case of severe amblyopia or uniocular dominance, this Lancet 360:621–622 movement may be lacking. If the angle of eccentric fi xa- 7. Van Leeuwen R, Eijkemans MJ, Vingerling JR, Hofman A, tion is identical with the angle of abnormal retinal corre- de Jong PT, Simonsz HJ (2007) Risk of bilateral visual spondence, there will also be no fi xation movement impairment in individuals with amblyopia: the Rotterdam [53, 54]. Th ese patterns are well known from cover test- study. Br J Ophthalmol 91:1450–1451 ing. Since it is possible that the child keeps fi xation of the 8. Webber AL, Wood JM, Gole GA, Brown B (2008) Th e ophthalmoscope with the dominant eye because this eye eff ect of amblyopia on fi ne motor skills in children. Invest is not occluded, cover testing is safer and certainly more Ophthalmol Vis Sci 49:594–603 sensitive to detect strabismus. 9. Archer SM (1988) Developmental aspects of the Brückner test. Ophthalmology 95:1096–1101 10. Barry JC (1999) Hier irrte Hirschberg: Der richtige Winkelfaktor beträgt 12°/mm Hornhautrefl exdezentrierung. Summary for the Clinician Geometrisch-optische Analyse verschiedener Methoden der Strabismometrie. Klin Monatsbl Augenheilkd 215: 104–113 ■ Brückner’s transillumination test allows for very 11. Barry JC, Eff ert R, Hoff mann N (1996) Detection and diag- sensitive detection of media opacity. Th erefore, nosis of small ocular misalignment with the Purkinje refl ex every ophthalmologic examination in early pattern method. Klin Monatsbl Augenheilkd 208:167–180 childhood should include the Brückner test. Th e 12. Coats D, Jenkins R (1997) Vision assessment of the pediat- test is suitable to detect strabismus but it does ric patient. Refi nements. Am Acad Ophthalmol 1:1 not allow for suffi cient detection of small angle 13. Donahue SP (2006) Relationship between anisometropia, strabismus. Reliable strabismus diagnostic in patient age, and the development of amblyopia. Am childhood requires additional cover testing and J Ophthalmol 142:132–140 testing of random dot stereopsis. Th e transillu- 14. Sjöstrand J, Abrahamsson M (1990) Risk factors in amblyopia. Eye 4:787–793 References 123

15. Ehrt O, Weber A, Boergen KP (2007) Screening for refrac- 31. Eff ert R, Barry JC, Colberg R, Kaupp A, Scherer G (1995) tive errors in preschool children with the vision screener. Self-assessment of angles of strabismus with photographic Strabismus 15:13–19 Purkinje I and IV refl ection pattern evaluation. Graefes 16. Schaeff el F, Mathis U, Brüggemann G (2007) Noncycloplegic Arch Clin Exp Ophthalmol 233:494–506 refractive screening in pre-school children with the 32. Eff ert R, Barry JC, Dahm M, Kaupp A (1991) A new pho- “power-refractor” in a pediatric practice. Optom Vis Sci tographic method for measuring squint angles in infants 84:630–639 and small children. Klin Monatsbl Augenheilkd. 198: 17. Zentralinstitut für kassenärztliche Versorgung (1991) 284–289 Hinweise zur Durchführung der Früherkennungsunter- 33. Becker R, Gräf M (2006) Systematische Fehler bei der suchungen im Kindesalter. Deutscher Ärzte Verlag, Köln fotografi schen Beurteilung der Hornhautspiegelbilder. 18. Brückner R (1962) Exakte Strabismusdiagnostik bei 1/2- Klin Monatsbl Augenheilkd 223:294–296 bis 3jährigen Kindern mit einem einfachen Verfahren, dem 34. Gräf M, Jung A (2008) Th e Brückner test: extended dis- “Durchleuchtungstest”. Ophthalmologica 144: 184–198 tance improves sensitivity for ametropia. Graefes Arch 19. Brückner R (1965) Praktische Übungen mit dem Clin Exp Ophthalmol 246:135–141 Durchleuchtungstest zur Frühdiagnose des Strabismus. 35. Roe LD, Guyton DL (1984) Perspectives in refraction. Surv Ophthalmologica 149:497–503 Ophthalmol 28:405–408 20. Gole GM, Douglas LM (1995) Validity of the Brückner 36. Roe LD, Guyton DL (1984) Th e light that leaks: Brückner refl ex in the detection of amblyopia. Aust N Z J Ophthalmol and the red refl ex. Surv Ophthalmol 28:655–670 23:281–285 37. Tongue AC, Cibis GW (1981) Brückner test. Ophthalmology 21. Kaufmann H (1995) Störungen des Binokularsehens. 88:1041–1044 Terminologie. In Kaufmann H (ed) Strabismus. Enke, 38. Griffi n JR, Cotter SA (1986) Th e Brückner test: evaluation Stuttgart, pp 162–165 of clinical usefulness. Am J Optom Physiol Opt 63: 22. Hirschberg J (1886) Beiträge zur Lehre vom Schielen 957–961 und von der Schieloperation. Zbl prakt Augenheilkd 39. Leff ertstra LJ (1977) Vergleichende Untersuchungen auf 10:5–9 unterschiedliche Refraktionsänderungen beider Augen bei 23. Smith P (1892) On the corneal refl ex of the ophthalmo- Patienten mit Strabismus convergens. Klin Monatsbl scope as a test of fi xation and deviation. Ophthalmic Rev Augenheilkd 170:74–79 11:37–42 40. Carrera A, Saornil MA, Zamora MI, Maderuelo A, 24. Brodie SE (1987) Photographic calibration of the Canamares S, Pastor JC (1993) Detecting amblyogenic dis- Hirschberg test. Invest Ophthalmol Vis Sci 28:736–742 eases with the photographic Brückner test. Strabismus 25. DeRespinis PA, Naidu E, Brodie SE (1989) Calibration of 1:3–9 Hirschberg test photographs under clinical conditions. 41. Noorden GKv, Campos EC (2002) Binocular vision and Ophthalmology 96:944–949 ocular motility. Mosby, St Louis 26. Kaakinen K, Tommila V (1979) A clinical study on the 42. Levatin P (1959) Pupillary escape in disease of the retina or detection of strabismus, anisometropia or ametropia of optic nerve. Arch Ophthalmol 62:768–779 children by simultaneous photography of the corneal and 43. Loewenfeld IE (1993) Th e pupil. Wayne State University, the fundus refl exes. Acta Ophthalmol 57:600–611 Detroit 27. Griffi n JR, McLin LN, Schor CM (1989) Photographic 44. Miller, NR (1995) Walsh and Hoyth’s clinical neuro- method for Brückner and Hirschberg testing. Optom Vis ophthalmology, Vol. I-V. Williams and Wilkins, Baltimore Sci 66:474–479 45. Miller JM, Leising Hall H, Greivenkamp JE, Guyton DL 28. Barry JC, Eff ert R, Kaupp A (1992) Objective measurement (1994) Quantifi cation of the Brückner test for strabismus. of small angles of strabismus in infants and children with Invest Ophthalmol Vis Sci 36:897–905 photographic refection pattern evaluation. Ophthalmology 46. Th ompson HS (1992) Th e pupil. In Hart WM Jr (ed) Adler’s 99:320–328 physiology of the eye. Mosby, St. Louis, pp 412–441 29. Barry JC, Eff ert R, Kaupp A, Burhoff A (1994) Measurement 47. Firth AY (1990) Pupillary responses in amblyopia. Br J of ocular alignment with photographic Purkinje I and IV Ophthalmol 74:676–680 refl ection pattern evaluation. Invest Ophthalmol Vis Sci 48. Greenwald MJ, Folk ER (1983) Aff erent pupillary defects 35:4219–4235 in amblyopia. J Pediatr Ophthalmol Strabismus 20: 30. Barry JC, Eff ert R, Reim M, Meyer-Ebrecht D (1994) 63–67 Computational principles in Purkinje I and IV refl ection 49. Kase M, Nagata R, Yoshida A, Hanada I (1984) Pupillary pattern evaluation for the assessment of ocular alignment. light refl ex in amblyopia. Invest Ophthalmol Vis Sci 25: Invest Ophthalmol Vis Sci 35:4205–4218 467–471 124 9 The Brückner Test Revisited

50. Portnoy JZ, Th ompson HS, Lennarson L, Corbett JJ (1983) 53. Rüssmann W, Fricke J, Neugebauer A (2004) Nachweis der Pupillary defects in amblyopia. Am J Ophthalmol 96: 609–614 Fehlstellung mit dem Ab- und Aufdecktest. In: Kaufmann 51. Gruber H, Lessel MR (1982) Modiﬁ kation des swinging H (ed) Strabismus. Th ieme, Stuttgart, pp 341–344 ﬂ ashlight tests. Klin Monatsbl Augenheilkd 181: 402–403 54. Rüssmann W, Kaufmann H (2008) Augenbewegungs- 9 52. Jiang MQ, Th ompson HS, Lam BL (1989) Kestenbaum’s störungen. In: Straub W, Kroll P, Küchle J (eds) number as an indicator of pupillomotor input asymmetry. Augenärztliche Untersuchungsmethoden. Enke, Stuttgart, Am J Ophthalmol 107:528–530 pp 637–643 Chapter 10

Amblyopia Treatment 2009 10 Michael X. Repka

Core Messages ■ Wearing optimum refractive correction before administered to the sound eye are equally initiation of patching or other amblyopia therapy eff ective. is associated with improvement in amblyopia in ■ For initial therapy of severe amblyopia for chil- about three quarters of children and a cure in dren 3 to less than 7 years of age, 6 h of daily about one fourth. Th is improvement may facili- patching and full-time patching appear to be tate subsequent treatment. equally eff ective. ■ For initial therapy of moderate anisometropic ■ Amblyopia therapy can be benefi cial for older and strabismic amblyopia among children children up to 17 years of age, especially if they 3–7 years of age, patching and atropine are equiv- have not been previously treated. alent. Atropine is slightly more acceptable ■ Th ere have not been any studies to date which than patching on the basis of parental ques- demonstrate the best therapy for patients with tioning. residual amblyopia following initial therapy. ■ For initial therapy of moderate amblyopia, 2 h of Th ere are also no studies that have identifi ed the daily patching or twice weekly topical atropine best treatments for deprivation amblyopia.

Th e strict age cut-off of 7 or 8 years for therapy has 10.1 Amblyopia Treatment 2009 been shown to be incorrect. Children through at least 13 years of age should be considered suitable for a trial of 10.1.1 Introduction amblyopia therapy, as a large proportion will experience Amblyopia management, long based on consensus or improvement [3]. Management of deprivation amblyo- clinical wisdom [1, 2], has been developing an evidence- pia, such as seen with unilateral aphakia or trauma, based foundation over the last decade. We have seen the remains diffi cult, frustrating to the families, and oft en completion of a series of randomized treatment trials and unsuccessful. Th ere is little new information on manage- prospective observational studies over the last 10 years. ment of these patients. Th ese studies have dealt solely with the most common forms of amblyopia, those due to anisometropia, strabismus or a combination. Spectacle correction is the base on 10.1.2 Epidemiology which all treatment for amblyopia must be built. Both patching and atropine penalization are eff ective as initial Amblyopia is considered the most common cause of management of moderate amblyopia. Initial dosages of monocular visual impairment in both children and 2 h daily of patching or twice weekly atropine have been young and middle-aged adults, in up to 4% of individu- shown to be eff ective and can be considered suitable for als [4].Simons, 1996 #181; [5]. It has been suggested that initial therapy. Severe amblyopia may be initially man- the prevalence is higher in underserved communities aged with 6 h of patching. Intensifi ed treatment for [6]. A study conducted by the National Eye Institute patients who are incompletely treated is logical to pre- found amblyopia to be the leading cause of monocular scribe, yet not proven in clinical trials. vision loss in the 20–70-year-old age group [4]. Th ese 126 10 Amblyopia Treatment 2009

estimates have been based on school- or clinic-based of strabismic and anisometropic amblyopia is slight in the studies. intermediate spatial frequencies tested with the low-con- Two very recent population-based studies from the trast letters of the Pelli-Robson charts [16, 17]. We have United States have reported prevalence estimates for ambly- recently confi rmed this fi nding of only a minimal defi cit 10 opia among preschool-aged children in urban areas. One with Pelli-Robson charts 3–7 years aft er enrollment in an study from Baltimore, Maryland, found the prevalence of amblyopia treatment trial [18]. amblyopia to be 1.8% in Whites and 0.8% in African- Most studies of reading ability of amblyopic patients Americans [7]. Th e authors extrapolated their fi nding to have tested the subjects binocularly, rather than monocu- suggest that there are approximately 271,000 cases of larly, generally over a wide range of ages. Some of these amblyopia among children 30–71 months of age in the studies have indicated that binocular reading ability in United States. Th e second study, completed in Los Angeles, children with amblyopia is impaired [19, 20], whereas California, detected amblyopia in 2.6% of Hispanic/Latino others have reported that reading ability is not aff ected children and 1.5% of African-American children, with 78% [21]. PEDIG recently reported the monocular oral read- of cases of amblyopia attributable to refractive error [8]. ing speed, accuracy, fl uency and comprehension of 79 A study of a birth cohort at age 7 years in the United children with previously treated amblyopia at a mean age Kingdom found 3.6% of children to have amblyopia [9]. of 10.3 years [22]. We found the amblyopic eyes to be Th ere was a suggestion in this latter study that amblyopia slightly slower and less accurate compared with fellow prevalence correlated mildly with lower socioeconomic eyes, while comprehension was similar. Because of our status. study design we could not compare these children to a Whatever the actual percentage of amblyopia in a non-amblyopic population, so the impact of the monocu- population, this disease remains a common ocular prob- lar loss of vision on the patient’s binocular reading ability lem among children. Th e causes of amblyopia depend on remains to be thoroughly explored. the population studied. In one treatment trial, amblyopia was associated with strabismus (37%), Anisometropia (38%) or both combined (24%) [10]. In another retro- 10.1.4 Diagnosis of Amblyopia spective series, amblyopia was associated with strabismus (57%), anisometropia (17%) or both (27%) [11]. Th e diagnosis of amblyopia requires detection of a diff erence in visual acuity between the two eyes while wearing a necessary spectacle correction. For children who can have optotype acuity accurately measured, this remains 10.1.3 Clinical Features of Amblyopia the method of choice, in fact arguably, the only method. Visual loss in amblyopia as measured with high-contrast Th e test should employ either crowded or line optotypes. opotoypes varies from mild to severe. Th e literature sug- Th e clinician should exercise caution when interpreting gests that about 25% of cases have visual acuity in the the results of optotype testing. Th e variability of the amblyopic eye worse than 20/100 and about 75%, 20/100 instrument needs to be considered. Specifi cally, what is or better [12, 13]. Th e more common causes of amblyo- the expected variability of a second measurement when pia are strabismus and moderate anisometropia, each there has been no actual change in the visual acuity? For accounting for about 35%, with 25% having both ani- the Amblyopia Treatment Study Visual acuity testing sometropia and strabismus [10, 11]. Much less common protocol of single surrounded HOTV, we found high is amblyopia related to high anisomyopia, bilateral high testability aft er age 3 years, with 93% of retests within 0.1 ametropia and disease of the anterior visual pathways logMAR. More importantly, the visual acuity needs to (e.g., optic nerve hypoplasia). Although good results diff er by more than 0.18 logMAR for the diff erence to have been occasionally reported with conventional treat- likely be true [23]. In my experience a one-line change ment, these cases are typically more diffi cult to treat from a prior visit nearly always led to a change in therapy successfully. prescribed, usually an escalation. In children the test– Other features of amblyopia include a reduction in retest variability is very high. For children 7–<13 years, a contrast sensitivity and possibly reading ability. Most stud- change in visual acuity must be at least 0.2 logMAR (ten ies have found a reduction in contrast sensitivity in eyes letters) from a previous acuity measure to be unlikely with amblyopia using sinusoidal gratings [14–16], whereas resulting from measurement variability [24]. Th ese two minimal loss has been reported with Pelli-Robson charts, studies of rigorously administered visual acuity testing which test intermediate spatial frequencies [16, 17]. protocols remind clinicians that substantial variability of Detection of a defi cit of contrast sensitivity aft er treatment visual acuity results is present in children and careful 10.2 Amblyopia Management 127 consideration of testing results before adjusting therapy suggested a tendency to spontaneous improvement of is warranted. the visual acuity defi cit associated with amblyopia [29, A recent article has also confi rmed that the visual acu- 30]. Alternatively, another research group found that ity may vary from test strategy to test strategy. Th e ATS- patients who did not comply with treatment deterio- HOTV protocol overestimated the visual acuity relative rated over time [31]. It is safe to comment that we do to the E-ETDRS protocol (0.68 lines for amblyopic eyes; not know enough about the natural history of this com- 0.25 lines for fellow eyes) [25]. mon condition. Fixation preference testing has long been the clinical method of choice (in fact the only method in widespread clinical use) for determining amblyopia in children unable to perform a quantitative acuity on an eye chart. Summary for the Clinician Th e examiner determines the preference for fi xation in a ■ Current estimates of the prevalence of amblyo- strabismic patient simply by determining the eye being pia among preschool aged children in the Unites used. For the orthotropic patient, a strabismus is created States range from 0.8 to 2.8%, with the highest with a 10- or 12-prism diopters vertical prism and the rate found among Hispanic Americans. Most assessment of fi xation preference is again made. If the cases are associated at least in part with refrac- patient alternated or at least could hold with the less- tive error. preferred eye through a blink or a pursuit movement, no ■ Fixation preference testing for amblyopia is amblyopia was felt present. Two recent reports using the unreliable for the detection of amblyopia. It also same testing protocol have found that the test is much less appears to not be suffi ciently reliable to guide reliable than we have thought. Th ese research groups amblyopia therapy in many children. tested children 30 to less than 72 months with fi xation ■ Care is needed when interpreting sequential preference testing and optotype acuity. Fixation prefer- measurements of visual acuity when made with ence testing identifi ed only 15% of preschool children diff erent instruments or testing paradigms. who had an IOD of two lines or more on visual acuity testing and 25% of those with an IOD of three lines or more [26]. Th ere were an insuffi cient number of children with strabismus to comment on that subgroup. 10.2 Amblyopia Management In the Multiethnic Pediatric Eye Disease Study (MEPEDS), the authors reported sensitivity of fi xation Best practice for management of amblyopia had been preference testing for amblyopia among children with based on clinician consensus [1]. However, no random- anisometropia was 20% (9/44), although specifi city was ized trial had ever been done comparing no treatment to 94% (102/109). Among strabismic children, sensitivity any amblyopia treatment. During the last 5 years, a large was 69% (9/13; worse in children 30–47 than 48–72 number of clinical trials assessing methods of amblyopia months old), and specifi city was 79% (70/89) [27]. treatment have allowed the incorporation of evidence- Hakim found that 75% of strabismic children had based information into the practice of amblyopia care positive test results by fi xation preference testing, but based on the earlier guidelines. only 13% had an IOD of two lines or more [28]. Th e obvious, albeit controversial confusion, is that fi xation preference testing misses most cases of amblyopia when used in 10.2.1 Refractive Correction a screening setting. In addition, the use of fi xation preference testing in a clinical setting for managing a patient Th e value of an accurate refraction can not be underesti- with strabismus would likely lead to substantial mated in the management of amblyopia. Th ese data are overtreatment. essential for both the diagnosis of amblyopia and the subsequent optimum treatment of the amblyopia. For secu- rity of the amblyopia diagnosis, the presence of an anisometropia helps substantiate the presence of amblyo- 10.1.5 Natural History pia. Th e refractive error requires a measurement obtained Limited natural history data are available for amblyopia under adequate cycloplegia, usually 1% cyclopentolate or as nearly all patients diagnosed are prescribed some ther- similar cycloplegic. Many clinicians instill a topical anes- apy. Although compliance is quite variable, most children thetic before the cycloplegic agent to prolong the reten- receive some intervention. Some authors have tion of the cycloplegic drug in the tear fi lm. 128 10 Amblyopia Treatment 2009

Prescribed glasses for ametropia are not controversial. amblyopic strabismic patients was not expected to occur Th e prescription for an esotropia patient should be full so oft en so PEDIG has launched an adequately powered plus power [32]. Even if this power slightly blurs distance prospective study of the impact of spectacle correction vision, it will not have a deleterious eff ect at the child’s alone to explore this result. 10 usual working distance. For the microstrabismic or orthotropic child, under correcting the hypermetropia symmetrically by up to 1.50 diopters avoids the problem of 10.2.2 Occlusion by Patching distance blur and does not seem to detract from the treatment outcome. For the exotropic patient, the anisometro- Th e benefi cial eff ect of occlusion with an adhesive patch in pia and any myopia need to be corrected. High the management of amblyopia has long been considered hypermetropia should be partially corrected. obvious. Some randomized-controlled treatment trials What has been controversial among clinicians is what have compared treatments, without an untreated control, to do (and when) once the eyeglasses prescription is writ- led to criticism that the improvements experienced were ten and spectacles obtained. Some clinicians have rou- due to age or learning eff ects or possibly the benefi ts of tinely started patching at the same time, while others spectacles alone as noted earlier [36]. To address that issue, have waited a variable amount of time. Recent research PEDIG conducted a RCT comparing occlusion to specta- has provided some guidance on this clinical decision, cles only. Before enrollment, the patients wore glasses until specifi cally the value of glasses alone in the management their vision stabilized between two consecutive visits. Th ey of amblyopia. In the United Kingdom, Stewart et al found were then randomized to continue spectacles alone com- a mean improvement of 2.4 lines in 65 children 3–8 years pared with 2 h of daily patching. Improvement in VA of the of age were treated with spectacles, taking an average of amblyopic eye from baseline to 5 weeks averaged 1.1 lines 14 weeks to reach best visual acuity [33]. Surprisingly, in the patching group and 0.5 lines in the control group improvement was noted among both anisometropic and (P = 0.006), and improvement from baseline to best mea- strabismic patients. Th ese authors have termed this eff ect sured VA at any visit averaged 2.2 lines in the patching refractive adaptation, although that term is potentially group and 1.3 lines in the control group (P < 0.001) [37]. confusing since the refraction does not actually adapt. Th us, occlusion was better but surprisingly there was con- Rather the improvement represents the remediation of tinuing benefi t of the spectacles alone, reinforcing how the amblyopia by optical correction alone. In a larger important this aspect of therapy must be. recent prospective study investigators in North America Th e dosage of occlusion therapy prescribed has his- enrolled 84 children 3 to <7 years old with untreated ani- torically ranged widely, from a few minutes to all waking sometropic amblyopia ranging from 20/40 to 20/250 [34]. hours per day. Some clinicians have prescribed fewer Optimal refractive correction was provided in accor- hours for fear of damaging the binocular visual system. dance with consensus guidelines similar to those above. In the initial PEDIG trial, comparing atropine to patch- VA was measured with the new spectacle correction at ing, both treatments were found to be equally eff ective baseline and at 5-week intervals until VA stabilized or [38]. Subgroup analysis of diff ering dosages from 6 h amblyopia resolved. VA improved with optical correction daily to full time (all waking hours less one daily) found alone by ≥2 lines in 77% of the patients and remarkably no advantage of prescribing more hours [39]. Th is led us resolved in 27% [34]. Although the study was designed to design two studies directed at exploring occlusion dos- and powered for children with anisometropia, strabismic age. In the fi rst trial, we compared 2 with 6 h daily for the and combined strabismic–anisometropic patients were initial treatment of moderate amblyopia, 20/40–20/80, enrolled in a parallel pilot study following the same pro- for a period of 4 months [40]. Visual acuity in the ambly- tocol to determine if such patients could respond to opic eye improved a similar amount in both groups. Th e spectacle correction alone [35]. Twelve patients with pre- improvement in the amblyopic eye from baseline to 4 viously untreated strabismic amblyopia were prescribed months averaged 2.40 lines in each group (P = 0.98). Th e spectacles and examined at 5-week intervals until visual 4-month visual acuity was ≥20/30 and/or improved from acuity was not improved from the prior visit. Amblyopic baseline by ≥3 lines in 62% in each group (P = 1.00). We eye acuity improved by ≥2 lines from spectacle-corrected did not follow and treat these patients aft er 4 months so baseline acuity in 9 (75%), resolving in three. Mean we do not know if a diff erence might develop. In the sec- change from baseline to maximum improvement was 2.2 ond trial of patching dosage, we compared 6 with full ± 1.8 lines. Improvement continued for up to 25 weeks. time or all waking hours less 1 h for severe amblyopia, Data on the ocular alignment aft er instituting the glasses 20/100–20/400 [41]. VA in the amblyopic eye improved were not available. Improvement in the visual acuity of to a similar extent in both groups. Th e improvement in 10.2 Amblyopia Management 129 the amblyopic eye acuity from the baseline to 17 weeks both groups: 2.84 lines in the atropine group and 3.16 averaged 4.8 lines in the 6-h group and 4.7 lines in the lines in the patching group. Th e patching group did get full-time group (P = 0.45). However, 75% of patients in better faster, but by 6 months, the diff erence of 0.034 was both groups were 20/40 or worse aft er therapy. Th ere is a clinically inconsequential. Both treatments were well tol- natural concern about amblyopia therapy, particularly erated, although the atropine was easier to administer with higher dosages, causing loss of vision in the sound based on parental questionnaires. eye. Th e sound eye lost two or more lines in 4% of the 6-h Th ese children were followed in the study for an addi- group and in 11% of the full-time group. Nearly all tional 18 months to describe prescribed treatment and patients returned to their baseline level with follow-up, stability of the improvement. Treatment was determined typically by just stopping all patching. by the investigator [42]. Remarkably, and at odds with Th ese patching dosage data show that for initial treat- clinical wisdom, nearly 90% received some treatment ment of amblyopia due to strabismus, anisometropia or during this period. Eighty percent received the same both combined, beginning with the lower dosage of treatment and 25% received the alternate treatment (some occlusion does not lessen the chance of success and may patients received both). At 2 years, visual acuity in the make the treatment more feasible. However, only about amblyopic eye improved a mean of 3.6 lines in the atro- one in four patients with moderate amblyopia was 20/25 pine group and 3.7 lines in the patching group. Th is dif- or better and one in four children with severe amblyopia ference in visual acuity between treatment groups was was 20/32 or better. small: 0.01 logMAR (95% confi dence interval, −0.02 to Th ese studies have taught much about initial patching 0.04). Th us, the relative equivalence of the techniques and therapy, but they have left substantial uncertainty about the persistence of the treatment benefi t were reaffi rmed. what to do for those children who are not completely cor- Stereoacuity outcomes were similar suggesting no untow- rected. Some clinicians have misinterpreted the results ard relative eff ect of either of the two treatments. and have recommended stopping therapy when the visual One concern regarding amblyopia therapy is the acuity ceases to improve with these prescribed doses. potential for inducing or worsening a strabismus. In addi- What needs to be explored is whether an increased dose tion, most authors have suggested treating amblyopia or a change in treatment approach will allow more com- before undertaking strabismus surgery. Th is study evalu- plete correction. At present, clinicians and parents will ated the chance of inducing a strabismus, but also the have to make that judgment without the results of a RCT chance of improving a strabismus with amblyopia treat- to guide the choice. Logically, some period of more ment. Of the 161 patients with no strabismus, similar intense therapy should be administered before discon- proportions initially assigned to the patching and atro- tinuing treatment. pine groups developed new strabismus by 2 years (18 vs. 16%, P < 0.84) [43]. Of the new cases of strabismus, only two patients in the patching group and three patients in the atropine group developed a deviation that 10.2.3 Pharmacological Treatment was greater than 8D. Perhaps surprisingly, of the 105 with Atropine patients with strabismus greater than 8D at enrollment, To fi nd an eff ective, yet easy to administer, treatment of 13% of those in the patching group and 16% of those in amblyopia has been a goal pursued by clinicians treating the atropine group improved to orthotropia without stra- amblyopia in response to the complaints and diffi culties bismus surgery. Th ese data show that strabismus may associated with occlusion therapy. Th is pursuit has led to develop or resolve with amblyopia therapy in about equal many failed treatments that were launched with great fan- proportions. fare, but ultimate abandonment. Th e dosage of atropine in the original PEDIG trial For more than a century, clinicians have used pharma- was once daily. Th is design was consistent with the cological penalization of the sound eye to make the child desire to maximize the likelihood of fi nding benefi t if use the amblyopic eye and thereby improve the visual acu- there was one. While that study was underway, the ben- ity of that eye. Most clinicians typically used this treat- efi t of less frequent administration was suggested by ment for patching failures or noncompliance. Case series Simons and coworkers [44]. Th ey reported reasonable reported eff ectiveness, but the common belief was that improvement from less frequent administration. Th is this was an inferior treatment. Th e largest prospective was plausible since the duration of cycloplegia was oft en study was completed in 2002, comparing once daily atro- more than 1 day. Th is fi nding led PEDIG to develop a pine to patching 6 or more hours per day for moderate clinical trial, which compared daily atropine to weekend amblyopia 20/30–20/100 [38]. Visual acuity improved in atropine. 130 10 Amblyopia Treatment 2009

Th e atropine dosage treatment trial included 168 chil- children, should easily be incorporated into a child’s daily dren younger than 7 years with amblyopia in the range of activities, and is likely to be attractive to a large propor- 20/40–20/80 associated with strabismus, anisometropia tion of parents. However, as with patching if the visual or both. Th ey were randomized to either daily or week- acuity improvement is not complete increasing the dos- 10 end atropine [45]. Th e improvement of the amblyopic eye age or changing to an alternative therapy should be con- from baseline to 4 months averaged 2.3 lines in each sidered. Th e eff ectiveness of such a treatment remains to group. Th e visual acuity of the amblyopic eye at study be proven. completion was either (1) at least 20/25 or (2) better than or equal to the sound eye in 39 children (47%) in the daily group and 45 children (53%) in the weekend group. Th e 10.2.4 Pharmacological Therapy Combined visual acuity of the sound eye at the end of follow-up was with a Plano Lens reduced by two lines in one patient in each group. Stereoacuity outcomes were similar in the two groups. Investigators have long looked for ways to intensify their Patients who were not cured continued on the ran- treatments, implicitly recognizing that the prescribed domized treatment beyond the 4-month outcome exam. therapy did not always have the desired eff ect. For atro- Th ey improved an average of 0.8 additional lines (0.7 lines pine penalization of the sound eye, it has been long noted among the 22 daily group patients and 0.8 lines among that adding optical penalization, by removing all hyper- the 31 weekend group patients). metropic correction from the sound eye, would add opti- At the time of study completion, 39 (47%) of the cal blur at distance to complement the cycloplegic blur patients in the daily group and 45 (53%) in the weekend provided at near. A retrospective report included 42 chil- group had an amblyopic eye acuity that was either (1) dren (mean age, 4.7 years) treated with daily atropine and 20/25 or better or (2) the same or better than the sound a plano lens for the sound eye [46]. Important caveats eye acuity, provided that the sound eye acuity had not were that eligible patients had failed patching treatment decreased from enrollment. Th e mean amblyopic eye and had at least 1.75 D of sound eye hypermetropia. acuity at study completion was 0.23 logMAR in the daily Surprisingly, they found a mean improvement in ambly- group and 0.21 logMAR in the weekend group (approxi- opic eye visual acuity from 20/113 to 20/37 aft er 10 weeks mately 20/32). Th e mean sound eye visual acuity at enroll- of treatment with atropine and a plano lens to the sound ment was 0.05 logMAR (approximately 20/25), with 81% eye. Th is was a remarkable achievement. However, of the sound eyes having acuity of 20/25 or better. Morrison and colleagues cautioned that this treatment Among patients who improved two or more lines resulted in a case of severe treatment-related amblyopia in from baseline during the study, 30% of patients achieved the sound eye when parental noncompliance occurs [47]. their best acuity at 5 weeks, 50% at 4 months, 7% at 6 To explore the value of this “augmented atropine months, 10% at 8 months and 3% at 10 months. Th ese approach,” PEDIG randomized 180 children with moder- results were similar in the two atropine treatment groups. ate amblyopia (visual acuities of 20/40–20/100) to week- Th us, a 4-month treatment period with atropine will end atropine use augmented by a plano lens or weekend treat most patients but is not suffi cient to complete treat- atropine use alone [48]. At 18 weeks, amblyopic eye ment for all. Th us, treatment should be continued until improvement averaged 2.8 lines in the group that received there is good evidence that a plateau in improvement has atropine plus a plano lens and 2.4 lines in the group that been achieved. received atropine alone (mean diff erence between groups Th ere is a chance of visual impairment of the sound adjusted for baseline acuity, 0.3 line; 95% confi dence eye so care needs to be taken. In this study 1% of sound interval, −0.2–0.8 line). Amblyopic eye visual acuity was eyes lost two or more lines of acuity at last follow up. As 20/25 or better in 24 patients (29%) in the group that expected, light sensitivity was common, reported by 16% received atropine only and 35 patients (40%) in the group of children. Facial fl ushing and fever, a more worrisome that received atropine plus a plano lens (P = 0.03). side eff ect, was reported by 1% of the children. However, more patients in the group that received atro- Summarizing, weekend atropine for moderate ambly- pine plus a plano lens had reduced sound eye visual acu- opia is eff ective in improving visual acuity. Th e amount of ity at 18 weeks; fortunately, there were no cases of improvement was comparable with that seen with 4 persistent reverse amblyopia. Th e important conclusion months of 2 or 6 h of daily patching [40]. Parents need to is that in spite of intuition, augmentation of weekend realize that most children will need at least 4 months of atropine use with a plano lens does not substantially treatment irrespective of which therapy and dosage. improve amblyopic eye visual acuity when compared Twice weekly atropine is fairly unobtrusive for preschool with weekend atropine use alone. 10.3 Other Treatment Issues 131

moderate amblyopia (20/40–20/80) and 6.3 lines (95% Summary for the Clinician CI, 5.1–7.5) for children with severe amblyopia (20/100– ■ A series of trials has shown that for amblyopia 20/320). Maximum improvement was achieved aft er 13 from anisometropia, strabismus or both com- weeks for some, yet only aft er a year for others. Th e obvi- bined, initial therapy should be refractive cor- ous conclusion is that glasses should be prescribed to rection with the expectation of substantial children at an early age and worn as much of the time as improvement. possible. ■ Occlusion is signifi cantly more eff ective than spectacles alone. ■ Atropine and patching are equally eff ective for initial treatment of mild amblyopia among chil- 10.3.2 Age Eff ect dren 3 to less than 7 years of age. Most clinicians have held that amblyopia treatment is ■ Initial dosages of 2 h of patching and weekend best accomplished when children are young and certainly atropine are similar in eff ectiveness to more before age 8 years. Among preschool children treated intensive therapy as initial treatment. with either patching or atropine there was no age eff ect ■ Expect that about 80% of the children will be identifi ed [53]. Th is fi nding along with case reports of 20/30 (6/9, 0.66) or better in the sound eye aft er effi cacy in older children, teens and even adults led treatment completion. PEDIG to undertake a treatment trial of subjects 7–17 ■ Augmented pharmacological treatment with a years of age [3]. In the 7 to 12-year-olds (n = 404), treat- plano lens for the sound eye is not associated ment was 2–6 h of patching daily plus daily atropine. with substantial benefi t as initial therapy, but is Fift y-three percent of the treatment group improved at associated with a risk of visual loss in the sound least ten letters compared with 25% of the optical correc- eye. tion group (P < 0.001). In the 13 to 17-year-olds (n = 103) treatment was 2–6 h of patching per day, improvement rates of ten letters or more were 25 and 23%, respectively (adjusted P = 0.22). More striking was the improvement 10.3 Other Treatment Issues among patients not previously treated; 47 and 20% of the two age groups, respectively. Most patients were left with 10.3.1 Bilateral Refractive Amblyopia a residual visual acuity defi cit. Th is means that older children who had never been treated should have a trial of Th e management of bilateral amblyopia from hyper- treatment. metropia and/or astigmatism has been the subject of several reports. Th e incidence was 4 of 830 (0.5%) children at entry into school in an older report [49]. Small case series have found substantial benefi t to treatment with spectacle 10.3.3 Maintenance Therapy correction. In one study, 10 of 12 children (83%) improved to 20/40 or better in both eyes with a mean follow-up of Clinical wisdom has suggested that amblyopia therapy 22 months [50]. A recent report study found that 21 of 36 should not be abruptly stopped, but rather needs to be children (58%) achieved a visual acuity of 20/25 or better continued for a period of time to reduce the chance of in at least one eye with a mean follow-up of 3.3 years [51]. recurrence [1]. Th is approach was indirectly studied by Neither study was suffi cient large to develop reasonable taking some patients from some of the early PEDIG trials estimates for the chance of success for these children. and whose therapy was being stopped or maintained on a PEDIG undertook a prospective observational study low dose of occlusion [54]. Th e recurrence rate was 24% of bilateral refractive amblyopia [52]. Inclusion criteria (35 of 145) (95% confi dence interval 17–32%). Th ere was included 20/40–20/400 best-corrected visual acuity in no diff erence between patching and atropine. In patients the presence of 4.00 diopters or more of hypermetropia treated with patching of 6–8 h per day, recurrence was by spherical equivalent, 2.00 diopters or more of astig- more common (42%) when treatment was abruptly matism, or both in each eye. Mean binocular visual acu- stopped compared with tapering to 2 h per day before ity improved from 0.50 logMAR (20/63) at baseline to cessation (14%, odds ratio 4.4, 95% confi dence interval 0.11 logMAR (20/25) at 1 year (mean improvement, 3.9 1.0–18.7). Absent additional data seems prudent to mon- lines; 95% confi dence interval, 3.5–4.2). Mean improve- itor all patients and to taper occlusion therapy (6 or more ment was 3.4 lines (95% CI, 3.2–3.7) for children with hours) and daily atropine therapy. 132 10 Amblyopia Treatment 2009

unable to fi nd signifi cant diff erences in educational, 10.3.4 Long-Term Persistence of an Amblyopia Treatment Benefi t social, or employment attainment between amblyopic and control subjects [68]. Conversely, a questionnaire- Th e longevity of the improvement in VA achieved with based study of adults with amblyopia and strabismus on 10 amblyopia treatment has been questioned. Short-term their quality of life found lifelong benefi ts as perceived recurrence and the need to repeat therapy is well known. by those patients [69]. Th e best estimates are about 25% will recur during the fi rst year aft er cessation of therapy [55–57]. Most of these cases will occur in the fi rst 6 months aft er cessation of Summary for the Clinician therapy. Based on clinical experience most of the recur- ■ Amblyopia therapy appears to lead to a persis- rences can be successfully treated, but prospective data tent improvement in visual acuity of the ambly- are needed. opic eye. Th e long-term benefi t of amblyopia therapy would ■ only be proven if the improvement in acuity experienced Amblyopia therapy for children from 7 to 17 by the amblyopic eye is maintained. Th ere are substantial years should be considered if there is no history data published in this area, which is quite troublesome. of an adequate trial of treatment. ■ Th e extent of deterioration reported in retrospective out- More research is needed to understand the eff ect come studies of children treated for amblyopia to be as of amblyopia on patient outcomes. high as 58% in spite of interim treatment, thereby reducing the actual benefi t of therapy [58–63]. To address this question, prospectively, children 3–<8 years enrolled in 10.4 Other Treatments our trial comparing patching to atropine were followed at 2 years aft er randomization, and a subgroup reexamined Clinicians have long known that the standard treatment at age 10 years, 3–7 years aft er randomization [64]. Two of patching and even atropine were not always successful. years aft er randomization visual acuity in the amblyopic Th ey have therefore sought alternatives to occlusion ther- eye improved a mean of 3.7 lines in the patching group apy as primary and secondary treatment of amblyopia. and 3.6 lines in the atropine group. In both treatment groups, the mean amblyopic eye acuity was approximately 20/32, 1.8 lines worse than the mean sound eye. 10.4.1 Filters At age 10 years, 169 patients had an amblyopic eye VA of 0.17 logMAR (approximately 20/32), and 46% of Bangerter foils were introduced nearly 50 years ago to amblyopic eyes had an acuity of 20/25 or better [65]. Age provide a graded reduction of image quality to the sound younger than 5 years at entry into the randomized trial eye [70]. Th e eight fi lter densities were designed to reduce was associated with a better visual acuity outcome visual acuity of the sound eye to a range of 20/25–20/300. (P < 001). Mean amblyopic and sound eye visual acuities Selecting the proper blur level would force the patient to at age 10 years were similar in the original treatment use the amblyopic eye. Th e fi lters are worn on the back groups (P = 0.56 and P = 0.80, respectively). Th e good surface of the spectacle lens are for the most part are not news here is that the visual acuity improvement was readily apparent. Proponents have suggested that the maintained. However, 88% of all of these patients were improved appearance compared with a patch would treated at least once between the primary 6-month out- increase patient compliance. In addition, fi lters do not come and the age 10 years evaluation. In addition, these cause skin irritation. Finally, one could postulate that children were part of a clinical trial, which may improve Bangerter foils are less disruptive to binocular function compliance with therapy and follow up compared with during treatment compared with patching. Th e key dis- the general population. advantage of Bangerter foils is that glasses must be worn Amblyopia treatment is considered cost-eff ective and the child must not look around the device. One small among the spectrum of eye and health care interven- uncontrolled case series on primary use of this treatment tions [66, 67]. However, there is substantial uncertainty comes from Iacobucci and associates [71]. Th ey treated concerning the eff ect of treatment on quality of life in 15 children, 3–8 years old, with amblyopia of 20/30–20/60 the future. Economic modeling cannot account for the for a mean duration of 9 months. Two thirds of patients impact of adaptation to the visual impairment from a (10 of 15) obtained amblyopic eye acuity of 20/20 or bet- young age compared with that of later onset. A large ter or equal to that of the sound eye. Of the remaining fi ve cohort study of adults in the United Kingdom was patients, four attained amblyopic eye acuity of 20/25 or 10.5 Controversy 133

20/30 or within a half line of the sound eye. Bangerter participating subjects. Subjects who received levodopa fi lters, as in this study, are prescribed for longer periods plus occlusion demonstrated signifi cant regression of than either patching or atropine because they are well visual acuity aft er stopping the medication. On average, tolerated. the amount of regression over 6 months of follow-up Bangerter fi lters have not been compared with patch- averaged 1.4 lines, similar to that experienced by those ing or atropine. PEDIG has completed a clinical trial com- receiving occlusion only [75]. paring Bangerter fi lters (0.2 and 0.3 densities) to 2 h of Forty children 6–<18 years were randomized to 4 daily occlusion. Th e results are currently being analyzed. weeks of levodopa (1.86 mg/kg/day (1.33–2.36 mg/kg/ day) plus full-time occlusion or full-time occlusion only [76]. No diff erence in visual acuity outcome was found. 10.4.2 Levodopa/Carbidopa Adjunctive Therapy Summary for the Clinician Levodopa is used to treat adults with Parkinson disease ■ Bangerter fi lters appear to be a useful option but and children with dopamine responsive dystonia. data compared with those of other treatments Dopamine is a neurotransmitter that does not cross the are not yet available. blood–brain barrier. However, levodopa administered ■ Many pilot studies have shown some improve- orally crosses the blood–brain barrier, where it is con- ment when patching is combined with levodopa/ verted to dopamine. Levodopa is typically used in combi- carbidopa for about 8 weeks. Durability of the nation with carbidopa, a peripheral decarboxylase treatment eff ect and a comparison with patching inhibitor that prevents the peripheral breakdown of alone needs to be completed. levodopa. Th is reduces the dose of levodopa and thereby reduces the primary side eff ects of nausea and emesis. A randomized longitudinal double masked placebo control trial of ten amblyopic children aged 6–14 years 10.5 Controversy [72]. Th e dosing averaged 0.5 mg/kg/tid and lasted for 3 weeks. Visual acuity of the amblyopic eyes improved by 10.5.1 Optic Neuropathy Rather 2.7 lines in the levodopa treated group, and by 1.6 lines in than Amblyopia the subjects treated with placebo. One month aft er the termination of treatment, the levodopa-carbidopa group Every clinician managing a child with amblyopia must be maintained a 1.2-line improvement in visual acuity. aware of the masquerade of an optic neuropathy as an A 1-week, randomized, placebo-controlled study was amblyopia. Careful attention to pupillary signs, appear- performed with 62 children with amblyopia who were ance of the optic nerve and response to therapy are between 7 and 17 years of age. Subjects were instructed to needed. An amblyopic patient who does not improve occlude the dominant eye for 3 h per day. Visual acuity (or deteriorates) with conventional therapies should be improved from 0.59 to 0.45 in the levodopa–carbidopa continually reassessed for the presence of an optic neu- group (average dose 0.51 mg/kg/tid) and from 0.69 to ropathy. Such a situation might be an optic neuropathy 0.63 in the control group (P = 0.023) [73]. related to compression or other progressive damage of In a prospective randomized trial, 72 subjects with the aff erent visual pathway, such as from an optic glioma amblyopia were distributed into three groups [74]. Group or a craniopharyngioma. A subjects received levodopa alone, group B received More controversially is the role of static optic nerve levodopa (0.50 mg/kg/t.i.d.) and part-time occlusion (3 h/ abnormalities in the genesis of visual loss diagnosed as day), and group C received levodopa and all waking horus amblyopia. It has been suggested by Lempert that these occlusion of the sound eye. Although 53/72 subjects fi ndings are very common. He has reported termed “dys- (74%) had an improvement in visual acuity (maximum = version” or hypoplasia in optic nerve photographs in 45% 4.6 Snellen lines; mean 1.6 Snellen lines, ≤10 years; mean of 205 amblyopic eyes [77, 78]. More recently, Lempert 1.1 Snellen lines, >10 years) aft er treatment, 52% of those has reported reduced optic disc rim areas for both ambly- who improved had regression in visual acuity when mea- opic and fellow eyes with the reduction most prominent sured aft er 1 year. in the amblyopic eyes [79]. If there was an abnormality of A follow-up report of three longitudinal studies (9–27 the optic nerve, we would expect that the retinal nerve months) using levodopa (0.55 mg/kg/t.i.d.) plus occlu- fi ber layer thickness would be reduced. Such investiga- sion for treatment of amblyopia included 30/33 (91%) of tions based on optical coherence tomography have not 134 10 Amblyopia Treatment 2009

found any substantive diff erence among amblyopic, fel- 11. Woodruff G, Hiscox F, Th ompson JR, et al (1994) Th e pre- low, and normal eyes [80–82]. In addition, it has never sentation of children with amblyopia. Eye 8:623–626 been clear why patients with an optic neuropathy would 12. Bray LC, Clarke MP, Jarvis SN, et al (1996) Preschool vision show the substantial improvement in visual acuity seen screening: a prospective comparative evaluation. Eye 10 during management of most cases of amblyopia. 10:714–718 13. Köhler L, Stigmar G (1973) Vision screening of four-year- old children. Acta Paediatr Scand 62:17–27 Summary for the Clinician 14. Hess RF, Howell ER (1977) Th e threshold contrast sensitiv- ■ Th e presence of an optic nerve abnormality in ity function in strabismic amblyopia: evidence for a two “typical” amblyopia remains controversial. type classifi cation. Vision Res 17:1049–1055 ■ Th e value of optic nerve head analysis in the 15. Howell ER, Mitchell DE, Keith CG (1983) Contrast thresh- management of most cases of amblyopia is not olds for sine gratings of children with amblyopia. Invest clarifi ed. Ophthalmol Vis Sci 24:782–787 16. McKee SP, Levi DM, Movshon JA (2003) Th e pattern of visual defi cits in amblyopia. J Vis 3:380–405 17. Moseley MJ, Stewart CE, Fielder AR, et al (2006) References Intermediate spatial frequency letter contrast sensitivity: its relation to visual resolution before and during amblyo- 1. American Academy of Ophthalmology pediatric ophthal- pia treatment. Ophthalmic Physiol Opt 26:1–4 mology panel (2002) Preferred practice pattern guidelines: 18. Repka MX, Kraker RT, Beck RW, et al (2009) Contrast sen- amblyopia. American Academy of Ophthalmology, San sitivity following amblyopia treatment in children. Arch Francisco, pp 1–25 Ophthalmol 127:1225–1227 2. American Optometric Association (1994) Care of the 19. Stift er E, Burggasser G, Hirmann E, et al (2005) Monocular patient with amblyopia, optometric clinical practice guide- and binocular reading performance in children with line. American Optometric Association, St. Louis, pp 1–51 microstrabismic amblyopia. Br J Opthalmol 89:1324–1329 3. Pediatric eye disease investigator group (2005) Randomized 20. Zurcher B, Lang J (1980) Reading capacity in cases of trial of treatment of amblyopia in children aged 7 to 17 ‘cured’ strabismic amblyopia. Trans Ophthalmol Soc U K years. Arch Ophthalmol 123:437–447 100:501–503 4. Simons K (1996) Preschool vision screening, methodology 21. Koklanis K, Georgievski Z, Brassington K, et al (2006) Th e and outcome. Surv Ophthalmol 41:3–30 prevalence of specifi c reading disability in an amblyopic 5. Williamson TH, Andrews R, Dutton GN, et al (1995) population: a preliminary report. Binocul Vis Strabismus Assessment of an inner city visual screening programme Q 21:27–32 for preschool children. Br J Ophthalmol 79:1068–1073 22. Repka MX, Kraker RT, Beck RW, et al (2008) Monocular 6. National eye institute offi ce of biometry and epidemiology oral reading performance aft er amblyopia treatment in (1984) Report on the national eye institute’s visual acuity children. Am J Ophthalmol 146:942–947 impairment survey pilot study. Department of Health and 23. Holmes JM, Beck RW, Repka MX, et al (2001) Th e amblyo- Human Services, Washington, DC pia treatment study visual acuity testing protocol. Arch 7. Friedman DA, Repka MX, Katz J, et al (2009) Prevalence of Ophthalmol 119:1345–1353 amblyopia and strabismus in white and African-American 24. Cotter SA, Chu RH, Chandler DL, et al (2003) Reliability children aged 6 through 71 months: the Baltimore pediat- of the electronic early treatment diabetic retinopathy study ric eye disease study. Ophthalmology 116:(in press) testing protocol in children 7 to <13 years old. Am 8. Multi-ethnic pediatric eye disease study group (2008) J Ophthalmol 136:655–661 Prevalence of amblyopia and strabismus in African 25. Birch EE, Strauber SF, Beck RW, et al (2009) Comparison American and hispanic children ages 6 to 72 months the of the amblyopia treatment study HOTV and the elec- multi-ethnic pediatric eye disease study. Ophthalmology tronic-early treatment of diabetic retinopathy study visual 115:1229–1236 acuity protocols in amblyopic children aged 5 to 11 years. 9. Williams C, Northstone K, Howard M, et al (2008) J AAPOS 13:75–78 Prevalence and risk factors for common vision problems in 26. Friedman DS, Katz J, Repka MX, et al (2008) Lack of con- children: data from the ALSPAC study. Br J Ophthalmol cordance between fi xation preference and HOTV optotype 92:959–964 visual acuity in preschool children: the Baltimore pediatric 10. Pediatric eye disease investigator group (2002) Th e clinical eye disease study. Ophthalmology 115:1796–1799 profi le of moderate amblyopia in children younger than 7 27. Cotter SA, Tarczy-Hornoch K, Song E, et al (2009) Fixation years. Arch Ophthalmol 120:281–287 preference and visual acuity testing in a population-based References 135

cohort of preschool children with amblyopia risk factors. 43. Repka MX, Holmes JM, Melia BM, et al (2005) Th e eff ect Ophthalmology 116:145–153 of amblyopia therapy on ocular alignment. J AAPOS 28. Hakim OM (2007) Association between fi xation prefer- 9:542–545 ence testing and strabismic pseudoamblyopia. J Pediatr 44. Simons K, Stein L, Sener EC, et al (1997) Full-time atro- Ophthalmol Strabismus 44:174–177 pine, intermittent atropine, and optical penalization and 29. Clarke MP, Wright CM, Hrisos S, et al (2003) Randomised binocular outcome in treatment of strabismic amblyopia. controlled trial of treatment of unilateral visual impairment Ophthalmology 104:2143–2155 detected at preschool vision screening. BMJ 327:1251 45. Pediatric eye disease investigator group (2004) A ran- 30. Hard AL, Williams P, Sjostrand J (1995) Do we have opti- domized trial of atropine regimens for treatment of mal screening limits in sweden for vision testing at the age moderate amblyopia in children. Ophthalmology 111: of four years? Acta Ophthalmol 73:483–485 2076–2085 31. Simons K, Preslan M (1999) Natural history of amblyopia 46. Kaye SB, Chen SI, Price G, et al (2002) Combined optical untreated owing to lack of compliance. Br J Ophthalmol and atropine penalization for the treatment of strabismic 83:582–587 and anisometropic amblyopia. J AAPOS 6:289–293 32. American academy of ophthalmology pediatric ophthal- 47. Morrison DG, Palmer NJ, Sinatra RB, et al (2005) Severe mology/strabismus panel (2007) Preferred practice pattern amblyopia of the sound eye resulting from atropine guidelines: pediatric eye evaluations. American Academy therapy combined with optical penalization. J Pediatr of Ophthalmology, San Francisco, California Ophthalmol Strabismus 42:52–53 33. Stewart CE, Moseley MJ, Fielder AR, et al (2004) Refractive 48. Pediatric eye disease investigator group (2009) adaptation in amblyopia: quantifi cation of eff ect and impli- Pharmacological plus optical penalization treatment for cations for practice. Br J Ophthalmol 88:1552–1556 amblyopia: results of a randomized trial. Arch Ophthalmol 34. Pediatric eye disease investigator group (2006) Treatment 127:22–30 of anisometropic amblyopia in children with refractive 49. Haase W (1978) Visual acuity in cases of monocular and correction. Ophthalmology 113:895–903 bilateral amblyopia: treatment during school age. Metabolic 35. Cotter SA, Edwards AE, Arnold RW, et al (2007) Treatment Ophthalmology 2:147–148 of strabismic amblyopia with refractive correction. Am 50. Schoenleber DB, Crouch ER (1987) Bilateral hyperme- J Ophthalmol 143:1060–1063 tropic amblyopia. J Pediatr Ophthalmol Strabismus 24: 36. Lempert P (2004) Th e eff ectiveness of patching for ambly- 75–77 opia should be tested with untreated control subjects (let- 51. Klimek DL, Cruz OA, Scott WE, et al (2004) Isoametropic ter to editor). Arch Opthalmol 122:423–424 amblyopia due to high hyperopia in children. J AAPOS 37. Pediatric eye disease investigator group (2006) A random- 8:310–313 ized trial to evaluate 2 hours of daily patching for strabismic 52. Pediatric eye disease investigator group (2007) Treatment and anisometropic amblyopia in children. Ophthalmology of bilateral refractive amblyopia in children three to less 113:904–912 than 10 years of age. Am J Ophthalmol 144:487–496 38. Pediatric eye disease investigator group (2002) A random- 53. Pediatric eye disease investigator group (2003) A compari- ized trial of atropine vs patching for treatment of moderate son of atropine and patching treatments for moderate amblyopia in children. Arch Ophthalmol 120:268–278 amblyopia by patient age, cause of amblyopia, depth 39. Pediatric eye disease investigator group (2003) Th e course of amblyopia, and other factors. Ophthalmology 110: of moderate amblyopia treated with patching in children: 1632–1638 experience of the amblyopia treatment study. Am 54. Pediatric eye disease investigator group (2004) Risk of J Ophthalmol 136:620–629 amblyopia recurrence aft er cessation of treatment. 40. Pediatric eye disease investigator group (2003) A random- J AAPOS 8:420–428 ized trial of patching regimens for treatment of moderate 55. Bhola R, Keech RV, Kutschke P, et al (2006) Recurrence of amblyopia in children. Arch Ophthalmol 121:603–611 amblyopia aft er occlusion therapy. Ophthalmology 41. Pediatric eye disease investigator group (2003) A random- 113(11):2097–2100 ized trial of prescribed patching regimens for treatment 56. Flynn JT, Schiff man J, Feuer W, et al (1998) Th e therapy of of severe amblyopia in children. Ophthalmology 110: amblyopia: an analysis of the results of amblyopia therapy 2075–2087 utilizing the pooled data of published studies. Trans Am 42. Pediatric eye disease investigator group (2005) Two-year Ophthalmol Soc 96:431–453 follow up of a six-month randomized trial of atropine ver- 57. Pediatric eye disease investigator group (2007) Stability of sus patching for treatment of moderate amblyopia in chil- visual acuity improvement following discontinuation of dren (reply to letter to editor). Arch Ophthalmol 123: amblyopia treatment in children aged 7 to 12 years. Arch 285–287 Ophthalmol 125:655–659 136 10 Amblyopia Treatment 2009

58. Fletcher MC, Silverman SJ, Boyd J, et al (1969) Biostatistical 70. Bangerter (1958) Orthoptische Behandlung des Begleit studies: comparison of the management of suppression schielens. pleoptik (monokulare orthoptik). Acta XVIII amblyopia by conventional patching, intensive hospital concilium ophthalmologica, Excerpta Medica Foundation pleoptics, and intermittent offi ce pleoptics. Am Orthopt 1:105–128 10 J 19:404–407 71. Iacobucci IL, Archer SM, Furr BA, et al (2001) Bangerter 59. Gregersen E, Rindziunski E (1965) “Conventional” occlu- foils in the treatment of moderate amblyopia. Am Orthopt sion in the treatment of squint amblyopia. a 10-year fol- J 54:84–91 lowup. Acta Ophthalmol (Copenh) 43:462–474 72. Leguire LE, Rogers GL, Bremer DL, et al (1993) Levodopa/ 60. Leiba H, Shimshoni M, Oliver M, et al (2001) Long-term carbidopa for childhood amblyopia. Invest Ophthalmol follow-up of occlusion therapy in amblyopia. Ophthal- Vis Sci 34:3090–3095 mology 108:1552–1555 73. Procianoy E, Fuchs FD, Procianoy L, et al (1999) Th e eff ect 61. Levartovsky S, Oliver M, Gottesman N, et al (1995) Factors of increasing doses of levodopa on children with strabis- aff ecting long term results of successfully treated amblyo- mic amblyopia. J AAPOS 3:337–340 pia: initial visual acuity and type of amblyopia. Br 74. Mohan K, Dhankar V, Sharma A (2001) Visual acuities J Ophthalmol 79:225–228 aft er levodopa adminstration in amblyopia. J Pediatr 62. Rutstein RP, Fuhr PS (1992) Effi cacy and stability of ambly- Ophthalmol Strabismus 38:62–67 opia therapy. Optom Vis Sci 69:747–754 75. Leguire LE, Komaromy KL, Nairus TM, et al (2002) Long- 63. Sparrow JC, Flynn JT (1977) Amblyopia: a long-term fol- term follow-up of l-dopa treatment in children with lowup. J Pediatr Ophthalmol 14:333–336 amblyopia. J Pediatr Ophthalmol Strabismus 39: 326–330 64. Pediatric eye disease investigator group (2005) Two-year 76. Bhartiya P, Sharma P, Biswas NR, et al (2002) Levodopa- follow-up of a 6-month randomized trial of atropine vs carbidopa with occlusion in older children with amblyo- patching for treatment of moderate amblyopia in children. pia. J AAPOS 6:368–372 Arch Ophthalmol 123:149–157 77. Lempert P (2000) Optic nerve hypoplasia and small eyes in 65. Pediatric eye disease investigator group (2008) A random- presumed amblyopia. J AAPOS 4:258–266 ized trial of atropine versus patching for treatment of mod- 78. Lempert P, Porter L (1998) Dysversion of the optic disc and erate amblyopia: follow-up at 10 years of age. Arch axial length measurements in a presumed amblyopic pop- Ophthalmol 126:1039–1044 ulation. J AAPOS 2:207–213 66. Konig HH, Barry JC (2004) Cost eff ectiveness of treatment 79. Lempert P (2008) Retinal area and optic disc rim area in for amblyopia: an analysis based on a probabilistic Markov amblyopic, fellow, and normal hyperopic eyes: a hypothesis model. Br J Ophthalmol 88:606–612 for decreased acuity in amblyopia. Ophthalmology 67. Membreno JH, Brown MM, et al (2002) A cost-utility 115:2259–2261 analysis of therapy for amblyopia. Ophthalmology 109: 80. Altintas O, Yuksel N, Ozkan B, et al (2005) Th ickness of the 2265–2271 retinal nerve fi ber layer, macular thickness, and macular 68. Rahi JS, Cumberland PM, Peckham CS (2006) Does volume in patients with strabismic amblyopia. J Pediatr amblyopia aff ect educational, health, and social outcomes? Ophthalmol Strabismus 42:216–221 Findings from 1958 British birth cohort. BMJ 332: 81. Repka MX, Goldenberg-Cohen N, Edwards AR (2006) 820–825 Retinal nerve fi ber layer thickness in amblyopic eyes. Am 69. van de Graaf ES, van der Sterre GW, van Kempen-du Saar J Ophthalmol 142:247–251 H, et al (2007) Amblyopia and strabismus questionnaire 82. Yen M, Cheng C, Wang A (2004) Retinal nerve fi ber layer (A&SQ): clinical validation in a historic cohort. Graefes thickness in unilateral amblyopia. Invest Ophthalmol Vis Arch Clin Exp Ophthalmol 245:1589–1595 Sci 45:2224–2230 Chapter 11 Best Age for Surgery for Infantile Esotropia: Lessons from the Early vs. Late Infantile Strabismus Surgery Study 11 H.J. Simonsz, G. H. Kolling, and the Early vs. Late Infantile Strabismus Surgery Study Group

Core Messages ■ Th e result of surgery for infantile esotropia (IE) 0.001) of those operated at approximately 49 can be described by the following outcome months recognized the Titmus Housefl y at the parameters: (1) the binocular vision conserved age of 6 years; there was no diff erence in stereop- or regained by early surgery, (2) the postopera- sis beyond Titmus Housefl y. tive angle of strabismus and the long-term stabil- ■ Reoperation rates were 28.7% in the early and ity of alignment, and (3) the number of operations 24.6% in the late group. 8.2% of the children needed to reach these goals or the chance of scheduled for early surgery and 20.1% of the spontaneous reduction of the strabismus into a children scheduled for late surgery had not been microstrabismus without surgery. To judge the operated at the age of 6 years; most developed best age for surgery in a specifi c child with IE, a microstrabismus. Esotropia less than 14° at the expected outcome of surgery should be esti- baseline at approximately 11 months of age had mated according to these parameters. not been operated at the age of 6 years in 35% of ■ Th ere have been no studies with prospectively the cases. Hypermetropia around spher. + 4 assigned early- and late-surgery groups and an increased the likelihood of regression without evaluation according to intention-to-treat, other surgery, underscoring the need of full refractive than the Early vs. Late Infantile Strabismus correction. Surgery Study (ELISSS). Th e primary outcome of ■ Findings of substantially fi ner stereopsis aft er very that study was that 13.5% of those operated at early surgery await confi rmation in a randomized approximately 20 months of age against 3.9% (P = controlled trial.

esotropia with and without nervous system impairment. 11.1 Introduction In a recent study among 627 consecutive strabismus patients younger than 19 years [6], 4.8% had congenital 11.1.1 Defi nition and Prevalence esotropia without and 7.0% congenital esotropia with Infantile esotropia (IE) is defi ned as an esotropia with an nervous system impairment, including any nervous sys- onset before the age of 6 months, with a large angle of tem impairment except speech delay. strabismus, no or mild amblyopia, small to moderate hypermetropia, latent nystagmus, dissociated vertical deviation, limitation of abduction, and absent or reduced 11.1.2 Sensory or Motor Etiology binocular vision, in the absence of nervous system disorders [1, 2]. IE may have diff erent causes, ranging from sensory to IE aff ects approximately 0.25% of the population motor defects. Prematurity, low birth weight, and low [3–5]. A higher prevalence has been found previously in Apgar scores are signifi cant risk factors for IE [5]. Motor studies where little distinction was made between fusion, i.e., translating image disparity information into a 138 11 Best Age for Surgery for Infantile Esotropia

vergence command to facilitate stereopsis, is a complex severity of the nasotemporal pursuit asymmetry [18] cerebral function that may well falter in nervous system and of the latent nystagmus [19]. damage, explaining the bad outcome of early surgery in In cats and macaque monkeys made to squint such cases [7, 8]. On the other hand, if esotropia results shortly aft er birth by cutting the medial rectus muscles 11 from some motor disorder, like a congenital palsy or an [10], cutting the lateral rectus muscles [20], or fi tting anatomical anomaly of an eye muscle or the bony orbit, with prism goggles [21, 22], there is a lack of binocular early surgery may well contribute to regain or conserve horizontal connections in the visual cortex, correlated binocular vision with fi ne stereopsis. with the duration of the lack of binocular vision [22]. As the cause of IE, whether sensory or motor, is the Th e restoration of binocular vision by removal of the predominant determinant of the degree of binocular prism goggles, simulating early surgery, demonstrated vision that may be conserved or regained by surgery, in these animals [18, 22], stresses the feasibility of early there is a strong need for fi ner distinction among the sub- surgery in IE cases when its cause is motor. In another types of IE. animal model, esotropia was found to occur naturally IE should be considered, similar to the working defi - in macaque monkeys [23]. Th is seems more like IE in nition formulated for congenital cerebral palsy [9], as a children than surgically induced esotropia [24], but group of permanent, but not unchanging, disorders with many of the macaques had high hypermetropia [23, strabismus and disability of fusional vergence and bin- 24], their accessory lateral rectus muscle was absent ocular vision, due to a nonprogressive interference, lesion, [25], or their horizontal recti were twice as large as or maldevelopment of the immature brain, the orbit, the those of, albeit younger, controls [24]. eyes, or its muscles, that can be diff erentiated according to location, extent, and timing of the period of development. Such an open matrix fi ts both congenital esotropia without nervous system impairment and congenital 11.1.4 History esotropia with nervous system impairment, and also Whatever its cause, whether sensory or motor, the end includes very early cases of accommodative esotropia state of IE is characterized by lack of binocular vision, that overlap with IE. first described by Claud Alley Worth in 1903 [26] when he wrote: “In the human infant the motor coordinations of the eyes are already partially developed at 11.1.3 Pathogenesis: Lack of Binocular birth. During the first few months of life these serve Horizontal Connections (in the absence of any disturbing influence) to main- in the Visual Cortex tain approximately the normal relative directions of the eyes. … When the fusion faculty has begun to develop, In IE, the horizontal binocular connections above and the instinctive tendency to blend the images formed in below the input layer in the visual cortex, which link ocu- the two eyes … will keep the eyes straight. When the lar dominance columns of the right and left eyes [10], do fusion faculty is fairly well developed, neither hyper- not develop (sensory cause) or cannot develop (motor metropia, nor anisometropia, nor heterophoria can cause). Th ey develop if the inputs from the right and left cause squint. … Sometimes, however, owing to a con- eye are obtained from corresponding images, facilitating genital defect, the fusion faculty develops later than it fusional vergence and stereopsis [11–13]. should, or it develops very imperfectly, or it may never At birth, each eye projects via both visual cortices to develop at all. Then, in this case, there is nothing but the contralateral middle temporal and medial superior the motor coordinations to preserve the normal rela- temporal area, sensitive to motion and disparity, and tive directions of the eyes, and anything which disturbs responsible for ipsiversive OKR, ipsiversive pursuit, ver- the balance of these coordinations will cause a perma- gence, and gaze holding. Accordingly, infants can follow nent squint.” objects moving towards the nose more easily, the so called nasotemporal OKR and pursuit bias. Th e ipsilateral middle temporal and medial superior temporal areas are accessed via the binocular horizontal connections in 11.1.5 Outcome Parameters V1 that only develop if binocular vision is possible. When these fail to develop, the nasotemporal bias per- Several case-series studies opposing this view reported sists and latent nystagmus develops [14–17]. Th e dura- stereopsis in 35–80% aft er surgery at the age of 0–6 tion of the lack of binocular vision determines the months [27–35]. Current US standard age of fi rst surgery 11.2 Outcome of Surgery in the ELISSS 139 is approximately 12–18 months of age, and in many Summary for the Clinician European countries, surgery for IE is performed at the age of 2 or 3 years. Th ere has been a call recently for ■ IE may have many causes, ranging from motor surgery within 2 months of the onset of esotropia [36]. to sensory. Whatever its cause, whether sensory However, there have been no randomized studies with or motor, the end state of untreated IE is charac- prospectively assigned early-surgery and late-surgery terized by lack of binocular vision. If its cause is groups and an evaluation according to intention-to- motor, loss of binocular vision can, in principle, treat. Elliot and Shafi q [37] concluded in their Cochrane be limited by early surgery. review: “As there are no randomised controlled trials in ■ Primary outcome measures of surgery are (1) bin- the area at present, it has not been possible to resolve ocular vision, (2) the angle and long-term stability the controversies regarding … age of intervention in of alignment, and (3) the number of operations or patients with IE. … Th ere is clearly a need for good the chance of spontaneous reduction of the stra- quality trials to be conducted in various areas of IE, in bismus into microstrabismus without surgery. order to improve the evidence base for the management of this condition.” Indeed, one cannot exclude the possibility that in the retrospective case-series studies, without a control group, an occasional child may have been operated that would 11.2 Outcome of Surgery in the ELISSS have straightened to 60˝ stereopsis without surgery. Th ree such cases occurred in the fi rst prospective study by Birch 11.2.1 Reasons for the ELISSS et al. [27] and two in the ELISSS. Th erefore, instead of providing the reader with a quick Early surgery may minimize further loss of the remaining recipe on whether to operate early or late, it seems more binocular vision. Th e fi rst prospective study of surgery appropriate to list and discuss the outcome measures that for IE Birch et al. [27] reported 35% random dot stereop- should be considered when contemplating early, very sis (disparity 400˝ or better) among 84 children operated early, or late surgery in a specifi c child. Th e primary out- at approximately 8.5 months. Sixty-three were aligned come measures are the following: within 5.7°. Th e average number of operations was 1.5. Th ree were not operated and had full stereopsis. Aft er this fi rst prospective study of surgery for IE had been pub- 1. Th e binocular vision conserved or regained by early lished, the need was felt in Europe for a large, prospective, surgery. controlled multicenter trial comparing early surgery for 2. Th e angle of strabismus aft er surgery and the long- IE with late surgery. term stability of alignment. 3. Th e number of operations to reach these goals or the chance of spontaneous reduction of the strabismus into a microstrabismus without surgery. 11.2.2 Summarized Methods of the ELISSS In the ELISSS, all children with IE were included who Th ere are other outcome parameters that should be confi rst presented to one of the participating clinics. Th e sidered. For instance, the child’s psychological and motor ELISSS study committee considered randomization development, and bonding between infant and parents impossible, because it was anticipated that the parents may be improved by early surgery. Th ese need evaluation would not cooperate: One fi rst would have had to inform within disciplines other than pediatric ophthalmology, the parents of the possibility of surgery next week, only however. to postpone surgery for 2 years when the randomization Endophthalmitis aft er strabismus surgery [38] occurs procedure prescribed late surgery [40]. Instead, each of preferentially in fi rst surgery in children under 6 years of the participating clinics chose beforehand whether to age, but it is not yet clear whether its prevalence in young operate all of their eligible patients in the recruitment children diff ers from that in very young children. Finally, period either early or late. Recruited children received general anesthesia may not be without risk in young chil- an extensive baseline examination at 6–18 months of dren. As a case in point, in a recent population-based, ret- age, were assigned to early surgery (6–24 months) or rospective birth cohort study, general anesthesia before late surgery (32–60 months), and were assessed at the the age of 4 years was signifi cantly correlated with learn- age of 6 years. All children who fi rst presented with con- ing disability [39]. vergent IE between 5 and 30° were included. However, 140 11 Best Age for Surgery for Infantile Esotropia

children with pre- or dysmaturity, nystagmus, nervous 60 system defi cit, retardation, dysmorphia or motility disorders other than up- or downshoot in adduction, V- or A-pattern, or limitation of abduction were excluded. 50 11 Following recruitment, the angle of strabismus, refraction, degree of amblyopia, and limitation of abduction were assessed in an extensive baseline examination, 40 based on a test–retest reliability study [41]. Orthoptic examinations, including angle and refraction, were repeated every 6 months. Cases with strongly established fi xation preference and/or signifi cant anisometro- 30 pia underwent appropriate and eff ective occlusion therapy to the point of near spontaneous alternation and central fi xation of the worse eye. Reoperation was 20 undertaken in cases with a residual esotropia of greater than 10°, or in case of overcorrection. Children were evaluated at the age of 6 years in the presence of inde- 10

pendent observers. Endpoints were level of binocular Percentage for unoperated and operated patients vision, manifest angle of strabismus at distance ﬁ xation, remaining amblyopia, number of operations, vertical 0 strabismus, angle at near, and inﬂ uence of surgical 1234567 1234567 technique. early late Degree of binocular vision

Fig. 11.1 Binocular vision at the age of 6 years aft er early or late surgery, stratifi ed according to whether the children had been oper- 11.2.3 Summarized Results of the ELISSS ated (black) or not (white) at the age of 6 years. Categories: (1) Bagolini A total of 58 clinics in 13 countries recruited 532 chil- negative, (2) Bagolini positive, (3) Housefl y positive, (4) Titmus circles 200˝–140˝, (5) Titmus circles 100˝–40˝, (6) all fi gures of Lang dren: 231 children at the age of 11.1 SD 3.7 months Test or TNO 480˝ and 240˝, (7) TNO 120˝–15˝ (See Ref. [57]) (baseline) for early surgery and 301 at the age of 10.9 SD 3.7 months for late surgery. An additional 442 patients screened for inclusion were excluded for various reasons, like prematurity (32), congenital nystagmus (49), or nervous system defi cit (99). No diff erences between groups 11.2.4 Binocular Vision at Age Six were found in the baseline examination apart from a At the age of 6 years, 51.2% of the early vs. 44.7% of the slightly larger angle in the early group [42]. Of 532 late group recognized Bagolini striated glasses, and 13.5% patients, 414 were evaluated at the age of 6 years in the of the early vs. 3.9% (P = 0.001) of the late group recog- presence of independent observers (82.7% of all forms nized the Titmus Housefl y; 3.0% of the early and 3.9% of were signed by the independent observer). Dropout rates the late group had stereopsis beyond Titmus Housefl y were 26.0% in the early and 22.3% in the late group, but (Fig. 11.1). Some children had been operated beyond the no diff erences existed between dropouts and completers set time frame (6–18 and 32–60 months), but “as treated” in the baseline examination, and clinics with many drop- analysis yielded the same result. outs did not have better results. Th e fi nal examinations were performed at the age of 6.8 SD 0.8 years, on average, in the early group and 6.8 SD 0.7 years in the late group. Th e interval between the last operation and the 11.2.5 Horizontal Angle of Strabismus fi nal examination was 4.4 SD 1.5 years in 157 children at Age Six from the early group, and 2.3 SD 1.1 years in 187 children from the late group. Th e number of orthoptic At the age of 6 years, the manifest horizontal angle during examinations in the early group was 11.3 SD 5.2 per fi xation at distance was 2.15° SD 5.45° in the early group patient, including all children who later became drop- (N = 167) and 3.21° SD 6.29° in the late group (N = 231), outs; in the late group, it was 11.4 SD 4.6. wearing full refractive correction. Surprisingly, 35.1% of 11.2 Outcome of Surgery in the ELISSS 141

30 20

10 20

10 Horizontal angle of strabismus –10 Percentage for unoperated and operated patients

–20 0 < = < = < = < = 0< = 4< = 8 < =12< =16< =20< =24> 24 < = < = < = < = 0< = < = 8 < =12< =16< =20< =24> 24 − − − 4 − − − 4 4 –10 01020 30 40 12 8 12 8 early late Horizontal angle of strabismus at baseline Horizontal angle of strabismus (deg)

Fig. 11.2 (Left ) Manifest horizontal angle of strabismus in degrees for both groups at the fi nal examination at the age of 6 years (N = 414), stratifi ed according to whether the children had been operated (black) or not (white). (Right) Relationship between horizontal angle at approximately 11 months and horizontal angle at the age of 6 years. Note that the variation of the horizontal angle of strabismus at approximately 11 months was similar to that at the age of 6 years. Note that one dot may represent more children (See Ref. [57]) the early-surgery group and 34.8% of the late-surgery IE” [45] among older children, 38.4% of the children had group were not aligned within 0–10°, despite the fact that a positive Bagolini test postoperatively, although all chil- the protocol prescribed to continue surgery until align- dren with any form of binocular vision preoperatively ment within 0–10° had been reached. Many children had had been excluded. Th ese children had signifi cantly bet- a small exotropia (especially in the early group), but in ter ocular alignment, which may have been either a cause other cases, a large esotropia existed that had not been or a consequence of the gain of binocular vision. considered a priority by the parents in the period preceding the fi nal examination. It was also surprising that the Summary for the Clinician variation of the angle of strabismus at age 6 was equal to ■ In the ELISSS, children with IE operated around its variation at baseline at 11 months (Fig. 11.2). Th ese the age of 20 months, achieved Bagolini striated fi ndings underscore that surgery for IE is elective and, as glasses or Titmus Housefl y stereopsis more fre- clinicians, we primarily see patients while they are being quently as compared to those operated around treated by us until they are straight. the age of 49 months. ■ No diff erence was found, however, for stereopsis beyond Titmus Housefl y. ■ Alignment was similar aft er early surgery, as 11.2.6 Alignment is Associated with Binocular Vision compared to that aft er late surgery, but a large variation of the angle of strabismus was found at Children with at least Titmus Housefl y stereopsis were the age of 6 years in both groups. better aligned (Fig. 11.3). Better alignment in case of bet- ■ Children with stereopsis were aligned better, which ter binocular vision has been found by Birch et al. [43] may have been either a cause or a consequence of and Fu et al. [44]. In the study “Randomized comparison the gain of binocular vision. of bilateral recession vs. unilateral recession-resection for 142 11 Best Age for Surgery for Infantile Esotropia

Fig. 11.3 Relation between TNO test 120” or better the level of binocular vision and angle of strabismus at distance ﬁ xation for both groups (N=414). Black dots Lang test (all) or TNO test 480” to 240” 11 represent the patients who had not been operated at the age of 6 years. One dot may represent more than one Titmus circles 100” to 40” child (See Ref. [57])

Titmus circles 200” to 140”

Housefly positive

Bagolini positive

Bagolini negative –15 –10 –5 0 510152025 Horizontal manifest angle of strabismus in degrees at age 6 in degrees for operated (grey circles) and unoperated (black) cases

80 11.3 Number of Operations and Spontaneous Reduction into Microstrabismus Without Surgery 70

11.3.1 The Number of Operations Per Child 60 and the Reoperation Rate in the ELISSS In the ELISSS, the number of operations among the chil- 50 dren who completed the study was 1.181 SD 0.67 per child 40

in the early group (N = 171) and 0.996 SD 0.64 in the late Percent group (N = 234), including children who were scheduled for surgery, but had not been operated at the age of 6 years. 30 Children scheduled for early surgery had been fi rst operated at 20.0 SD 8.4 months, but 8.19% (14) had not been 20 operated at the age of 6 years. Children scheduled for late surgery had been fi rst operated at 49.1 SD 12.7 months, but 10 20.09% (47) had not been operated at the age of 6 years. Accordingly, the reoperation rates were 1.181/(1–0.0819)–1 = 28.7% in the early group and 0.996/(1–0.2009)–1 = 24.6% 0 01234 01234 in the late group, including second and third reoperations. early late Among the children operated 2 or 3 times, only a few were operated for consecutive divergence, although consecutive Number of operations Surgery Group divergence occurred frequently (Fig. 11.4). Fig. 11.4 Number of operations per child. Among the children operated 2 or 3 times, only a few were operated for consecutive divergence (black), although consecutive divergence occurred 11.3.2 Reported Reoperation Rates frequently. One child from the early group was operated twice for consecutive divergence (striated). Note that 8.2% from the Reported reoperation rates range from 11% aft er early early group and 20.1% from the late group had not been oper- surgery to 70% aft er very early surgery [46–54]. Studies ated at the age of 6 years (See Ref. [57]) 11.3 Number of Operations and Spontaneous Reduction into Microstrabismus Without Surgery 143 with follow-up between 1 and 2 years [7, 48, 51–53] have regression analysis showed no statistically signifi cant dif- reported reoperation rates between 8 and 35%. Studies ference between clinics concerning chance of reoperation. with 7 or 8 years of follow-up have reported 33% for late To test whether the large diff erences between reported [47], 11% for early [54], and 70% for very early [49] sur- reoperation rates aft er early surgery, mentioned earlier, gery. In a recent population study by Louwagie et al. [4] were due to the diff erences in the duration of follow-up, a over a period of 30 years in Olmsted County, the 130 cases meta-regression was performed. For each study, the mean of IE that had occurred underwent a mean of 1.80 opera- duration of follow-up, the mean age at operation, and the tions during a mean follow-up period of 13.5 years from reoperation rate were obtained from the publication or their date of diagnosis, i.e., a 80% reoperation rate, includ- original data. Th e mean duration of follow-up and mean ing second and third reoperations. Th e median age at age at operation were regressed on the logistically trans- operation was 14 months, the average age was 18 months. formed reported reoperation rate. Th e meta-regression In a multicenter study by Van de Vijver-Reenalda model had an R-squared value of 0.44. Th e infl uence of this et al. [55], reoperation rates were assessed 6–23 years aft er confounding factor was estimated in a multivariate logistic fi rst surgery had taken place among 181 patients. Th ese model. Reoperation rates were adjusted for duration of fol- patients were consecutive cases of the registries of surgery low-up with the meta-regression model and plotted against in each of the seven participating university clinics. Nine the mean age at operation for each study (Fig. 11.5). patients could not be contacted by telephone, and in six Aft er adjustment of the reoperation rates reported patients, the postoperative angle of strabismus 3 months aft er short follow-up periods, reoperation rates became postoperatively was unknown. Of the remaining 166 more similar to the rate reported by Helveston et al. [49] patients, on average 4.33 years old at surgery, 32 had a aft er a long follow-up period. A trend for more reopera- reoperation, in 60% of cases within 2 years aft er the fi rst tions aft er early surgery when compared with that aft er operation. Average reoperation rate was 19.3%. Logistic late surgery can be noted (Fig. 11.5).

100%

Louwagie [4] 80%

Helveston [49] Stager [64] Charles [7] Keenan [51] 60% Kushner [52]

Early [57] Nelson [53] 40% Bartley [47] Late [57]

Helveston [48] Vijver [55] 20% Tolun [92] Reoperation rate

0% 012Age in months 24 36 48 60

Fig. 11.5 Exploratory meta-analysis of studies reporting reoperation rates (closed circles) aft er surgery for IE. “Early” and “Late” refer to the early and late groups of the ELISSS. Th e reoperation rates aft er shorter follow-up periods were corrected for the duration of follow-up with a multivariate logistic model (closed black squares) 144 11 Best Age for Surgery for Infantile Esotropia

three examiners examining one infant, 1.0 signifying 11.3.3 Test-Retest Reliability Studies complete agreement) was 0.80. Th e distribution for the One of the reasons contributing to a higher reoperation largest diff erence between any two of the three measured rate aft er early surgery is the inaccuracy in measuring the angles averaged 6.5°. In 10% of the infants, the largest dif- 11 angle of strabismus in young children. In a test-retest reli- ference between any two of the three measured angles ability study [41] preceding the ELISSS a total of 190 exceeded 10°. Standard deviations and intraclass correla- infants of the age of 12.1 SD 2.5 (range 9–15) months tion coeffi cients were the same for both the methods of were examined in ten university clinics on one day by measurement (Fig. 11.6). three orthoptists. Fift een parameters of the orthoptic In a recent similar study [56], 143 children aged 22.2 examination were assessed that were considered to be of SD 15.0 months (range 2.1–60.2) with esotropia were prognostic importance and, hence, suited to detect and examined by two masked examiners on one or two occa- correct for disparities between the groups in the ELISSS. sions yielding 199 test-retest pairs for prism and alternate In 144 of the 190 infants, the manifest horizontal angle of cover test at distance fi xation and 239 at near fi xation. For strabismus was estimated, either with prisms and corneal angles greater than 11.3°, the 95% limits of agreement on refl exes during fi xation of an object with a light at 50 cm a measurement and on a diff erence between two mea- or by estimation of the location of the corneal refl ex rela- surements were ±4.2 and ±5.9° for prism and alternate tive to the pupil during fi xation of an object with a light at cover test at distance and ±4.7 and ±6.7° at near. For 50 cm. Th e angle of strabismus averaged 21° for the fi rst, angles of 5.7–11.3°, they were ±2.3 and ±3.3° at distance second, and third examinations, with approximately and ±1.9 and ±2.7° at near. equal standard deviations for all three examinations. Th e From these two studies, it is evident that one of the intraclass correlation coeffi cient (diff erences between reasons contributing to a higher reoperation rate aft er

10 Angle (degrees) measured by second or third orthoptist

0 0 10 20 30 40 50 Angle (degrees) measured by first or second orthoptist (N=144 children x 3 pairs of measurements)

Fig. 11.6 144 infants at approximately one year of age were examined in ten university clinics on one day by three orthoptists or, rarely, by a strabismologist. Th e horizontal angle of strabismus was measured, either with prisms and corneal refl exes or by estimation of the location of the corneal refl ex relative to the pupil. Larger circles represent more measurements 11.3 Number of Operations and Spontaneous Reduction into Microstrabismus Without Surgery 145 early surgery is the inaccuracy in measuring the angle of fi xation, 3 months postoperatively. Th ey estimated the strabismus in young children. reoperation rate at almost double, probably because of an observer bias, as patients who come for reoperation are more vividly remembered (Fig. 11.7).

11.3.4 Relation Between the Postoperative Angle of Strabismus and the Reoperation Rate 11.3.5 Scheduled for Surgery, but no Surgery Done at the End Variance in the preoperative measurement of the hori- of the Study at the Age of Six Years zontal angle results in variance of the postoperative angle. However, does postoperative variance of the angle cause In the ELISSS, children scheduled for early surgery had additional reoperations? been fi rst operated at 20.0 SD 8.4 months, but 8.19% (14) In the study by Van de Vijver-Reenalda et al. [55] on had not been operated at the age of 6 years. Children reoperation rates 6–23 years aft er fi rst surgery in children scheduled for late surgery had been fi rst operated at 49.1 operated at 4.33 years, on average, the average reoperation SD 12.7 months, but 20.09% (47) had not been operated rate was 19.3%. Th e reoperation rate was only 7.3%, how- at the age of 6 years. ever, for those with a residual angle of −4 to +4° (82, In his analysis of 500 children with IE [1], Costenbader 49.4%), 3 months postoperatively. Th e reoperation rate identifi ed the size and variability of the angle, onset at was 25% for children who were divergent in excess of 5° birth, duration of strabismus, age at presentation, age at and 29% for children between 10 and 14° convergent, 3 surgery, hypertropia, and amblyopia as “factors that infl u- months postoperatively. ence cure”. Ahead of his time, Costenbader included 80 For comparison, eight strabismologists, the heads of cases in his analysis who had not been operated at all. He the departments where the retrospective study had been analyzed his data truly in accordance with the “intention done, were asked to give their estimations of the reopera- to treat” principle. Th ese 80 children had “alignment and tion rate based on the angle of strabismus at distance fusion” in 76% of the cases, when compared with 38.4%

90 N=12 80

N=17 30 N=4 N=51 20

10 N=62 N=20 Observed reoperation rate and experts' estimates (%) 0 < -5° -4 to -1° 0 to 4° 4 to 9° 10 to 14° > 14° Postoperative angle of strabismus in degrees, 3 months postoperatively

Fig. 11.7 Observed reoperation rate in relation to angle of strabismus 3 months postoperatively in 166 patients operated between 6 and 23 years previously (black) and average estimates by eight strabismologists (white) 146 11 Best Age for Surgery for Infantile Esotropia

of children operated once and 36% of children operated 11.3.7 Predictors of Spontaneous twice. In the studies by Costenbader [1], by Birch (1990) Reduction into Microstrabismus and in the ELISSS [57], children who had been scheduled for surgery but who had not been operated at fi nal assess- In the ELISSS, of all parameters assessed in the baseline 11 ment had better binocular vision than those who had examination at approximately 11 months, only the angle been operated. of strabismus at baseline predicted, to some extent, Spontaneous resolution of infantile strabismus has fi rst whether a child had been operated at the age of 6 years or been reported by Clarke & Noel [58]. In a study by the not (Fig. 11.9). Among children with an angle equal or Pediatric Eye Disease Investigator Group [59], among 170 smaller than 13° at baseline at approximately 11 months, children with IE (age ±3 months at recruitment), of those 34.9% had not been operated at the age of 6 years. who had had an angle of strabismus >21.8° during two Hypermetropia around spher. + 4 increased the likeli- examinations at least one week apart, 2.4% had an angle hood of regression without surgery, emphasising the <4.6° at ±7 months. Among those children who had had need for full refractive correction (there may have been an angle of strabismus >11.3° during two examinations at some very early cases of accommodative esotropia). Age recruitment, 27% had an angle <4.6° at ±7 months. at recruitment, age that strabismus reportedly had started Reduction of the angle within 5° frequently results in and degree of amblyopia at baseline examination seemed microstrabismus with peripheral fusion, central sup- not predictive. pression, and a favorable appearance. Due to the peripheral fusion, the strabismus remains stable and rarely needs additional surgery, as has been found for small 11.3.8 Random-Eff ects Model Predicting angles postoperatively in the study by Van de Vijver the Angle and its Variation et al. [55]. In the 532 children of the ELISSS, the angle of strabismus, refraction, and visual acuity was assessed at baseline at approximately 11 months and every 6 months thereaft er, 11.3.6 Spontaneous Reduction of the Angle until the fi nal evaluation at the age of 6 years. Th e result- In the ELISSS, more than half of the children who were ing, slightly more than 6,000, orthoptic exams were used scheduled for surgery, but had not been operated at the to construct a random-eff ects model [61] that forecasts age of 6 years, had a spontaneous reduction of the strabis- the expected angle and its variation years ahead, on the mus into a microstrabismus (Fig. 11.8). basis of one or more measurements of the angle and Th ere are few studies with similar longitudinal mea- refraction in infancy. surements of the angle of strabismus in a large group of Angles of strabismus measured at diff erent ages and children. In a recent study by Pediatric Eye Disease the refraction of the patient can be entered in the model. Investigator Group [60], the angle of strabismus was mea- On entering successive measurements of the angle of sured in 81 children with IE aged 6.0 ± 1.7 months (range strabismus, the model adjusts the slope, i.e., yearly 2.4–9.5) at baseline and at 6-week intervals for 18 weeks, increase or decrease of the expected angle, according to using prism and alternate cover test at near (70% of the the trend. Th e uncertainty about the slope decreases children) or a modifi ed Krimsky at near (30%). In 20%, with additional measurements because the random all four measurements were within 2.9° or less than one eff ect of the slope of the lines decreases. Th e uncertainty another. In 46%, any two of the four measurements dif- about the slope is compounded by additional variation fered by 8.5° or more. of the angle around this slope for an individual child Could we have distinguished the ELISSS children who (Fig. 11.10). were scheduled for surgery but, in the end, were never In simulations with the random-eff ects model, it was operated, at an early age? In other words, can the reduc- found that the chance of a spontaneous reduction of a tion of the angle be predicted and, hence, unnecessary strabismus into a microstrabismus is considerable when operations be avoided in individual cases by waiting? Th is an angle of strabismus 14° or less is found repeatedly at line of reasoning only pertains to the majority of cases the age of 1 or 2 years. In the ELISSS, esotropia 13° or less where microstrabismus with peripheral fusion is the best at baseline at approximately 11 months of age had not possible result. One cannot exclude the rare possibility been operated at the age of 6 years in 35% of the cases that an occasional child, with a pure motor cause of IE, (Fig. 11.7). If the angle is large on multiple measurements, would achieve full binocular vision with 60 arc seconds the chance that the esotropia will decrease into a stereopsis by very early surgery. microstrabismus spontaneously is very small. 11.3 Number of Operations and Spontaneous Reduction into Microstrabismus Without Surgery 147

10 Horizontal angle of strabismus 0

–10 0 12 24 36 48 60 72 84 96 Age at examination (months)

10 Horizontal angle of strabismus

–10 0 12 24 36 48 60 72 84 96 Age at examination (months)

Fig. 11.8 Th e upper panel shows the 6-monthly measurements of the angle of strabismus in those ELISSS children who had been scheduled for early surgery at baseline at approximately 11 months of age, but had not been operated at the age of 6 years (14, 8.2%). Th e lower panel shows these measurements for the children who had been scheduled for late surgery, but had not been operated at the age of 6 years (47, 20.1%). Th ese children correspond to the white bars in Figs. 11.1, 11.2, and 11.9 148 11 Best Age for Surgery for Infantile Esotropia

30 In the model, refractive error exerted its largest inﬂ uence, i.e., causing the largest chance of spontaneous reduction into a microstrabismus, at a spher. + 4. Some children in the ELISSS study population may 11 actually have been very early cases of accommodative 20 esotropia. In case of hypermetropia, especially with convergence excess, a large reduction in the angle may occur aft er ﬁ tting full correcting glasses, thereby avoiding surgery.

10 Summary for the Clinician ■ Th e chance of a spontaneous reduction of the

Percentage for unoperated and operated patients esotropia into microstrabismus is considerable when an angle of strabismus of 13° or less is 0 £ £ £ £ £ £ £ > 29 £ £ £ £ £ £ £ > 29 = 5 = 9 = 13 = 17 = 21 = 25 = 29 = 5 = 9 = 13 = 17 = 21 = 25 = 29 found repeatedly at the age of 1 year. £ ■ Early Late Fit full-correcting glasses in case of hypermetropia Horizontal angle of strabismus (degrees) accompanying esotropia at an early age because a large reduction of the angle of strabismus can Fig. 11.9 Angle of strabismus at baseline at approximately 11 be achieved without surgery and with better bin- months for all 414 operated (black) and unoperated (white) ocular vision. patients who underwent the ﬁ nal examination at the age of 6 years (same group as in Figs. 11.1 & 11.2). Children who had not been operated at the age of 6 years (white bars) had had smaller angles at baseline (See Ref. [57])

40 40

35 35

30 30

25 25

20 20

15 15 Measured angle (deg.) Measured angle (deg.) 10 10

5 5

0 0 0 122436486072 0 122436486072 Age (months) Age (months)

Fig. 11.10 Random-eff ects model predicting the angle and its variation based on one or more measurements of the angle and refraction in infancy. For the construction of this model, the random eff ect for a patient was defi ned as the deviation of the average angle, the fi xed eff ect. A vector was defi ned based on age and spherical equivalent of the patient. A covariance matrix of the random- eff ects estimations was defi ned and fi lled with the values from the approximately 6,000 orthoptic exams in 532 children. Th e model predicts the average angle in relation to age. A linear relation suffi ced. Th e variance around the prediction (curved lines represent one and two standard deviations) consists of uncertainty in the estimations, random eff ects and the residuals. Left : an example prediction based on three increasing angles measured at 9, 12 and 15 months. Right: an example prediction where the angle decreases in successive measurements; the chance that spontaneous reduction into a microstrabismus occurs is considerable References 149

population-based study from Olmsted County, Minnesota, Appendix 1965 to 1994. Arch Ophthalmol 127:200–203 Members of the Early vs. Late Infantile Strabismus Surgery 5. Mohney BG, Erie JC, Hodge DO, Jacobsen SJ (1998) Study Group were: (Austria) A. Langmann, S. Lindner, Congenital esotropia in Olmsted County, Minnesota. S. Priglinger, M. Raab, H. Th aller-Antlanger, D. Koschkar- Ophthalmology 105:846–850 Moser, H. Gruber-Luka, R. Führer, S. Harrer, K. Rigal, 6. Mohney BG (2007) Common forms of childhood strabis- R. Pelz, B. Puchhammer, A. Th aler, E. Moser, K. Schmidt, mus in an incidence cohort. Am J Ophthalmol 144: (Belgium) M. Spiritus, M. van den Broeck, S. Vandelannoitte, 465–467 A. Finck, P. Evens, D. Godts, (France) M. Bourron- 7. Charles S, Moore A (1992) Results of early surgery for Madignier, S. Vettard, O. Benhadj, (Germany)E-Ch. infantile esotropia in normal and neurologically impaired Schwarz, G. Wunsch, C. Jandeck, S. Lutt-Freund, D. Jüptner- infants. Eye 6:603–606 Johanning, E. Sommer, G. Hochmuth, G. Gusek-Schneider, 8. Holman RE, Merritt JC (1986) Infantile esotropia: Schürhoff , A. Boss, A. Zubcov, B. Herrmann, G. Kommerell, results in the neurologic impaired and “normal” child at B. Lieb, R. Weidlich, U. Wittenbecher, E. Schulz, K. Rettig, NCMH (six years). J Pediatr Ophthalmol Strabismus 23: G. Kolling, B. Stoll, B. Käsmann, E. Grintschuk, A. Kirsch, 41–45 T. Schmidt, M. Klopfer, C. Ecker, K.P. Boergen, O. Ehrt, 9. Surveillance of cerebral palsy in Europe (SCPE) (2002) H.D. Schworm, B. Lorenz, B. Derr, (Great Britain) Prevalence and characteristics of children with cerebral C.J. McEwen, I. Marsh, L. Gannon, C. Timms, D. Taylor, palsy in Europe. Dev Med Child Neurol 44:633–640 P. Fells, J.P. Lee, (Italy) R. Frosini, L. Campa, F. Carta, 10. Hubel DH, Wiesel TN (1965) Binocular interaction in A. Carta, (Netherlands) L. Wenniger-Prick, Y Everhard- striate cortex of kittens reared with artifi cial squint. Halm, A.G. Tjiam, M. van Duuren, H.J. Simonsz, H.M. J Neurophysiol 28:1041–1059 van Minderhout, (Norway) G. Hanken, A. Angermeier, 11. Tychsen L, Burkhalter A (1995) Neuroanatomic abnor- O.H. Haugen, L. Steene Eriksen, B. A. Olsen, E. Dueland, malities of primary visual cortex in macaque monkeys W. Evans Lothe, T. Bulie, H.P. Brinck, T. Kalseth, (Sweden) with infantile esotropia: preliminary results. J Pediatr G. Ladenvall, A.B. Edvinsson, A. Wallin, R. Alvarado, Ophthalmol Strabismus 32:323–328 M. Fornander, U. Lidén, L. Lindberg, I. Wiklund, 12. Tychsen L, Burkhalter A (1997) Nasotemporal asymme- G. Lennerstrand, B. Derouet-Eriksson, B. Sunnqvist, tries in V1: ocular dominance columns of infant, adult, G. Gunnarssen, P. Jakobsson, G. Kvarnström, M. Lindberg, and strabismic macaque monkeys. J Comp Neurol 388: D. Grandell, K. Johansson, A.-L. Galin, I. Axelsson, B.-M. 32–46 Petersson, (Switzerland) G. Klainguti, J. Strickler, 13. Tychsen L, Wong AM, Foeller P, Bradley D (2004) Early K. Landau, B. Baerlocher, (Turkey) G. Haciyakupoglu, versus delayed repair of infantile strabismus in macaque A. Sefi k Sanaç, E. Cumhur Sener, S. Demirci, N. Erkam, monkeys: II. Eff ects on motion visually evoked responses. Huban Atilla, N. Erda, A. Tulin Berk. Statististical analy- Invest Ophthalmol Vis Sci 45:821–827 sis of the ELISSS was performed by K. Unnebrink of the 14. Hoff mann KP (1983) Eff ects of early monocular depriva- Coordination Center for Clinical Trials, University tion on visual input to cat nucleus of the optic tract. Exp Hospital Heidelberg. All other statistical analyses were Brain Res 51:236–246 performed by M.J.C. Eijkemans of the Department of 15. Hoff mann KP, Distler C (1986) Th e role of direction selec- Public Health, Erasmus Medical Center, Rotterdam. tive cells in the nucleus of the optic tract in the cat and monkey during optokinetic nystagmus. In: Keller EL, Zee DS (eds) Adaptive processes in visual and oculomotor systems. Pergamon: New York, pp 261–266 References 16. Hoff mann KP, Bremmer F, Th iele A, Distler C (2002) 1. Costenbader FD (1961) Infantile esotropia. Trans Am Directional asymmetry of neurons in cortical areas MT Ophthalmol Soc 59:397–429 and MST projecting to the NOT-DTN in macaques. 2. Von Noorden GK. (1988) A reassessment of infantile J Neurophysiol 87:2113–2123 esotropia. XLIV Edward Jackson memorial lecture. Am 17. Tychsen L, Lisberger SG (1986) Maldevelopment of visual J Ophthalmol 105:1–10 motion processing in humans who had strabismus with 3. Greenberg AE, Mohney BG, Diehl NN, Burke JP (2007) onset in infancy. J Neurosci 6:2495–2508 Incidence and types of childhood esotropia: a population- 18. Hasany A, Wong A, Foeller P, Bradley D, Tychsen L (2008) based study. Ophthalmology 114:170–174 Duration of binocular decorrelation in infancy predicts the 4. Louwagie CR, Diehl NN, Greenberg AE (2009) Mohney severity of nasotemporal pursuit asymmetries in strabis- BG. is the incidence of infantile esotropia declining? A mus macaque monkeys. Neuroscience 156:403–411 150 11 Best Age for Surgery for Infantile Esotropia

19. Richards M, Wong A, Foeller P, Bradley D, Tychsen L 34. Ing MR, Okino LM (2002) Outcome study of stereopsis in (2008) Duration of binocular decorrelation predicts the relation to duration of misalignment in congenital esotro- severity of latent (fusion maldevelopment) nystagmus in pia. J AAPOS 6:3–8 strabismic macaque monkeys. Invest Ophthalmol Vis Sci 35. Wright KW, Edelman PM, McVey JH, Terry AP, Lin M 11 49:1872–1878 (1994) Highgrade stereo acuity aft er early surgery for con- 20. Crawford ML, von Noorden GK (1979) Th e eff ect of short- genital esotropia. Arch Ophthalmol 112:913–919 term experimental strabismus on the visual system in 36. Tychsen L (2005) Can ophthalmologists repair the brain in macaca mulatta. Invest Ophthalmol Vis Sci 18:496 infantile esotropia? Early surgery, stereopsis, monofi xation 21. Crawford ML, von Noorden GK (1980) Optically induced syndrome, and the legacy of Marshall Parks. J AAPOS concomitant strabismus in monkeys. Invest Ophthalmol 9:510–521 Vis Sci 19:1105–1109 37. Elliott S, Shafi q A (2005) Interventions for infantile esotro- 22. Wong AM, Foeller P, Bradley D, Burkhalter A, Tychsen L pia. Cochrane Database of Systematic Reviews Issue 1. Art. (2003) Early versus delayed repair of infantile strabismus No.: CD004917. DOI: 10.1002/14651858.CD004917.pub2 in macaque monkeys: I. ocular motor eff ects. J AAPOS 38. Simonsz HJ, Batstra MR, Mooy CM, van Leeuwen WB, 7:200–209 Löffl er KU, Hartwig NG (2009) Age-related immune 23. Kiorpes L, Boothe R (1981) Naturally occurring strabis- defects causing endophthalmitis aft er strabismus surgery mus in monkeys (Macaca nemestrina). Invest Ophthalmol in young children or in elderly. Invest Ophthalmol Vis Sci Vis Sci 20:257–263 50:ARVO E-Abstract 1134 24. Tychsen L, Richards M, Wong A, Foeller P, Burhkalter A, 39. Wilder RT, Flick RP, Sprung J, Katusic SK, Barbaresi WJ, Narasimhan A, Demer J (2008) Spectrum of infantile Mickelson C, Gleich SJ, Schroeder DR, Weaver AL, Warner esotropia in primates: behavior, brains, and orbits. J DO (2009) Early exposure to anesthesia and learning dis- AAPOS 12:375–380 abilities in a population-based birth cohort. Anesthesiology 25. Boothe RG, Quick MW, Joosse MV, Abbas MA, Anderson 110:796–804 DC (1990) Accessory lateral rectus orbital geometry in nor- 40. Early vs late infantile strabismus surgery study group mal and naturally strabismic monkeys. Invest Ophthalmol (1993a) Th e protocol for the Early vs. Late infantile strabis- Vis Sci 31:1168–1174 mus surgery study. Strabismus 1:135–157. 26. Worth CA (1903) Squint: its causes, pathology and treat- 41. Early vs late infantile strabismus surgery study group ment. John Bale, Sons and Danielsson, London (1993b) How accurate is orthoptic examination at age one? 27. Birch EE, Stager DR, Berry P, Everett ME (1990) Prospective Strabismus 1:75–83 assessment of acuity and stereopsis in amblyopic infantile 42. Meyer K, Breitschwerdt H, Kolling GH, Simonsz HJ (1998) esotropes following early surgery. Invest Ophthalmol Vis Th e early vs. late infantile strabismus surgery study: do Sci 31:758–765 sources for bias exist in this non-randomized trial? Br 28. Birch EE, Stager DR, Everett ME (1995) Random dot stere- J Ophthalmol 82:934–938 oacuity following surgical correction of infantile esotropia. 43. Birch EE, Felius J, Stager DR Sr, Weakley DR Jr, Bosworth J Pediatr Ophthalmol Strabismus 32:231–235 RG (2004) Pre-operative stability of infantile esotropia 29. Birch E, Stager D, Wright K, Beck R (1998) Th e natural his- and post-operative outcome. Am J Ophthalmol 138: tory of infantile esotropia during the fi rst six months of 1003–1009 life. Pediatric eye disease investigator group. J AAPOS 44. Fu VL, Stager DR, Birch EE (2007) Progression of inter- 2:325–328 mittent, small angle, and variable esotropia in infancy. 30. Birch EE, Stager DR Sr (2006) Long-term motor and sen- Invest Ophthalmol Vis Sci 48:661–664 sory out comes aft er early surgery for infantile esotropia. 45. Polling JR, Eijkemans MJ, Esser J, Gilles U, Kolling GH, J AAPOS 10:409–413 Schulz E, Lorenz B, Roggenkämper P, Herzau V, Zubcov A, 31. Ing MR, Costenbader FD, Parks MM, Albert OO (1966) Ten Tusscher MP, Wittebol-Post D, Gusek-Schneider GC, Early surgical treatment for congenital esotropia. Am J Cruysberg JR, Simonsz HJ (2009) A randomised compari- Ophthalmol 652:1419–1427 son of bilateral recession vs. unilateral recession-resection as 32. Ing MR (1981) Early surgical alignment for congenital surgery for infantile esotropia. Br J Ophthalmol 93:954–957 esotropia. Trans Am Ophthalmol Soc 79:625–663 46. Arnoult JB, Yeshurun O, Mazow ML (1976) Comparative 33. Ing MR (1995) Surgical alignment prior to six months of study of the surgical management of congenital esotropia age for congenital esotropia. Trans Am Ophthalmol Soc of 50 prism diopter or less. J Pediatr Ophthalmol 93:135–146 13:129–131 References 151

47. Bartley GB, Dyer JA, Ilstrup DM (1985) Characteristics of scheelzien: een retrospectief onderzoek. Ned Tijdschr recession-resection and bimedial recession for childhood Geneesk 143:2121 esotropia. Arch Ophthalmol 103:190–195 56. Pediatric Eye Disease Investigator Group (2009) 48. Helveston EM, Ellis FD, Schott J, Mitchelson J, Weber JC, Interobserver reliability of the prism and alternate cover Taube S, Miller K (1983) Surgical treatment of congenital test in children with esotropia. Arch Ophthalmol 127: esotropia. Am J Ophthalmol 96:218–228 59–65 49. Helveston EM, Neely FN, Stidham DB, Wallace DK, Plager 57. Simonsz HJ, Kolling GH, Unnebrink K (2005) Final report DA, Sprunger DT (1999) Results of early alignment of con- of the early vs. late infantile strabismus surgery study genital esotropia. Ophthalmology 106:1716–1726 (ELISSS), a controlled, prospective, multicenter study. 50. Hiles DA, Watson BA, Biglan AW (1980) Characteristics of Strabismus 13:169–199, Erratum (2006) Strabismus 14: infantile esotropia following early bimedial rectus reces- 127–128 sion. Arch Ophthalmol 98:697–703 58. Clarke WN, Noel LP (1982) Vanishing infantile esotropia. 51. Keenan JM, Willshaw HE (1992) Outcome of strabismus Can J Ophthalmol 17:100–102 surgery in congenital esotropia. Br J Ophthalmol 76: 59. Pediatric Eye Disease Investigator Group (2002) 342–345 Spontaneous resolution of early onset-esotropia: experi- 52. Kushner BJ, Morton GV (1984) A randomized comparison ence of the congenital esotropia observational study. Am J of surgical procedures for infantile esotropia. Am J Ophthalmol 133:109–118 Ophthalmol 98:50–61 60. Pediatric Eye Disease Investigator Group, Christiansen SP, 53. Nelson LB, Calhoun JH, Sion JW, Wilson T, Harley RD Chandler DL, Holmes JM, Arnold RW, Birch E, Dagi LR, (1987) Surgical management of large angle congenital Hoover DL, Klimek DL, Melia BM, Paysse E, Repka MX, esotropia. Br J Ophthalmol 71:380–383 Suh DW, Ticho BH, Wallace DK, Weaver RG Jr (2008) 54. Tolun H, Dikici K, Ozkiris A (1999) Long-term results of Instability of ocular alignment in childhood esotropia. bimedial rectus recessions in infantile esotropia: J Pediatr Ophthalmology. 115:2266–2274 Ophthalmol Strabismus 36:201–205 61. Simonsz HJ, Eijkemans MJC, Early vs Late Strabismus 55. Van de Vijver-Reenalda H, Polling JR, Simonsz HJ, Surgery Study Group (2006). Natural course of infantile Cruysberg JRM, Kommerell G, Schulz E, Wenniger-Prick esotropia: angle of strabismus and refraction in the Early LJJM (1999) Cumulatieve kans op heroperatie gerelateerd vs. Late Strabismus Surgery Study. Invest Ophthalmol Vis aan de postoperatieve scheelzienshoek bij congenitaal Sci 47:ARVO E-Abstract 2934 Chapter 12 Management of Congenital Nystagmus with and without Strabismus 12 Anil Kumar, Frank A. Proudlock, and Irene Gottlob

Core Messages ■ Congenital nystagmus consists of involuntary spectacles, contact lenses (CL), or low visual periodic to-and-fro oscillations of the eye, which aids. are usually horizontal and present within the fi rst ■ Recently, medical treatment for congenital nys- 3 months of life. tagmus with memantine and gabapentin has been ■ Congenital nystagmus can be idiopathic or occur shown to reduce nystagmus intensity and to in association with defects in the aff erent visual increase visual acuity. Baclofen is benefi cial in the system such as albinism, congenital retinal dystro- management of congenital PAN. phies or congenital retinal dysfunction disorders ■ Surgery in congenital nystagmus is used to cor- (such as achromatopsia and congenital stationary rect the anomalous head posture (AHP) and to night blindness (CSNB) ), congenital optic atrophy, dampen the nystagmus. optic nerve hypoplasia, and congenital cataracts. ■ For Anderson−Kestenbaum- like procedures var- ■ Congenital nystagmus need to be diff erentiated ious extents of surgery have been proposed by from manifest latent nystagmus (MLN) and con- diff erent surgeons. However, if the head turn is genital periodic alternating nystagmus (PAN) as signifi cant, only limitation of motility due to a the management of these conditions diff ers. large extent of surgery will correct the head turn. ■ Several compensatory mechanisms exist in con- ■ If the patient has a squint, care needs to be taken genital nystagmus, which tend to decrease the that Anderson−Kestenbaum-like procedures are nystagmus and thus improve the visual acuity. performed on the dominant or fi xing eye. Th ese mechanisms need to be analyzed carefully Strabismus correction is best planned during the because their understanding is important for the same surgical session on the non-fi xing eye. patient’s management. ■ Surgery causing artifi cial divergence (exophoria) ■ Various modes of management are available for is benefi cial in patients with binocular vision patients with congenital nystagmus such as opti- and damping of nystagmus on convergence. cal, medical, and surgical treatment. A combina- Combination of Anderson−Kestenbaum-like pro- tion of treatment options might be helpful to cedures and artifi cial divergence surgeries have achieve the best outcome. been shown to be benefi cial. ■ Th e incidence of signifi cant refractive errors in ■ Recently, tenotomies of extraocular muscles have patients with congenital nystagmus is around been advocated for dampening nystagmus and 85%. Hence, correcting refractive errors improves for increasing the null region. However, the exact visual acuity and is important at an early age to mechanism is not fully understood and further prevent ambylopia. Optical treatment can involve studies are needed. 154 12 Management of Congenital Nystagmus with and without Strabismus

Before During 12.1 Overview Treatment Treatment Th e management of congenital nystagmus presents a complex problem, which requires the accurate diagnosis 12 of the underlying causes of congenital nystagmus and an CIN understanding of the compensatory mechanisms used. Diagnosis can involve detailed clinical examination with ancillary testing such as the eye movement recordings and electrodiagnostics. It is important to delineate Memantine SN between the diﬀ erent forms of congenital nystagmus such as congenital periodic alternating nystagmus (PAN) and manifest latent nystagmus (MLN) before treatment is considered. Treatment of congenital nystagmus is rapidly evolv- ing, with new methods of treatment emerging which are CIN now proving to be beneﬁ cial. Th e armamentarium of treatment of congenital nystagmus includes optical, medical, and surgical treatments. Currently, in most nystag-

mus forms there is no defi nite answer as to which is the Gabapentin SN best treatment option. Th is chapter highlights the diff erent modes of treatment. Th e fi rst section of this chapter discusses in detail the clinical characteristics of patients with congenital nystagmus with and without sensory defi cit, MLN, and PAN. In the second section, the compensatory mechanism CIN involved and methods to identify them are considered. Th e third section discusses the treatment options available for congenital nystagmus. Placebo

SN 12.1.1 Congenital Nystagmus with and Without Sensory Deﬁ cits

Congenital nystagmus consists of involuntary periodic 3º to-and-fro oscillations of the eye. It usually presents 1sec within the fi rst 3 months of life; however, onset as late as 12 months to 10 years has been reported [1]. Th e inci- Fig. 12.1 Original horizontal eye movement recordings of right dence of congenital nystagmus is estimated to be 1 in eyes of (fi rst row) a patient with congenital idiopathic nystagmus 2,000, in a population-based survey done in UK. (CIN) and (second row) a patient with secondary nystagmus (SN) Th e eye movements in congenital nystagmus are associated with albinism before and during memantine treatment; (third row) a patient with CIN and (fourth row) a patient with SN mainly in the horizontal plane, although they can be ver- associated with achromatopsia before and during gabapentin treat- tical or torsional, or in a combination of diff erent planes. ment; (fi ft h row) a patient with SN and (sixth row) a patient with SN Congenital nystagmus is oft en described in the literature associated with albinism before and during placebo treatment at as being a jerk nystagmus with accelerating slow phase; examinations one and four. Eye movements to the right are repre- however, IIN may show diff erent waveforms that usually sented by an upward defl ection, and eye movements to the left by a downward defl ection. Th e eye movement recordings show the vari- vary with eccentricity. Frequently, congenital nystagmus ability in waveforms with the common occurrence of an underly- consists of underlying pendular oscillations interrupted ing pendular waveform. Th ey also show reduction of intensity aft er by regularly occurring foveating saccades (quick phases) treatment with memantine and gabapentin but not with placebo as shown in Fig. 12.1. Nystagmus intensity oft en changes with the direction of gaze. Th e region of lowest nystag- fi xation for optimal vision with the head position being mus intensity and longest foveation periods is known as used to maintain vision in the null region. Consequently, the “null region.” Th is is oft en the preferred region of patients oft en exhibit an anomalous head posture (AHP) 12.1 Overview 155 if the null region is eccentric. Typically, the oscillation retinal dysfunction disorders (such as achromatopsia and drift s toward the null region with the drift becoming congenital stationary night blindness (CSNB) ), and con- accentuated further away from the null region. Th is genital cataracts. To assess visual potential when treating results in the quick phases usually beating away from the a patient, it is important to carefully diagnose whether an null region with slow phases oft en accelerating toward aff erent visual defect is present. Ocular albinism is fre- the null region. quently misdiagnosed as idiopathic nystagmus as the Congenital nystagmus can be idiopathic with the most phenotypical characteristics might be subtle. Figure 12.2 likely cause being abnormal development of the brain shows clinical signs seen in a patient with oculocutane- areas controlling eye movements and gaze stability. It can ous albinism as well as in a patient with ocular albinism. also occur in association with defects in the aff erent Th e patient with ocular albinism has dark hair and skin, visual system such as albinism, congenital optic atrophy, very mild iris transillumination, but a hypopigmented optic nerve hypoplasia, congenital retinal dystrophies or fundus. Both patients have foveal hypoplasia to varying

Oculocutaneous Ocular Albinism (OCA) Albinism (OA)

a b

Appearance

c d

Iris transillumination

Fig. 12.2 Phenotypical e f characteristics of patients with oculocutaneous (a, c, e, g) and ocular (b, d, f, h) albinism. Th e patient with Fundoscopy oculocutaneous albinism has light hair and more prominent iris transillumination than the patient with ocular albinism. Both g h patients have fundus hypopigmentation, macular hypoplasia, and small optic Optical nerves. Optical coherence Coherence tomography (OCT) shows Tomography foveal thickening in both patients (g, h) with total absence of foveal pit in the patient with oculocutaneous albinism (g) 156 12 Management of Congenital Nystagmus with and without Strabismus

degrees as shown using optical coherence tomography 12.1.1.1 The Clinical Characteristics (OCT). Both patients had increased crossing of optical of Congenital Nystagmus nerve ﬁ bers in the chiasm shown on visual evoked potential examination (see Fig. 12.3d). ■ Onset in infancy 12 Th e diﬀ erent causes of nystagmus can be diagnosed by ■ Nystagmus is mainly horizontal and conjugate detailed clinical examination aided by electrodiagnostics ■ Eye movement recordings are usually horizontal (electroretinograms (ERGs) and visual evoked potentials waveforms (both pendular and jerk) that vary with (VEPs) ) (Fig. 12.3). eccentricity

Electroretinogram

a Normal Scotopic Photopic Flicker

b Congenital Stationary Night Blindness

Fig. 12.3 Examples of c Achromatopsia scotopic, photopic, and ﬂ icker electroretinograms (ERGs) of (a) a normal 500µv subject, (b) a patient with 20µv 10µv congenital stationary night 20ms 20ms 20ms blindness (CSNB) with a negative scotopic ERG, and F (c) a patient with achro- d Visual Evoked Potentials z matopsia with extinguished photopic ERG and ﬂ icker Albinism Normal O O 1 O 2 ERG. (d) Visual evoked z potential of a patient with O - F albinism showing asymme- O1 - Fz 1 z try between recording from Oz - Fz the right and left hemisphere Oz - Fz (see placements of electrodes O2 - Fz on scalp in upper right O2 - Fz corner) when the right and

left eye are individually Right Eye Open O - O stimulated. Owing to O1 - O2 1 2 increased crossing of optic nerve ﬁ bers in the chiasm, the evoked potentials are O1 - Fz more pronounced in the O1 - Fz contralateral hemisphere Oz - Fz (O1, O2, and O3 are Oz - Fz electrodes placed over the back of the head (near the O2 - Fz O2 - Fz occipital pole of the cortex)

in left , central, and right Left Eye Open positions, respectively; F is Z O - O the reference electrode) 1 2 O1 - O2 12.1 Overview 157

■ Possible presence of AHP, strabismus, and refractive eye is occluded. Th e AHP changes to the other side in an errors alternating monocular occlusion, which helps in the diag- ■ Decreased amplitude of nystagmus in null point nosis of MLN. If patients with MLN have alternating fi xa- ■ Dampening of nystagmus on convergence tion the head turn can change spontaneously, depending ■ Th e intensity of nystagmus increases with fi xation, on which eye is fi xing. Figure 12.5e, f shows an alternating decreases with sleep or inattention AHP to the right and left in one of our patients who had fusional maldevelopment syndrome with latent nystagmus confi rmed on eye movement recordings. Th e patient has exotropia and is freely alternating. He is always keep- 12.1.2 Manifest Latent Nystagmus (MLN) ing the fi xing eye in adduction and therefore his head pos- MLN is most commonly associated with infantile or ture is alternating with a turn to the right with the right childhood onset esotropia as well as ambylopia. MLN is eye fi xing and left with the left eye fi xing. When one eye defi ned as jerk nystagmus that develops at an early age was patched his head turn was unidirectional in the direc- and increases with monocular viewing, triggered by tion of the open eye. Th e cause of MLN appears to be due occlusion of one eye. Previously latent nystagmus was to disruption of binocular vision during visual develop- distinguished from MLN where no nystagmus was ment, especially when the motion sensitive areas of the detected when both eyes were open. However, it has been middle temporal and medial superior temporal cortex do shown that in cases clinically diagnosed as “latent nystag- not develop binocular function. mus,” nystagmus is seen on eye movement recordings Patients can have a combination of congenital and even when both eyes are open. Hence MLN/latent nys- latent nystagmus. According to Dell’Osso [2], 80% of nystagmus is considered as a single entity (MLN). tagmus is congenital nystagmus, 15% is MLN, and 5% is Characteristically, the amplitude of MLN decreases in a combination of both forms. adduction and increases in abduction, with the fast phase of the nystagmus beating toward the side of the fi xating 12.1.2.1 Clinical Characteristics of Manifest eye or open eye. MLN has a distinctive slow phase with an Latent Nystagmus (MLN) exponentially decreasing or linear velocity in all positions of gaze as shown in Fig. 12.4. As nystagmus decreases in ■ Onset in infancy adduction in patients with MLN, they frequently develop ■ Nystagmus is horizontal and conjugate an AHP toward the side of the fi xating eye when the fellow ■ Associated with strabismus and amblyopia

LEFT EYE RIGHT EYE LEFT EYE BOTH EYES COVERED COVERED COVERED UNCOVERED Right Eye

RIGHT BEATING

Left Eye 10º 0.5 sec L

Fig. 12.4 Original horizontal eye movement recordings of both eyes of a patient with manifest latent nystagmus (MLN) and exotropia during an alternating cover test. Eye movements to the right are represented by an upward deﬂ ection, and eye movements to the left by a downward deﬂ ection. Th e fast phase is always beating toward the open eye (to the right with the left eye covered and to the left with the right eye covered). When both eyes are open the direction of the fast phase is toward the dominant left eye. Th e velocity of the slow phase is decelerating or linear. Arrows indicate blinks 158 12 Management of Congenital Nystagmus with and without Strabismus

Anomalous Head Posture in Idiopathic Correction of Anomalous Head Posture in Infantile Nystagmus Idiopathic Infantile Nystagmus with Anderson- Kesternbaum Surgery

Child Horizontal head turn 12 Without visual effort With visual effort Before surgery After surgery a bgh

Adult Vertical and horizontal head turn with esotropia Without visual effort With visual effort Before surgery After surgery cdij

Bi-directional Alternating Head Turn in MLN Measurement of head turn using Harms wall Right head turn Left head turn ef k

Fig. 12.5 Abnormal head posture (AHP) of a child with idiopathic congenital nystagmus (a) without visual eff ort and (b) with increased head turn while pointing at pictures on the Lang stereo test. Panel (c) shows a patient with idiopathic congenital nystagmus without head posture when there is no visual eff ort and (d) a prominent abnormal head posture when reading at distance. Spontaneous alternating head turn to the right (e) and left (f) in a patient with MLN. Panel (g) shows a patient with idiopathic congenital nystagmus with approximately 45° head turn to the left before surgery and with straight head position (h) aft er Anderson– Kestenbaum procedure. A patient with oculo-cutaneous albinism and chin depression, face turn to the right and left esotropia before surgery (i) and aft er surgery (j). An accurate method of measuring AHP is achieved by using the Harms Wall (k) where the degree of head turn is measured by the amount of displacement of the cross observed on the tangent screen. Th e cross is projected from a light source fi xed on the head

■ Eye movement recordings have a characteristic 12.1.3 Congenital Periodic Alternating slow phase with exponentially decreasing or linear Nystagmus (PAN) velocity ■ Amplitude of nystagmus decreases in adduction and Congenital PAN is classifi ed as a variant of congenital increases in abduction, with the fast phase of nystag- nystagmus according to the CEMAS classifi cation. mus toward the side of fi xating eye Congenital PAN is discussed as a separate entity because 12.1 Overview 159 it has specifi c implications for management which are PAN. Absence of alternating AHP in congenital PAN diff erent from other forms of nystagmus. is possibly due to the asymmetry of the PAN cycle, Th e frequency of congenital PAN is variably reported nystagmus beating longer in one direction than the in the literature. Gradstein et al. [3] in a retrospective other, and also the unequal intensities of nystagmus in analysis of approximately 200 congenital nystagmus the two phases. patients with and without sensory defi cits found 18 patients (9%) with a diagnosis of PAN. Five of these 18 patients had albinism. AHP was seen in 16 of the 18 12.1.3.1 Clinical characteristics of congenital patients. Shallo-Hoff man et al. [4] in a prospective study periodic alternating nystagmus involving 18 patients with congenital nystagmus without ■ Onset in infancy. sensory defi cits found that seven patients (39%) had PAN. ■ Nystagmus horizontal and conjugate. Abadi and Pascal [5] found 12 patients with PAN in 32 ■ Eye movement recording shows a characteristic active patients with oculocutaneous albinism (37.5%). Th ese 12 phase with right/left beating nystagmus followed by a patients did not exhibit AHP nor had dampening of nys- quite transition phase and then an active left /right tagmus on convergence (Fig. 12.6). beating nystagmus. Congenital PAN is most oft en missed or misdiagnosed ■ Th e AHP is usually bidirectional. if not properly investigated. Th e main reasons for diffi culties in recognizing PAN are: Summary for the Clinician ■ Long cycle duration: Th e cycle duration of the congenital PAN is variable lasting mostly between 2 and 7 min. ■ Familiarity with the clinical characteristics of Th us, ocular motility examination (clinical or with eye congenital nystagmus, MLN, and congenital movement recordings) must extend over a prolonged PAN will minimize the chances of misdiagnos- time period. ing these conditions and plan proper manage- ■ Th e absence of alternating head turn: Classically, a clin- ment of these conditions. ical sign assisting in the diagnosis of congenital PAN is ■ Electrodiagnostics: both ERG and VEP should the alternating or bidirectional head turn. Gradstein be done in all patients with congenital nystagmus et al. [3], on the contrary, have reported that the major- to fi nd a cause for the congenital nystagmus. ity of patients with congenital PAN used a predomi- ■ Eye movement recording aids in diff erentiating nant head posture rather than an alternating head congenital nystagmus from MLN and congenital posture. Abadi and Pascal [5] also reported the absence PAN. of AHP in all the 12 patients diagnosed with congenital Right Eye R

LEFT BEATING 5º 3 sec RIGHT BEATING L Left Eye

Fig. 12.6 Original eye movement recordings of a patient with idiopathic congenital periodic alternating nystagmus (PAN) of the right and left eye showing left beating nystagmus, a quiet phase and right beating nystagmus. Eye movements to the right are represented by an upward deﬂ ection, and eye movements to the left by a downward deﬂ ection. Arrows indicate blinks 160 12 Management of Congenital Nystagmus with and without Strabismus

NBS occurs with the waveform characteristics of increas- 12.2 Compensatory Mechanisms ing velocity slow phase and variable angle esotropia Several compensatory mechanisms exist in congenital (Fig. 12.7a–d). MLN is also frequently associated with nystagmus which tend to decrease the nystagmus and infantile esotropia. Most cases diagnosed as nystagmus 12 thus improve the visual acuity. Th ese compensatory blockage syndrome in the past probably corresponded mechanisms are achieved with superimposed vergence to infantile esotropia associated with MLN. and version movements. Diff erent compensatory mechanisms may coexist in Summary for the Clinician the same patient with congenital nystagmus. Th ese mech- ■ anisms need to be analyzed carefully both to plan the Compensatory mechanisms are seen in patients treatment and also to make prognostic predictions. with congenital nystagmus to increase visual acuity by decreasing the intensity of nystagmus. ■ Compensatory mechanism can be achieved by convergence or version movements in case of 12.2.1 Dampening by Versions eccentric null region. Compensatory mecha- Version eye movements are used in some patients as a nisms by versions lead to AHP. compensatory mechanism to reduce congenital nystag- ■ Several compensatory mechanisms usually exist mus. Sustained contractions of yoke muscles help main- in the same patient. tain the eyes in a peripheral lateral, vertical, or oblique gaze, depending on the position of the null region, leading to dampening of nystagmus. Th ese versions are 12.2.3 Anomalous Head Posture (AHP) oft en accompanied, and consequently identifi ed, by an AHP. An eccentric horizontal null zone leads to hori- AHP in children could be due to abnormalities of the zontal head turn and an eccentric vertical null zone oculomotor system, neck muscles, or the central nervous leads to chin elevation or depression. For example, in a system. Th e ocular causes of AHP include strabismus, patient who has null position in the laevoversion, the nystagmus, refractive errors, and ptosis. Although clini- compensatory head position is face turn to right, for cal diff erentiation of these disorders is accurately accom- null zone in elevation the compensatory mechanism is plished aft er thorough history and ocular examination, chin down position. Compensatory cycloversion leads the exact mechanism of AHP is oft en diffi cult to deter- to head tilt. A right head tilt corresponds to blocking mine in patients with combination of strabismus and nys- incyclotorsion of the right eye and excyclotorsion of the tagmus. It is important to delineate the cause of AHP and left eye. the amount of AHP before considering treatment in patients with congenital nystagmus.

12.2.2 Dampening by Vergence 12.2.3.4 Measurement of AHP Th ere are two distinct clinical conditions which use An AHP typically becomes progressively larger with dampening by convergence as a compensatory mecha- increased visual eff ort. Hence, quantifi cation of the sur- nism to reduce the amplitude and frequency of nystag- gery must be based on an appropriate eff ort of fi xation, mus. Th ese are MLN and nystagmus blockade syndrome usually achieved by testing visual acuity at distance and (NBS). near. Figures 12.5a, b show a child with no AHP when no Adelstein and Cüppers [6] coined the term “nystag- visual eff ort is needed. However, when he identifi es a ste- mus blockage syndrome” as having the following clinical reoptic stimulus on the Lang test at near he is using a features: head turn to the right. Similarly, Figs. 12.5c, d show a patient with no head turn without visual eff ort. However, ■ Esotropia with sudden onset in early infancy, oft en he uses a very large chin elevation and head turn to the preceded by nystagmus right when he is asked to read small letters at distance. ■ Pseudoparalysis of both abducens nerves AHP can be measured objectively, while reading ■ Th e appearance of manifest nystagmus as the fi xating small optotypes at distance and near, using calipers or eye moves from adduction toward abduction the Harms wall (Fig. 12.5k). For diff erential diagnosis, it ■ Increase in the angle of the convergent squint when a is important to record visual acuity with both eyes open base-out prism is put in front of the fi xating eye as well as with each eye occluded. It is also useful 12.2 Compensatory Mechanisms 161

Fig. 12.7 A patient with During nystagmus Blocking with convergence nystagmus blockage a b syndrome (a) with straight eyes, (b) when dampening nystagmus with right esotropia, (c) wearing Fresnel prisms for surgical evaluation, which showed dampening of nystagmus and (d) aft er bimedial medial rectus recessions. Original eye movement recordings show periodic convergence to dampen the nystagmus before surgery and quieter eye With prisms After surgery movements aft er surgery (e) cd

e Eye movement recordings

BEFORE SURGERY AFTER SURGERY nystagmus blockage

Right Eye R

10º 2 sec L Left Eye

clinically to look at the eff ects of straightening the head ■ No AHP: Th is could indicate that either the patient is on nystagmus. using vergence as a compensatory mechanism, that the null region is in the primary position, or that no 12.2.3.5 Eff ect of Monocular and Binocular compensatory mechanism is being used by the Visual Acuity Testing on AHP patient ■ A horizontal AHP consisting of a face turn to the right or left Testing Visual Acuity with Both ■ Eyes Open A vertical AHP consisting of a chin elevation or depression AHP should be fi rst assessed testing visual acuity with ■ A bidirectional or alternating AHP both eyes open to determine the existence and the type of ■ A head tilt to the right or left AHP naturally adopted by the patient. Th e patient could ■ A combination of AHP in diff erent planes have one of the following: 162 12 Management of Congenital Nystagmus with and without Strabismus

Testing Visual Acuity with Either 12.3 Treatment Eye Covered Testing AHP under monocular conditions using occlusion Various modes of treatment are available for patients with helps to diff erentiate between congenital nystagmus and congenital nystagmus. However, it is necessary to decide 12 MLN, since in congenital nystagmus the AHP is usually con- the best method to treat these patients in the light of cordant (i.e., usually does not change position when cover- understanding the type of congenital nystagmus and the ing one eye), whereas in MLN nystagmus the AHP is compensatory mechanism being used. Sometimes a com- discordant. Th is is because in MLN the intensity of the nys- bination of treatment options might be needed to achieve tagmus tends to be least in adduction. Consequently, in a better outcome. MLN the head turn and the nystagmus direction reverse Th e main aim of treatment of congenital nystagmus is: when fi xation shift s from one eye to the other (Fig. 12.5e, f). 1. To improve visual acuity 2. To diminish the amplitude and frequency of nysta- 12.2.3.6 Testing AHP at Near gmus Since convergence has an eff ect on nystagmus, AHP 3. To shift the null position to primary position with the should also be tested when measuring visual acuity or aim of correcting an AHP reading at near (e.g. at 33 cm). All the observations noted 4. To correct the strabismus if present regarding the position of AHP and the nystagmus intensity for distance should also be evaluated for near vision. Th e main categories of treatment of nystagmus are optical, medical, and surgical although other forms of treatment have been attempted such as acupuncture, 12.2.3.7 The Eff ect of Straightening the biofeedback, and use of botulinum toxin-A. Head in Patients with AHP On straightening the head, if the nystagmus increases, then the cause of the AHP is almost certainly due to the 12.3.1 Optical Treatment nystagmus. If there is no change in the nystagmus, the AHP is either due to other ocular causes, a structural Th e incidence of signifi cant refractive errors in patients anomaly of the head or neck, CNS anomalies, or because with congenital nystagmus has been estimated to be as of strabismus. Since strabismus in presence of nystagmus high as 85% [7]. Th e importance of correcting refractive can be responsible for AHP thorough examination for errors besides improving visual acuity is to prevent amby- comitant or incomitant squint is important in all patients lopia and to treat the associated strabismus, commonly with nystagmus. If the strabismus increases with head seen in patients with congenital nystagmus. Optical treat- straightening, it indicates that an incomitant deviation is ment can involve spectacles, contact lenses (CL), or low responsible for the AHP. However, if the strabismus visual aids. improves with straightening of the head, the AHP is more likely associated with the nystagmus or some other cause. 12.3.1.1 Refractive Correction Summary for the Clinician A full cycloplegic refraction should be performed in chil- ■ It is important to delineate the cause of AHP dren. A simple correction of refraction is the easiest way and the amount of AHP before considering of improving the visual acuity in congenital nystagmus. treatment in patients with congenital nysta- Hence, all patients with congenital nystagmus should gmus. have precise refraction with appropriate correction before ■ AHP typically becomes progressively larger with attempting other modalities of treatment. increased visual eff ort. Hence quantifi cation of the head turn for surgical assessment must be based on measurement during maximal visual 12.3.1.2 Spectacles and Contact Lenses (CL) eff ort. ■ In patients with combination of strabismus and Several studies have suggested that CL improve visual nystagmus, the cause of AHP needs to be carefully function better than spectacles in patients with congeni- analyzed. tal nystagmus [8, 9]. Th e possible mechanisms underlying this are that CL reduces the chromatic and spherical 12.3 Treatment 163 aberration, together with the prismatic eff ect, compared 12.3.1.4 Low Visual Aids to spectacles [8–10]. Since CL move with the eyes, the patient permanently looks along the visual axis of the Th e use of telescope, magnifi cation glasses, large print correcting lens unlike with spectacles. CL also have the books, computer with large fonts, and other low vision aids additional advantage of inducing convergence and are valuable refractive adjuncts that can be used in patients accommodative eff ort, which both decrease congenital with low vision associated with congenital nystagmus. nystagmus in some patients [8, 11]. It has been suggested that CL reduce the intensity of the nystagmus by providing sensory feedback through the eye lid [8, 11]. Tinted Summary for the Clinician CL have also been used to reduce photophobia in patients ■ All children with congenital nystagmus must with achromatopsia [12]. have cycloplegic refraction and appropriate full refractive correction. ■ Th e importance of correcting refractive errors 12.3.1.3 Prisms besides improving visual acuity is to prevent ambylopia and to treat the associated strabismus In 1950, Metzger [13] was the fi rst to describe the treat- commonly seen in patients with congenital ment of congenital nystagmus by using prisms in specta- nystagmus. cles in four patients with nystagmus. Prisms are used to ■ improve visual acuity by reducing the intensity of nystag- Refractive correction could be achieved by mus and also to correct the AHP. glasses, CL, or low visual aids. ■ Base-out prisms are prescribed to induce fusional A trial of CL should be off ered to suitable patients convergence, which may be eff ective in decreasing the as they have shown to improve visual acuity better amplitude of nystagmus, thus improving visual acuity than spectacles. [13]. Presence of binocular vision is a prerequisite for the use of base-out prisms since fusional convergence in response to prism-induced retinal disparity cannot be 12.3.2 Medication expected in patients without fusion. Prism adaptation for both distant and near vision helps to determine the larg- Medications such as baclofen, cannabis, gabapentin, or est amount of prism-induced convergence that dampens memantine were fi rst trialed in acquired nystagmus. nystagmus without creating diplopia. Th ese studies led to the use of several of these drugs for Prisms can also be used in preoperative evaluation or congenital nystagmus as well. However, most of the as a non-surgical treatment to correct AHP in patients reports in the literature consist of single cases or small with congenital nystagmus and eccentric null points. Th e case series. Because of the prolonged treatment required base of the prism is inserted opposite to the preferred and the side eff ects of medications, one needs to weigh direction of gaze. For instance, in patients with head turn the benefi ts of pharmacological treatment in comparison to right, the null zone is in laevoversion, and prisms base- with the other treatment modalities. out in front of right eye and base-in in front of left eye will Hertle et al. [15] reported a case study of a patient with correct the head turn. Likewise, chin elevation or depres- congenital nystagmus, who showed improvement in fove- sion can be corrected by prism base-up or prism base- ation time with broadening of null zone and increased down, respectively, in front of both eyes. Godde-Jolly and visual acuity aft er the use of an anti-anorexic drug (diethyl Larmande [14] advocate the use of a combination of hori- proprionate). Pradeep et al. [16] reported reduction in zontal and vertical prisms when the null zone is in an nystagmus intensity and improvement in visual acuity in oblique position of gaze. a patient with congenital nystagmus aft er smoking can- Since the visual acuity is oft en decreased with the nabis. Th ere are a number of other reports suggesting the use of Fresnel prisms and prisms incorporated in use of tranquilizers and the anti-epileptic phenobarbital glasses, this method is not eff ective to treat larger com- in the treatment of congenital nystagmus with reported pensatory head posture in patients with congenital nys- improvement in the visual acuity. Sarvananthan et al. [17] tagmus. Nonetheless, it can be useful for preoperative reported a case study of a patient, with congenital nystag- assessment of the amount of AHP in terms of prism mus and corneal dystrophy being treated with gabapentin, diopters, and also the response of the patients to prisms, which showed decrease in nystagmus and improvement which form a guide for planning the surgical treatment in visual acuity. Shery et al. [18] showed a reduction in of nystagmus. nystagmus amplitude and increase in visual acuity in 164 12 Management of Congenital Nystagmus with and without Strabismus

seven patients (three with congenital idiopathic nystag- nystagmus who underwent acupuncture, Blekher et al. mus and fi ve with associated ocular defects) treated with [23] showed an increased foveation time in four patients. gabapentin. McLean et al. [19] conducted the fi rst randomized, 12 controlled, double-masked trial of memantine and gaba- 12.3.4 Biofeedback pentin in the treatment of congenital nystagmus. A total of 48 patients with congenital nystagmus with and with- Auditory feedback is a method that was fi rst introduced out sensory defi cits were included in the study. Sixteen to treat patients with congenital nystagmus in 1980 in patients in each group received memantine, gabapentin, which the patient hears a sound cue representing the or placebo treatment. Th e maximum dose of memantine intensity of the nystagmus [24]. Auditory feedback has was up to 40 mg/day and gabapentin up to 2,400 mg/day. been shown to be eff ective in decreasing the amplitude of Results showed reduction in nystagmus using eye move- nystagmus in patients with congenital nystagmus; how- ment recordings (see Fig. 12.4) and increase in visual ever, Sharma et al. [25] have shown that the action is not acuity in both treatment groups with memantine and sustained being present only during the duration of the gabapentin showing a signifi cant improvement compared biofeedback therapy. with the placebo-controlled group. Th ere are several case reports of patients with congenital PAN being treated with baclofen with some suc- 12.3.5 Botulinum Toxin-A (Botox) cess [4, 20]. In 2002, Solomon et al. [21] reported a reduction in nystagmus with improved reading ability in Carruthers et al. [26] studied four patients with congeni- a single case of congenital PAN treated with baclofen. tal nystagmus treated by botox injected into multiple Comer et al. [22] did a retrospective review of eight horizontal rectus muscles. Th ree of the four patients were patients diagnosed with congenital PAN and treated with reported to have achieved a signifi cant improvement in baclofen. AHP improved in four of the eight patients the visual acuity. However, the botox injection needs to treated with four patients improving in Snellen visual be repeated every 3–4 months. acuity by one line. Th e dose of baclofen was initially Oleszczynska-Prost et al. [27] in a case series of 32 started at 15 mg/day with a weekly increase in the dose to patients with congenital nystagmus treated with botox up to 120 mg/day. showed an improvement in visual acuity in all the patients. Th e amplitude of nystagmus decreased by 29–50%. Th e Summary for the Clinician head turn was corrected in few patients. Th e common complications of repeated botox injection are ptosis, ret- ■ Recently, medical treatment has been used for robulbar hemorrhage, and spread of the toxin to other congenital nystagmus. horizontal or vertical muscles resulting in palsies of these ■ In an RCT [19] of medical treatment of congeni- muscles. tal nystagmus, both memantine and gabapentin showed reduction in nystagmus and improvement in visual acuity. 12.3.6 Surgical Treatment of Congenital ■ Th e dosage of memantine used to treat congeni- Nystagmus tal nystagmus was up to 40 mg/day, and that of Gabapentin 2,400 mg/day. Th e surgical principles for correction of the AHP and ■ Th e decision to treat patients medically should dampening of nystagmus uses the basic strabismus pro- be individualized given the long-term treatment, cedure involving either the recession, resection proce- the benefi ts, and side eff ects of medications. dures, or both. Th e aim is to move the eyes conjugately in the opposite direction to the gaze angle of the null region, or to artifi cially create an exotropia in patients with good binocular fusion in the presence of conver- 12.3.3 Acupuncture gence null. Newer surgical procedures such as tenotomy Acupuncture of the sternoclenoidmastoid muscle of the of extraocular muscles have now been developed based neck has been shown to reduce the frequency of nystag- on the benefi cial secondary eff ects noted in patients who mus and improve the visual acuity by increasing the were earlier treated with the strabismus procedure length of foveations, although the exact mechanism is not (Anderson−Kestenbaum procedure) to dampen the known. In a case series of six patients with congenital congenital nystagmus. 12.3 Treatment 165

Th e importance of diagnosing congenital PAN and example, 40% augmentation of the Parks procedure cor- MLN preoperatively is crucial as the surgical manage- responds to 7, 8.4, 9.8, and 11.2 mm. Nelson et al. [32] ment diff ers from congenital nystagmus in these cases. In found that a more sustained correction of the AHP in addition to the nystagmus, a detailed examination evalu- congenital nystagmus was obtained by an augmented ating the presence or absence of strabismus is also impor- modifi ed Kestenbaum procedure. Th ey suggested 40% tant. Th e common strabismus forms seen in association augmentation of modifi ed Kestenbaum procedure for with nystagmus are esotropia, exotropia, dissociated ver- patients with 30° of head turn, and 60% augmentation for tical deviation, and dissociated horizontal deviation. A patients with 45° of head turn. Taylor recommended that proper surgical plan should be made to either correct this recession of 8–9 mm of the lateral rectus muscle and strabismus along with the nystagmus as a single proce- 6 mm recession of the medial rectus muscle be performed dure or in two stages. Th e patient should, however, be in conjunction with 6 mm resections of the respective informed that a second procedure might be necessary in antagonists [33]. case of residual strabismus or AHP, which needs to be De Decker [34] advocated the modifi cation of addressed. Anderson procedure to correct the AHP. In this procedure, only the yoke muscles are recessed, to as much as 10–12 mm, rather than 4–5 mm as suggested by Anderson. Since the recession of medial rectus is more eff ective than 12.3.6.1 Management of Horizontal AHP recession of lateral rectus, the medial rectus is recessed A face turn to right or left is the most common compen- 2 mm less than the lateral rectus muscle. For example, in satory posture encountered in patients with nystagmus patients with a face turn to right, the right medial rectus with an eccentric null position. Various surgical proce- is recessed 10 mm, and the left lateral rectus is recessed dures are used to correct this AHP and shift the null zone 12 mm. As only the two yoke muscles are operated on, it into primary position. spares the other two horizontal muscles, which could be Anderson, Goto, and Kestenbaum in 1950s indepen- available if further surgery is required. dently reported the surgical procedures for the correction Flynn and Dell’Osso [35] confi rmed the initial fi nd- of AHP in patients with congenital nystagmus [20, 28, 29]. ings described by Kestenbaum of an increase in the visual Anderson postulated that the muscles acting during the acuity aft er the Kestenbaum-type procedure. Th ey also slow phase of the nystagmus were overacting. He conse- demonstrated that the Anderson-Kestenbaum procedure quently treated the nystagmus using a recession or weak- does not alter the binocular function in those patients ening procedure of the two yoke muscles involved. Goto, with intact binocular function before surgery. on the contrary, believed that there was underaction of It is very diffi cult to advocate a rigid dosage scheme the muscles acting during the fast phase of the nystag- for all patients. Each surgeon adopts his own nomogram mus, and advocated strengthening or resection of these to correct the amount of AHP. two muscles. Kestenbaum advocated a combined resec- With very large head turns of 40–45°, in our experi- tion and recession procedure on all the four horizontal ence, very large amounts of surgery is needed. Restriction rectus muscles. He recessed or resected the two horizon- of eye movements is oft en a necessary consequence of tal muscles of each eye. He also suggested performing the large Kestenbaum procedures but is necessary to reduce same quantity of surgery for both weakening and large AHPs. strengthening procedures (5 mm). Parks [30] made mod- In Fig. 12.5g, h an example of a child who underwent ifi cations in the Kestenbaum technique and proposed horizontal Anderson−Kestenbaum procedure is shown. that, to obtain symmetrical horizontal ductions of the She was fi rst examined at 1 year of age because of nystag- two eyes, surgery should be a 5 mm recession of medial mus since birth. A diagnosis of congenital idiopathic nys- rectus and a 8 mm resection of the lateral rectus for the tagmus (CIN) was made aft er detailed clinical examination eyes in adduction, and 6 mm resection of medial rectus and electrodiagnostic tests. At 2 years of age, she started and a 7 mm recession of the lateral rectus of the fellow to develop an AHP. A refractive error of −4D cyl. in the eye. Th is became the classical “5, 6, 7, 8” measurements right eye and −2D cyl. in the left eye was detected, but she for the Kestenbaum procedure modifi ed by Parks. was unable to wear glasses owing to the large AHP. Th e Because of the high rates of recurrence and undercor- child was reassessed at the age of 3 years. Her visual acu- rection following the modifi ed Kestenbaum procedure, ity was 6/24 with both eyes open. She had an AHP of Calhoun and Harley [31] recommended augmentation of about 45° (Fig. 12.5g). No squint was detected. We per- the original Parks modifi cation of Kestenbaum procedure formed an augmented Anderson−Kestenbaum procedure by 40–60% depending on the amount of head turn. For to correct the AHP (recession of right lateral rectus and 166 12 Management of Congenital Nystagmus with and without Strabismus

left medial rectus and resection of right medial rectus and 12.3.6.2 Management of Vertical AHP left lateral rectus by 12 mm each). Postoperatively, the AHP was corrected without residual AHP. Th e child was Chin elevation or chin depression are compensatory able to wear glasses, which improved the visual acuity to mechanisms for a null position with eyes in down or 12 6/9 with both eyes open (Fig. 12.5h). Postoperatively, the upgaze, respectively. Vertical or torsional AHP to dampen child had limitation on right gaze, which is necessary to the nystagmus is seen less frequently than horizontal AHP. avoid recurrence of head turn. Pierse [36] in 1959 was the fi rst to attempt to correct verti- In the presence of strabismus, the amount of the sur- cal AHP. He reported two cases with chin-up position for gery performed on each muscle is modifi ed to correct the which he did bilateral inferior rectus recession and supe- strabismus in addition to the head turn. In patients with rior oblique tenectomies, with marked improvement of strabismus or amblyopia, the surgery for AHP must be vision in primary position and improvement in the AHP. planned on the fi xing eye or non-amblyopic eye. If neces- Schlossman [37] reported a patient with chin-down pos- sary, eso- or exotropia can be corrected by performing ture for which he resected the inferior rectus and recessed diff erent amounts of recess–resect procedures on the the inferior oblique. Parks [30] suggested operation on all non-fi xing eye simultaneously. For example, in a patient four vertical rectus muscles for chin elevation or depres- with a head turn to the right and left esotropia, surgery sion greater than 25°. He recommended 4 mm resection for the AHP needs to be performed on the right eye and recession for these patients. For patients with chin (medial rectus recess and lateral rectus resect). Th is will elevation or depression less than 25°, only 4 mm recession reduce the esotropia. Depending on the amount cor- of the appropriate vertical muscle without resection was rected for the AHP, the amount of surgery on the left eye recommended. Taylor and Jesse [38] recommended supe- needs to be reduced (i.e., smaller than the amount cor- rior rectus recession and inferior oblique myectomy for rected for AHP on the right eye) to correct the squint. If chin-down posture, inferior rectus recession and superior the esotropia and the head turn are approximately of oblique tenotomy for chin-up position. equal size, it is suffi cient to correct the head position on In 1990, Sigal et al. [39] conducted a poll of AAPOS the fi xating eye. If a patient has a right AHP with left members to fi nd the methods used to correct vertical AHP. exotropia, the amount of the left squint surgery needs to Two surgical procedures were used by most of the respon- be increased (i.e., larger than the surgery for AHP on the dents to correct vertical AHP. While 44% of the respon- right eye). dents preferred recession surgery alone, 55% preferred Surgical decision for the child shown in Fig. 12.5e, f both recession and resection procedure on all four vertical was a challenge. Since he adducted each eye to dampen rectus muscles. Recession only consisted of bilateral aver- his latent nystagmus, bimedial rectus recession would age vertical muscle recession of 4.8 mm for 10°, 5.9 mm for have been the ideal surgery. However, he also had a large 20°, and 7.3 mm for 30° AHP. Average amount of surgery exotropia, which would have increased with bimedial for both recession and resection of bilateral vertical rectus recessions. We performed, therefore, large bilateral medial muscle were 4.5 mm recession and 4.3 mm resection for rectus recessions (12 mm) and even larger bilateral rectus 10° AHP, 5.3 mm recession and resection for 20° AHP, recessions (16 mm). Postoperatively, the head turn was 7.7 mm recession and 6.4 mm resection for 30° AHP. improved signifi cantly and he remained with moderate Robert and colleagues [40] described a series of seven exotropia. Alternatively, one could have performed a patients with vertical AHP, three of whom underwent com- Faden procedure on both medial recti combined with lat- bined bilateral inferior rectus recession and bilateral supe- eral recti recessions. rior rectus resection for chin-up AHP. Four patients NBS can also be treated surgically. Figure 12.7 shows an underwent superior rectus recession and inferior oblique example of a patient with congenital nystagmus and NBS anteriorization for chin-down AHPs. Based on their results, before surgery. Th e patient complained of one eye moving they recommended a minimum combined bilateral 8 mm inward intermittently. To dampen his nystagmus, he devel- recession and 8 mm resection of the vertical rectus muscles, oped large intermittent right esotropia (Fig. 12.7a, b). Eye should be performed for chin-up AHP greater than 30°. movement recordings (Fig. 12.7e) show large convergent Yang et al. [41] conducted a retrospective review of 20 movements in the right eye which dampened the nystag- patients who underwent surgery for vertical AHP. Th ey mus. With a trial of Fresnel prisms (20 base out on each found that recession alone caused either no change side), the eyes remained esotropic and the nystagmus or worsening of the vertical AHP, while the recession- dampened. Aft er bimedial rectus recession, he developed resection procedure of all four vertical rectus muscle pro- a small constant esotropia (Fig. 12.7d). Th e nystagmus duced excellent results in correcting the vertical AHP. was signifi cantly reduced (Fig. 12.7e). Th ey recommended 12 mm of combined recession and 12.3 Treatment 167 resection for each pair of vertical rectus muscles for 10–15° vertical transposition of the horizontal rectus muscles to AHP, 16 mm for 20–25°, and 20 mm for more than 30° correct the head tilt. For example, transposing the medial AHP. For example, for 10° chin-down posture, 6 mm rectus downward and the lateral rectus upward causes resection of inferior rectus and 6 mm recession of superior excycloduction in the right eye. rectus should be performed of both eyes. Von Noorden et al. [45] proposed the horizontal trans- In Fig. 12.5i, j an example of a patient who underwent position of the vertical rectus muscles to correct the head simultaneous Anderson procedure for vertical and hori- tilt. For example, to achieve excyclotorsion of the right zontal AHP and correction of squint is shown. Th is eye and incyclotorsion of the left eye in case of right head patient was diagnosed as having oculocutaneous albinism tilt, the right superior rectus muscle is transposed nasally, with nystagmus. She had a visual acuity of 6/36 with both and the right inferior muscle inferiorly, and in the left eye, eyes open. She had a chin-down position of approxi- the superior rectus muscle is transposed temporally, and mately 20° and face turn to right of approximately 20°, the left inferior muscle nasally. Th is surgery has been more at near than at distance (Fig. 12.5i shows head posi- found to be eff ective when operated on both eyes, in tion at distance). She had left esotropia of 35 prism patients with no fi xation preferences or with binocularity diopters. She underwent Anderson procedure (bilateral and also on the fi xating eye alone in monocular fi xation. superior rectus recession of 12 mm) and correction of Spielmann [46] recommended slanting the insertions squint on the dominant right eye to correct simultane- of all four rectus muscles. For example, excycloduction of ously the horizontal AHP and the squint (right eye medial the right eye can be achieved by recessing the temporal rectus recession of 9 mm). Postoperatively, her AHP and part of the superior rectus, inferior part of the lateral, squint were well corrected (Fig. 12.5j). nasal part of the inferior and superior part of the medial Operating on the oblique muscles to correct the vertical rectus muscle insertions. Sigal et al. [39] found fi ve diff er- AHP harbors a potential complication of iatrogenic ent surgical procedures used by AAPOS members to treat cyclotropia in patients with binocularity. As the vertical torsional AHP: muscles also contribute to the torsional status of the eye, one could expect torsional problems with large amounts of sur- 1. Bilateral vertical rectus muscle recession gery on the vertical muscles as well. Th is can be counter- 2. Bilateral vertical rectus muscle recess−resect acted by shift ing the insertion of the vertical rectus muscles 3. Bilateral oblique muscle weakening laterally. For example, a large recession of the superior rec- 4. Bilateral oblique muscle recess−resect tus causes excylcotropia. Moving the insertion of the supe- 5. Bilateral oblique muscle weakening and vertical rectus rior rectus temporally reduces the induced excylcotropia. muscle recession

When dealing with moderate to severe AHP, 88% of surgeons preferred operating on at least one oblique muscle. 12.3.6.3 Management of Head Tilt Head tilt is due to compensatory cycloversion. A right head tilt corresponds to blocking incyclotorsion in the right eye 12.3.6.4 Artifi cial Divergence Surgery and of excyclotorsion in the left eye. Based on the Kestenbaum principle to shift the muscle in the direction of Patients suitable for artifi cial divergence surgery should be the AHP, Conrad and de Decker [42, 43] in a review of 66 orthotropic with convergence as the compensatory mech- cases with head tilt suggested rotating both eyes around the anism used to dampen the nystagmus. Binocular fusion is sagittal axis toward the shoulder to which the head is tilted. necessary to achieve this eff ect. Th e vergence dampens the Th ey combined a recession−resection procedure at the nystagmus regardless of the stimulus inducing the conver- anterior portions of the oblique muscles with transposi- gence. Th is principle has been used optically (base-out tions of their insertion toward the posterior−anterior pole. prisms) and surgically (artifi cial divergence) to dampen Th ey had a success rate of 54%; while some improvement the nystagmus. was seen in 25% of cases, 21% of cases showed no improve- Cüppers [47] proposed the concept of artifi cial diver- ment. Although surgery of oblique muscles is technically gence in patients with convergence dampening of nys- more complex than surgery of horizontal muscles, De tagmus. In this procedure, an exodeviation is induced, Decker advocated this surgery because it avoids disturbing which can be compensated by fusional convergence. Th is the vascular supply through horizontal muscles. causes the patients to have convergence innervations In cases where horizontal surgery is also necessary for even at distance. Th e acceptability and eff ectiveness strabismus or horizontal AHP, De Decker [44] suggested of artifi cial divergence surgery should be evaluated 168 12 Management of Congenital Nystagmus with and without Strabismus

preoperatively by using the prism adaptation test. A 12.3.6.5 Surgery to Decrease base-out prism is prescribed to induce artifi cial diver- the Intensity of Nystagmus gence. Inducing divergence with a base-out prism causes the patient to converge, and therefore decreases the nys- In patients who do not exhibit any compensatory mecha- 12 tagmus, which can then be followed by the correspond- nism to dampen the nystagmus, various surgeries have ing amount of recession−resection procedure [48]. Th e been done to dampen the congenital nystagmus. Th ese amount of surgery is based on the prism diopters toler- procedures were referred by Crone [52] as immobiliza- ated by the patient preoperatively. tion procedures. Various surgical procedures have been Spielmann [49] in a retrospective study of 120 patients mentioned in the literature. Von Noorden summarized who underwent artifi cial divergence surgery found 93 these surgical principles, including large recession of all (77.5%) of the patients were orthophoric, 18 patients had horizontal rectus muscles, the tenotomy procedure, fi xa- exophoria postoperatively, and 9 patients had exotropia. tion of the extraocular muscles to the periosteum of the Exotropia was found to be associated with hypermetro- lateral orbital wall, retro-equatorial myopexy of all hori- pia. Spielmann proposed bilateral recession of medial zontal rectus muscles, placement of retro-equatorial rectus muscle by 5–13 mm depending on the amount of encircling silicone band over rectus muscles in both eyes prism determined preoperatively by the prism adaptation and extirpation of horizontal rectus muscles. test. She recommended 5 mm recession if the fusion was Both retro equatorial recession of horizontal rectus tolerated with 30–40 PD, 7 mm for 50–60 PD, and 8 mm muscle and tenotomy procedure have been used more if fusion exceeds 60 PD. frequently and will be discussed in detail. Some patients have a convergence null in addition to the gaze angle null causing the AHP. If the amount of Retro-Equatorial Recession of Horizontal divergence induced by base-out prisms did not satisfacto- Rectus Muscles rily correct the AHP, these patients benefi tted by a combi- Bietti and Bagolini [53], in 1956, fi rst described retro- nation of artifi cial divergence and Anderson–Kestenbaum equatorial recession of all four horizontal rectus muscles. procedure [48, 50]. Th e amount of surgery is done for the Von Noorden and Sprunger [54] performed this procedure total prism diopters tolerated by artifi cial divergence pro- on three patients and reported increased acuity in two cedure and then the remaining AHP is corrected using patients and correction of head posture in one patient. the Anderson–Kestenbaum procedure. Helveston et al. [55] performed this procedure in ten patients Zubcov et al. [48] compared pre- and postoperative and reported dampening of nystagmus and improvement of eye movement recording and binocular visual acuities of visual acuity in 80% of patients. All his patients also reported patients who underwent the Anderson−Kestenbaum improvement in visual acuity and head posture. Datta et al. procedure (n = 7), artifi cial divergence procedure (n = 6), [56] performed surgery on nine patients and reported and a combination of both procedures (n = 5) in patients decreased amplitude in 15 eyes and increased visual acuity with congenital nystagmus. In patients who underwent in 12 eyes. Boyle et al. [57] in a retrospective review of 18 artifi cial divergence surgery, only one patient developed patients who underwent retro-equatorial recession surgery 4 PD esophoria postoperatively. Stereopsis improved in of horizontal muscle, 50% of patients showed improvement four patients. Four patients had a head turn of less than in visual acuity by at least one Snellen line. All patients 5°. Binocular visual acuity improved in 50% of the underwent medial rectus recession of 8–10 mm, and bilat- patients by 1–2 Snellen lines. Eye movement recordings eral lateral rectus muscle recession of 8–12 mm. showed broadening of the null zone. In patients who Bagheri et al. [58] reported results of 20 patients who underwent a combined procedure, stereopsis improved underwent horizontal rectus recession surgery. Th irteen in two patients and no residual head turn greater than 5° patients (76.5%) improved in visual acuity from one to was found. Binocular visual acuity improved by two or three Snellen lines. AHP improved in most of the patients. more Snellen lines in four of the fi ve patients. Broadening Similar results were also documented by other authors, of the null zone was noticed in all patients. Davis et al. [59] and Atilla et al. [60]. Th ey calculated the Graf et al. [51] in a retrospective study to analyze the amount of recession individually depending on the angle of eff ects of Kestenbaum surgery and artifi cial divergence deviation, head position, and amount of strabismus if pres- surgery found that artifi cial divergence surgery when ent. Recessions performed on the medial rectus were more performed alone off ers better correction of AHP than eff ective than recession on the lateral rectus. Th us surgery with the Kestenbaum surgery. However, in patients with is planned based on the eff ect of recession of the medial large AHP, combining both artifi cial divergence surgery rectus muscle rather than the lateral rectus recession. To and Kestenbaum surgery gives better results. correct the associated strabismus, the surgical plan is References 169 revised by increasing the recession of medial rectus muscles Acknowledgments We acknowledge support from Shery in case of esotropia, and recession of lateral rectus muscles Thomas, Chris Degg, Nagini Sarvananthan, Rebecca McLean, in case of exotropia. Similar adjustments can be made to Mervyn Thomas, Mylvaganam Surendran, and Shegufta Farooq. We thank the Nystagmus Network for their continued correct the AHP for example in patients with left face turn, interest in and support for nystagmus research. We acknowl- the right lateral rectus and left medial rectus is recessed edge the fi nancial support of Ulverscroft Foundation, more than the right medial rectus and left lateral rectus. Medisearch, National Eye Research Centre, and Nystagmus Network. The Tenotomy Procedure Advancements in understanding secondary mechanisms involved in the reducing nystagmus amplitude in patients who underwent recession−resection surgery for congenital nystagmus mainly to correct the AHP has led to a new References surgical procedure “tenotomy” of extraocular muscle. 1. Gresty MA, Bronstein AM, Page NG, Rudge P (1991) Th is procedure has been reported to be benefi cial in Congenital-type nystagmus emerging in later life. Neurology patients without compensatory mechanisms, also in 41:653–656 patients with a null region at or near primary position 2. Dell’Osso LF (1985) Congenital, latent and manifest latent and in patients with a non-stationary null region (PAN) nystagmus–similarities, diff erences and relation to strabis- [61].Th e tenotomy procedure can be done on both hori- mus. Jpn J Ophthalmol 29:351–368 zontal and vertical rectus muscles based on the dominant 3. Gradstein L, Reinecke RD, Wizov SS, Goldstein HP (1997) plane of the nystagmus. Congenital periodic alternating nystagmus. Diagnosis and Following the initial success of the tenotomy proce- management. Ophthalmology 104:918–928; discussion dure in an animal model [62], clinical trials [63, 64] were 928–919 performed on patients with congenital nystagmus with 4. Shallo-Hoff mann J, Riordan-Eva P (2001) Recognizing and without sensory defi cits including asymmetric con- periodic alternating nystagmus. Strabismus 9:203–215 genital PAN. In the fi rst trial, involving ten patients, bin- 5. Abadi RV, Pascal E (1994) Periodic alternating nystagmus ocular visual acuity increased in fi ve patients and in humans with albinism. Invest Ophthalmol Vis Sci 35: remained unchanged in the remaining patients. Th e eye 4080–4086 movement recording data showed an increase in the aver- 6. Adelstein F, Cuppers C (1966) On the problem of true and age foveation times in all nine patients’ fi xating eyes. In apparent abducens paralysis (so-called “blocking syn- the second trial, tenotomy was performed on fi ve patients drome”). Buch Augenarzt 46:271–278 with congenital nystagmus. Visual acuity improved in 7. Hertle RW, Zhu X (2000) Oculographic and clinical char- four of the fi ve patients, but did not improve in a patient acterization of thirty-seven children with anomalous head with retinal dystrophy. postures, nystagmus, and strabismus: the basis of a clinical algorithm. J AAPOS 4:25–32 8. Abadi RV (1979) Visual performance with contact lenses Summary for the Clinician and congenital idiopathic nystagmus. Br J Physiol Opt 33: ■ Various surgical procedures are used to treat 32–37 both the AHP and strabismus seen in patients 9. Allen ED, Davies PD (1983) Role of contact lenses in the with congenital nystagmus. Surgical consists management of congenital nystagmus. Br J Ophthalmol mostly of recessions alone or the combination of 67:834–836 recessions and resections depending on the 10. Hertle RW (2000) Examination and refractive manage- amount of head turn and strabismus. ment of patients with nystagmus. Surv Ophthalmol 45: ■ Th e surgical plan depends on whether patient has 215–222 horizontal or vertical AHP or head tilt and the 11. Dell’Osso LF (2002) Development of new treatments for presence or absence of strabismus. Other compen- congenital nystagmus. Ann N Y Acad Sci 956:361–379 satory need to be taken into consideration before 12. Schornack MM, Brown WL, Siemsen DW (2007) Th e use deciding on the type of surgery. For example, if of tinted contact lenses in the management of achromatop- there is dampening of nystagmus mechanisms on sia. Optometry 78:17–22 convergence, artifi cial divergence surgery alone 13. Metzger EL (1950) Correction of congenital nystagmus. can be performed, or it can be combined with Am J Ophthalmol 33:1796–1797 Anderson−Kestenbaum like procedures. 14. Goddé-Jolly D, Larmande A (1973) Les nystagmus. Paris, Masson 170 12 Management of Congenital Nystagmus with and without Strabismus

15. Hertle RW, Maybodi M, Mellow SD, Yang D (2002) Clinical 33. Taylor JN (1973) Surgery for horizontal nystagmus– and oculographic response to Tenuate Dospan (diethyl- Anderson-Kestenbaum operation. Aust J Ophthalmol propionate) in a patient with congenital nystagmus. Am 1:114–116 J Ophthalmol 133:159–160 34. De Decker W (1987) Kestenbaum transposition in nystag- 12 16. Pradeep A, Th omas S, Roberts EO et al (2008) Reduction mus theraphy. Transposition in horizontal and torsional of congenital nystagmus in a patient aft er smoking canna- plane. Bull soc Belge Ophthalmol 221–222 bis. Strabismus 16:29–32 35. Flynn JT, Dell’Osso LF (1979) Th e eff ects of congenital nys- 17. Sarvananthan N, Proudlock FA, Choudhuri I et al (2006) tagmus surgery. Ophthalmology 86:1414–1427 Pharmacologic treatment of congenital nystagmus. Arch 36. Pierse D (1959) Operation on the vertical muscles in cases Ophthalmol 124:916–918 of nystagmus. Br J Ophthalmol 43:230–233 18. Shery T, Proudlock FA, Sarvananthan N et al (2006) Th e 37. Schlossman A (1972) Nystagmus with strabismus: surgical eff ects of gabapentin and memantine in acquired and con- management. Trans Am Acad Ophthalmol Otolaryngol genital nystagmus: a retrospective study. Br J Ophthalmol 76:1479–1486 90:839–843 38. Taylor JN, Jesse K (1987) Surgical management of congeni- 19. McLean R, Proudlock F, Th omas S et al (2007) Congenital tal nystagmus. Aust N Z J Ophthalmol 15:25–34 nystagmus: randomized, controlled, double-masked trial 39. Sigal MB, Diamond GR (1990) Survey of management of memantine/gabapentin. Ann Neurol 61:130–138 strategies for nystagmus patients with vertical or torsional 20. Anderson JR (1953) Causes and treatment of congenital head posture. Ann Ophthalmol 22:134–138 eccentric nystagmus. Br J Ophthalmol 37:267–281 40. Roberts EL, Saunders RA, Wilson ME (1996) Surgery for 21. Solomon D, Shepard N, Mishra A (2002) Congenital peri- vertical head position in null point nystagmus. J Pediatr odic alternating nystagmus: response to baclofen. Ann N Y Ophthalmol Strabismus 33:219–224 Acad Sci 956:611–615 41. Yang MB, Pou-Vendrell CR, Archer SM et al (2004) Vertical 22. Comer RM, Dawson EL, Lee JP (2006) Baclofen for patients rectus muscle surgery for nystagmus patients with vertical with congenital periodic alternating nystagmus. Strabismus abnormal head posture. J AAPOS 8:299–309 14:205–209 42. Conrad HG, de Decker W (1978) “Kestenbaum’s surgical 23. Blekher T, Yamada T, Yee RD, Abel LA (1998) Eff ects of rotation of the eyes” in patients with head tipped to the acupuncture on foveation characteristics in congenital shoulder (author’s transl). Klin Monatsbl Augenheilkd nystagmus. Br J Ophthalmol 82:115–120 173:681–690 24. Abadi RV, Carden D, Simpson J (1980) A new treatment 43. De Decker W, Conrad HG (1988) Torsional shift opera- for congenital nystagmus. Br J Ophthalmol 64:2–6 tion, a tool in complete early childhood strabismus. Klin 25. Sharma P, Tandon R, Kumar S, Anand S (2000) Reduction Monatsbl Augenheilkd 193:615–621 of congenital nystagmus amplitude with auditory biofeed- 44. De Decker W (1990) Rotatorischer Kestenbaum an geraden back. J AAPOS 4:287–290 Augenmuskeln. Z Prakt Augenheilkd 11:111 26. Carruthers J (1995) Th e treatment of congenital nystag- 45. von Noorden GK, Jenkins RH, Rosenbaum AL (1993) mus with Botox. J Pediatr Ophthalmol Strabismus 32: Horizontal transposition of the vertical rectus muscles for 306–308 treatment of ocular torticollis. J Pediatr Ophthalmol 27. Oleszczynska-Prost E (2004) Botulinum toxin A in the Strabismus 30:8–14 treatment of congenital nystagmus in children. Klin Oczna 46. Spielmann A (1987) Th e “oblique” Kestenbaum procedure 106:625–628 revisited. In: Lenk-Schafer M (ed) Orthoptic horizons. 28. Goto N (1954) A study of optic nystagmus by the electro- Transactions of the sixth international orthoptic congress. oculogram. Acta Soc Ophthalmol Jap 58:851–865 Harrogate, UK, pp 433 29. Kestenbaum A (1953) New operation for nystagmus. Bull 47. Cuppers C (1971) Problems in the surgery for ocular nys- Soc Ophtalmol Fr 6:599–602 tagmus. Klin Monatsbl Augenheilkd 159:145–157 30. Parks MM (1973) Symposium: nystagmus. Congenital 48. Zubcov AA, Stark N, Weber A et al (1993) Improvement of nystagmus surgery. Am Orthopt J 23:35–39 visual acuity aft er surgery for nystagmus. Ophthalmology 31. Calhoun JH, Harley RD (1973) Surgery for abnormal head 100:1488–1497 position in congenital nystagmus. Trans Am Ophthalmol 49. Spielmann A (1993) La mise en divergence artifi cielle dans Soc 71:70–83; discussion 84–77 les nystagmus congénitaux. A propos de 120 cas. Bull Soc 32. Nelson LB, Ervin-Mulvey LD, Calhoun JH et al (1984) Fr Ophtalmol 6/7:571–578 Surgical management for abnormal head position in nys- 50. Sendler S, Shallo-Hoff mann J, Muhlendyck H (1990) tagmus: the augmented modifi ed Kestenbaum procedure. Artifi cial divergence surgery in congenital nystagmus. Br J Ophthalmol 68:796–800 Fortschr Ophthalmol 87:85–89 References 171

51. Graf M, Droutsas K, Kaufmann H (2001) Surgery for nys- tion or head posture in patients with nystagmus. J AAPOS tagmus related head turn: Kestenbaum procedure and arti- 9:433–437 fi cial divergence. Graefes Arch Clin Exp Ophthalmol 59. Davis PL, Baker RS, Piccione RJ (1997) Large recession 239:334–341 nystagmus surgery in albinos: eff ect on acuity. J Pediatr 52. Crone RA (1971) Th e operative treatment of nystagmus. Ophthalmol Strabismus 34:279–283; discussion 283–275 Ophthalmologica 163:15–20 60. Atilla H, Erkam N, Isikcelik Y (1999) Surgical treatment in 53. Bietti GB (1956) Notes on ophthalmological surgical tech- nystagmus. Eye 13(Pt 1):11–15 nics. Boll Ocul 35:642–656 61. Dell’Osso LF (1998) Extraocular muscle tenotomy, dissec- 54. von Noorden GK, Sprunger DT (1991) Large rectus muscle tion, and suture: a hypothetical therapy for congenital nys- recessions for the treatment of congenital nystagmus. Arch tagmus. J Pediatr Ophthalmol Strabismus 35:232–233 Ophthalmol 109:221–224 62. Dell’Osso LF, Hertle RW, Williams RW, Jacobs JB (1999) A 55. Helveston EM, Ellis FD, Plager DA (1991) Large recession new surgery for congenital nystagmus: eff ects of tenotomy of the horizontal recti for treatment of nystagmus. on an achiasmatic canine and the role of extraocular prop- Ophthalmology 98:1302–1305 rioception. J AAPOS 3:166–182 56. Datta H, Prasad S (1994) Postequatorial horizontal rectus 63. Hertle RW, Dell’Osso LF, FitzGibbon EJ et al (2004) recession in the management of congenital nystagmus. Horizontal rectus muscle tenotomy in children with infan- Indian J Ophthalmol 42:203–206 tile nystagmus syndrome: a pilot study. J AAPOS 8: 57. Boyle NJ, Dawson EL, Lee JP (2006) Benefi ts of retroequa- 539–548 torial four horizontal muscle recession surgery in congeni- 64. Hertle RW, Dell’Osso LF, FitzGibbon EJ et al (2003) tal idiopathic nystagmus in adults. J AAPOS 10:404–408 Horizontal rectus tenotomy in patients with congenital 58. Bagheri A, Farahi A, Yazdani S (2005) Th e eff ect of bilateral nystagmus: results in 10 adults. Ophthalmology 110: horizontal rectus recession on visual acuity, ocular devia- 2097–2105 Chapter 13 Surgical Management of Dissociated Deviations 13 Susana Gamio

Core Messages ■ Dissociated deviation (DD) manifests as a slow, for cases with bilaterally symmetric DVD. Cases intermittent, and variable vertical (DVD), hori- with asymmetric DVD are more common. Th ese zontal (DHD), and torsional (DTD) movement. cases require asymmetrical techniques. It is usually found in patients with early onset ■ Dissociated horizontal deviation (DHD): Th e strabismus and profound sensorial anomalies. main diagnostic sign of DHD is the presence of a ■ Th e treatment for patients with DD requires a horizontal deviation, esotropia (ET), or exotropia specifi c surgical approach to improve the vertical, (XT) that changes with fi xation of each eye, unre- horizontal, and torsional misalignment simulta- lated to diff erent accommodation, muscle weak- neously. ness, or restriction. Th e technique most used for ■ DVD neither disappears nor improves over time; DHD is unilateral lateral rectus (LR) recession. the aim of treatment is to obtain a latent deviation. Retroequatorial myopexy (posterior fi xation) of ■ Symmetric dissociated vertical deviation (DVD), the LR with recession of this muscle is recom- with good bilateral visual acuity (VA), without mended by certain authors. Bilateral LR recession oblique muscle dysfunction: four surgical alter- is indicated when XT is bilateral; unilateral or natives: (1) Bilateral large superior rectus (SR) bilateral medial rectus (MR) recession when the recession. (2) Bilateral retroequatorial myopexy patient exhibits ET instead of XT. Performing an (posterior fi xation) of the SR combined with LR recession added to MR advancement is a valid or without recession of these muscles. (3) Four alternative in cases with previous surgery on the oblique muscles weakening procedure. (4) Bilateral medials. inferior rectus (IR) resection. ■ Dissociated torsional deviation (DTD): Children ■ Bilateral DVD with deep unilateral amblyopia: with DD frequently have head turn but they also three available procedures: (1) Unilateral SR have head tilt. Th e head tilt can be toward the recession, (2) Unilateral inferior oblique anterior shoulder of the fi xing eye (direct tilt) or toward transposition (IOAT), and (3) Unilateral IR resec- the contralateral side (inverse tilt). We have to tion or tucking. take into account the head tilt to attempt to ■ DVD with inferior oblique overaction (IOOA) improve the head position when performing and V pattern: (1) Bilateral IOAT. (2) Bilateral SR surgery. recession added to bilateral inferior oblique (IO) ■ Obtaining long-term control of the deviation in recession. patient with DD is diffi cult; a successful out- ■ DVD with superior oblique overaction (SOOA) come in the postoperative period does not guar- and A pattern: (1) Bilateral SR recession, (2) antee the fi nal alignment. In treated patients Bilateral SR recession + superior oblique (SO) with DD, some kind of movement is always posterior tenectomy, or (3) Four oblique muscles detected when performing the cover test. DVD weakening procedure. never disappears completely and the dissociated ■ Symmetric vs. Asymmetric surgeries for DVD: behavior in DHD also persists when testing Bilateral symmetric procedures are performed under slow cover test. 174 13 Surgical Management of Dissociated Deviations

Vertical manifestation of DD is known as DVD and is 13.1 Dissociated Deviations characterized for being a slow, intermittent, variable, and Dissociated deviation (DD) Represents a Challenge for bilateral movement of elevation, abduction, and extor- Diagnosis and Surgical Treatment. It is known to exhibit sion of the nonfi xating eye. Th e downward vertical drift 13 a slow, variable, and intermittent movement with vertical, of the hypertropic eye takes place together with intorsion horizontal, and torsional components. It is commonly and adduction. found in patients with early onset strabismus and pro- Th e horizontal component has recently been found sensorial anomalies [1–5]. described and is called dissociated horizontal deviation Diagnosis is not easy because the movement is slow (DHD) [5, 8–10]. Even though most papers consider and needs a more prolonged occlusion to appear; the DHD as a variable, intermittent exodeviation with a dif- amount of deviation is variable, intermittent, and depends ferent magnitude according to fi xating eye, there exist on attention. Th ese patients usually show horizontal, ver- cases that exhibit an esodeviation with the same charac- tical, and torsional movements when performing the teristics of variability and intermittence [11, 12]. Th ere cover test and have diff erent amounts of deviation when are also patients who manifest esotropia (ET) when fi x- fi xing with each eye. Th ey also have latent nystagmus ating with one eye and exotropia (XT) when fi xating (LN), head tilt, and associated oblique muscles dysfunc- with the other eye [9]. tion in many cases. Th e torsional component of this entity, named dissoci- A distinctive feature of dissociated strabismus is the ated torsional deviation (DTD), occurs simultaneously response to changes in light density; these changes impact with vertical movement: extorsion of the elevating eye and on the deviation amount. When neutral fi lters of increasing intorsion of the fi xating eye. Th e vertical movement always density (bagolini fi lter bar) are placed before the fi xating cooccurs with extorsion of the elevating eye and intorsion eye, the hypertropic eye falls (Bielschowsky’s phenomenon) of the descending eye. Th is is infl uenced by oblique mus- [6]. Conversely, increasing light in the hypertropic eye will cles dysfunction also causing incomitance of vertical and cause an increase in upward deviation. A further peculiar torsional deviation in lateroversions [11, 13, 14]. behavior of patients with DD is evidenced by Posner’s Measuring horizontal and vertical DD is complicated maneuver [7]: when occluding one eye, the eye moves because we need to superimpose horizontal and vertical upwards, when occluding the contralateral eye (keeping prisms over each eye. In addition, it is necessary to mea- the other eye occluded), the second eye moves upwards sure DVD and DHD with each eye fi xating in all gaze posi- and the fi rst one downwards, becoming aligned in the ver- tions (including head tilts) to have the necessary panorama tical plane (Fig. 13.1). to choose the best surgical procedure for each case. Red glass testing yields particular results in dissoci- Th erefore, surgical treatment of patients with DD ated vertical deviation (DVD). Regardless of whether the requires a specifi c surgical approach. Long-term surgical red fi lter is placed before the right eye or the left one, the results and recommendations for these cases remain patient sees the red light below the white one. sparse in literature. Th e purpose of this chapter is to men- Th ese maneuvers attest to the tight interocular inter- tion the surgical alternatives tailored to treat each partic- relation of this particular form of strabismus. ular case.

Fig. 13.1 Posner’s maneuver: when occluding one eye, the eye moves upwards; when occluding the contralateral eye (keeping the other eye occluded), the second eye moves upwards and the ﬁ rst one downwards, becoming aligned in the vertical plane 13.2 Surgical Alternatives to Treat Patients with DVD 175

retraction and lid fi ssure asymmetry. Th is technique Summary for the clinician may limit elevation, especially in abduction (Pseudo ■ DD have three components: vertical (DVD), hori- inferior oblique over action (IOOA) ). It should be noted zontal (DHD), and torsional (DTD) movements. that weakening of SR modifi es horizontal deviation in ■ Surgical plan requires taking into account the PP, causing a 6 PD exodeviation, which should be taken three components and must be tailored to treat into account when planning surgery. each particular case. Conventional recession (3–5 mm) of SR together with retroequatorial myopexy (12–15 mm of original insertion) is used by several author successfully [64]. Th e posterior fi xation suture must be placed at least 20 mm, and preferably 23–25 mm from the limbus, which oft en is 13.2 Surgical Alternatives to Treat technically troublesome. Patients with DVD Th e four oblique weakening procedures proved to be Patients with DVD are usually asymptomatic, but in those an eff ective technique to treat these cases. Th is procedure cases where signifi cant hypertropia is manifested sponta- is especially useful in cases that underwent surgery on neously, or those associated with horizontal misalign- two horizontal rectus muscles in each eye and in those ment, surgical treatment should be considered knowing where operating on the SR implies a risk of anterior seg- that the problem will not always be completely solved. ment ischemia. DVD neither disappears nor improves over time [15]. IR resection: Although this technique has been pro- Treatment is focused on obtaining a latent vertical devia- posed as a primary procedure, we believe that it should be tion, only present with occlusion and to a lesser amount. reserved for reoperation in the case of failure of SR reces- Multiple techniques have been developed for DVD sion. It creates a marked restriction of elevation and in treatment; the most successful ones are those that limit some cases alterations in the lid fi ssures. Its additional elevation to a greater degree. horizontal eff ect, ET on PP, should also be considered. To choose the surgical procedure, the following should be taken into account: (1) visual acuity (VA) (2) degree of non-DVD incomitance (3) oblique muscles dysfunction with A or V pattern (4) Degree of DVD symmetry. 13.2.2 Bilateral DVD with Deep Unilateral Amblyopia DVD cases with deep monocular amblyopia are usually characterized by great asymmetry in vertical deviation, 13.2.1 Symmetric DVD with Good Bilateral even simulating monocular DVD. Visual Acuity, with No Oblique Monocular surgery is possible in patients with a devi- Muscles Dysfunction ating eye with no possibilities of becoming fi xating eye Th e following are the most used procedures in these cases: due to deep amblyopia. Th ere are four procedures that may be used in these 1. Bilateral large superior rectus (SR) recession (7–12 mm) cases: [16–20] 2. Bilateral retro-equatorial myopexy (posterior fi xation) 1. Unilateral SR recession [16, 33]. of the SR combined with or without recession of these 2. Unilateral inferior oblique anterior transposition muscles [18, 21–24] (IOAT) [34, 35]. 3. Four oblique muscles weakening procedure (superior 3. Unilateral IR resection or tucking [36]. oblique (SO) recession or tenectomy and inferior 4. Unilateral SR retroequatorial myopexy (posterior fi xa- oblique (IO) recession or anterior transposition tion) combined with or without recession of this (IOAT) ) [25–28] muscle [18]. 4. Bilateral inferior rectus (IR) resection [16, 29–32] When unilateral SR recess is decided, the amount of such Large SR recession with hang-loose technique is one must be moderate (5–7 mm) to avoid postoperative of the mostly used in these cases. Extensive dissection hypotropia. Th is technique is chosen in cases showing is required to clean attachments off the SR to avoid comitant vertical deviation in lateroversions. 176 13 Surgical Management of Dissociated Deviations

Many authors express concern that unilateral SR IOAT remained with postoperative vertical deviation. recession might also result in an unacceptable postopera- 10/20 of such cases had preoperative asymmetric DVD. tive hypotropia in the operated eye or in a large hypertro- Although late development of a postoperative A pat- pia in the contralateral eye, if the patient were to switch tern strabismus does not appear to be a problem even in 13 fi xation [20]. For this reason, unilateral surgery is reserved patients with modest preoperative V patterns, the true for patients with dense amblyopia, who would have little incidence of the development of A pattern have not been or no chance of changing fi xation aft er surgery. In addressed to date. Schwartz and Scott’s paper [33], postoperative hypotropia Bradley Black [39] reported that aft er the operation, 50% developed in the operated eye in 12 patients (21%). Nine of his patients had experienced neither A nor V pattern. of these patients had deviations less than 10 PD. In Th irty-three percent had a V pattern averaging 4 PD (2–8 Helveston’s study [3], only 5 out of 33 patients undergo- PD). Seventeen percent had a postoperative A pattern. ing unilateral surgical correction of DVD developed a In our series, 4/20 patients with bilateral IOAT had signifi cant deviation in the unoperated eye. Duncan and postoperative A pattern (20%) over 36-month follow-up von Noorden [21] demonstrated the development of con- on average. tralateral DVD postoperatively in 8/35 cases. When there is a remaining postoperative vertical devi- In those cases manifesting incomitance in laterover- ation aft er the IOAT, a unilateral SR recession can be per- sions: greater hypertropia in adduction, unilateral IOAT formed according to the amount of vertical deviation in is chosen. PP. Th is procedure proved eff ective in obtaining good Bothun and Summers [34] proved that unilateral vertical alignment and has apparently given a predictable IOAT is an eff ective treatment for unilateral or markedly and stable result with low incidence of postoperative asymmetric DVD in patients with a strong, contralateral complications. fi xation preference. Th is surgery reduces IOOA, but may Several studies have attempted to obtain better also cause an ipsilateral hypotropia. Ipsilateral DVD in surgical outcomes in asymmetric DVD with IOOA by PP decreased from a mean of 20.2 to 3.7 PD in their performing asymmetric procedures. Th ere are several series. Ninety percent of the patients had an excellent surgical alternatives: postoperative result. Goldchmit et al. [35] found that the unilateral IOAT ■ Combined unilateral IO resection and bilateral IOAT. produces a mean correction of 18.1 PD (range, 4–33) in ■ Graded bilateral IOAT (1, 2, or 3 mm anterior to the PP, directly proportional to the size of the hypertropia IR muscle insertion). before surgery. ■ Graded bilateral IOAT (1, 2, or 3 mm posterior to the IR muscle insertion). ■ Symmetric and bilateral IOAT + SR recession of the most hypertropic eye. 13.2.3 DVD with Inferior Oblique Overaction (IOOA) and V Pattern Burke et al. [40] suggested a graded procedure to eff ec- When DVD is associated with IOOA, the hypertropia is tively treat coexisting DVD and IOOA. It has signifi cantly greater in adduction and a V pattern may be observed. In reduced the mean DVD from 13.4 PD to 6.7 PD. In cases extreme adduction, a true hypertropia may be seen in of asymmetric DVD, unequal transpositions were per- addition to the DVD. formed: IOAT in the eye with the larger DVD can be placed up to 2 mm anterior to the temporal pole of the IR. 1. Bilateral IOAT has become a popular surgical treat- Th e DVD remained controlled in 86% of their cases aft er ment for DVD with IOOA. a 2-year follow-up. Th e best results were obtained in those 2. Th e second alternative is to perform a bilateral SR patients with a preoperative DVD of less than 15 PD. recession added to bilateral IO recession [37]. Mims and Wood [41] also performed bilateral graded displacement of the IO tendon, attaching the muscle at a Th e IOAT reduces the hypertropia to an acceptable point 2–4 mm anterior to the lateral end of the IR inser- amount, and eliminates the IOOA and the V pattern with tion. Th ese authors reported low residual IOOA in 11/61 a low incidence of recurrence. However, this surgical patients. Only one patient required reoperation for mani- procedure has yielded poor results in patients with fest DVD. asymmetric DVD and IOOA [38]. Kratz et al. [42] compared two groups of patients with Nine out of 20 consecutive patients in our series with DVD who underwent standard or graded IOAT. In the DVD and IOOA who underwent bilateral and symmetric graded group, the IO tendon was placed in one of the 13.2 Surgical Alternatives to Treat Patients with DVD 177 three stations: 1 mm posterior or 1 mm anterior to the IR Th e weakening of both elevators (IO and SR) always insertion or at the level of the IR insertion. In the stan- results in an elevation defi ciency, that could be acceptable dard group, the IO tendon was positioned 1 mm anterior in cases with large hypertropia, but it could induce a to the IR insertion for all degrees of DVD. Th e residual noticeable and undesirable chin-up head position. postoperative DVD was 1.15 PD in the graded group compared with 2.44 PD in the standard group. Th is difference was statistically signifi cant. 13.2.4 DVD with Superior Oblique Finally, Snir et al. [43], to improve the postoperative Overaction (SOOA) and A Pattern outcome in patients with asymmetric DVD with IOOA, In these cases, DVD is greater in abduction of the nonfi x- augmented the functional change in the IO induced by ating eye than in PP. Th e SOOA causes incomitance in IOAT by resecting the IO muscle in the eye with greater DVD and A pattern [14, 44, 45] (Fig. 13.2). vertical deviation before displacing it anterior to the IR In this group, when A pattern anisotropia is small not insertion. Th e IO resection was graded according to the over 14 PD diff erence in the preoperative vertical deviation between the eyes: 3 mm for a diff erence of up to 10 PD and 5 mm 1. Bilateral SR improves DVD and controls A pattern for a diff erence of 11–20 PD. Th ese authors compared the [46]. postoperative outcomes of six consecutive patients who underwent combined graded monocular resection and If the A pattern is larger, undercorrection is obtained; bilateral ATIO with six consecutive historical control therefore, other alternatives should be used. patients who underwent equal IOAT. Th e mean diff er- 2. Bilateral SR recession + bilateral SO posterior tenec- ence of the asymmetric DVD in the primary position was tomy or [44, 47, 48]. reduced from 13.3 to 2.2 PD in the study group and from 3. Four oblique weakening procedure [27, 28]. 13.3 to 10.2 PD in the control group (P = 0.004). In conclusion, for patients with asymmetric DVD and Simultaneous weakening of SO and SR may cause an coexisting IOOA and V pattern, we recommend bilateral inversion of vertical incomitance, transforming the A IOAT combined with monocular graded IO resection in pattern into V pattern. Th us, it is benefi cial to carry out the eye with greater DVD or bilateral but graded IOAT to the four oblique weakening procedure in these patients prevent the postoperative vertical deviation. [28, 44].

Fig. 13.2 Dissociated vertical deviation (DVD) with SOOA and A pattern: DVD is greater in abduction of the nonﬁ xating eye 178 13 Surgical Management of Dissociated Deviations

It may be a quite complex and lengthy procedure for tropia, or it can remain aligned when the DVD is of a nonexperienced surgeons; it produces a symmetric out- similar magnitude to that of the vertical tropia. Th is come and so it is not the preferred option in a markedly situation may be erroneously interpreted as monocular asymmetrical case. It could also produce a vertical devi- DVD. 13 ation. When this complication occurs, a simple SR reces- Asymmetric DVD will oft en appear to be unilateral. sion of the hypertropic eye can be performed according However, by performing the proper maneuvers, the bilat- to the hypertropia amount in PP, thus solving the erality of most cases can be detected. Th e objective eye problem. movement recording clearly demonstrates that DVD is Th ere are several surgical alternatives to treat asym- bilateral in almost all cases. metric cases with A pattern. A graded bilateral IOAT or a Bilateral symmetric procedures are performed for SR recession of the most hypertropic eye can be added to cases of bilaterally symmetric DVD (within ± 7 PD), but the usual SO weakening. asymmetric DVD is more common, and larger DVD can Th e size of the A pattern and the presence of asym- be found in the nonfi xating eye or even in the fi xating metry are important when deciding the technique to be eye. employed. Determining the diff erence in the amount of SR recession in these asymmetric cases remains challenging. Th e maximum diff erence allowed to obtain a good outcome remains controversial. 13.2.5 Symmetric vs. Asymmetric Surgeries for DVD DVD is often perceived as a bilateral condition; however, many cases are markedly asymmetric. These 13.2.6 DVD with Hypotropia cases are usually found associated with unilateral deep of the Nonfi xating Eye amblyopia. DVD usually manifests as an intermittent hypertropia, Just as oblique muscle dysfunction makes DVD but there are certain cases with hypotropia of the nonfi x- incomitant in diff erent gaze positions, the presence of a ating eye. Although rare, these cases are identifi ed in dif- true vertical deviation (hypo or hypertropia) makes it ferent reports under the labels of Dissociated hypotropia asymmetric. [49, 50], Hypotropic DVD, Hypotropic Dissociated Th e nondissociated vertical tropia can be lesser or Deviation [51], or Inverse DVD (Fig. 13.3). larger than the amplitude of the DVD. Yet, we are not going to refer to patients with this con- When the nondissociated hypertropia is larger than dition, but to those with DVD and a hypotropic nonfi xat- the magnitude of the DVD, the hypotropic eye is never ing eye. We can distinguish two groups: the higher eye. Despite the fact that the greater amplitude of DVD is 1. Consecutive cases: cases secondary to surgical usually seen in the nonfi xating eye, cases with greater overcorrection (previous vertical acting muscles DVD in the fi xating eye do exist and may show hypotro- surgery). pia of the fellow eye in binocular conditions. When the 2. Primitive cases: patients with asymmetric DVD (greater cover test is performed, this hypotropic eye can either in the fi xating eye), with associated nondissociated verti- become hypertropic if DVD is larger than the vertical cal tropia or with unilateral deep amblyopia.

Fig. 13.3 Bilateral DVD with left hypotropia in primary position 13.3 Dissociated Horizontal Deviation 179

Th ree situations can lead to hypotropia of the nonfi xating 13.3 Dissociated Horizontal Deviation eye in a patient with DVD: DHD has become a more recognized entity in the last few 1. Hypertropia in the nondominant eye: the patient years and is usually related to the horizontal deviation appears to have greater DVD amplitude in the non- associated with DVD in patients with early onset strabis- dominant eye: when he changes the fi xation and fi x- mus history. Th e main diagnostic sign of DHD is the ates with that eye, despite its own DVD, hypotropia in presence of a horizontal variable deviation, ET, or XT that the other one becomes evident. changes with fi xation of each eye, unrelated to diff erent 2. True hypotropia of the nondominant eye. When the accommodation or presence of primary and secondary occlusion of this eye is performed, the magnitude of deviation due to weakness or restriction. DVD will determine the position reached by the eye: it It is a slow and variable horizontal movement, similar can be aligned, hypo, or hypertropic. to the intermittent hypertropia that characterized the 3. Nondissociated hypertropia in the dominant eye lead- DVD. Commonly both conditions coexist; both are vari- ing to hypotropia of the fellow eye in binocular condi- able and diffi cult to measure and are also more prominent tions. Th ese patients seem to have greater DVD during inattention. amplitude in the dominant eye. In DHD, we cannot neutralize the horizontal devia- Most cases of DVD that show hypotropia are due to sur- tion by the classical prism and alternating cover test. gical overcorrection, but other causes such as asymmetric Alternate cover testing must be performed slowly allow- DVD associated with vertical deviation or deep unilateral ing the nonfi xating eye time for the slow drift to fully amblyopia may be responsible for this clinical feature. manifest. It is also necessary to make the right eye fi xate Accurate diagnosis is essential for correct surgical man- fi rst and neutralize with prism the left eye deviation, and agement [52]. then let the left eye fi xate and neutralize the right eye deviation. Summary for the clinician Th e reversed fi xation test (RFT) [53] is useful to diagnose DHD. During this test, the patient is asked to ■ To choose the surgical procedure for DVD, we fi xate through the prism that neutralizes the deviation need to take into account: (1) VA; (2) vertical devi- of one of his eyes and then the occluder is shift ed to the ation incomitance; (3) oblique muscles dysfunc- uncovered eye without the prism and it is observed for tion with A or V pattern; (4) DVD symmetry. any refi xation movement when the cover test is per- ■ Symmetric DVD with good bilateral VA, with- formed. Th e test is positive when a refi xation movement out oblique muscle dysfunction: four surgical which can be measured placing prisms in front of this alternatives: (1) Bilateral large SR recession. (2) eye is observed. Bilateral retroequatorial myopexy (posterior fi x- Brodsky et al. [54] found that 50% of his patients with ation) of the SR combined with or without reces- consecutive XT had DHD demonstrated by a positive sion of these muscles. (3) Four oblique muscles RFT. Seven of the 14 patients with DHD had a greater weakening procedure. (4) Bilateral IR resection. exodeviation when fi xating with the preferred eye. In our ■ Bilateral DVD with deep unilateral amblyopia: series, seven patients had greater exodeviation when fi x- three available procedures: (1) Unilateral SR ating with the dominant eye, seven patients had greater recession. (2) Unilateral IOAT. (3) Unilateral esodeviation when fi xating with the nondominant eye, IR resection or tucking. and three cases had XT when fi xating with the dominant- ■ DVD with IOOA and V pattern: (1) Bilateral eye and ET when fi xating with the nonpreferred eye. Only IOAT. (2) Bilateral SR recession added to bilat- one patient had greater ET when fi xating with the domi- eral IO recession. nant eye. Th ese fi ndings seem to support his hypothesis ■ DVD with SOOA and A pattern: (1) Bilateral SR that the exodeviation is usually smaller with the nonpre- recession. (2) Bilateral SR recession + SO posterior ferred eye fi xating (Fig. 13.4). tenectomy. (3) Four oblique weakening procedure. DHD is oft en observed to be larger with visual inat- ■ Symmetric vs. Asymmetric surgeries for DVD: tention than when the prisms measurements are done, Bilateral symmetric procedures are performed for and the eye position under general anesthesia (GA) usu- cases with bilaterally symmetric DVD. Asymmetric ally shows greater deviation than the measured angle in DVD is more common and these cases require the awake state. asymmetrical techniques. Examining the patient under GA [55] is extremely useful to decide the amount of surgery to be done. Th e 180 13 Surgical Management of Dissociated Deviations

Fig. 13.4 Dissociated horizontal deviation (DHD). She has greater exodeviation when ﬁ xating with the dominant eye

eye position under GA used to show greater exodeviation Summary for the Clinician when the innervational forces are abolished. Th e forced duction can diagnose a restriction and the spring back ■ Th e main diagnostic sign of DHD is the presence test can determine a medial rectus (MR) muscle weak- of a horizontal variable deviation, ET, or XT that ness when it was previously recessed. changes with fi xation of each eye, unrelated to Wilson and McClatchey, in 1991 [5], recommended diff erent accommodation or presence of primary graded unilateral lateral rectus (LR) recession for the and secondary deviation due to weakness or treatment of DHD, and this was the most common restriction. method to treat it when surgery is indicated. ■ Th e technique most used for DHD was unilat- It was said that bilateral surgery is less oft en required eral LR recession. Bilateral LR recession is indi- for DHD than for DVD. However, DHD is almost cated when XT is bilateral; unilateral or bilateral always associated with DVD, so we consider that bilat- MR recession when the patient exhibits ET eral surgery to treat both is a good option in many instead of XT. Performing a LR recession added patients [56]. to an MR advancement is a valid alternative in All our patients had DHD coexisting with DVD; ten cases with previous surgery on the medials. cases received bilateral surgery to treat both conditions, fi ve underwent surgery just for the DVD because the horizontal deviation was small, and two patients received surgery for the horizontal deviation alone despite having DVD as well. 13.4 Dissociated Torsional Deviation. Head tilts in patients with Th e most used technique for DHD was unilateral LR Dissociated Strabismus recession. Retroequatorial myopexy (posterior fi xation) of the LR with a recession of this muscle is recom- Th ere is very little information on DTD in literature. mended by certain authors [12]. Bilateral LR recession Torsional movements are involved in the genesis of this is indicated when XT is bilateral, unilateral, or bilateral form of strabismus and oblique muscles are the main MR recession when the patient exhibits ET instead of oculomotor muscles with torsional action [2, 58, 59]. XT. Performing a LR recession added to MR advance- DVD mechanism has been elucidated recently by means ment is a valid alternative in cases with previous surgery of ocular movement recording techniques. DVD would on the medials. be mediated primarily by the SO in the fi xating eye and DVD and DHD usually coexist. When the vertical or the IO in the fellow eye, added to a bilateral supraversion the horizontal deviation manifests frequently, a surgical required for the maintenance of fi xation with the fi xating plan to fi x the drift of the eyes is needed. Bilateral sur- eye. In the latter eye, only an intorsional movement is gery is proposed to address both conditions simultane- observed, because the vertical components of SO and SR ously [57]. are annulled. A movement of elevation, abduction and 13.4 Dissociated Torsional Deviation. Head tilts in patients with Dissociated Strabismus 181 extorsion characteristic of DVD produced by SR and OI Direct tilt is observed in patients without horizon- is observed in the fellow eye. In this case, the vertical vec- tal alignment and with a head turn and fixation in tors would be added while the extorsion and abduction adduction. On tilting the head toward the fixating eye produced by the IO in upgaze would prevail on intorsion side, they are demanding more vestibular innervation and adduction of the SR. to increase adduction and therefore, they could Children with DD frequently have head turn; they usu- improve their monocular fixation. ally fi xate in adduction but they also have head tilts. Th e Th e most patients who adopt inverse tilt can obtain head tilt can be toward the shoulder of the fi xating eye (direct better vertical alignment in that position. tilt) or toward the contralateral side (inverse tilt) [60, 61]. Out of 50 consecutive patients in our series who Th is head tilt has been thought to be related to the underwent surgical treatment for DVD, only 54% (27/50) presence of DVD, but there is no evidence confi rming the had head tilt. Of 27 cases, 14 had direct tilt (51%); the relationship between these two fi ndings. head tilt did not improve vertical alignment. Th ey usually Guyton [58] claims that adopting an anomalous head obtain improvement of the head position by means of the posture can infl uence latent and manifest LN in some bilateral SR recession surgery. cases. Th e head tilt would damp the pattern of LN associ- Direct tilt improves the vertical alignment in two ated with the fi xing eye, and therefore, surgery on the fi x- situations: when a contracture of the SR of the nonfi xating eye is practically always necessary to abolish head tilts. ing eye exists or in asymmetric DVD cases, larger in the Brodsky et al. [62] proposed that direct tilt is not com- fi xating eye. pensatory for binocular vision, while a head tilt toward We found inverse head tilt, which improved the ver- the hyperdeviated eye (inverse tilt) serves to neutralize tical alignment, in 13/27 (49%) cases. Many of these the hyperdeviation and stabilizes binocular vision. patients had vertical deviation in PP and it was not rare According to Jampolsky’s description of Bielschowsky to fi nd SR contracture of the fi xating eye. When fi xing head tilt test (BHTT) response in DVD [63], there is an with either eye, the head tilt improved the vertical increased hyperdeviation of the contralateral eye on alignment. tilting to either side, the exactly inverse behavior to that When we have a patient with DD who needs surgery, of SO palsy or SR overaction/contracture syndrome the head tilt should be taken into account to attempt to (Fig. 13.5). improve the head position.

Fig. 13.5 Bielschowsky head tilt test (BHTT) response in DVD: there is an increased hyperdeviation of the contralateral eye on tilting to either side 182 13 Surgical Management of Dissociated Deviations

Finally, we want to point out that a great number of 8. Romero-Apis D, Castellanos-Bracamontes A (1992) patients with DD do not have head tilt. Th is fact makes Dissociated horizontal deviation: clinical fi ndings and sur- evident that there are other nonelucidated factors that gical results in 20 patients. Binocul Vis 7:135–138 determine such a particular clinical sign. 9. Wilson ME, Saunders RA, Berland JE (1995) Dissociated 13 horizontal deviation and accomodative esotropia: treatment options when an eso and exodeviation co-exist. Summary for the Clinician J Pediatr Ophtahlmol Strabismus 32:228 ■ When we have a patient with DD who need sur- 10. Zubcov AA, Reinecke RD, Calhoun JH (1990) Asymmetric gery, we have to take into account the presence horizontal tropias, DVD, and manifest laternt nystagmus: of head tilt to attempt to improve the head an explanation of dissociated horizontal deviation. J Pediatr position. Ophtahlmol Strabismus 27:59 ■ Direct tilt (toward the fi xing eye) is not compen- 11. Spielmann A (1990) Vertical and torsional deviations satory for binocular vision, while a head tilt in early strabismus. Bull Soc Ophtalmol Fr. 90(4):373–378; toward the hyperdeviated eye (inverse tilt) serves 381–384 to improve the vertical alignment. 12. von Noorden GK (1996) Cyclovertical deviations. In: Binocular vision and ocular motility: theory and management of strabismus, 5th edn. Mosby-Year Book, St Louis, pp 360 13. Berard PV, Reydy R, Berard PV Jr (1990) Symptomatologic 13.5 Conclusions value of dissociated vertical divergence in concomitant strabismus. Bull Soc Ophtahlmol Fr 90(1):31–38 Obtaining long-term control of the deviation in patient 14. McCall LC, Rosenbaum AL (1991) Incomitant dissociated with dissociated strabismus is diffi cult; a successful out- vertical deviation and superior oblique overaction. come in the postoperative period does not guarantee the Ophthalmolgy 98:911 fi nal alignment. In treated patients with DD, we will 15. Harcourt B, Mein J, Johnson F (1980) Natural history and always see some kind of movement when performing the associations of dissociated vertical divergence. Trans cover test. DVD never disappears completely and the dis- Ophtahlmol Soc UK 100:495 sociated behavior in DHD also persists when testing 16. Braverman DE Scott WE (1977) Surgical correction of under slow cover test. dissociated vertical deviations. J Pediatr Ophtahlmol Strabismus 14:337–342 17. Jampolsky A (1986) Management of vertical strabismus. Trans New Orleans Acad Ophtahlmol 34:141 References 18. Lorenz B, Raab I, Boergen KP (1992) Dissociated vertical 1. Guyton DL (2000) Dissociated vertical deviation: etiology, deviation: what is the most eff ective surgical approach? mechanism, and associated phenomena. J AAPOS 4: J Pediatr Ophtahlmo Strabismus 29:21 131–144 19. Magoon E, Cruciger M, Jampolsky A (1982) Dissociated 2. Guyton DL, Cheeseman EW Jr., Ellis FJ, Straumann D, Zee vertical deviation: an asymmetric condition treated DS (1998) Dissociated vertical deviation: an exaggerated with large bilateral superior rectus recession. J Pediatr normal eye movement used to damp cyclovertical latent Ophtahlmol Strabismus 19:152 nystagmus. Trans Am Ophthalmol Soc 96:389–429 20. Scott WE, Sutton VJ, Th alacker JA (1982) Superior rectus 3. Helveston EM (1980) Dissociated vertical deviation: a clin- recessions for dissociated vertical deviation. Ophtahlmology ical and laboratory study. Trans Am Ophthalmol Soc 78: 89:317–322 734–779 21. Duncan LF, von Noorden GK (1984) Surgical results in 4. Raab EL (1970) Dissociative vertical deviation. J Pediatr dissociated vertical deviations J Pediatr Ophthalmol Ophthalmol Strabismus 7:146–151 Strabismus 21:25–27 5. Wilson ME, McClatchey SK (1991) Dissociated horizontal 22. Hiles DA, Baybars I, Biglan AW (1986) Long-term stability deviation. J Pediatr Ophthalmol Strabismus 28:90–95 of the superior rectus recession Faden operation for dissocia- 6. Bielschowsky A (1938) Lectures on motor anomalies: II. tive vertical deviation. In: Campos ED (ed) Proceedings of Th e theory of heterophoria. Am J Ophtahlmol 21:1129 ISA V. Athena Scientifi c, Rome, Modena, Italy, pp 403–412 7. Posner A (1944) Noncomitant hyperphorias: considered as 23. Sprague JB, Moore S, Eggers H et al (1980) Dissociated ver- aberrations of the postural tonus of the muscular apparatus tical deviation: treatment with the fadenoperation of Am J Ophtahlmol 27:1275 Cuppers. Arch Ophtahlmol 98:465 References 183

24. von Noorden GK (1978) Posterior fi xation suture in stra- 39. Black BC (1997) Results of anterior transposition of the bismus surgery. In: Symposium on strabismus. Trans new inferior oblique muscle in incomitant dissociated vertical Orleans acad ophtahlmol. CV Mosby, St. Louis, pp 307 deviation. JAAPOS 1(2):83–87 25. Acosta Silva MA, Campomanes G (2000) Cirugia de cua- 40. Burke JP, Scott WE, Kutshke PJ (1993) Anterior transposi- tro oblicuos para Desviacion Vertical Disociada y sin- tion of the inferior oblique muscle for dissociated vertical drome em A. CLADE anais 2000 del XIV Congreso del deviation. Ophthalmology 100:245–250 CLADE. São Paulo, pp 359–360 41. Mims JLIII, Wood RC (1989) Bilateral anterior transposi- 26. Gamio S (2002) A surgical alternative for dissociated verti- tion of the inferior obliques. Arch Ophthalmol 107:41–44 cal deviation based on new pathologic concepts: weaken- 42. Kratz RE, Rogers GL, Bremer DL, Leguire LE (1989) ing all four oblique eye muscles. Outcome and results in 9 Anterior tendon displacement of the inferior oblique for cases. Binocul Vis Strabismus Q 17(1):15–24 DVD J Pediatr Ophtahlmol Strabisumus 26:212–217 27. Gamio S (2006) Weakening the four oblique muscles in the 43. Snir M, Axer-Siegel R, Cotlear D, Sherf I, Yassur Y (1999) tereatment of DVD. In: Proceedings of the joint congress: Combined resection and anterior transposition of the infe- the Xth meeting of ISA and the fi rst extraordinary meeting rior oblique muscle for asymmetric double dissociated ver- of CLADE. São Paulo, Brazil pp 97–100 tical deviation. Ophtahlmology 106(12):2372–2376 28. Texeira Krieger F, Caron Lambert A (2000) Efeito do debili- 44. Velez G, Velez F, Ela-Dalman N (2008) Surgical manage- tamento do músculo Oblicuo superior hiperfuncionante ment of dissociated vertical deviation associated with A associado a anteriorizacao do músculo oblicuo inferior na pattern strabismus. Poster presented at the 34th AAPOS Divergencia Vertical Dissociada. CLADE anais 2000 del Annual Meeting. Washington XIV Congreso del CLADE. Sao Paulo, pp 447–450 45. Velez G (2000) A clinical classifi cation of DVD for a better 29. Esswein Kapp MB, von Noorden GK (1994) Treatment of surgical approach. Fetscrif for Arthur Jampolsky. Th e residual dissociated vertical deviation with inferior rectus Smith Kettlewell Eye Research Institute, pp 59–63 resection. J Pediatr Ophtahlmol Strabismus 31:262 46. Melek N, Mendoza JC, Ciancia AO (1998) Bilateral reces- 30. Noel LP, Parks MM (1982) Dissociated vertical deviation: sion of the superior rectus: its infl uence in A and V pattern associated fi ndings and results of surgical treatment. Can strabismus. J AAPOS 2:61 J Ophtahlmol 17:10 47. Prieto-Diaz J (1979) Posterior tenectomy of the superior 31. Parks MM (1975) Dissociated hyperdevitions. In: Ocular oblique. J Pediatr Ophtahlmol Strabismus 16:321 motility and strabismus. Harper and Row, Hagerstown, 48. Shin GS, Elliott RL, Rosenbaum AL (1996) Posterior supe- MD, pp 149 rior oblique tenectomy at the scleral insertion for collapse 32. Sargent RA (1979) Dissociated hypertropia: surgical treat- of A pattern strabismus. J Pediatr Ophthalmol Strabismus ment. Ophtahlmology 86:1428 33:211 33. Schwartz T, Scott W (1991) Unilateral superior rectus 49. Kraft SP Long QB, Irving EL (2006) Dissociated hypotro- recession for the treatment of dissociated vertical devia- pia: clinical features and surgical management of two cases. tion. J Pediatr Ophtahlmol Strabismus 28:219 JAAPOS 10(5):389–393 34. Bothun ED, Summers CG (2004) Unilateral inferior 50. Greenberg MF, Pollard ZF (2001) A rare case of bilateral oblique anterior transposition for dissociated vertical devi- dissociated hypotropia and unilateral dissociated esotro- ation. JAAPOS 8(3):259–263 pia. JAAPOS 5(2):123–125 35. Goldchmit M, Felberg S, Souza-Dias C (2003) Unilateral 51. Kraft SP, Irving EL, Steinbach MJ, Levin AV (2000) A case anterior transposition of the inferior oblique muscle for of hypotropic dissociated vertical deviation: surgical man- correction of hypertropia in primary position. JAAPOS agement. In: Spiritus M (ed) Transactions of the 25th meet- 7(4):241–243 ing of the European strabismological association. Aeolus, 36. Arroyo Yllanes ME, Escanio Cortes ME, Perez Perez JF, Lisse, Th e Netherlands, pp 93–95 Murillo Murillo L (2007) Unilateral tucking of the inferior 52. Gamio S (2007) Hypotropia in patients with dissociated rectus muscles for dissociated vertical deviation. Cir Cir vertical deviation. Transactions of the 31th ESA meeting. 75(1):7–12 Mykonos, Greece, pp 337–340 37. Varn MM, Saunders RA, Wilson ME (1997) Combined 53. Brodsky MC, Gräf MH, Kommerell G (2005) Th e reversed bilateral superior rectus muscle recession and inferior fi xation test: a diagnostic test for dissociated horizontal oblique muscle weakening for dissociated vertical devia- deviation. Arch Ophthalmol 123(8):1083–1087 tion. J Am Assoc Pediatr Ophthalmol Strabismus 1:134 54. Brodsky MC, Fray KJ (2007) Dissociated horizontal devia- 38. Del Monte MP (1993) Atlas of pediatric ophthalmology tion aft er surgery for infantile esotropia: clinical character- and strabismus surgery. Churchill-Livingstone, New istics and proposed pathophysiologic mechanisms. Arch York, pp 9 Ophthalmol 125(12):1683–1692 184 13 Surgical Management of Dissociated Deviations

55. Apt L, Isenberg S (1977) Eye position of strabismus patients 60. Bechtel RT, Kushner BJ, Morton GV (1996) Th e relation- under general anesthesia. Am J Ophthalmol 84(4): ship between dissociated vertical divergence (DVD) and 574–579 head tilts. J Pediatr Ophtahlmol Strabismus 33:303 56. Wilson ME, Hutchinson AK, Saunders R (2000) Outcomes 61. Santiago AP, Rosenbaum AL (1998) Dissociated vertical 13 from surgical treatment for dissociated horizontal devia- deviation and head tilts. J Am Assoc Pediatr Ophtahlmol tion. J AAPOS 4(2):94–101 Strabismus 2:5 57. S. Gamio, MD (2008) Diagnosis and surgical treatment of 62. Brodsky MC, Jenkins R, Nucci P (2004) Unexplained head dissociated horizontal deviation (DHD). In: Transactions tilt following surgical treatment of congenital esotropia of the 32nd Meeting of the European Strabismological A postural manifestation of DVD. Br J Ophthalmol 88(2): Association. Edited Rosario Gomez de Liano. European 268–272. Erratum in: Br J Ophthalmol 2004 Apr;88 (4):599 Strabismological association. Depósito legal: M-14174- 63. Jampolsky A (1994) A new look at the head tilt test In: 2009. Madrid, Spain, 37 pp 113–115 Fuchs AF, Brandt TH, Buttner U, Zee DS (eds) Contemporary 58. Guyton DL (2004) Dissociated vertical deviation: an ocular motor and vestibular research A tribute to David A acquired nystagmus-blockage phenomenon. Am Orthoptic Robinson. Springer, Sttuttgart, pp 432–439 Journal 54:77–87 64. De Decker W, Conrad HG (1975) Fadenoperation nach 59. Guyton DL (2008) Ocular torsion reveals the mechanisms Cuppers bei komplizierten Augenmuskelstorungen und of cyclovertical strabismus: the Weisenfeld lecture. Invest nichtakkommodativem Konvergenzexzess. Klin Monatsbl Ophthalmol Vis Sci 49(3):847–857; 846 Augenheilkd 167:217 Chapter 14 Surgical Implications of the Superior Oblique Frenulum 14 Burton J. Kushner and Megumi Iizuka

Core Messages ■ Th e superior oblique (SO) tendon is attached to ■ An intact frenulum can result in the SO tendon the undersurface of the superior rectus muscle by scarring into the superior rectus insertion when an areolar frenulum. the latter is resected. ■ Th e frenulum, if left intact, causes the SO tendon to ■ Th e posterior SO tenectomy procedure is eﬀ ec- move posteriorly with the superior rectus muscle tive in collapsing small A patterns but oft en does when it is recessed. Th is can prevent the SO from not eliminate overdepression in adduction. Th is becoming scarred into the superior rectus insertion apparent contradiction can be explained by the when the latter is recessed. It can, however, prevent change in SO vector force that results from cut- the superior rectus muscle from taking up slack ting the frenulum, which is unavoidable with this when recessed with a suspension technique. surgical procedure.

potential complication of the SO tendon becoming 14.1 Introduction scarred into the insertion of the superior rectus muscle. Th e superior oblique (SO) muscle is adherent to the Recently, studies have suggested that scarring of the SO undersurface of the superior rectus muscle by an areolar muscle in this way can produce a complication referred to connective tissue. Jampolsky was the fi rst to describe the as the SO tendon incarceration syndrome [2]. Th is syn- surgical signifi cance of this local adherence, which he drome is a restrictive strabismus characterized by a referred to as a frenulum [1]. Th e term frenulum can be hypertropia with incyclotropia of the aff ected eye that is defi ned as a membranous fold of skin that supports or associated with scarring of the SO tendon to the nasal restricts the movement of an organ, such as the small corner of the insertion of the superior rectus muscle. It is band of tissue connecting the tongue to the fl oor of the a very diffi cult surgical problem to correct and hence mouth. Jampolsky stated that when the frenulum is left should be avoided if possible. intact, the SO tendon moves with the superior rectus Th e frenulum may also be an important structure to muscle. Hence, when the superior rectus muscle is consider during SO surgery as well. Prieto Diaz advo- recessed, the SO tendon will not only retract with it but cated cutting the frenulum to obtain maximal weakening may also constrain the posterior movement of the muscle of the SO muscle by the temporal approach [3, 4]. On the if the superior rectus is recessed, using an adjustable other hand, excessive stripping of the frenulum may also suture or suspension (A.K.A. hang-back) technique. He be an additional cause of SO tendon incarceration syn- found that if the frenulum is left intact, the SO tendon via drome when weakening procedures are performed on the the frenulum will prevent the superior rectus muscle SO tendon [2, 5]. from achieving a recession of greater than 10 mm. Several authors, cited above, have alluded to the Th erefore, Jampolsky recommended severing the frenu- importance of the proper handling of the frenulum for lum if a recession of greater than 10 mm of the superior both superior rectus surgery and SO surgery. Th eir state- rectus is desired to obtain the desired amount of reces- ments appear logical, but it is only recently that the eff ect sion. He also recommended cutting the frenulum during of severing the frenulum on the position of both the SO superior rectus resections, so as to avoid pulling the SO tendon and superior rectus muscle at surgery has been tendon forward with the resection, resulting in the quantifi ed [2]. In addition, it has been observed that the 186 14 Surgical Implications of the Superior Oblique Frenulum

posterior partial tenectomy procedure on the SO tendon is eff ective in collapsing of A patterns that measure less than 20 PD (prism diopters); however, it is less eff ective in decreasing the overdepression in adduction [3, 5]. Th is 14 residual overdepression in adduction has been described as pseudo-SO overaction (pseudo-SOOA) [3, 5]. It appears that the inevitable severing of the SO frenulum that occurs with this surgical procedure can explain the persistence of the overdepression in adduction in spite of its eff ectiveness in collapsing the pattern, as described in Sect. 10.2.3 of this chapter.

14.2 Clinical and Theoretical Investigations Fig. 14.1 Photograph of right eye at surgery as seen from below. A series of clinical in vivo investigations of the eff ect of Th e needle of a 6–0 Polyglactin 910 suture is being passed diff erent methods of handling the SO tendon frenulum, through anterior aspect of the superior oblique (SO) tendon as well as some theoretical calculations made from scale midway between the nasal and temporal edge of the superior rectus muscle with the superior rectus muscle disinserted and modeling shed important light on how the SO frenulum refl ected upward. Th e small arrow denotes the SO tendon; the should be handled when surgery is performed on the SO large arrow denotes the refl ected superior rectus muscle. tendon or superior rectus muscle. (Reprinted from [6] Elsevier Press)

reference knot distance. Th e superior rectus muscle was then suspended 6, 8, 10, 12, and 14 mm for a total of three 14.2.1 The Eff ect of Superior Rectus Muscle recessions at each distance in a randomly generated order Recession on the Location of the to avoid any infl uence of tissue hysteresis or tissue mem- Superior Oblique Tendon Before ory. Th e temporary suspension of the muscle was accom- and After Cutting the Frenulum plished by grasping the sutures in the superior rectus with Th is experiment consisted of measuring the posterior dis- forceps at the desired distance from the superior rectus placement of the SO tendon with recession of the superior and then holding this point on the sutures at the superior rectus muscle before and aft er cutting the SO frenulum in rectus insertion. Th e eye was then rotated to the primary three patients (2, 8, and 25 years of age) who were under- position and the conjunctiva was lift ed to verify if the going enucleation for unrelated reasons [6]. At the time of muscle had completely taken up the slack in the suspen- surgery but before the globe was enucleated, the position sion suture. If the slack had not been spontaneously taken of the SO tendon was measured before and aft er cutting up for the desired amount of recession, the superior rec- the frenulum in the eye undergoing enucleation while tus muscle was reposited with instruments and the occur- suspending the superior rectus muscle at various dis- rence thereof noted. Th e eff ect of the superior rectus tances. Th is was performed as follows: Th e superior rectus suspension on the position of the SO tendon was recorded muscle was isolated on a muscle hook, imbricated with using calipers to measure the distance from the reference two double-armed 6–0 Polyglactin 910 sutures, the check knot to the insertion of the superior rectus muscle. Th is ligaments were cut in the usual manner, and the superior was referred to as the second reference knot distance. A rectus muscle was disinserted. Th e underlying SO tendon masked assistant (resident, fellow, or scrub nurse) then insertion was identifi ed without cutting the frenulum. A read the caliper distance using a straight ruler to the near- single-armed 6–0 Polyglactin 910 suture was sewn into est 0.5 mm. By subtracting the second reference knot dis- the anterior aspect of the SO tendon midway between the tance from the initial reference knot distance, the amount nasal and temporal edge of the superior rectus muscle and of posterior movement of the SO tendon was calculated knotted in place (Fig. 14.1). A reference knot was tied in for each successive suspension of the superior rectus mus- this suture approximately 15–20 mm from the knot placed cle (Fig. 14.2). in the SO tendon. Next, with the superior rectus held at Th e SO frenulum was then completely severed under the original insertion, the distance between the reference direct visualization by elevating the superior rectus knot and the superior rectus muscle insertion was muscle and lysing the connection between it and the recorded. Th is distance was referred to as the initial underlying SO tendon using sharp and blunt dissection. 14.2 Clinical and Theoretical Investigations 187

Fig. 14.2 Axial view of the left eye as viewed from superiorly in the orbit illustrating location of SO tendon before cutting the frenulum while suspending the superior rectus muscle at various distances. Superior rectus suspended at (a) Original insertion, (b) 6 mm, (c) 14 mm. (Reprinted from [6] Elsevier Press)

All the above measurements were repeated, again with muscle can be recessed using a suspension. It appears, three measurements for each superior rectus suspension however, that this should result in a substantial alteration distance performed in a randomly determined sequence. of the force of the SO muscle. Yet clinically, we do not Th ere was essentially a one-to-one correlation between observe such a profound change in the SO muscle func- the amount of superior rectus recession and posterior tion. One explanation may be that the frenulum allows movement of the SO tendon for superior rectus reces- some movement of the SO tendon relative to the superior sions up to 10 mm. Aft er severing the frenulum, there was rectus muscle during active contraction. Our studies were negligible movement of the SO tendon reaching a maxi- all done with the patients anesthetized and consequently mum of only 1.7 mm in only one patient for a superior did not address that possibility. rectus recession of 14 mm. Aft er cutting the frenulum, the SO muscle moved For superior rectus recessions between 10 and 14 mm, minimally when the superior rectus muscle was recessed. the suspended superior rectus typically would not take up Because the anterior border of the SO tendon is approx- the slack to achieve the desired amount of recession prior imately 8 mm posterior to the superior rectus when the to severing the frenulum without being manually repos- globe is rotated in the downward position, an 8 mm ited. Th is confi rms that the frenulum intimately links the recession of the superior rectus muscle would place its superior rectus muscle and the SO tendon. Th e fact that new insertion overlying the SO tendon if the frenulum the superior rectus muscle did not consistently take up is severed. Th e SO insertion is broad and underlies a the slack for large suspension recessions (10–14 mm) relatively large area beneath the superior rectus muscle. with the frenulum intact, but did so more oft en when the Consequently, cutting the frenulum may result in diffi - frenulum was severed, is probably due to a constraining culty with suturing the superior rectus to the sclera eff ect of the frenulum. Th e frenulum is attached to the SO without incorporating some of the SO insertion whose tendon, which in turn has limited amount of slack to diaphanous nature can make it diffi cult to visualize. We allow the tendon to continue to move freely posteriorly. therefore agree with Jampolsky’s recommendations to Hence, at these large recession values, the frenulum may preserve the frenulum for superior rectus recessions prevent adequate weakening unless the superior rectus that are 10 mm or less to insure that the SO tendon will muscle is sutured in place. We therefore advocate cutting move posteriorly with the recessed superior rectus mus- the frenulum for superior rectus muscle recessions that cle and not get scarred into the new superior rectus are larger than 10 mm, especially when using a suspen- insertion [1, 7]. Furthermore, for recessions greater than sion technique. 10 mm we advocate lysing this areolar connection owing In theory, when the frenulum is intact the orientation to its constraining eff ect [6]. of the SO tendon would bow backwards as illustrated in Although we did not study superior rectus resections Fig. 14.2c when a large recession of the superior rectus [6], we speculate that with the frenulum intact, the SO muscle is performed. Th is graphically illustrates why an tendon would be pulled anteriorly with the superior rec- intact frenulum will limit the amount the superior rectus tus muscle as previously stated by Jampolsky, and the SO 188 14 Surgical Implications of the Superior Oblique Frenulum

tendon may therefore be at risk of being sutured into was measured and recorded in the aforementioned the insertion site of the superior rectus muscle [1, 7]. masked manner. Th is was recorded as the initial reference Consequently, for superior rectus resections, we also knot distance. Th e SO tendon was then disinserted, and advocate separating the frenulum. two successive forced ductions to rotate the eye maxi- 14 mally up and in were performed. With the eye returned to the primary position, the distance between the initial reference knot and the temporal superior rectus edge was 14.2.2 The Eff ect of the Frenulum remeasured with calipers to give the second reference on Superior Oblique Recession knot distance. Th e masked assistant then read the caliper Using a Suspension Technique distance using a straight ruler to the nearest 0.5 mm. Th e Th is experiment consisted of assessing how far the SO amount of recession of the SO tendon was calculated to tendon retracted (recessed) aft er disinsertion to simulate the nearest 0.5 mm by subtracting the second reference what happens with either a recession with a suspension knot distance from the initial reference knot distance. technique or a free disinsertion. Th is was done both Th is was repeated for three sets of measurements. before and aft er separating the frenulum in a second Traction was then placed on the SO tendon, to pull it series of four patients (ages 8, 17, 22, and 47 years) who approximately 12–14 mm out from under the superior rec- were undergoing bilateral SO recession using a suspen- tus muscle temporally (Fig. 14.4). Th is movement essen- sion technique. Th e position of the SO was measured tially brought all of the tendon that is normally under the before and aft er cutting the frenulum in the following superior rectus muscle out temporal to it, and eff ectively manner: Th e SO tendon’s insertion was isolated through severed the frenulum connection. Th is maneuver is similar a superotemporal incision aft er fi rst hooking the superior to what frequently occurs if one just exerts substantial trac- rectus muscle. Th e SO tendon was hooked at its insertion tion on the SO tendon when weakening it at the insertion with care to avoid pulling the tendon from under the or during a SO tendon tucking procedure. Two forced duc- superior rectus muscle, thus preserving the frenulum. tions were again performed to rotate the eye up and in. Th e Th is was done by refl ecting the superior rectus nasally as distance between the knot and the superior rectus edge was minimally as possible but suffi cient to allow for visualiza- measured with calipers in the same manner as when the tion of the insertion of the SO tendon. A 6–0 Polyglactin frenulum was intact. Again, using simple subtraction, 910 suture was woven through the tendon near the inser- the amount of recession of the SO tendon aft er the frenu- tion and knotted (Fig. 14.3). A reference knot was tied in lum was stripped was calculated using our masked mea- the suture 15–20 mm from the distal end of the SO ten- surement technique for three successive measurements. don and the superior rectus muscle was set back in its To control the possibility that the amount of recession unrefl ected position. Th e distance from the reference simply increased with the multiple forced ductions that were knot to the temporal edge of the superior rectus muscle needed to obtain multiple measurements, a single set of

Fig. 14.3 Axial view of the right eye viewed from superiorly in the orbit illustrating movement of the SO tendon. (a) A 6–0 Polyglactin 910 suture woven through the insertion, just aft er hooking the SO tendon. Th e frenulum is intact. (b) Th e SO tendon disinserted with the frenulum intact. A relatively small amount of recession occurs. (c) Aft er stripping the frenulum a much larger amount of recession of the SO tendon occurs than prior to stripping the frenulum. (Reprinted from [6] Elsevier Press) 14.2 Clinical and Theoretical Investigations 189

achieved by cutting the frenulum [4]. It also suggests that asymmetric eﬀ ects may occur with bilateral SO recession using a suspension technique, if there is asymmetric stripping of the frenulum. On the other hand, stripping the frenulum may allow the disinserted SO tendon to migrate forward resulting in the SO tendon incarceration syndrome [2]. Th us how the frenulum is handled with these procedures may be a matter of tradeoﬀ s.

14.2.3 The Theoretical Eff ect of the Superior Oblique Frenulum on the Posterior Partial Tenectomy of the Superior Oblique Th e threefold function of the SO muscle includes intorsion, depression, and to a lesser degree, abduction. Th ese actions are uniquely related to its anatomy and the angle Fig. 14.4 Surgical photograph of the right eye rotated down- the tendon makes with the anterior–posterior axis. Th e ward as viewed from below; superior muscles are at the top in SO tendon makes an angle of approximately 54° with the the photograph. Traction is placed on the SO tendon pulling it anterior–posterior axis. Th e anterior fi bers of the SO ten- 12–14 mm out from under the superior rectus muscle tempo- don make a relatively large angle with the anterior–poste- rally to eff ectively sever the frenulum. Small arrow denotes SO tendon; large arrow denotes suture tied to the cut end of the SO rior axis and therefore are thought to primarily have a tendon. (Reprinted from [6] Elsevier Press) torsional action, and only a small vertical action. Prieto Diaz calculated the relative vertical and torsional actions of the anterior and posterior fi bers of the SO tendon using measurements was taken prior to and aft er stripping the computer-aided design soft ware and determined that the frenulum on the other (control) eye in the same manner as vertical action is approximately 1/3 of the torsional action in the fi rst (study) eye. In two patients, the study procedure [8]. Th e posterior fi bers of the SO tendon make a smaller was performed in the right eye fi rst, and in the other two angle with the anterior–posterior axis than the anterior patients, the study procedure was performed in left eye fi rst. fi bers. He concluded they therefore contribute approxi- Th e mean distance that the SO tendon recessed was mately 50% less torsion than the anterior fi bers but twice 2.4 ± 0.4 mm before cutting the frenulum and 8.5 as much vertical action. ± 0.7 mm aft er cutting the frenulum. Th ere was a statisti- These anatomic considerations of the differential cally signifi cant diff erence between the two measure- effects of the anterior and posterior fibers of the SO ments (P = 0.0011, paired two-tailed student’s t-test). tendon have given rise to different surgical procedures Th e same procedure was followed in the fellow control depending on whether one wants more torsion vs. eye for one set of measurement. For the control eyes the vertical correction. For example, the Harada–Ito mean recession prior to stripping the frenulum was 2.4 operation tightens the anterior fibers and primarily ± 0.3 mm and aft er stripping the frenulum was 8.0 provides torsional changes [9]. Conversely, the poste- ± 0.8 mm (P = 0.0004, paired two-tailed student’s t-test). rior partial tenectomy primarily weakens the more Th ese values for the amount of recession obtained in the posterior fibers of the SO tendon and thus gives more control eyes before and aft er stripping the frenulum vertical correction with minimal change in torsion. were essentially identical to the values for the study eyes, Prieto–Diaz first described this procedure, which despite the control eyes only having a single measure- consists of cutting the posterior 4/5 or 7/8 of the SO ment. Th is confi rms that taking multiple measurements tendon at its insertion and then excising a posterior prior to stripping the frenulum was not a confounding triangle of tendon extending about 8–12 mm toward factor on the amount that the SO moved aft er stripping the trochlea [10, 11]. He proposed this operation to the frenulum. surgically treat A-patterns without affecting torsion. It Th e results of this experiment are consistent with the has been reported to be effective in collapsing A pat- observation that the maximal eff ect of a recession of terns of up to 20 PD; however, it is not effective in the SO tendon using a suspension technique can only be decreasing the overdepression in adduction resulting 190 14 Surgical Implications of the Superior Oblique Frenulum

Fig. 14.5 Th is patient underwent bilateral posterior tenectomy of the SO tendon combined with bilateral 5 mm lateral rectus muscle recessions to treat an exotropia associated with 18PD of A pattern. Before surgery he had +2 bilateral SO overaction. Th e surgery not only eliminated the A pattern but overcorrected it resulting in a small V pattern, yet his SO overaction persisted

in a pseudo-SOOA [3, 5] (Fig. 14.5). Why this proce- have kept the distance between the anterior edge of the dure fails to address the overdepression in adduction SO tendon and the SR insertion the same, implying that has not been adequately explained. We feel that some the constraining property of the frenulum completely unique considerations about the SO frenulum as well prevents the SO tendon from slipping anteriorly. In this as some anatomic considerations of the SO tendon scenario, the original angle made by the anterior fi bers of explain why the posterior partial tenectomy operation the SO tendon and the anterior–posterior axis is approxi- does not eliminate the overdepression in adduction. mately the same. As seen in Fig. 14.6b, the anterior fi bers To study this, we used scale fi gures of the anatomy of of the SO tendon still make an angle of 75° with the ante- the SO and SR obtained from Orbit™1.8 (Eidactics, San rior–posterior axis. Consequently, in the normal nonop- Francisco, CA) to determine the angles made by the ante- erated eye, the contribution of the SO forces of intorsion, rior and posterior fi bers of the SO tendon with the ante- abduction, and depression remain relatively unchanged rior–posterior axis when the eye was in the primary in adduction compared with the primary position. position, as well as in adduction. We then modifi ed those Figure 14.6c illustrates the situation aft er a posterior fi gures to assume that the frenulum constrained the SO partial tenectomy procedure. Th e excised portion of the tendon to the SR muscle and recalculated the same angles. posterior four fi ft hs of the SO tendon insertion is out- Th e contribution of the net force directed parallel to the lined in black. Th is surgical procedure necessitates that anterior–posterior axis represents the force that creates the frenulum be excised, which allows the SO tendon to depression, and the contribution of the net force directed move forward. Th is substantially decreases the angle perpendicular to the anterior–posterior axis represents between the anterior fi bers of the SO tendon and the the torsional force. Th e percentage of original SO force anterior–posterior axis. In Fig. 14.6c, we measured this that is directed vertically and torsionally is the cosine and angle to be approximately 40°. In this position, the sine of the angle made by the SO tendon and the ante- depressor action of the SO tendon is increased compared rior–posterior axis, respectively, multiplied by 100. with that found in the unoperated state. Th e magnitude Figure 14.6a shows the eye in primary position. Th e of depression is the sine of 40° or 77% of the total net anterior fi bers of the SO tendon make an angle of 75° with force as compared with only 26% prior to the surgical the anterior–posterior axis. Th us, the torsional force vec- procedure. Th is may be one explanation why overdepres- tor of these fi bers is the sine of 75°, or 0.97 times the mag- sion in adduction persist aft er posterior partial tenec- nitude of the net force. Or in other words, the torsional tomy. Th is residual abnormality of versions may be due force vector equals 97% of the net force. Similarly, the to the unavoidable excision of the SO frenulum, which vertical force vector is the cosine of 75° multiplied by 100, occurs with this surgical procedure, and the eff ect this or 26% of the net force. has on the subsequent distribution of vertical force of the When the eye is adducted 35°, and if one assumes the SO tendon. Persistent overdepression in adduction has frenulum constrains the tendon to the SR muscle, the ten- been reported as occurring in 40.4% [12]–57% [5] of don will bow backwards as shown in Fig. 14.6b. In this patients aft er posterior partial SO tenectomy. Despite this picture, which is modifi ed from the Orbit™1.8 model, we unwanted overdepression in adduction, weakening of the 14.2 Clinical and Theoretical Investigations 191

Fig. 14.6 Th ree-dimensional scale fi gure of the anatomy of the SO modifi ed from Orbit™1.8 program seen from above. (a) Representation of an unoperated eye in the primary position. Th e anterior fi bers of the SO tendon make an angle of 75° with the anterior–posterior axis. Th e magnitude of the force vector for depression of the SO tendon is 26% of the total net force. (b) Representation of an unoperated eye in adduction. Th is is modifi ed from Orbit™1.8 to assume the frenulum completely constrains the tendon to the SR muscle. Th e original angle made by the anterior fi bers of the SO tendon and the anterior–posterior axis is preserved measuring 75°. Th e magnitude of the force vector for depression of the SO tendon remains unchanged at 26% C) Representation of the eye in adduction following posterior partial tenectomy procedure of the SO tendon. Th e absence of the constraining eff ect of the frenulum allows the SO tendon to slide forward. Th is decreases the angle between the anterior fi bers of the SO tendon and the anterior–posterior axis to 40°. Th e magnitude of the force vector for depressor of the SO tendon increases to 77% of the total net force

SO with posterior partial tenectomy eff ectively reduces Th is results in a pseudo-SOOA in the ipsilateral eye by the exo-shift in down gaze and thus reduces the A pattern Herring’s law [5, 8]. Th ere are several theories as to the [5, 10–12]. Th is may be due to the ability of the adducting cause of this limitation. For example, anteriorization of power of the inferior rectus muscle to prevail over any the SO tendon insertion to a preequatorial location aft er residual abducting power of the weakened SO in the a posterior partial tenectomy has been theorized. Using adducted and depressed position (unpublished written the Orbit™ 1.8 model, Castanera simulated that an ante- personal communication from A. Castanera de Molina, rior shift of the muscle insertion centroid of 4.45 mm July 18, 2007). However, overdepression occurs even aft er a posterior partial tenectomy would cause a reduc- when the A-pattern is eff ectively collapsed, suggesting tion in the vertical force of the SO tendon [13]. He also that this motility pattern is not simply due to a surgical modeled the situation in which the cut end of the SO undercorrection. Castanera considers this common post- tendon could inadvertently be reattached to the sclera, operative complication of downshoot in adduction to be thus simulating a recession plus resection procedure. a direct consequence of the surgery itself (unpublished Both simulations show a similar change in the vertical written personal communication from A Castanera de force component such that the SO tendon becomes an Molina, July 18, 2007). Th is would be consistent with our elevator in abduction with no change of depression in hypothesis that excision of the frenulum can result in for- adduction. Another cause of the limitation to depres- ward slippage of the remaining fi bers of the SO when the sion in abduction of the contralateral eye may due to eye is adducted, thus increasing their vertical force. iatrogenic incarceration of the SO tendon to the SR Some investigators have speculated that the down- insertion [2, 5, 13]. Th is complication also places the shoot in adduction seen aft er partial posterior SO tenec- eff ective insertion of the SO tendon to a preequatorial tomy occurs secondary to a limitation of depression in position. One further mechanism could be the presence abduction of the contralateral eye aft er bilateral surgery. of underlying occult SR contracture [7]. We feel that 192 14 Surgical Implications of the Superior Oblique Frenulum

contralateral restriction of depression in abduction can- Summary for Clinicians not fully account for the persistence of overdepression in adduction aft er partial posterior SO tenectomy, ■ Th e SO frenulum is an important structure. How because we have seen this occur in the operated eye it is handled with superior rectus and SO surgery 14 aft er unilateral surgery. Also, we have observed that this may aff ect the surgical outcome. fi nding is oft en present immediately aft er surgery. Th is ■ Th e frenulum should be severed for superior would tend to rule out postoperative iatrogenic mechan- rectus recessions that exceed 10 mm, to allow for ical restriction in the contralateral eye as the cause. We the desired recession eff ect. do recognize, however, that since most SO weakening ■ Th e frenulum should be severed for all superior procedures are bilateral, both residual overdepression rectus resections to prevent the SO tendon incar- in adduction of the ipsilateral eye and limitation to ceration syndrome. depression in abduction of the contralateral eye could ■ Th e frenulum should be left intact for superior occur. Furthermore, these two conditions would be rectus recessions that are less than 10 mm to pre- additive with respect to their eff ect on versions in vent the SO tendon incarceration syndrome. adduction. ■ With SO recessions using a suspension technique We considered the anatomical eff ects of the SO the handling of the frenulum is a matter of trade- frenulum on the vertical and torsional force vectors of off s. Severing the frenulum will involve a greater the SO tendon using basic two-dimensional trigonom- amount of recession, but may predispose to the etry. We recognize that there are some obvious oversim- SO tendon incarceration syndrome. Leaving the plifi cations in our theoretical analysis. Th e geometric frenulum intact will prevent that restrictive stra- angles drawn on the scaled model are somewhat arbi- bismic syndrome but will limit the amount of trary. For example, our modeling of the anterior fi bers recession obtained. Asymmetric handling of the of the unoperated SO tendon when the eye is adducted frenulum with bilateral SO recession may predis- (see again Fig. 14.6b) assumes that the frenulum com- pose to an asymmetric response. pletely constrains the tendon. In reality, there is proba- ■ Th e posterior tenectomy operation of the SO is bly some elasticity of the frenulum that allows at least eff ective in collapsing up to 20 PD of A pattern but some forward slippage [6]. We assume this to be the case is less eff ective in eliminating the overdepression as common clinical observations confi rm that the SO in adduction. has a greater vertical and lesser torsional action in adduction than in the primary position. Nevertheless, prior investigation on the constraining eff ect of the SO References tendon frenulum suggests that our model is at least qualitatively sound, even if it is not exactly quantita- 1. Jampolsky A (1981) Superior rectus revisited. Tr Am tively accurate [6, 7]. In addition, we reduced a complex Ophth Soc 79:233 three-dimensional situation into a two-dimensional 2. Kushner BJ (2007) Superior oblique tendon incarceration construct, and the abducting contribution of the SO syndrome. Arch Ophthalmol 125:1070–1076 tendon was ignored. We feel, however, that this would 3. Prieto-Diaz J (1988) Management of superior oblique have minimal impact on our conclusions, as the abduct- overaction in A-pattern deviations. Graefes Arch Clin Exp ing force of the SO muscle is relatively small. Th us, Ophthalmol 226:126–131 although the actual numbers we calculated are approxi- 4. Prieto-Diaz J (1989) Superior oblique overaction. Int mate, our qualitative analysis confi rms what seems logi- Ophthalmol Clin 29:43–50 cal. Specifi cally, if we assume that the SO tendon is 5. Castanera de Molina A, Fabiani R, Giner MG (1998) constrained by the frenulum in the primary and Downshoot in infra-adduction following selected superior adducted fi elds of gaze, cutting the frenulum aft er a pro- oblique surgical weakening procedures for A-pattern stra- cedure such as a partial posterior tenectomy would col- bismus. Binocul Vis Strabismus Q 13:17–28 lapse the angle the anterior fi bers make with the 6. Iizuka M, Kushner B (2008) Surgical implications of the anterior–posterior axis. Th is reduction in the angle superior oblique frenulum. J AAPOS 12:27–32 makes the SO tendon a more eff ective depressor in the 7. Jampolsky A (1986) Management of vertical strabismus. adducted position. Th is may be an explanation for the Symposium on pediatric ophthalmology: transactions of residual overdepression in adduction in the ipsilateral the new Orleans acad ophthalmol. Raven, New York, pp eye aft er posterior partial tenectomy of the SO tendon. 141–171 References 193

8. Prieto-Diaz J (1996) Selective and moderated weakening 12. Shin GS, Elliott RL, Rosenbaum AL (1996) Posterior supe- of the superior oblique muscle. Memorias del IV Congresso rior oblique tenectomy at the scleral insertion for collapse del Consejo Latinoamericano de Estrabismus. Mayo, of A-pattern strabismus. J Pediatr Ophthalmol Strabismus Buenos Aires, pp. 535–541 33:211–218 9. Harada M, Ito Y (1964) Surgical correction of cyclotropia. 13. Castanera de Molina A, ML GM (1997) Persistent SO Jap J Ophthalmol 8:88–96 “overaction” aft er surgical treatment of A-pattern anisot- 10. Prieto-Diaz J (1976) Tenectomia parcial posterior del obli- ropies. In: M. Spiritus (ed) Transactions 24th meeting cuo superior. Arch Oft almol B Aires 51:267–271 European strabismological association; Vilamoura, 11. Prieto-Diaz J (1979) Poseterior partial tenectomy of the Portugal. Aeolus, Buren, Th e Netherlands SO. J Pediatr Ophthalmol Strabismus 16:321–323 Chapter 15 Pearls and Pitfalls in Surgical Management of Paralytic Strabismus 15 Seyhan B. Özkan

Core Messages ■ Careful preoperative assessment and a correct ■ The major pitfall in paralytic strabismus is the diagnosis of the problem are the essential factors coexistence of a restrictive element. The sec- for a successful outcome of surgical treatment. ondary restrictions may mask the partial func- ■ Th e pearl to go through the correct route in surgi- tional recovery in a paretic extraocular muscle cal management of paralytic strabismus is to (EOM), and sometimes they may become a know the questions that need to be answered dur- more prominent problem than the paralytic ing the preoperative assessment. Th e correct condition itself. answers for these questions clarify the method of ■ Th e restoration of ocular alignment should be appropriate surgical treatment. planned to create a new balance in both eyes. ■ During the preoperative assessment, the potential Paralytic strabismus is a binocular problem even for fusion must be carefully evaluated. Acquired in cases with unilateral involvement. Th ere should loss of fusion or, in other words, central fusion be no hesitation to operate the sound eye where disruption may coexist in acquired paralytic ocu- necessary. lar motility problems. In such cases, restoration ■ Th e methods of surgical treatment primarily aim of the ocular alignment may make the symptoms to weaken the unopposed overaction of the worse because of the increased awareness of antagonist, then to strengthen the paretic muscle diplopia with two overlapping images. where possible or to create a mechanical eff ect by ■ Th e aims of surgical treatment are primarily to transposition, and fi nally to weaken the yoke obtain a diplopia-free fi eld, to achieve symmetric muscle in the sound eye. In certain cases like ocular motility and a good looking eye that will complete third nerve palsy, creating a restriction allow eye contact, and to correct the abnormal with surgery may be required to keep the eye in head posture, if any. primary position.

15.1 General Principles of Surgical 15.1.1 Aims of Treatment Treatment in Paralytic Strabismus Th e major aims of treatment are enlargement of diplopia- Paralytic strabismus is one of the most challenging areas free ﬁ eld, restoration of ocular alignment, and restoration of in strabismus practice. In other types of strabismus, the the appearance of the patient, to correct abnormal head pos- ophthalmic surgeon considers to operate six muscles for ture, and to improve the ductions. Th e last one is the concern each eye to restore the ocular alignment. However, in of the strabismus surgeon, and the patients usually do not paralytic strabismus, the ocular alignment needs to be complain of limited ductions and are mostly not even aware restored with limited number of muscles, sometimes of the limitation of their ductions if it is not very severe. even with only one functioning extraocular muscle (EOM). In this chapter, the general principles of surgi- 15.1.2 Timing of Surgery cal treatment will be reviewed ﬁ rst and then the treatment strategies in third, fourth, and sixth cranial nerves In all types of paralytic strabismus, the stability of the devi- will be evaluated. ation must be observed before considering any surgical 196 15 Pearls and Pitfalls in Surgical Management of Paralytic Strabismus

intervention. Th e time period that the spontaneous recovery ■ Electromyography (EMG) occurs is usually accepted as 6 months; however, this period ■ Increase of intraocular pressure with positions of may last longer, especially in third nerve palsies. A waiting gaze period of 12 months is recommended for third nerve palsies ■ Measurement of saccadic eye movements 15 and spontaneous recovery may occur even in a longer period ■ Botulinum toxin A (BTXA) injection into the antago- of time in some cases [1]. As a general rule, one must con- nist EOM sider that if the deviation is still unstable following consecutive examinations aft er 6 months, surgical treatment must be Among those methods, the saccadic eye movement postponed till the deviation becomes stable. recordings provide very reliable information. However, in most of the clinics, saccadic eye movement recording is not available as a routine clinical method. 15.1.3 Preoperative Assessment BTXA may also be used as a diagnostic tool in paralytic strabismus [2]. Th e secondary unopposed contrac- Prior to any treatment, one must be sure about the diag- ture of the antagonist EOM may not allow the eye to move nosis. Restrictive motility problems may simulate para- toward the direction of the aff ected muscle despite some lytic conditions and sometimes both restrictive and spontaneous recovery. For diagnostic purpose, BTXA is paralytic problems occur at the same time making the injected into the antagonist EOM. An improvement of the clinical picture more complicated. Th e combination of movement toward the functional area of the paretic mus- restrictive and paralytic problems mostly occurs in orbital cle indicates that there is some residual function of the blow-out fractures and in long-standing paralytic prob- paretic muscle [3] (Figs. 15.1 and 15.2). However, it must lems. Th e combination of restrictive element has a nega- be kept in mind that in presence of severe contracture tive eff ect on the predictability of surgical results, so the with fi brosis BTXA injection does not give reliable results, presence of any restrictive factors must be carefully evalu- as BTXA cannot eliminate the fi brotic tissue eff ect. ated in all cases with paralytic strabismus. Despite the numerous methods for preoperative For a correct surgical planning, the following ques- assessment of the restrictive forces, the surgeon may have tions need to be answered preoperatively in cases with to change the surgical plan depending upon the traction paralytic strabismus: test results under general anesthesia. In long-standing 1. Is the problem partial (paresis) or total (paralysis)? paralytic strabismus, the contracture and fi brosis may not 2. Are there any restrictive factors? only aff ect the EOMs but also the fascial structures and 3. Is the problem congenital or acquired? EOM pulleys and an orbital fi brosis develops [4, 5]. In 4. Is there “acquired loss of fusion” or in other words such cases, the traction test will be found positive despite “central fusion disruption?” the disinsertion of the EOM. Th ese cases represent the most challenging paralytic ocular motility problems. Th e answers for the fi rst two questions will be discussed ■ together. Is the problem congenital or acquired?

■ Is the paralytic problem partial or total? In congenital paralytic disorders, there may be some ■ Are there any restrictive factors? developmental abnormalities like the tendon abnormalities in congenital superior oblique palsy, EOM fi brosis, or If there are no restrictive forces, it is not diffi cult to assess orbital fi brosis. Most of the congenital cases do not com- whether the paralytic condition is partial or total. Th ese plain of diplopia. Th e exception of this is decompensated factors may be primary as it is the case in blow-out frac- congenital fourth nerve palsy presenting with vertical ture or secondary as the contracture of the antagonist diplopia. muscle(s) in long-standing paralytic problems. For a correct evaluation of the role of accompanying ■ Is there “acquired loss of fusion (central fusion restrictive factors and the residual function of the paretic disruption)?” EOM, the following tests may be used: Acquired loss of fusion or central fusion disruption may ■ Measurement of the deviation in nine positions of occur in paralytic strabismus cases especially the post- gaze traumatic ones. In these cases, because of the involve- ■ Assessment of the ocular rotations ment of the fusional areas which are supposed to be ■ Traction test located in the midbrain, the previously healthy fusional ■ Active forced generation test ability is lost causing intractable diplopia. When the 15.1 General Principles of Surgical Treatment in Paralytic Strabismus 197

Fig. 15.1 Use of botulinum toxin A (BTXA) for assessment of the function of the paretic muscle. If the paretic muscle has some residual function the eye moves toward the functional area of the paretic extraocular muscle (EOM) following injection of BTXA into the antagonist muscle [3]

Paretic EOM Partially recovered Contracture of paretic EOM the antagonist Paralysis of the antagonist with BTXA a

Fig. 15.2 In a patient with left sixth nerve palsy (a) the improvement of abduction of the left eye aft er injection of BTXA into the medial rectus muscle is shown (b) [3]

deviation is neutralized by prisms or synoptophor, these two close images cannot be tolerated and cause more patients typically describes a vertical sliding of the images symptoms compared with the two far away images in a when the two images were overlapped and were just patient with a large deviation. about to appear single. Th e diagnosis of this challenging problem preoperatively is very important. If the patient has an acquired loss of fusion and intractable diplopia, 15.1.4 Methods of Surgical Treatment the deviation should better be corrected temporarily by prisms or BTXA to allow the assessment of the tolerance ■ Decreasing the strength of the antagonist: Recession of diplopia [2, 6]. In some cases during this period, the or disinsertion of the antagonist is the preferred fusional ability may be regained and in those ones sur- method. If it will be combined with full tendon trans- gery may be performed safely. Our preferred method is position, BTXA injection instead of surgical recession BTXA injection in such cases to provide a temporary should be preferred for the risk of anterior segment period of orthophoria under real-life conditions. Th e ischemia. decreased contrast sensitivity and the loss of image qual- ■ Strengthening the paretic EOM: Resection or tendon ity related to Fresnel prisms may have a negative eﬀ ect on tuck could be performed. For strengthening proce- recovery of fusion. If the patient cannot overcome or tol- dures, the paretic muscle is preferred to have some erate diplopia with the use of BTXA or prisms, surgical residual function. Th e exception of this is superior correction of the deviation may cause an increase of the oblique palsy. Because of the tendon length and the complaint of diplopia. Orthophoria in a patient with anatomical characteristics, superior oblique tendon intractable diplopia is much more bothersome compared tuck may be performed in a superior oblique muscle with the diplopia with a large deviation. Th e overlapping with no residual function. 198 15 Pearls and Pitfalls in Surgical Management of Paralytic Strabismus

■ Weakening the yoke muscle in the sound eye: recession enough to allow the passive adduction of the Recession or faden operation of the yoke muscle in the eye, orbital wall periost fi xation of the lateral rectus mus- unaff ected eye is the preferred method to increase the cle, and BTXA injection in residual deviations [7–9]. fi eld of binocular diplopia-free fi eld. Orbital wall periost fi xation is a recently described 15 method for the inactivation of lateral rectus muscle that Th ese are the general principles that the strabismus sur- we found useful in our clinical practice. Posterior Tenon geon needs to consider in all types of paralytic strabismus fi xation is proposed to be an alternative method to periost cases. Th e cranial nerve palsies will be evaluated individ- fi xation [10]. Th e potential reversibility of the procedure ually during the rest of the manuscript. is the advantage of both of these methods. Medial rectus resection: Although the resection of a paralytic muscle is not so eff ective, some authors prefer to perform a large resection to obtain a mechanical resis- 15.2 Third Nerve Palsy tance against abduction. In our experience, this eff ect Th ird nerve palsy may aff ect the third nerve in total, or the does not last long and we do not prefer to resect medial superior or inferior branches of the nerve as well as the rectus muscle. isolated EOM involvement. All these types of third nerve Superior oblique tendon transposition: Th e aims of palsy may present with a total or partial involvement, and superior oblique tendon transposition is to correct the they represent a wide range of ocular motility problems. hypotropia, making the superior oblique an adductor, Th e involvement of the inferior branch of the third nerve creating a mechanical barrier against abduction, and thus aff ects medial rectus, inferior rectus, and inferior oblique preventing the recurrence of the exodeviation. Superior muscles, whereas the superior branch aff ects the superior oblique tendon transposition may work if and only if the rectus and levator palpebrae superioris muscle. superior oblique muscle has some function. Especially, in long-standing ones, it may be diffi cult to assess the function of the superior oblique muscle while the eye is fi x- ated in an abducted position. In such patients with no 15.2.1 Complete Third Nerve Palsy apparent hypotropia or intorsion in ocular motility exam- In complete third nerve palsy, the major problem is the ination, slit lamp observation may be very helpful. Any unopposed contracture of the antagonist lateral rectus attempt of intorsion of the eye can easily be observed muscle. Th ere is a small hypotropia with a large angle under slit lamp. Superior oblique tendon transposition exodeviation and ptosis due to the involvement of levator may be performed by trochlear luxation and superior palpebrae superioris muscle. If the pupillary fi bers are oblique tendon resection or with Scott’s method by cut- aff ected, a mydriatic pupilla will be observed. In congeni- ting the superior oblique tendon via nasal approach and tal and long-standing cases, fi brosis of the intraorbital suturing the tendon 2 mm anterior and nasal to the supe- structures develops. Th e aims of treatment in complete rior rectus tendon without destroying the trochlea [7, 11]. third nerve palsy are to obtain an improvement of the Th e latter is our preferred method for superior oblique appearance of the patient, orthophoria in primary posi- tendon transposition, which is a less invasive one. tion, and a fi eld of binocular single vision in a very lim- Th e procedures to keep the eye in passive adduction: For ited area. Prior to any surgical intervention, the patient a permanent eff ect fascia lata, silicone band or superior must be informed about the goals of surgery and the pos- oblique tendon may be used to fi xate the globe to the sibility of a more bothersome diplopia with the decrease orbital periosteum [12, 13]. Traction sutures are used to of the proximity of the two images in primary position. keep the eye in passive adduction for a transient period to Th e surgical treatment modalities in complete third increase the eff ect of surgery [14, 15]. Th ese sutures are nerve palsy may be summarized as follows: kept in place for 6 weeks. Th is is our method of choice in total third nerve palsy [3] (Figs. 15.3–15.5). Th e other ■ Weakening of the lateral rectus muscle. methods are usually performed in secondary cases with a ■ Resection of the medial rectus muscle. failure of a previous operation. ■ Superior oblique tendon transposition. Th e major problems in total third nerve palsy are lat- ■ Th e procedures that keep the eye in passive eral rectus contracture that cannot be overcome by any adduction. methods, orbital fi brosis in long-standing cases, recurrence of exodeviation, and the more bothersome diplopia Weakening of the lateral rectus muscle: Th e methods following a successful surgery that provides orthophoria of weakening are supramaximal recession, hang back in a very limited area. 15.2 Third Nerve Palsy 199

Fig. 15.3 Preoperative right exo and hypotropia in a patient with right congenital third nerve palsy [3]

and in that case, the treatment should be modiﬁ ed depending upon the severity of the involvement of the EOM(s). As the goal is to enlarge the diplopia-free ﬁ eld, the sound eye may be operated where necessary. In that case, faden operation or recession of the yoke muscle in the sound eye may be used.

Summary for the Clinician ■ Th e correct evaluation of a complete or incomplete third nerve palsy (to diagnose the number of aff ected muscles) and assessment of a total or partial involvement (the residual function of the aff ected muscles) are the pearls for an appropriate surgical planning. ■ In complete third nerve palsy, superior oblique Fig. 15.4 In the case with congenital third nerve palsy traction function may easily be overlooked. Th e pearl is to sutures are seen in upper and lower eyelid to keep the eye in use slit lamp for a precise evaluation to see the adducted position [3] tiny intorsion. ■ In incomplete or partial third nerve palsy, the aim is to provide a functional diplopia-free area; 15.2.2 Incomplete Third Nerve Palsy however, in complete third nerve palsy, the aim is to fi xate the aff ected eye in primary position. In incomplete third nerve palsy with a superior or infe- ■ Orbital fi brosis is the bad prognostic sign for any rior branch or isolated EOM involvement, the treatment type of surgery. Th e pearl is to create surgically should be planned depending upon the aff ected EOM(s). induced restriction that provides a mechanical Recess-resect or transposition with a recession or BTXA pulling eff ect. A temporary pulling by traction injection may be preferred. In isolated inferior oblique sutures is very eff ective that allows the develop- palsy, transposition of horizontal recti perfectly works ment of the scar tissue while the globe was fi xated without weakening the superior rectus muscle. Complete on adduction. third nerve palsy may present with partial involvement 200 15 Pearls and Pitfalls in Surgical Management of Paralytic Strabismus

Fig. 15.5 Postoperative appearance of the patient aft er removal of the traction sutures 6 weeks aft er surgery. Orthophoria is obtained in primary position [3]

What is the amount of the deviation in primary position? 15.3 Fourth Nerve Palsy If the vertical deviation in primary position is exceeding In fourth nerve palsy hypertropia, inferior oblique overac- 15 prism diopters, two muscle surgeries need to be tion and superior oblique underaction is observed in the considered. aff ected eye. In long-standing unilateral cases, a secondary What is the position of gaze with the largest deviation? contracture of the superior rectus develops and a pseudo Th e surgical treatment should be planned on the EOMs overaction of the superior oblique muscle in the sound eye functioning in the fi eld of gaze with the largest deviation. is observed. Abnormal head posture and a positive To obtain a reliable data, the measurement of the devia- Bielschowsky head tilt test are the other fi ndings of fourth tion should be done in nine diagnostic positions of gaze. nerve palsy. In unilateral cases, the typically observed Is it congenital or acquired? Th e reply to this question abnormal head posture is chin down with head tilt toward has a specifi c importance in fourth nerve palsy. Congenital the unaff ected side. In bilateral cases, the abnormal head cases may present with superior oblique tendon abnor- posture may be as in unilateral cases if there is marked malities, such as abnormal tendon laxity, tendon inser- asymmetry. If the bilaterality is symmetrical, then the tion abnormalities, and sometimes even agenesis of the abnormal head posture aims to compensate the “V” pat- tendon [16–19]. Because of the frequent tendon abnor- tern. In acquired cases, vertical or torsional diplopia is the malities in congenital cases, it was proposed that these main complaint of the patients. Congenital cases do not cases might have primary developmental abnormality of usually complain about diplopia; however, in decompen- the superior oblique tendon rather than fourth nerve sated congenital fourth nerve palsy, the patient has vertical palsy [16]. However, in a previous MRI study where we diplopia. Some patients may benefi t from prisms but most looked for the superior oblique muscle size in congenital of the patients require surgical treatment. and acquired cases, we demonstrated that congenital For a correct surgical plan, one needs to have the cor- cases with abnormal tendon laxity may have denervation rect answers for the following questions: atrophy in the superior oblique muscle bulk and our fi ndings were confi rmed in other recent studies [20, 21]. If the ■ What is the amount of deviation in primary position? abnormality would only be limited with the tendon itself, ■ What is the position of gaze with the largest denervation atrophy would not be expected to develop in deviation? those cases with congenital fourth nerve palsy. Th e dif- ■ Is it congenital or acquired? ferential diagnosis in congenital and acquired cases is not ■ Is there any superior oblique tendon laxity? only important for the etiological investigation but also ■ Is there any superior rectus contracture? for surgical planning. Th e clinical clues suggesting that ■ Is it unilateral or bilateral? the patient has a congenital superior oblique palsy may be ■ Is there any torsional diplopia? summarized as follows: 15.3 Fourth Nerve Palsy 201

■ History, old photos palsy. In these cases, traction test is positive in depression ■ Absence of a preceding event on adduction. In motility examination, a limitation of ■ Prominent abnormal head posture depression on adduction and a pseudo overaction of the ■ Facial asymmetry superior oblique muscle in the sound eye are the clues for ■ Coexistence of amblyopia superior rectus contracture (Fig. 15.7). Recession of supe- ■ Signifi cant superior oblique underaction rior rectus muscle is advised in those cases with superior ■ Large vertical fusional amplitude rectus contracture [23, 24] (Fig. 15.8). ■ Coexisting horizontal deviation Is it unilateral or bilateral? Especially in traumatic ■ Absence of subjective torsion cases, masked bilaterality is very common. All of the cases with fourth nerve palsy should be carefully evaluated for Is there any superior oblique tendon laxity? Superior the clues of bilateral involvement [25, 26]. Th e bilateral oblique tendon laxity can be assessed prior to surgery involvement may be asymmetric but even with marked with traction test that was described by Guyton [22] and asymmetry surgery should be planned in both eyes. Th e modifi ed by Plager [17]. Th e globe is fi xated by two for- clinical clues suggesting bilateral involvement are as ceps at inferior nasal and superior temporal areas and follows: with retropulsion the globe is elevated on adduction. With this maneuver, the globe is pushed against the supe- ■ Bilateral inferior oblique overaction. rior oblique tendon, and with back and forth movements, ■ Bilateral superior oblique underaction. the globe the tendon can easily be felt (Fig. 15.6). If there ■ Positive Bielschowsky head tilt test with the head tilted is an agenesis of the superior oblique tendon, the tendon on both sides. In case of a marked asymmetry, cannot be felt and the globe is totally free with back and Bielschowsky head tilt test may be positive on the side forth movements. As an additional fi nding when the with marked involvement. globe is elevated on adduction cornea disappears in total ■ “V” pattern deviation. if there is a tendon laxity. ■ Abnormal head posture to compensate the “V” Is there any superior rectus contracture? Superior rectus pattern. contracture may develop in long-standing fourth nerve ■ Objective torsion exceeding 10° [40].

Fig. 15.6 Steps of superior oblique tendon tuck in abnormally lax superior oblique tendon in the right eye. (1) Th e globe is grasped with retropulsion. (2)Th e globe is moved superonasally and the cornea disappears in total, the back and forth movements indicate superior oblique tendon laxity. (3) Superior oblique muscle is found abnormally lax. (4)Tucking is performed with non absorbable sutures. (5) Superior oblique tendon is ﬁ xated on the sclera. (6) Traction test is repeated aft er tucking. Note the diﬀ erence of the position of the cornea 202 15 Pearls and Pitfalls in Surgical Management of Paralytic Strabismus

Fig. 15.7 Preoperative appearance of a patient with right long-standing fourth nerve palsy with ipsilateral superior rectus contracture. Note the limitation of depression in the right eye and the pseudo overaction of the left superior oblique muscle

Fig. 15.8 Postoperative appearance of the patient with right long-standing fourth nerve palsy following inferior oblique disinsertion and adjustable superior rectus recession of the right eye

Is there any torsional diplopia? Torsional diplopia is a although an excyclotorsion is observed in fundus exami- symptom that occurs in acquired fourth nerve palsy. Th e nation, and this is one of the clues for diff erential diagno- patients with a decompensated congenital fourth nerve sis of a congenital and acquired fourth nerve palsy. Some palsy has vertical diplopia without a torsional element, patients may not describe torsional diplopia properly 15.3 Fourth Nerve Palsy 203 unless asked specifi cally and may complain about “blur- muscle [26, 30]. Th is procedure is usually performed ring” in certain gaze positions. bilaterally and has a minimal eff ect on the vertical devia- Surgical methods of treatment may be summarized as tion in primary position and does not alter the esodevia- follows: tion on downgaze. So, it is only indicated if there is subjective torsional complaint that need to be corrected. ■ Inferior oblique weakening procedures Superior rectus recession in the aff ected eye: Th e indica- ■ Superior oblique strengthening procedures tion for superior rectus recession is a vertical deviation ■ Superior rectus recession in the aff ected eye exceeding 15 prism diopters in combination with supe- ■ Inferior rectus recession in the contralateral eye rior rectus contracture [23, 24]. It should be considered as an additional surgery with inferior oblique weaken- Inferior oblique weakening procedures: Inferior oblique ing. Th e predictability of the recession in a restricted weakening procedures are the most commonly performed superior rectus muscle will be low and adjustable reces- operations for treatment of fourth nerve palsy [41, 42]. sion should better be preferred in those cases. In cases Th e weakening procedures are disinsertion, myectomy, with agenesis of the superior oblique tendon, superior recession, and anteroposition of the inferior oblique mus- rectus recession is the procedure of choice with inferior cle. Inferior oblique weakening should be performed in oblique weakening. all cases with inferior oblique overaction. Our preferred Inferior rectus recession of the contralateral eye: Th e method for inferior oblique weakening is disinsertion. If cases that do not fi t any of the indications specifi ed above the deviation in primary position is more than 15 prism and where there is a vertical deviation exceeding 15 prism diopters, inferior oblique weakening will not be enough diopters are the candidates for contralateral inferior rec- to correct the deviation [27]. Anteroposition of inferior tus recession. It can be performed in combination with oblique muscle should be regarded with caution as it may inferior oblique weakening of the aff ected eye or as a sec- cause asymmetrical results because of the limitation of ondary procedure in cases with residual deviation. elevation and it is not recommended in unilateral cases Progressive overcorrection and lower eyelid retraction [24]. Anterior and nasal transposition of inferior oblique are well recognized problems with inferior rectus reces- muscle is a recently described method to be used in ones sion [31]. with congenital absence of superior oblique tendon [28]. In summary for an appropriate surgical plan for the Superior oblique strengthening procedures: Superior individual patient, the diagnosis of a congenital or oblique strengthening procedures are superior oblique acquired palsy, the deviation in nine positions of gaze, tendon tuck and Fells modifi ed Harada-Ito operation. abnormal head posture, the subjective characteristics of Superior oblique tendon tuck has a high risk of iatrogenic diplopia, and the traction test results are required. Brown syndrome in acquired cases with a normal tendon. In some particular cases BTXA may be used. Some However, it is a safe and very eff ective procedure in con- authors reported encouraging results with BTXA injec- genital cases with abnormal tendon laxity [18, 29]. In tion of ipsilateral inferior oblique muscle [32]. Contralateral cases with marked hypertropia and marked abnormal inferior rectus injection may be performed during acute head posture, superior oblique tendon tuck may be per- or chronic superior oblique palsies. Botulinum toxin is formed alone or usually in combination with inferior helpful to control postoperative over and undercorrec- oblique weakening. If there is no apparent inferior oblique tions; ipsilateral inferior rectus injection in the former and overaction, superior oblique tendon tuck may be per- contralateral inferior rectus injection in the latter [33]. In formed without weakening the inferior oblique muscle. our clinical practice, we use BTXA only for inferior rectus To reduce the risk of iatrogenic Brown syndrome, trac- muscle in fourth nerve palsy and the patient benefi ts with tion test must be performed aft er tucking with loop BTXA injection if there is no signifi cant torsional sutures (Fig. 15 6). If the traction test is positive then the element. amount of tuck should be reduced. Th e triad of indications for superior oblique tendon tuck is large angled ver- Summary for the Clinician tical deviation, prominent abnormal head posture, and superior oblique tendon laxity. ■ Th e pearl is the correct evaluation of a congenital In acquired cases with marked torsional diplopia, Fells and acquired case. Large vertical fusional ampli- modifi ed Harada-Ito procedure is the method of choice tudes, facial asymmetry, and absence of torsional that strengthens the anterior torsional fi bers. Th e anterior diplopia are the major clues for congenital fourth fi bers of superior oblique muscle are transposed lateral nerve palsy. and anteriorly at the upper border of the lateral rectus 204 15 Pearls and Pitfalls in Surgical Management of Paralytic Strabismus

■ Th e major pitfall is to overlook masked bilateral- ■ What is the amount of the measurement of the devia- ity. Presence of a “V” pattern and a large extor- tion in primary position? sion indicates bilaterality. Consider bilateral ■ Is the paralysis total or partial? surgery in such cases despite the absence of ■ Are there any medial rectus contracture? 15 apparent inferior oblique overaction and superior oblique underaction. Surgical methods of treatment may be summarized as ■ Inferior oblique weakening alone provides satis- follows: factory outcome in most of the cases if the vertical deviation does not exceed 15 prism diopters. ■ Medial rectus recession and lateral rectus resection. ■ Ipsilateral superior rectus and contralateral infe- ■ Medial rectus weakening of the sound eye. rior rectus weakening procedures should always ■ BTXA injection into the medial rectus muscle + be considered in combination with inferior vertical rectus muscle transposition. oblique weakening. ■ Medial rectus recession + vertical rectus muscle trans- ■ Do not consider superior oblique tuck surgery in position: Th is method carries a risk of anterior seg- acquired ones. Th e risk for symptomatic iatro- ment ischemia. Th at risk may be reduced by ciliary genic Brown syndrome is very high. Superior artery preserved full tendon transposition, perform- oblique tendon tuck should be reserved for con- ing the surgery in two divided sessions leaving at least genital cases with abnormal tendon laxity and a 3 months between two operations, or by performing a large vertical deviation. partial vertical rectus transposition. ■ Fells modifi ed Harada-Ito procedure is a surgery ■ If there is bilateral involvement, surgery should be for acquired bilateral cases with marked torsional performed in both eyes. component. Medial rectus recession and lateral rectus resection: Recess–resect should be reserved only for those with a good residual function of the aff ected lateral rectus muscle. If the residual function of the lateral rectus muscle is very limited, then transposition will work better than recess–resect procedure. Th e correct surgical decision 15.4 Sixth Nerve Palsy for a recess–resect or a transposition procedure is highly Lateral rectus underaction, esotropia, and a horizontal important. A wrong decision for a recess–resect proce- diplopia, which is more prominent at distance, and abnor- dure in an old patient makes the patient lose his or her mal head posture in unilateral cases keeping the aff ected chance to have a transposition procedure because of the eye in adduction are the clinical features of sixth nerve signifi cant risk of anterior segment ischemia. To obtain palsy. Lateral rectus underaction may be very subtle in a more reliable assessment for the residual lateral rectus partially aff ected cases and it is essential to measure the function, BTXA injection is recommended as a fi rst line deviation in nine positions of gaze. Partially aff ected cases treatment and the rest of the treatment plan is made benefi t from prisms. Addition of prisms only on distance according to the results that are obtained by BTXA glasses are enough in most of the cases. injection [3, 39] (Fig. 15.9). Botulinum toxin has a major role in treatment of sixth In cases with a signifi cant limitation of ocular motility, nerve palsy both for diagnostic and therapeutic purposes. BTXA provides the assessment of the residual function of During acute stage, injection of BTXA into the medial the paretic muscle in the absence of secondary fi brotic rectus muscle of the aff ected eye provides a symptomatic changes in medial rectus muscle. If there is no improve- relief. Although it was previously proposed that BTXA ment in abduction following a relaxation of the medial increased the possibility of spontaneous recovery, ran- rectus muscle by BTXA, it indicates that lateral rectus domized clinical trials demonstrated that BTXA injection muscle is totally dead and a transposition is required. We does not alter the chance of spontaneous recovery, but evaluate the ocular motility 1 week aft er the BTXA injec- provides a rapid symptomatic relief of diplopia [34–38]. tion and if there is no improvement on abduction, we In chronic stage in mild partial cases BTXA injection perform full tendon width vertical rectus muscle transpo- alone may provide a satisfactory improvement of the sition during the maximal BTXA eff ect. Th is method deviation. reduces the risk for anterior segment ischemia. For a correct surgical plan, one needs to have the cor- Medial rectus weakening of the sound eye: Medial rectus rect answers for the following questions: recession or faden operation of the medial rectus muscle References 205

• Botulinum toxin injection References as the first line treatment 1. Golnik KC, Miller NR (1991) Late recovery of function aft er oculomotor nerve palsy. Am J Ophthalmol 111: • Cure-no further treatment 566–570 2. Ansons AM, Davis H (2001) Diagnosis and management • Patient satisfied - of ocular motility disorders, 3rd edn. Oxford, Blackwell regular injections • Unsatisfactory result - necessary information Science, Paris Berlin Tokyo, pp 143–162 for recess-resect or 3. Özkan SB (2006) Strategies of treatment in paralytic stra- transposition surgery bismus. Türkiye Klinikleri J Surg Med Sci 2:58–65 4. Demer JL, Miller JM, Poukens V, et al (1995) Evidence for Fig 15.9 Th e use of BTXA for planning of treatment in sixth fi bromuscular pulleys of the recti extraocular muscles. nerve palsy [3] Invest Ophthalmol Vis Sci 36:1125–1136 5. Demer JL, Miller JM, Poukens V (1996) Surgical implications of the rectus extraocular muscle pulleys. J Pediatr of the sound eye increases the area of binocular diplopia- Ophthalmol Strabismus 33:208–218 free fi eld. A combination of recession and resection of the 6. Özkan SB, Dayanir V, Kir E, et al (2001) Role of botulinum medial rectus muscle provides an adjustable faden eff ect in toxin A in management of acquired loss of fusion. In: de the medial rectus muscle and may prove to be useful to Faber JT (ed) Transactions 27th meeting of the European reduce the symptoms of the patient with more control strabismological association. Swets and Zeitlinger, Th e compared with conventional faden operation [43]. Netherlands, pp 195–198 Th e problems of treatment in sixth nerve palsy are the 7. Gottlob IG, Catalano R, Reinecke RD (1991) Surgical man- anterior segment ischemia risk and the insuffi cient cor- agement of oculomotor nerve palsy. Am J Ophthalmol rection because of a recess–resect procedure in a non 111:71–76 functioning lateral rectus muscle. 8. Morad Y, Kowal L, Scott AB (2005) Lateral rectus muscle disinsertion and reattachment to the lateral orbital wall. Br J Ophthalmol 89:983–985 9. Velez FG, Th acker N, Britt MT, et al (2004) Rectus muscle orbital wall fi xation: a reversible profound weakening pro- Summary for the Clinician cedure. J AAPOS 8:473–480 ■ Th e correct diagnosis of partial and total sixth 10. Heo H, Park SW (2008) Rectus muscle posterior tenon nerve palsy is the pearl for a successful outcome fi xation as an inactivation procedure. Am J Ophthalmol of surgery. 146:310–317 ■ Th e major pitfall is the misinterpretation of the 11. Young TL, Conahan BM, Summers CG, et al (2000) lateral muscle function because of the second- Anterior transposition of the superior oblique tendon in ary medial rectus restriction in long-standing the treatment of oculomotor nerve palsy and its infl uence cases. on postoperative hypertropia. J Pediatr Ophthalmol ■ BTXA has major role both for surgical planning Strabismus 37:149–155 and as an adjunct to surgery. 12. Salazar Leon JA, Ramirez-Ortiz MA, Salas-Vargas M ■ Recess–resect procedure works only in ones (1998) Th e surgical correction of paralytic strabismus with good residual function of the lateral rectus using fascia lata. J Pediatr Ophthalmol Strabismus 35: muscle. Consider vertical rectus transposition 27–32 without augmentation sutures in ones with 13. Villasenor Solares J, Riemann BI, Romanelli Zuazo AC, very limited evidence of lateral rectus muscle et al (2000) Ocular fi xation to nasal periosteum with a function. Augmentation sutures increases the superior oblique tendon in patients with third nerve palsy. eff ect of transposition and should better be J Pediatr Ophthalmol Strabismus 37:260–265 used in ones with a totally dead lateral rectus 14. Daniell MD, Gregson RM, Lee JP (1996) Management of muscle. fi xed divergent squint in third nerve palsy using traction ■ To reduce the problems of vertical rectus muscle sutures. Aust N Z J Ophthalmol 24:261–265 transposition procedure keep parallel to the spiral 15. Khaier A, Dawson E, Lee J (2008) Traction sutures in the of Tillaux. management of long standing third nerve palsy. Strabismus 16:77–83 206 15 Pearls and Pitfalls in Surgical Management of Paralytic Strabismus

16. Helveston EM, Krach D, Plager DA, et al (1992) A new 30. Roberts C, Dawson E, Lee J (2002) Modifi ed Harada-Ito classifi cation of superior oblique palsy based on congenital procedure in bilateral superior oblique paresis. Strabismus variations of the tendon. Ophthalmology 99:1609–1615 10:211–214 17. Plager DA (1990) Traction testing in superior oblique 31. Sprunger DT, Helveston EM (1993) Progressive overcor- 15 palsy. J Pediatr Ophthalmol Strabismus 27:136–140 rection aft er inferior rectus recession. J Pediatr Ophthalmol 18. Plager DA (1992) Tendon laxity in superior oblique palsy. Strabismus 30:145–148 Ophthalmology 99:1032–1038 32. Lozano-Pratt A, Estanol B (1994) Treatment of acute paral- 19. Wallace DK, von Noorden GK (1994) Clinical characteris- ysis of the fourth cranial nerve by botulinum toxin A tics and surgical management of congenital absence of the chemodenervation. Binocul Vis Strabismus Q 9:155–168 superior oblique tendon. Am J Ophthalmol 118:63–69 33. Garnham L, Lawson JM, O’Neill D, et al (1997) Botulinum 20. Özkan SB, Aribal ME, Sener EC, et al (1997) Magnetic toxin in fourth nerve palsies. Aust N Z J Ophthalmol resonance imaging in evaluation of congenital and acquired 25:31–35 superior oblique palsy. J Pediatr Ophthalmol Strabismus 34. Holmes JM, Beck RW, Kip KE, et al (2000) Botulinum toxin 34:29–34 treatment versus conservative management in acute trau- 21. Sato M. Magnetic resonance imaging and tendon anomaly matic sixth nerve palsy or paresis. J AAPOS 4:145–149 associated with congenital superior oblique palsy (1999) 35. Lee J, Haris S, Cohen J, et al (1994) Results of a prospective Am J Ophthalmol 127:379–387 randomized trial of botulinum toxin therapy in acute uni- 22. Guyton DL (1981) Exaggerated traction test for the oblique lateral sixth nerve palsy. J Pediatr Ophthalmol Strabismus muscles. Ophthalmology 88:1035–1040 31:283–286 23. Aseff AJ, Munoz M (1998) Outcome of surgery for supe- 36. Metz HS, Masow M (1988) Botulinum toxin treatment of rior oblique palsy with contracture of ipsilateral superior acute sixth and third nerve palsy. Graefe’s Arch Clin Exp rectus treated by superior rectus recession. Binocul Vis Ophthalmol 226:141–144 Strabismus Q 13:177–180 37. Murray ADN (1991) Early botulinum toxin treatment of 24. Mims JL (2003) Th e triple forced duction test(s) for diag- acute sixth nerve palsy. Eye 5:45–47 nosis and treatment of superior oblique palsy with an 38. Repka MX, Lam GC, Morrison NA (1994) Th e effi cacy of updated fl ow chart for unilateral superior oblique palsy. botulinum neurotoxin A for the treatment of complete and Binocul Vis Strabismus Q18:15–24 partially recovered chronic sixth nerve palsy. J Pediatr 25. Kushner BJ (1988) Th e diagnosis and treatment of bilateral Ophthalmol Strabismus 31:79–83 masked superior oblique palsy. Am J Ophthalmol 105: 39. Riordian PR, Lee JP (1992) Management of VIth nerve 186–194 palsy – avoiding unnecessary surgery. Eye 386–390 26. Price NC, Vickers S, Lee JP, et al (1987) Th e diagnosis and 40. Kraft SP, O’Reilly C, Quigley PL, et al (1993) Cyclotorsion surgical management of acquired bilateral superior oblique in unilateral and bilateral superior oblique paresis. J Pediatr palsy. Eye 1:78–85 Ophthalmol Strabismus 30:361–367 27. Hatz KB, Brodsky MC, Killer HE (2006) When is isolated 41. von Noorden GK, Murray E, Wong SY (1986) Superior inferior oblique muscle surgery an appropriate treatment oblique paralysis: a review of 270 cases. Arch Ophthalmol for superior oblique palsy? Eur J Ophthalmol 16:10–16 104:1771–1776 28. Hussein MA, Stager DRSr, Beauchamp GR, et al (2007) 42. von Noorden GK, Campos EC (2002) Binocular vision and Anterior and nasal transposition of the inferior oblique ocular motility, 6th edn. Mosby, St. Louis, USA pp 559–565 muscle. J AAPOS 11:29–33 43. Dawson E, Boyle N, Taherian K, et al (2007) Use of a com- 29. Özkan SB, Can D, Demirci S, et al (1995) Surgical treat- bined recession and resection of a rectus muscle procedure ment in congenital superior oblique palsy. Türkiye in the management of incomitant strabismus. J AAPOS Klinikleri. J Surg Med Sci 4:223–226 11:131–134 Chapter 16 Modern Treatment Concepts in Graves Disease 16 Anja Eckstein and Joachim Esser

Core messages ■ Graves orbitopathy (GO) is part of an autoim- dently. To restrict damage, anti-infl ammatory mune systemic disease, which is composed of therapy (e.g., systemic steroids or orbital radio- hyperthyroidism, orbitopathy, dermopathy, and therapy) is indicated in moderate to severe active acropachy. disease stages. ■ Stimulating antibodies against the TSH receptor ■ Patients with sight-threatening GO should be are directly involved in the pathogenesis of hyper- treated with i.v. steroids as fi rst-line treatment; if thyroidism; their role is less clear with regard to the response is poor aft er 1 to 2 weeks, they the other manifestations. However, high TSH should be immediately referred for surgical receptor antibody concentrations are associated decompression. with a higher prevalence and more severe course ■ In patients with mild GO, local measures and an of extra-thyroidal symptoms. expectant strategy are usually suffi cient, but treat- ■ Main symptoms of GO are orbital soft tissue ment may be justifi ed if quality of life is reduced infl ammation, proptosis due to increase (mainly signifi cantly. through adipogenesis) of orbital volume and ■ In the inactive disease stages, proptosis can impairment of ocular and lid motility due to be alleviated through orbital decompression; infl ammation, and scarring of chiefl y the levator, restricted ocular and lid motility can be improved inferior, and medial rectus muscles. In severe by muscle recession and appearance can be cases, vision-threatening compression of the optic improved by blepharoplasty of lower and upper nerve can occur. lids. ■ Infl ammatory phase is self-limiting but may ■ Important for the successful treatment of GO is relapse, in most cases, owing to insuffi ciently continuous and stable sustenance of euthyroidism controlled thyroid disease, but also indepen- and smoking cessation.

receptor (TSHR) can be measured in the serum as indica- 16.1 Graves Orbitopathy (GO): Pathogenesis and Clinical Signs tors of the failed immune system. Th ose antibodies stimulate the TSHR in an uncontrolled manner and are 16.1.1 Graves Orbitopathy is Part of a directly responsible for the development of hyperthyroid- Systemic Disease: Graves Disease (GD) ism. Whether the TSHR alone or in combination with other antigens is responsible for the extra-thyroidal Graves orbitopathy is a part of a systemic autoimmune aspects of GD is of considerable research interest. disease. Th e full clinical picture is composed of hyperthy- Symptoms of GO are caused by infl ammation in the con- roidism, orbitopathy, pretibial myxedema, and acropachy. nective tissue of the orbit, an increase of intraorbital vol- Th e full symptom complex is very rare – Myxedema and ume due to enhanced adipogenesis, overproduction of acropachy occur only in 3–5%. With a prevalence of glycosaminoglycanes (GAG), and fi brosis of the extraoc- 0.5–2%, GD is a relatively common autoimmune disease ular muscles [2]. Orbital fi broblasts are pivotal to these [1]. In nearly all patients, antibodies against the TSH pathologic processes. Cultured orbital fi broblasts can be 208 16 Modern Treatment Concepts in Graves Disease

stimulated by patient IgG, several cytokines, and autolo- ■ Proptosis (exophthalmos) with possible concomitant gous lymphocytes. Stimulation by autologous lymphocytes lower lid retraction (mainly due to increased adipo- is antigen-dependent, as direct cell−cell contact, MHC genesis, but also due to enlargement of extraocular class II, and CD40–CD154 signaling are necessary [3]. In muscles and infl ammatory swelling) 16 addition, orbital fi broblasts may diff erentiate to preadi- ■ Ocular surface lesions (due to lagophthalmos, pocytes, which are accompanied by an increase in TSHR increased lid width, impaired Bell’s phenomenon, and expression [4]. Th us, the shared candidate autoantigen reduced tear secretion and deteriorated composition) between thyroid and orbita is the TSHR. ■ Restriction of ocular excursions – most oft en upgaze Clinically, high serum levels of TSHR antibodies and abduction (due to fi brosis of inferior and medial (TRAb) are associated with higher prevalence and rectus muscles) increased severity of GO. However, presence of TRAb ■ In rare cases (about 5%), dysthyroid optic neuropathy alone does not cause the complete symptom complex. (DON) (due to apical crowding) Neonates of mothers with TRAb positive GD usually develop hyperthyroidism, which gradually dwindles as antibodies are cleared from the child’s body, yet only few 16.1.2.1 Clinical Changes Result develop eye signs (mainly proptosis). Immunization of in Typical Symptoms mice against the TSHR does generate TRAb and hyperthyroidism but no associated orbital infl ammation [5]. ■ Change of facial appearance Th us, factors other than the presence of TRAb are prob- ■ Symptoms related to infl ammation: painful, oppres- ably involved in the development of GD. In GD, there is a sive feeling on or behind the globe, pain of attempted strong genetic component [see Chap. 16.5.2] involving up-, lateral, or downgaze immunoregulatory and thyroid-specifi c genes [6]. ■ Symptoms related to ocular surface irritation: gritty In most patients, there is a close temporal relationship sensation, light sensitivity, excess tearing, and reduced between the onset of hyperthyroidism and orbitopathy. visual acuity Orbitopathy usually manifests within 6 months before or ■ Symptoms related to restricted ocular motility: diplo- aft er the fi rst clinical signs of hyperthyroidism. MRI pia, abnormal head positure images of patients who suff er from hyperthyroidism but ■ Symptoms related to DON: reduced visual acuity, not from clinically overt orbitopathy reveal orbital mani- restricted visual fi eld, and desaturated color percep- festation in more than two-thirds of those patients [7]. Th e tion development of GO is a marker for a more severe course of GD and associated with signifi cantly lower remission rates of hyperthyroidism [8]. However, GO can also occur many 16.1.3 Clinical Examination of GO years aft er the onset of thyroid disease or − in rare cases − long before or even without overt thyroid disease [9]. In Determining the phase of GO at each clinical assessment 75% of euthyroid GO patients, thyroid-specifi c antibodies [14] is fundamental to the establishment of an appropri- can be detected as indicators of associated thyroid disease ate management plan (Fig. 16.2). Immunomodulatory [10]. About half of those patients will develop thyroid dys- therapies can only be eff ective in the presence of active function within the following 18 months [11]. infl ammation. Certain surgical treatments, on the other hand, (orbital, lid, or strabism surgery) should only be performed when GO has been constantly inactive for at least 6 months. 16.1.2 Graves Orbitopathy−Clinical Signs Graves Orbitopathy is typically characterized by the following clinical characteristics (Fig. 16.1) [12, 13]: 16.1.3.1 Signs of Activity ■ Most frequent sign (in 90–98% of patients): upper lid Th e active phase of the disease is the period when the retraction, oft en with lateral fl are and lid lag on verti- patient is most likely to be symptomatic: gaze evoked or cal downward pursuit, lagophthalmos (due to fi brosis spontaneous grittiness, light sensitivity, and excessive of the levator palpebrae muscle) orbital aching – gaze evoked or spontaneous. Patients ■ Other common signs: soft tissue signs, e.g., periorbital notice change of severity over the previous 3 months. swelling and redness, conjunctival swelling and injec- Classical signs of infl ammation are used as surrogate tion, prominent glabellar rhytids (due to infl ammation) markers to evaluate the degree of orbital infl ammation: 16.1 Graves Orbitopathy (GO): Pathogenesis and Clinical Signs 209

a b c

d e f

g h i

Fig. 16.1 Patient examples of typical symptoms of GO: 1A–1C Patient with mild GO, the only sign is upper lid retraction at the right eye. 1D–1F Patient with typical impairment of motility: 1D the patients developed a vertical squint of 22° (+VD), the upgaze of the left eye with 0°, 1E the coronary MRI scans show the enlargement of the inferior rectus muscle of the left eye. Th e other muscles are almost normal. 1G–1I Patients with full picture of GO with DON: marked signs of soft tissue inﬂ ammation (conjunctival injection and chemosis, caruncle inﬂ ammation, redness and swelling of the lids), marked proptosis, severe impairment of ocular motility right and dysthyroid optic neuropathy both eyes. Enlargement of all extraocular muscles was seen in the coronary MRI apical crowding in the orbital apex and intracranial fat prolapse in the axial MRI

■ Eyelid redness was 80% in estimating the response to immunomodula- ■ Conjunctival injection tion. Patients with disease duration of more than 18 ■ Chemosis (conjunctival edema) months are less likely to respond to immunomodulation. ■ Eyelid swelling A-mode ultrasound, T2 weighted or STIR sequence MRI ■ Infl ammation of caruncle or plica images, and serum or urine levels of a number of infl ammatory markers including IL-6, and urine GAG excretion All features of soft tissue infl ammation can be assessed by provide only little additional benefi t in predicting the comparison with standard patient photographs available response to anti-infl ammatory therapy [17]. at www.eugogo.eu. Studies show that reproducibility of patient assessment can be improved by the use of this atlas 16.1.3.2 Assessing Severity of GO and careful methodology (interobserver agreement − 86%). Photographic documentation is a reliable method Th e following features are quantifi ed to assess severity: for assessing soft tissue signs for follow-up. Signs of activity are summarized in the clinical activity score (CAS) ■ Lid fi ssure width (distance between the lid margins in (maximal seven points at the fi rst visit and maximal ten mm with the patient in primary position; sitting, points at follow-up) (Table 16.1) [16]. Using a cut-off of at relaxed, with distant fi xation) least four (fi rst visit three), the positive predictive value ■ Swelling of the eyelids (absent/moderate/severe) 210 16 Modern Treatment Concepts in Graves Disease

All patients with GO

• Restore euthyroidism • Urge smoking withdrawal 16 • Refer to specialist centers, except for the mildest cases • Local measures

Mild Moderate to severe Sight-threatening (DON)

i.v. GCs Local measures Active Inactive wait and see Progression Poor response (2 weeks)

Stable and i.v. GCS inactive (± OR) Prompt decompression

Still active Stable and inactive Rehabilitative Stable and surgery Rehabilitative inactive (if needed) surgery i.v. GCs Rehabilitative (± OR) surgery

Fig. 16.2 Management of Graves’ orbitopathy. Anti-inﬂ ammatory therapy in the active phase includes: intravenous glucocorticoids (i.v. GCs) and orbital radiotherapy (OR); Rehabilitative surgery includes orbital decompression, squint surgery, lid lengthening, and blepharoplasty/browplasty. Sight threatening GO (with dysthyroid optic neuropathy (DON) demands rapid decompression in case of poor response to i.v. GCs within 2 weeks. For the deﬁ nitions of GO severity and activity, see Chap. 16.1.3

■ Redness of the eyelids (absent/present) If vertical strabism is present, the contralateral eye should ■ Conjunctival injection (absent/present) be occluded. To evaluate upper and lower lid retraction, ■ Conjunctival chemosis (absent/present) eyelid position is measured in relation to the respective ■ Infl ammation of the caruncle or plica (absent, present) limbus. ■ Exophthalmos (measured in mm using the same Proptosis is usually measured with an exophthalmom- Hertel exophthalmometer and same intercanthal dis- eter. Numerous diff erent makes are available with diff er- tance for an individual patient) ent scales, so for each patient the same exophthalmometer ■ Subjective diplopia score (0 no diplopia; 1 intermit- with identical intercanthal distance should always be used tent, i.e., diplopia in primary position of gaze, when for follow-up. Proptosis is defi ned as a reading 2 mm tired or when fi rst awakening; 2 inconstant, i.e., diplo- greater than the upper normal limit for that patient’s age, pia at extremes of gaze; 3 constant, i.e., continuous gender, and “race.” More important, however, is the mea- diplopia in primary or reading position) sured change during follow-up. ■ Eye muscle involvement (duction in degrees) Th ere are numerous ways of assessing extraocular ■ Corneal involvement (absent/punctate lesions/corneal muscles. Subjective diplopia scores are simple but only of ulcer) limited help, since signifi cant changes in limitation of ■ Dysthyroid optic neuropathy (DON) (best-corrected motility may go unnoticed, when bilateral symmetrical visual acuity, color (de-) saturation, optic disk, relative reduction of upgaze results in no noticeable double vision. aff erent pupillary defect (absent/present), visual fi elds, Th e measurement of monocular excursions is a more visually evoked potentials) exact way to assess restricted excursions of each eye separately. Excursions are best measured using a bowl Examination of lid fi ssure width should be performed or arc perimeter, but so-called “Kestenbaum glasses” or with the head in a stationary position and under fi xation. the position of light refl exes may be used as well. Normal 16.1 Graves Orbitopathy (GO): Pathogenesis and Clinical Signs 211

Table 16.1. Clinical activity score (CAS), maximal 7 points at very large muscles in the orbital apex, fat herniation the fi rst visit and maximal 10 points at follow-up, active disease through the superior orbital fi ssure, and tense ballotte- ≥ CAS 4 (three fi rst visits) ment of the globe and venous stasis. DON is insidious as Clinical activity score CAS (one point is given its onset is rarely obvious and visual acuity is long pre- for each feature) served. Color vision disturbances are present in most patients. Only 30–40% of the patients present with swell- Subjective signs of activity ing of the optic disc. Visual fi eld defects are most com- Painful, oppressive feeling on 1 monly paracentral or inferior. VEP amplitudes are or behind the globe reduced and latency periods can be delayed [14]. Pain of attempted up-, side-, 1 Severity can be scored using the NOSPECS classifi ca- or downgaze tion, which provides in its slightly modifi ed version a Objective signs of activity maximal score of 14 (Table 16.3) for patients with all manifestations of GO in its most active stage [19]. Redness of the eyelids 1 Redness of the conjunctiva 1 Chemosis 1 16.1.3.3 Imaging Infl ammatory eyelid swelling 1 Infl ammation of the caruncle 1 Orbital imaging can be necessary for diff erential diag- or plica nosis as well as, in special situations, to facilitate treat- Sum score (at fi rst consultation no Maximal 7 ment decisions. If the patient presents with asymmetrical evaluation of progression possible) symptoms (usually unilateral proptosis), infl ammatory orbital disease of nonthyroidal etiology or orbital Signs of progression tumors have to be ruled out. Orbital imaging is neces- Increase of 2 mm or more in 1 sary for all clinical treatment decisions in Dysthyroid proptosis in the last 1–3 months optic neuropathy. Signal intensity in T2-weighted MRI Decrease in eye movements of 5° 1 scans corresponds to infl ammatory edema and can be or more in the last 1–3 months used to ease treatment decisions in diffi cult clinical Decrease in visual acuity in the 1 situations. Orbital ultrasound is only informative if last 1–3 months performed and evaluated by experienced clinicians Sum score Maximal 10 [20]. values are given in Table 16.2. Th e prism cover test (separate measurement of the squint angles in primary posi- 16.1.4 Classifi cation of GO tion for far and near distances) and the fi eld of binocular single vision are used to fi t corrective prisms and to plan Members of EUGOGO recommend to classify patients ≥ squint surgery. according to activity (active disease CAS 4, inactive dis- Of outstanding importance is the evaluation of the ease CAS < 4) and according to severity to manage corneal surface. Th is requires slit lamp examination to patients with GO [21]. detect punctate fl uoresceine staining or ulceration; the Severity classifi cation: latter constitutes an ophthalmologic emergency. Th ere is no single test that proves DON. DON occurs 1. Sight-threatening GO: Patients with dysthyroid optic bilateral in 70% of the patients. Anatomical indicators are neuropathy (DON) or corneal breakdown. Th is category warrants immediate intervention. Table 16.2. Normal values for monocular excursions (aft er 2. Moderate-to-severe GO: Patients without sight-threat- Mourits et al. [18]) ening GO whose eye disease has suffi cient impact on daily life to justify the risks of immunosuppression (if Direction of gaze Monocular excursion (°) active) or surgical intervention (if inactive). Patients Abduction 46 with moderate-to-severe GO usually present with one Upgaze 90° 34 or more of the following: lid retraction >2 mm, mod- Adduction 47 erate or severe soft tissue involvement, exophthalmos >3 mm above normal for “race” and gender, intermit- Downgaze 270° 58 tent, or constant diplopia. 212 16 Modern Treatment Concepts in Graves Disease

Table 16.3. Modiﬁ ed NOSPECS score for quantiﬁ cation of severity, maximal score of 14 NOSPECS score 0 1 2 3

Lid retraction No Yes 16 Soft tissue inﬂ ammationa 0 1–4 5–8 >8 Proptosis and or Site Diﬀ erence <17 mm 17–18 mm 19–22 mm >22 mm <1 mm 1–2 mm 3–4 mm >4 mm Extraocular muscle involvement No >20° upgaze ≤20° upgaze >35°abduction but not ≤35°abduction normal Corneal defects No Yes Optic nerve compression No Yes aUpper lid edema 0–2; Lower lid edema 0–2; conjunctival injection 1; conjunctival chemosis 1

3. Mild GO: patients whose features of GO have only a the active phase and rehabilitative surgical treat- minor impact on daily life, insuﬃ cient to justify ments in the inactive phase of the disease. immunosuppressive or surgical treatment. Th ey usu- ■ According to its grade, GO can be classiﬁ ed as mild, ally have only one or more of the following: minor lid moderate to severe, and sight threatening. Mild retraction (<2 mm), mild soft tissue involvement, GO permits a “wait and see” approach, moderate- exophthalmos <3 mm above normal for “race” and to-severe GO requires immunosuppressive treat- gender, transient or no diplopia, and corneal exposure ment in the active phase, and sight-threatening GO responsive to lubricants. demands immediate treatment with i.v. steroids/ orbital decompression/treatment of ocular surface Treatment decision can be made with the help of a damage. detailed management plan (see Fig. 16.2)

Summary for the clinician 16.2 Natural History ■ Graves’ Orbitopathy is part of an autoi mmune systemic disease encompassing hyperthyroid- Control of thyroid function infl uences the course of GO ism, orbitopathy, dermatopathy, and acropachy. (see Chap. 4). Patient with mild-to-moderate GO, moni- ■ TSHR receptor antibodies (TRAb) are indica- tored over 1 year without treatment, improved in 22%, tors of the failed immune system and direct showed minor improvement or no change in 42 and 22%, pathomechanism for hyperthyroidism. Th eir role respectively, and deteriorated in 14% [22]. With or within the pathogenesis of orbitopathy is less clear, out treatment, there are oft en residual symptoms of GO though patients with high serum TRAb levels in the form of lid retraction, proptosis, and muscle dys- have a higher prevalence of GO and develop more function. Th e outcome is signifi cantly better in patients severe disease stages. Orbital fi broblasts play a who have been diagnosed early and treatment started pivotal role in the pathologic changes in the orbit promptly. (release of chemokines, production of glycoseam- inoglycanes/fi brosis, and diff erentiation into adi- pose tissue). Summary for the Clinician ■ Assessment of activity (clinical activity score) and severity is necessary for disease manage- ■ Spontaneous improvement of GO with restora- ment: immunomodulation is performed during tion of euthyroidism occurs in more than 60% of the patients. 16.3 Treatment of GO 213

which is preventable if corticosteroids are administered 16.3 Treatment of GO simultaneously. Data on long-term safety are reassuring, but theoretical concerns about carcinogenesis remain for 16.3.1 Active Infl ammatory Phase younger patients, particularly those under the age of 35 Treatment is indicated in patients mainly with active years. Retinal microvascular abnormalities have been moderate-to-severe GO with a clinical activity score of detected in a minority of patients, mostly in those with four or more. concomitant severe hypertension or diabetic retinopathy. Consequently, these two comorbidities are considered absolute contraindications to OR. It is possible that dia- 16.3.1.1 Glucocorticoid Treatment betes, even in the absence of retinopathy, represents a risk factor for the development of retinal changes aft er OR, Glucocorticoids (GC) have been used in the management but the evidence is less persuasive [21, 27]. of GO administered locally, orally, or through i.v. [23]. Oral GC therapy (starting dose, 80–100 mg or 1 mg/kg body weight) requires high doses for prolonged periods of time. No randomized, placebo-controlled study, evalu- 16.3.1.3 Combined Therapy: Glucocorticoids ating oral glucocorticoid treatment was ever performed. and Orbital Radiotherapy Open trials or randomized studies, in which oral GC were Combination of systemic GC (either orally or locally) compared with other treatments, show a favorable with OR is more eff ective than either treatment alone. It is response in about 33–63% of patients, particularly con- unclear whether combining i.v. GCs with OR is more cerning soft tissue signs, eye muscle involvement of recent eff ective than i.v. GCs alone [28]. Representative studies onset, and DON. Eye disease frequently fl ares up on are summarized in Table 16.4. tapering out or withdrawing of oral GC therapy. Side eff ects are frequent. Local retrobulbar or subconjunctival administration 16.3.1.4 Other Immunosuppressive of glucocorticoids is less eff ective than oral GC. Treatments and New Developments Intravenous GC pulse therapy is more eff ective than One major problem is recurrent activity of GO aft er max- oral GC (dose: 250 mg–1 g/week, over 6–12 weeks or imal doses of i.v. glucocorticoid therapy and orbital radio- 500 mg–1 g for 3 consecutive days, followed by oral GCs); therapy. In most of the cases, poor control of thyroid response rates of about 80% are reported [24]. Evidence function, high TSH-receptor-antibody levels, and nico- for the superiority of any of the diff erent i.v. GC schedules tine abuse are among the underlying reasons. A thyroid as well as studies on the optimal cumulative dose is still specialist should always be consulted. In cases of expected lacking. Although i.v. GCs are tolerated better than oral low chance of remission or uncontrolled thyroid func- GCs, life-threatening liver failure has been reported in tion, defi nitive therapy of the thyroid has to be initiated. association with very high cumulative doses in 0.8% of Th yroidectomy is preferred because radioiodine therapy patients. Intravenous administration appears to be safe, if carries a risk of deterioration of active GO. In patients the cumulative dose is below 8 g methylprednisolone in with marked proptosis, orbital decompression has to be each course of therapy. considered because apart from proptosis reduction, decompression may also silence orbital infl ammation − probably due to improvement of orbital lymphatic and 16.3.1.2 Orbital Radiotherapy venous drainage. If activity still does not decline, other Th e reported response rate to orbital radiotherapy (OR) immunomodulatory agents have to be considered. Two in open trials is about 60%. Total doses between 10 and studies have shown the superiority of the combination of 20 Gy are commonly absorbed per orbit, fractionated in oral GCs and cyclosporine over either treatment alone. single doses between 1 and 2 Gy over a 2–20 week period. Recent treatment studies of GO patients with the Higher doses are no more eff ective. Th e response to OR B-lymphocyte depleting monoclonal antibody Rituximab did not diff er from oral prednisone in a randomized con- have shown promising results. Administered together trolled trial (RCT), but glucocorticoids are faster acting. with standard methimazole-therapy, it prolongs remis- Two recent RCTs have shown that OR is more eff ective sion of thyroid function in comparison with methimazole than sham irradiation in improving diplopia and eye monotherapy. Also, the stimulatory capacity of TRAbs muscle motility [25, 26]. OR is usually well tolerated, but was reduced markedly. Clinical activity of GO signifi - may cause transient exacerbation of ocular symptoms, cantly decreased aft er injection of 1,000 mg i.v. Rituximab 214 16 Modern Treatment Concepts in Graves Disease

Table 16.4. Representative results of randomized clinical trails of anti-inﬂ ammatory therapy for active GO

Randomization Response rates P values Authors

Group A Group B Group A Group B

16 i.v. methylprednisolonea Oral Prednisonecc 88% « 63% <0.02 Marcocci Radiotherapyb Radiotherapyb (n = 41) (n = 41) « «

i.v. methylprednisoloned oral prednisonee 77% « 51% <0.01 Kahaly (n = 35) (n = 35)

Comparison between i.v. and oral glucorticoid therapy is marked with horizontal arrows and comparison of single vs. combined (with orbital radiotherapy) therapy is marked with vertical arrows ([24, 29] Doses for glucocorticoid and radiotherapy: a15 mg/kgKG for four cycles, then 7.5 mg/kgKG for four cycles; each cycle consisted of two infusions on alternate days at 2-week intervals b20 Gy in ten daily doses of 2 Gy over 2 weeks c100 mg daily for 1 week, then weekly reduction until 25 mg daily, and then tapering by 5 mg every 2 weeks d500 mg once weekly for 6 weeks, 250 mg once weekly for 6 weeks, total treatment period: 12 weeks e100 mg daily starting dose, tapering by 10 mg/week, total treatment period: 12 weeks

twice at 2-week interval. Even proptosis was signifi cantly Frequent topical lubricants, moisture chambers, tars- reduced. Subsequent randomized controlled trials with orrhaphy, amnion epithelium membrane as shield, and Rituximab need to be performed [30–32]. Th e anti-TNF botulinum toxin injections in the levator muscle (doses a drug Etanercept is described as eff ective as well in an for therapeutic ptosis: e.g., 30 IE Dysport®) should be open trial [33]. applied immediately. Surgical decompression or lid Treatments of marginal or unproven value include lenghthening a chaud should be considered when the somatostatin analogs, azathioprine, ciamexone, and i.v. above measures alone are ineff ective [21]. immunoglobulins.

16.3.1.6 Other Simple Measures 16.3.1.5 Therapy of Dysthyroid Optic Neuropathy that may Alleviate Symptoms (DON) and Sight-Threatening Corneal Breakdown Th e symptoms of corneal exposure (grittiness, watering, and photophobia) should be treated with lubricant eye- High-dose i.v. GCs are the preferred fi rst-line treatment drops. Nocturnal ointment is of great benefi t if eyelid clo- for DON (3 × 500 mg–1 g at consecutive days within 1 sure is incomplete. week, if necessary repeated the following week). If the Prisms may correct intermittent or constant diplopia. response to i.v. GCs is absent or poor aft er 1–2 weeks, or Sleeping with the head in an upright position may the dose/duration of steroid required induces signifi cant improve lymphatic drainage and alleviate early morning side eff ects, orbital decompression should be carried out eyelid swelling. Diuretics are rarely useful. Upper lid promptly. Orbital decompression should be recom- retraction can be reduced by injecting botulinum toxin mended promptly to patients with DON or corneal (e.g., 5–15 IU Dysport®) subconjunctivally in the tarsal breakdown who cannot tolerate glucocorticoids. Both muscle (Mueller muscle). Full eff ect is evident aft er 2–3 i.v. GC therapy and orbital decompression surgery should days and persists for about 4–6 weeks. Th e outcome is only be performed in clinical centers with the appropri- variable and the dose of botulinum toxin must be adjusted ate expertise. individually. Transient double vision and ptosis may Sight-threatening corneal breakdown must be treated occur in 10–20%. Th is procedure should be carried out in as an emergency as well. specialized centers [34]. 16.3 Treatment of GO 215

aft er cessation of antithyroid drug therapy. Because of its Summary for the Clinician infl uence on ocular motility and lid width, decompression ■ Patients with active moderate-to-severe GO or surgery should be performed fi rst. Vertical squint correc- active mild GO with suffi cient impairment tion may then be performed. Pseudoretraction will resolve on daily life should receive anti-infl ammatory postoperatively but lower lid retraction can occur aft er treatment. inferior rectus recession. Small medial rectus recessions ■ Glucocorticoids are applied most effi ciently i.v. can be combined with lid surgery; larger recessions should 250 mg–1 g weekly over 6–12 weeks or at con- be performed separately [35, 36]. secutive days within 1 week (cumulative dose: 1.5–3g) followed by an oral regime (response 16.3.2.1 Orbital Decompression rate about 80%). Cumulative doses of 8 g should not be exceeded to prevent liver damage and A wide range of surgical approaches is used to reduce other severe side eff ects. disfi guring proptosis in patients with GO. Th e amount of ■ Orbital radiotherapy is indicated primarily for proptosis reduction depends on the number of walls patients with impaired motility. Fractionated removed and whether or not fatty tissue is removed. doses between 10 and 20 Gy are applied to each Serious complications are rare. Common surgical orbit (response rate about 60%). approaches for orbital decompression are: coronal, via ■ Combined therapy (glucocorticoids and orbital the upper skin crease, the lateral canthus, or the inferior radiotherapy) is more effi cient than each therapy fornix (both together = swinging eyelid), sub-ciliary, alone. directly through the lower lid, transcaruncular, transna- ■ Patients with dysthyroid optic neuropathy should sal, and transanthral. Further restriction of ocular motil- be treated with i.v. steroids as fi rst-line treat- ity is still a major complication; this mainly occurs with ment; if the response is poor aft er 1–2 weeks, medial wall decompression. Th e risk is much lower with they should be referred for immediate surgical removal of the lateral wall alone. Clinically obvious decompression. In case of marked proptosis or impairment of motility increases the risk of postopera- severe corneal exposure, surgical decompression tive diplopia signifi cantly. can be immediately performed. At present, the medial, inferior, and lateral walls are ■ New therapeutic strategies for patients with addressed during bony orbital decompression (Fig. 16.3), severe GO are being tested – most promising is B while the orbital roof is neglected due to potential com- cell depletion, which inactivates GO and sup- plications. Minimally invasive approaches and hidden ports remission of thyroid dysfunction. incisions are preferred. Decompression of the medial ■ Simple measures like topical lubricants, botuli- orbital wall is necessary to decompress the optic nerve in num toxin for retracted lids and prisms for com- patients with DON. pensation of double vision are important for the Th e transnasal endoscopic procedure addresses the quality of life of the patients. medial and inferior orbital walls. Th e advantage of a con- venient scarless procedure is opposed by the relative high risk of decreased ocular motility and inferior and nasal dislocation of the globe. Proptosis may be reduced by 2–5 mm. 16.3.2 Inactive Disease Stages With the coronary approach, all orbital walls can be Rehabilitative surgery includes one or more of the follow- accessed and proptosis reduction up to 10 mm can be ing procedures: (a) orbital decompression, the usual indi- achieved. Th is is, however, an elaborate procedure. cation for surgery being disfi guring exophthalmos with or To enhance the eff ect of lateral wall decompression, without keratopathy; (b) squint correction; (c) lid length- the procedure can be combined with removal of its deep ening; and (d) blepharoplasty/browplasty. Prerequisite for portion or with additional fat removal (Fig. 16.3). Th e lat- successful surgery is a minimum of 6 months of stable eral wall has, due to a very low risk of diplopia, increas- inactive ophthalmologic and thyroid disease. Concerning ingly become the fi rst choice for orbital decompression thyroid disease, this means either constant doses of (traditional concept – inferior-medial decompression Levothyroxin aft er defi nitive therapy (thyroidectomy/ fi rst) in cases of rehabilitative surgery. Th e approach to radioiodine therapy) or stable remission at least 6 months the lateral wall is variable via the upper skin crease, 216 16 Modern Treatment Concepts in Graves Disease

a b

16 1 1 3 2

Fig. 16.3 Surgical approaches for orbital decompression in coronar and axial view. All orbital walls except the roof are addressed. Th e lateral wall can be removed conservatively (A1), until is deep portion (A2) or completely (A3). Various surgical approaches are possible to decompress the inferior (B2) and medial (B1) orbita. Th e inferior-lateral region of the orbit is the most common zone for fat removal (B3)

swinging eyelid, sub-ciliary, or directly through the lower ■ Horizontal squint <10°: unilateral medial rectus reces- lid. Average proptosis reduction ranges between 2 and sion (side: eye with least abduction), dose 1 mm reces- 5 mm. (Literature is reviewed in [37].) sion per 1.75° of intended squint angle reduction, maximal recession distance 6–7 mm ■ Horizontal squint ≥10°: bilateral medial rectus reces- 16.3.2.2 Extraocular Muscle Surgery sion, dose 1 mm recession per 1.6° of intended squint Th e basic concept for eye muscle surgery in GO is reces- angle reduction (dose side diff erent, when side diff er- sion of the fi brotic muscle. Th e approach is diff erent for ence in monocular abduction), maximal recession inferior and medial rectus muscles. Vertical deviation distance per eye 6–7 mm increases with side diff erences in monocular upward ■ Combined horizontal and vertical squint: small verti- excursions. Bilateral symmetric restrictions of inferior rec- cal angles disappear aft er correction of horizontal tus muscles cancel each other out and cases with abnormal squint; a two-step procedure (large angle fi rst) is more head posture need to be corrected by symmetric inferior precise; if all in one procedure is preferred (only rec- rectus recession. Bilateral restriction of abduction adds up. ommended for unilateral procedures): consider higher Diff erent concepts for surgical strabism correction are dose eff ect for vertical squint 2.1° per mm recession available: preoperatively determined recession distances ■ Lower lid retraction aft er inferior recession can be according to dose eff ect curves, and intraoperative deter- prevented through dissection of the capsulopalpebral mination of recession distance via active or passive motil- ligament. Upper lid retraction of the eye with eleva- ity and adjustable sutures (literature is reviewed in [38]). tion defi cit (“pseudoretraction”) will disappear aft er Principles for extraocular muscle surgery in patients inferior recession with GO: ■ Convergent squint correction aft er decompression: consider lower dose eff ects: unilateral medial rectus ■ Vertical squint – no head tilt when covering the eye recession: 1 mm recession per 1.2° of intended squint with more limited upgaze: Recession of inferior rectus angle reduction; bilateral rectus recession: 1 mm reces- muscle: dose: 1 mm recession per 2° of intended squint sion per 1.0° of intended squint angle reduction; con- angle reduction, maximal recession distance 7–8 mm, sider medial rectus tendon elongation with a spacer persisting vertical squint: second step: recession of the for very large angles: 1 mm elongation per 0.9° of contralateral superior rectus muscle dose 1 mm/per 2° intended squint angle reduction of intended squint angle reduction ■ Vertical squint – head tilt when covering the eye with Dose eff ect data are summarized in Table 16.5 [38, more limited upgaze: asymmetric bilateral inferior 40–42]. rectus recession (side diff erence in mm depends on In most cases, it is possible to improve the fi eld of bin- the squint angle, measured with head tilt: 1 mm reces- ocular single vision. Over-corrections occur more oft en sion per 2° of intended squint angle reduction) when the muscle is not directly fi xed to the sclera but is 16.3 Treatment of GO 217

Table 16.5. Extraocular muscle surgery: dose eff ect coeffi cients: squint angle reduction (°)/per mm muscle recession (source: [38–41]) Muscle Dose eff ect: angle [°] Authors reduction/ mm recession

Inferior rectus muscle 2.0 Esser et al., 1999 2.1 Krizok et al., 1993 Medial rectus muscle unilateral 1.75 Eckstein et al., 2004 Medial rectus muscle bilateral 1.6 Eckstein et al., 2004 Combined Eckstein et al., 2004 unilateral inferior rectus muscle 2.1 unilateral medial rectus muscle 1.9 Aft er orbital decompression Eckstein et al., 2008 Medial rectus muscle unilateral 1.2 Medial rectus muscle bilateral 1.0 Tendon elongation with interponate 0.9

adjusted on the following day. Th is probably occurs due Upper lid lengthening: Many diff erent techniques for to adaptation of the muscles to changed tension during lenghthening the upper eyelids have been described. the operation. Post-operative tone increase occurs in Among these are techniques with or without implants. structures that were previously relaxed, e.g., the antago- In most cases, use of implants is not necessary. Th ese are nist and the “passive orbital tissue.” Th ey return to their Müllerotomy or recession, medial or lateral levator original tension, which leads to a further globe rotation aponeurosis recession, lateral horn cut (important for against the direction of the recession. Th erefore, the eff ect lateral fl are), medial and lateral full thickness levator-, of squint angle reduction increases signifi cantly within Müller-muscle-, and conjunctival recession. Since lateral the fi rst postoperative month. retraction (temporal fl are) is the most important aspect Persistent diplopia in extreme gaze is common, which of upper lid retraction in patients with Graves orbitopa- is usually tolerable in upgaze, since the used gaze fi eld is thy, division of the lateral horn of the aponeurosis is nec- larger in downgaze than in upgaze. essary in most cases. Sutures may be placed between the Success rates (ocular alignment within about 2–3° in tarsal plate and the detached aponeurosis to prevent primary position) are similar for the diff erent approaches spontaneous disinsertion. When sutures are used, it is and vary mainly between 60 and 80% for horizontal important to protect the cornea, e.g., using the conjunc- squint and up to 90% for vertical squint. tiva as a cover. Myotomies without spacers (graft s) require patient cooperation. If compliance is poor or marked fi brosis is present, spacers may be used. Th e vertical height of the implant should be approximately twice 16.3.2.3 Lid Surgery the measured eyelid retraction or measured eyelid Th e most common indication for lid surgery in GO is retraction +2 mm, respectively. Patients examples before upper lid retraction due to levator muscle fi brosis. and aft er upper lid lengthening without and with implant Genuine lid retraction has to be discriminated from are shown in Fig. 16.4. Th e implant is used in a patient pseudo-lid retraction due to fi brosis of the inferior rectus with severe GO (aft er three wall decompression for muscle. Th e latter resolves aft er inferior rectus recession. DON) with marked fi brosis of levator palpebrae muscle. Lower lid lengthening is indicated in lower lid retraction Correction of upper lid retraction is successful when following inferior rectus recession. Bilateral lower lid 1–2 mm of the superior cornea is covered, the lid margin retraction with proptosis should primarily be referred for contour is smooth, when upper lid skin crease is between orbital decompression. Another indication for eyelid sur- 7 and 10 mm, and lids are symmetric. Most of the surgi- gery is increased preaponeurotic and subdermal fat, cal procedures are ascribed success rates of about resulting in bulging eye lids. Th is may be treated during 70–80%. Asymmetry can occur due to over- or under- blepharoplasty when redundant lid skin is excised (review correction, lid crease recession, and a thickened eyelid of the literature: [35, 43]). aft er use of a graft . 218 16 Modern Treatment Concepts in Graves Disease

a e

b f

Fig. 16.4 Upper lid lenghthening in GO. 4A–4D In most of the cases upper lid retraction does not exceed 2 mm and levator muscle desinsertion (4D scheme from [15]) will suﬃ ce. Patient example with upper lid retraction right eye in primary position (4A), in downgaze showing the lid lag on vertical downward pursuit (4B) and aft er lid lenghthening (4C). In rare cases with marked retraction (especially aft er decompression), the use of an implant is necessary (4E–4G). Patient example before (4E) and aft er lid lengthening with an implant (5 mm Tutopatch®) (4F) and intraoperative situation (4G)

Lower lid lengthening: To correct lid retraction lateral tarsal strip or tarsorrhaphy. Undercorrection is exceeding 1 mm, a “spacer” between lower lid retrac- common. tors and tarsus is required (Fig. 16.5). Various organic Upper and lower lid blepharoplasty: Upper lid deb- and anorganic materials have been used as spacers. ulking and blepharoplasty is the ﬁ nal surgical proce- Th ese include auricular cartilage, hard palate mucosa, dure in the functional and cosmetic rehabilitation of expanded polyethylene Medpor microplates, autoge- the GO patient. Redundant skin and fat can be excised nous tarsus transplants, porcine acellular dermal using scissors and bipolar cauterant, laser, or monopo- matrix, and donor sclera or pericardium. Th e vertical lar cauterization needle. In the lower lid, the skin exci- expansion of the spacer should amount to 3 times the sion should be modest to avoid lower lid retraction or lid retraction in mm. Most spacers, except hard palate ectropion. It is important to remove preaponeurotic fat mucosa, need to be covered with conjunctiva. Th e lower (Fig. 16.6) and even subdermal fat together with the lid retractors are accessible either by anterior subciliary orbicularis muscle. Prolapsing lower lid fat can also be or posterior subtarsal transconjunctival approach. Th e removed transconjunctivally in patients without excess eﬀ ect of lower lid lengthening can be increased by skin. 16.3 Treatment of GO 219

a Sutures for stabilisation of the interponate

d c

e Tarsus f

Interponate

Lid retractors Lig. capsulopalp.

Inferior rectus muscle

Fig. 16.5 Lower lid lengthening in GO. Lower lid retraction can occur aft er large inferior rectus muscle recession if the ligamentum capsulopalpebrale cannot be suﬃ ciently detached from the inferior rectus muscle. Patient example: 5A before inferior rectus muscle recession of 7.5 mm, vertical squint: −VD15°. 5B lower lid retraction aft er inferior recession. 5C intraoperative situation: size and position of the implant. 5D patient situation 1 day postoperative. 5E cross section of the lower lid with implant (black), F ﬁ nal result aft er lower lid lengthening with an implant and lateral tarsorrhaphy of 5 mm

40,0 grey TBII values below: 2.3-15.6x better Summary for the Clinician 35,0 Zone: chance for a good course of GO 30,0 no ■ Disfi guring proptosis can be reduced through pre- TBII values above: 8.7-31.1x higher 25,0 diction risk of a severe course orbital decompression. Various surgical tech- 20,0 possible niques are available. Th e amount of reduction 15,0 depends on the number of walls removed and 8,8 TBII [IU/l] whether or not fat is removed. Removal of the 10,0 5,1 4,8 2,9 2,8 5,0 medial wall is accompanied with the highest and 5,7 0,0 removal of the lateral wall with the lowest risk of 2,6 1,5 1,5 1,5 1,5 -5,0 postoperative diplopia. If muscle restriction is 1-4 5-8 9-12 13-16 17-20 20-24 present preoperatively, the risk of postopera- Months after first symptoms of GO tively deteriorated ocular motility is increased. ■ Th e basic concept for eye muscle surgery in GO Fig. 16.6 Cut off TBII levels for the prediction of a good course of is recession of the fi brotic muscle. Diff erent GO (grey line) and for the prediction of a severe course of GO (black line). For patients with TBII level within the grey zone no prognostic approaches are possible: preoperatively deter- statement for the course of their GO is possible. Example: A GO mined recession distances according to dose– patient presenting at 1–4 months aft er onset of the disease with TBII eff ect curves and intraoperative determination of values below 5.7 IU/L has a 13.9-fold higher chance of a mild curse recession distance via active or passive motility of GO than a patient with TBII values above this cut off . Otherwise, and adjustable sutures. Success rates are high. when TRAb are still above 8.8 IU/l 6 months aft er the beginning of GO the odds ratio to develop a severe course of GO is 18 220 16 Modern Treatment Concepts in Graves Disease

■ Upper lid retraction can be corrected in most [51] and relapse of hyperthyroidism can be accompanied patients without the use of a spacer through by worsening/reactivation of pre-existing GO. recession of levator palpebrae and Müller mus- Regular consultations with a thyroid specialist are cle. Implants have to be inserted for successful necessary. 16 lower lid lengthening. Th e eff ect can be enhanced with a lateral tarsal strip or tarsorrhaphy. Th e last step in surgical rehabilitation is blepharoplasty 16.4.2 Relationship Between of upper and lower lids. TSH-Receptor-Antibody (TRAb) Levels and Orbitopathy Th e relation between TRAb and GO was for a long time subject to debate and became evident with modern, more 16.4 Thyroid Dysfunction and GO sensitive second-generation TRAb assays. Th e prevalence of GO among patients with Graves’ hyperthyroidism 16.4.1 Association Between Treatment increases with higher serum TRAb levels [52]. Th ere is a of Hyperthyroidism and Course of GO signifi cant correlation of clinical activity [53] and severity [54] with TRAb levels in untreated individuals. In late Th e main goal during the early stage of thyroid disease is to stages, non-responders to anti-infl ammatory therapy reveal achieve euthyroidism. Not only does this alleviate most higher TRAb levels [55]. Patients with moderate-to-severe thyroid symptoms, it is also benefi cial for the further course GO have signifi cantly higher TRAb levels over the whole of GO. Antithyroid drug therapy seems to prevent most course of the disease (24 months follow-up) (Fig. 16.6). Cox effi ciently further deterioration of GO in comparison with regression analysis 6 months aft er disease onset revealed a thyroidectomy and radioiodine therapy [44]. Observational hazard ratio of 1.27 to incur severe GO per every unit trials showed that thyroidectomy in the intermediate phase increase of TRAb [56]. When TRAb are still above 8.8 IU/l (6–12 months aft er fi rst symptoms of GO) may positively 6 months aft er beginning of GO, the odds ratio to develop a infl uence the clinical course of GO. In later, inactive stages, severe course of GO is 18. Patients with TRAb levels in the this benefi cial eff ect is lost [45]. Leaving too large thyroid risk zone (see Fig. 16.6) should have short control intervals, remnants increases the risk of recurrence of hyperthyroid- treated with anti-infl ammatory therapy in cases of doubt ism and reactivation of GO [46]. Radioiodine therapy car- and treated longer with higher doses. ries a small but not inconsiderable (about 15%) risk of inducing or worsening GO in the intermediate phase [47]. Radiogenic infl ammation of the thyroid during and aft er radioiodine application may reinforce the autoimmune Summary for the Clinician reaction in the thyroid and activate or induce GO. It does not, however, infl uence inactive GO [48]. ■ Restoration of euthyroidism is benefi cial for the Patients with poor prognosis for remission should course of GO. receive defi nitive therapy as a prerequisite for surgical ■ Radioiodine therapy carries a small but not con- rehabilitation of GO. Poor prognosis for hyperthyroid- siderable (about 15%) risk of inducing or wors- ism can be expected with persisting high TSH-receptor- ening GO. antibodies (TRAb) during the course of antithyroid ■ Patients with poor prognosis for remission of drug therapy. Remission rates are about 3% if TRAb are hyperthyroidism should receive defi nitive therapy still above 10 IU/l aft er 6 months, above 7.5 IU/l aft er 12 as a perquisite for surgical rehabilitation of GO. months, and 3.9 IU/l aft er 15 months of antithyroid ■ Th e overall relapse rate of hyperthyroidism aft er drug therapy (TRAb levels must be measured with a cessation of antithyroid drug therapy is 50%. second generation assay for these statements to be Th erefore, surgical rehabilitation of GO should valid). Remission rates are low (about 8%) in cases of only be started aft er a 6 months period of stable moderate-to-severe GO [8, 49, 50]. remission. Relapses of hyperthyroidism can be Patients with a chance for remission of thyroid disease accompanied by worsening or reactivation of GO. (non-smoker, low TRAb levels, small thyroid, mild hyper- ■ TSH-receptor autoantibodies are independent thyroidism at manifestation) should be followed for at risk factors for GO and help to predict severity least 6 months aft er cessation of antithyroid drug therapy and outcome of the disease. Certain cut off levels before surgical rehabilitation (if necessary) is initiated. can be used for treatment decisions. Th e overall relapse rate for hyperthyroidism is about 50% 16.5 Environmental and Genetic Infl uence on the Course of GO 221

genetically infl uenced. Th e concordance rate for clinically 16.5 Environmental and Genetic Infl uence on the Course of GO overt Graves disease is 35% for monozygotic twins (MZ) and 3% for dizygotic twins (DZ). Model-fi tting analysis on 16.5.1 Relationship Between Cigarette the pooled twin data showed that 79% of the disposition Smoking and Graves Orbitopathy for the development of GD is attributable to genetic factors [60]. Approximately, half of the patients show a positive Th ere is a strong and consistent association between smok- family history of thyroid dysfunction with a higher fre- ing and GO. Smoking increases the prevalence of GO quency among females in comparison with males. Positive among patients with Graves’ hyperthyroidism. Smokers family history is also more common in maternal than in suff er from more severe GO than non-smokers. A dose– paternal relatives. Th e reporting of a parent with thyroid response relationship between the amount of cigarettes dysfunction is associated with a lower median age at diag- smoked daily and the risk of developing GO has been nosis for GD. Th ere is an inverse relationship between the demonstrated (Fig. 16.7). Smoking increases the risk of number of relatives with thyroid dysfunction and age at extraocular muscle fi brosis sevenfold [57]. Smoking diagnosis [61]. Frequently, identical susceptibility genes increases the likelihood of progression of GO aft er radio- are designated for Graves and Hashimoto’s disease (sum- iodine therapy. Th ere is also evidence that smoking either marized in [6, 62]). Within monozygotic twins, it is possi- delays response or impairs the outcome of treatment for ble for one twin to develop typical Graves disease while the GO [58]. As to the thyroid, smoking is a similarly inde- other suff ers from Hashimoto’s thyroiditis without orbit- pendent risk factor for relapse of hyperthyroidism aft er opathy [63]. Th us, there is clear evidence for genetic sus- antithyroid drug treatment [51]. In vitro models (orbital ceptibility to develop thyroid autoimmunity. Th e disease fi broblast cell cultures) have been used to illustrate the phenotype, however, appears to be determined by environ- impact of smoke constituents on GO, which were found to mental factors, for instance, smoking behavior. enhance two of the central processes in GO: adipogenesis In the meantime, linkage and candidate gene analyses and GAG production in a dose-dependent manner [59]. have revealed more than 50 genes, which may contribute Th e eff ect is markedly enhanced in the presence of the to autoimmune thyroid disease. However, essential genes proinfl ammatory cytokine IL-1. Th e synergy between cig- which are crucial for disease development remain to be arette smoke and cytokine action may have potential for identifi ed. Th e genes identifi ed to this day comprise thy- therapeutic implications. roid specifi c genes (TSHR, Th yroglobulin) and immune modulating genes (among them: HLA class II, CTLA-4, PTPN22, CD40). Important for the disease phenotype are functional consequences of these gene variants. Table 16.5.2 Genetic Susceptibility 16.6 displays the most important susceptibility genes, Th ere are a number of epidemiological and twin studies including possible functional consequences (modifi ed which clearly indicate that autoimmune thyroid disease is from Jacobson et al. [6]).

Summary for the Clinician Severity of GO and Smoking ■ Graves’ disease arises owing to interaction 100 90 Prevalence of GO between environmental and genetic factors. 80 Proptosis ■ Smoking is associated with a higher prevalence 70 Diplopia of GO, the development of more severe disease 60 stages of GO, reduced eﬀ ectiveness of treatments 50 for GO, and with the progression of GO aft er 40 30 radioiodine treatment. Th erefore, the patient

% of the GO patients 20 should be advised to stop smoking. 10 ■ Immune regulatory and thyroid-speciﬁ c genes 0 contribute to the disease. Th e risk for ﬁ rst-degree Nonsmoker 1-10 10-20 > als 20 Cigarettes Cigarettes Cigarettes relatives is 3%. About 50% of patients report a positive family history, more common in the Fig. 16.7 Association of GO symptoms with the number of maternal than in the paternal trait. smoked cigarettes 222 16 Modern Treatment Concepts in Graves Disease

Table 16.6. Susceptibility genes and possible functional consequences in Graves’ disease (slightly modiﬁ ed from Jacobson et al. [6])

Gene Associated variants Potential mechanisms

Immune response 16 modulating genes HLA DR DR3 Alteration in autoantigen presentation CTLA-4 Several SNP’s (A/G49, Reduction of suppression of T-cell activation CT60, 3’UTR AT) (CTLA-4 = negative regulator of T-cells) CD 40 Kozak sequence SNP Alteration of translational effi ciency of CD40 in CD40 expressing tissues (APC, thyrocytes, orbital fi broblasts) PTPN22 R620W Inhibition of T-cell activation IL23R Several SNP (rs11209026, rs7530511, Reduction of activation of T cells, natural rs2201841, rs10889677) killer (NK) cells, monocytes, and dendritic cells “protecting factor”, expansion of Th 17 subset Th yroid specifi c genes Th yroglobulin Several SNP Alteration in thyroglobulin peptide presentation by HLA DR to T-cells TSHR 28 SNPs revealed association Alteration in TSHR peptide presentation by HLA DR to T-cells, alterations in Auto AB binding

glucocorticoids should be avoided unless the patient 16.6 Special Situations suff ers from optic neuropathy. Orbital radiotherapy is contraindicated in children. Orbital surgery may be 16.6.1 Euthyroid GO necessary in cases of severe exophthalmos, but for most Patients with euthyroid GO developed less severe symp- patients a conservative and expectant approach is most toms, especially fewer soft tissue signs and more asym- appropriate [65]. metric disease (unilateral proptosis) than hyperthyroid patients. Levels of thyroid-specifi c antibodies are lower and less prevalent. However, they occur in at least 75% of the patients; therefore, the application of sensitive assay 16.6.3 GO and Diabetes technology is of utmost concern [64]. Systemic glycocorticoids may induce or exacerbate diabetes or hypertension. However, indications for glucocorticoid use in patients with diabetes or hypertension are no diff erent than in other patients. Close monitoring 16.6.2 Childhood GO of blood sugar levels and blood pressure is important. GO is rare in childhood because of the low incidence of Th iazide or loop diuretics should be used cautiously dur- Graves disease in this age group. Th e eye disease is usu- ing high-dose steroid therapy to avoid hypokalemia. Th e ally milder in children than in adults and oft en stabi- same principle applies to surgical treatment. Orbital lizes and eventually resolves without intervention. Soft radiotherapy may increase the risk of retinopathy in tissue infl ammation is rare in childhood GO. Achieving diabetic and hypertensive patients. Diabetes or hyper- and maintaining euthyroidism are as important objec- tension are no contraindication to surgical orbital tives as in adult patients. Exposure to smoking (active decompression or other surgical treatments. Optic neu- and, possibly even passive) is probably as detrimental as ropathy occurs signifi cantly more oft en in diabetic in adults. Because of their eff ect on growth, patients (reviewed in [21]). References 223

nd thyrotropin receptor antibody levels. Th yroid 10: Summary for the Clinician 1093–1100 ■ Patient with euthyroid GO develop less active/ 12. Bartley GB, Fatourechi V, Kadrmas EF, et al (1996) Clinical severe and more asymmetric GO symptoms. features of Graves’ ophthalmopathy in an incidence cohort. ■ If present at all GO is mild in childhood and Am J Ophthalmol 121:284–290 rarely needs treatment. 13. Wiersinga W, Kahaly G (2007) Graves’s ophthalmopathy. A ■ Orbital irradiation is possibly contraindicated in multidisciplinary approach. Karger patients with diabetic retinopathy and DON 14. Wiersinga WM, Perros P, Kahaly GJ, et al (2006) Clinical occurs more oft en. assessment of patients with Graves orbitopathy: the European Group on Graves Orbitopathy recommendations to generalists, specialists and clinical researchers. Eur J Endocrinol 155:387–389 15. Hintschich C, Haritoglou C (2005) Full thickness eyelid References transsection (blepharotomy) for upper eyelid lengthening 1. Wiersinga WM, Bartalena L (2002) Epidemiology in lid retraction associated with Graves disease. Br J and prevention of Graves’ ophthalmopathy. Th yroid 12: Ophthalmol 89:413–416 855–860 16. Mourits MP, Prummel MF, Wiersinga WM, et al (1997) 2. Gianoukakis AG, Smith TJ (2008) Recent insights into the Clinical activity score as a guide in the management of pathogenesis and management of thyroid-associated oph- patients with Graves’ ophthalmopathy. Clin Endocrinol thalmopathy. Curr Opin Endocrinol Diabetes Obes (Oxf) 47:9–14 15:446–452 17. Terwee CB, Prummel MF, Gerding MN, et al (2005) 3. Feldon SE, Park DJ, O’Loughlin CW, et al (2005) Autologous Measuring disease activity to predict therapeutic outcome T-lymphocytes stimulate proliferation of orbital fi broblasts in Graves’ ophthalmopathy. Clin Endocrinol (Oxf) derived from patients with Graves’ ophthalmopathy. Invest 62:145–155 Ophthalmol Vis Sci 46:3913–3921 18. Mourits MP, Prummel MF, Wiersinga WM, et al (1994) 4. Smith TJ (2004) Novel aspects of orbital fi broblast pathol- Measuring eye movements in Graves ophthalmopathy. ogy. J Endocrinol Invest 27:246–253 Ophthalmology 101:1341–1346 5. McLachlan SM, Nagayama Y, Rapoport B (2005) Insight 19. Eckstein AK, Plicht M, Lax H, et al (2006) Th yrotropin into Graves’ hyperthyroidism from animal models. Endocr receptor autoantibodies are independent risk factors for Rev 26:800–832 Graves’ ophthalmopathy and help to predict severity and 6. Jacobson EM, Tomer Y (2007) Th e genetic basis of thyroid outcome of the disease. J Clin Endocrinol Metab autoimmunity. Th yroid 17:949–961 91:3464–3470 7. Villadolid MC, Yokoyama N, Izumi M, et al (1995) 20. Kahaly GJ (2004) Recent developments in Graves’ ophthal- Untreated Graves disease patients without clinical ophthalmopathy imaging. J Endocrinol Invest 27:254–258 mopathy demonstrate a high frequency of extraocular 21. Bartalena L, Baldeschi L, Dickinson A, et al (2008) muscle (EOM) enlargement by magnetic resonance. J Clin Consensus statement of the European Group on Graves Endocrinol Metab 80:2830–2833 orbitopathy (EUGOGO) on management of GO. Eur J 8. Eckstein AK, Lax H, Losch C, et al (2007) Patients with Endocrinol 158:273–285 severe Graves’ ophthalmopathy have a higher risk of relaps- 22. Perros P, Crombie AL, Kendall-Taylor P (1995) Natural ing hyperthyroidism and are unlikely to remain in remis- history of thyroid associated ophthalmopathy. Clin sion. Clin Endocrinol (Oxf) 67:607–612 Endocrinol (Oxf) 42:45–50 9. Bartley GB, Fatourechi V, Kadrmas EF, et al (1996) 23. Kahaly G (2007) Management of moderately severe Graves Chronology of Graves’ ophthalmopathy in an incidence orbitopathy. In: Wiersinga WM, Kahaly GJ (eds) Graves cohort. Am J Ophthalmol 121:426–434 orbitopathy: a multidiscplinary approach. Karger, Basel, pp 10. Eckstein AK, Loesch C, Okundia H, et al (2009) Patients 120–152 with euthyroid and primarily hypothyroid status have low 24. Kahaly GJ, Pitz S, Hommel G, et al (2005) Randomized, TRAb levels and develop a tendentious milder and signifi - single blind trial of intravenous versus oral steroid mono- cantly more asymmetric Graves’ Ophthalmopathy. Br J therapy in Graves orbitopathy. J Clin Endocrinol Metab Ophthalmol in print 90:5234–5240 11. Khoo DH, Eng PH, Ho SC, et al (2000) Graves’ ophthal- 25. Mourits MP, van Kempen-Harteveld ML, Garcia MB, et al mopathy in the absence of elevated free thyroxine and (2000) Radiotherapy for Graves orbitopathy: randomised triiodothyronine levels: prevalence, natural history, a placebo-controlled study. Lancet 355:1505–1509 224 16 Modern Treatment Concepts in Graves Disease

26. Prummel MF, Terwee CB, Gerding MN, et al (2004) A ran- 40. Eckstein A, Schulz S, Esser J (2004) Is combined surgical domized controlled trial of orbital radiotherapy versus correction of horizontal and vertical squint of value in sham irradiation in patients with mild Graves’ ophthalmo- graves’ ophthalmopathy? Klin Monatsbl Augenheilkd pathy. J Clin Endocrinol Metab 89:15–20 221:769–775 16 27. Bradley EA, Gower EW, Bradley DJ, et al (2008) Orbital 41. Krzizok T, Efi nger K, Kaufmann H (1993) Dosierbarkeit radiation for graves ophthalmopathy: a report by the von Augenmuskeloperationen bei endokriner Orbitopathie. American academy of ophthalmology. Ophthalmology Z prakt Augenheilk 14:273–279 115:398–409 42. Schittkowski M, Fichter N, Guthoff R (2004) Strabismus 28. Marcocci C, Bartalena L, Panicucci M, et al (1987) Orbital surgery in Graves disease–dose-eff ect relationships and cobalt irradiation combined with retrobulbar or systemic functional results. Klin Monatsbl Augenheilkd 221: corticosteroids for Graves’ ophthalmopathy: a comparative 941–947 study. Clin Endocrinol (Oxf) 27:33–42 43. Neoh C, Eckstein A (2007) Eyelid surgery. In: Wiersinga 29. Marcocci C, Bartalena L, Tanda ML, et al (2001) Comparison WM, Kahaly GJ (eds) Graves orbitopathy: a multidisc- of the eff ectiveness and tolerability of intravenous or oral plinary approach. Karger, Basel, pp 188–198 glucocorticoids associated with orbital radiotherapy in the 44. Tallstedt L, Lundell G, Torring O, et al (1992) Occurrence management of severe Graves’ ophthalmopathy: results of a of ophthalmopathy aft er treatment for Graves’ hyperthy- prospective, single-blind, randomized study. J Clin roidism. Th e thyroid study group. N Engl J Med 326: Endocrinol Metab 86:3562–3567 1733–1738 30. El Fassi D, Banga JP, Gilbert JA, et al (2009) Treatment of 45. Marcocci C, Bruno-Bossio G, Manetti L, et al (1999) Th e Graves disease with rituximab specifi cally reduces the pro- course of Graves’ ophthalmopathy is not infl uenced by duction of thyroid stimulating autoantibodies. Clin near total thyroidectomy: a case-control study. Clin Immunol 130:252–258 Endocrinol (Oxf) 51:503–508 31. Heemstra KA, Toes RE, Sepers J, et al (2008) Rituximab in 46. Dralle H, Sekulla C (2004) Morbidity aft er subtotal and relapsing Graves disease, a phase II study. Eur J Endocrinol total thyroidectomy in patients with Graves disease: the 159:609–615 basis for decision-making regarding surgical indication 32. Salvi M, Vannucchi G, Campi I, et al (2007) Treatment of and extent of resection. Z Arztl Fortbild Qualitatssich Graves disease and associated ophthalmopathy with the 98(Suppl 5):45–53 anti-CD20 monoclonal antibody rituximab: an open study. 47. Bartalena L, Marcocci C, Bogazzi F, et al (1998) Relation Eur J Endocrinol 156:33–40 between therapy for hyperthyroidism and the course of 33. Paridaens D, van den Bosch WA, van der Loos TL, et al Graves’ ophthalmopathy. N Engl J Med 338:73–78 (2005) Th e eff ect of etanercept on Graves’ ophthalmopa- 48. Perros P, Kendall-Taylor P, Neoh C, et al (2005) A pro- thy: a pilot study. Eye 19:1286–1289 spective study of the effects of radioiodine therapy for 34. Traisk F, Tallstedt L (2001) Th yroid associated ophthalmo- hyperthyroidism in patients with minimally active pathy: botulinum toxin A in the treatment of upper eyelid graves’ ophthalmopathy. J Clin Endocrinol Metab 90: retraction–a pilot study. Acta Ophthalmol Scand 5321–5323 79:585–588 49. Carella C, Mazziotti G, Sorvillo F, et al (2006) Serum thy- 35. Baldeschi L (2008) Correction of lid retraction and rotropin receptor antibodies concentrations in patients exophthalmos. Dev Ophthalmol 41:103–126 with Graves disease before, at the end of methimazole 36. Garrity JA, Bahn RS (2006) Pathogenesis of graves oph- treatment, and aft er drug withdrawal: evidence that the thalmopathy: implications for prediction, prevention, and activity of thyrotropin receptor antibody and/or thyroid treatment. Am J Ophthalmol 142:147–153 response modify during the observation period. Th yroid 37. Mourits MP, Bijl HM, Baldeschi L, et al (2008) Outcome of 16:295–302 orbital decompression for disfi guring proptosis in patients 50. Schott M, Morgenthaler NG, Fritzen R, et al (2004) Levels with Graves orbitopathy using various surgical procedures. of autoantibodies against human TSH receptor predict Br J Ophthalmol relapse of hyperthyroidism in Graves disease. Horm Metab 38. Esser J, Eckstein A (1999) Ocular muscle and eyelid sur- Res 36:92–96 gery in thyroid-associated orbitopathy. Exp Clin Endocrinol 51. Orgiazzi J, Madec AM (2002) Reduction of the risk of Diabetes 107(Suppl 5):S214–S221 relapse aft er withdrawal of medical therapy for Graves dis- 39. Eckstein A, Holdt M, Johnson KTM, et al (2008) Tendon ease. Th yroid 12:849–853 elongation: a new surgical technique for large convergent 52. Khoo DH, Ho SC, Seah LL, et al (1999) Th e combination of squint aft er three wall orbital decompression in thyroid absent thyroid peroxidase antibodies and high thyroid- associated ophthalmopathy. ESA proceedings stimulating immunoglobulin levels in Graves disease iden- References 225

tifi es a group at markedly increased risk of ophthalmopathy. 59. Cawood TJ, Moriarty P, O’Farrelly C, et al (2007) Smoking Th yroid 9:1175–1180 and thyroid-associated ophthalmopathy: a novel explana- 53. Gerding MN, van der Meer JW, Broenink M, et al (2000) tion of the biological link. J Clin Endocrinol Metab Association of thyrotrophin receptor antibodies with 92:59–64 the clinical features of Graves’ ophthalmopathy. Clin 60. Brix TH, Kyvik KO, Christensen K, et al (2001) Evidence Endocrinol (Oxf) 52:267–271 for a major role of heredity in Graves disease: a population- 54. Noh JY, Hamada N, Inoue Y, et al (2000) Th yroid- based study of two Danish twin cohorts. J Clin Endocrinol stimulating antibody is related to Graves’ ophthalmopathy, Metab 86:930–934 but thyrotropin-binding inhibitor immunoglobulin is 61. Manji N, Carr-Smith JD, Boelaert K, et al (2006) Infl uences related to hyperthyroidism in patients with Graves disease. of age, gender, smoking, and family history on autoim- Th yroid 10:809–813 mune thyroid disease phenotype. J Clin Endocrinol Metab 55. Eckstein AK, Plicht M, Lax H, et al (2004) Clinical results 91:4873–4880 of anti-infl ammatory therapy in Graves’ ophthalmopa- 62. Ban Y, Tomer Y (2005) Genetic susceptibility in thyroid thy and association with thyroidal autoantibodies. Clin autoimmunity. Pediatr Endocrinol Rev 3:20–32 Endocrinol (Oxf) 61:612–618 63. Aust G, Krohn K, Morgenthaler NG, et al (2006) Graves 56. Eckstein AK, Plicht M, Lax H, et al (2006) TSH-receptor disease and Hashimoto’s thyroiditis in monozygotic twins: autoantibodies are independent risk factors for Graves’ case study as well as transcriptomic and immunohistologi- ophthalmopathy and help to predict severity and outcome cal analysis of thyroid tissues. Eur J Endocrinol 154:13–20 of the disease. J Clin Endocrinol Metab 91(9):3464–3470 64. Eckstein A, Loesch C, Glowacka D, et al (2009) Euthyroid 57. Pfeilschift er J, Ziegler R (1996) Smoking and endocrine and primarily hypothyroid patients develop milder and ophthalmopathy: impact of smoking severity and current signifi cantly more asymmetric Graves ophthalmopathy. Br vs lifetime cigarette consumption. Clin Endocrinol (Oxf) J Ophthalmol 93:1052–1056 45:477–481 65. Krassas GE, Gogakos A (2006) Th yroid-associated oph- 58. Eckstein A, Quadbeck B, Mueller G, et al (2003) Impact of thalmopathy in juvenile Graves disease–clinical, endocrine smoking on the response to treatment of thyroid associ- and therapeutic aspects. J Pediatr Endocrinol Metab ated ophthalmopathy. Br J Ophthalmol 87:773–776 19:1193–1206 Index

A refractive adaptation, 103–104 Abducens palsy, 71 refractive error, 97 Abnormal central nervous system (CNS) esotropia, 2 sensitivity, 100 Abnormal central nervous system (CNS) exotropia, 3 stereoacuity test, 101 Acquired motor neuropathy, 71–72 strabismus, 97 Acquired nonaccommodative esotropia, 2 treatment compliance, 105 Acquired pulley heterotopy, 63–64 type of treatment, 103 Amblyopia treatment 2009 vision in preschoolers study age eff ect, 131 (VIP), 100, 101 amblyopia management vision tests, 100 patch occlusion, 128–129 vs. diagnostic test, 97 pharmacological therapy, plano lens, 130 Anisometropia, 33 pharmacological treatment, atropine, 129–130 Anisometropic amblyopia, 2, 3 refractive correction, 127–128 Anomalous head posture (AHP) Bangerter fi lters, 132–133 Anderson–Kesternbaum surgery, 158 bilateral refractive amblyopia, 131 binocular visual acuity testing, 161–162 clinical features, 126 horizontal management, 165–166 deep unilateral amblyopia, 175–176 idiopathic infantile nystagmus, 158 diagnosis, 126–127 measurement, 160–161 epidemiology, 125–126 monocular eff ect, 161–162 levodopa/carbidopa adjunctive therapy, 133 straightening eff ect, head, 162 long-term persistence, 132 testing, near vision, 162 maintenance therapy, 131 vertical management, 166–167 natural history data, 127 Anomalous retinal correspondence (ARC), 34 optic neuropathy, 133–134 Atropine, 129–130 spectacle correction, 125 Amblyopia, screening B Child Health Promotion Program (CHPP), 96 Bagolini test, 141 classifi cation, 95 Bangerter foils, 132 conventional occlusion, 104 Bell’s phenomenon, 88 cover-uncover test, 100–101 Bielschowsky head tilt test (BHTT), 181 defi nition, 95–96 Bilateral feedback control Duane’s/Brown’s syndrome, 97 applications, 21–22 justifi cation, 98 muscle lengths, 19–21 lay screeners, 102 Bilateral posterior tenectomy, 190 older children, 104–105 Bilateral refractive amblyopia, 131 optical penalization, 104 Binocular alignment system orthoptists, 102 control system pharmacological occlusion, 104 A-/V-pattern strabismus, 14 photorefractive keratectomy (PRK), 105 basic muscle length, 15–16 photoscreening/autorefraction, 101–102 bilateral phenomena, 14–15 pre-school vision screening, 98–99 breakdown, 14 quality of life fi nal common pathway, 17–18 emotional well-being, 107 perturbation, 13 impact of treatment, 108 sensory torsion, 14 impact on education, 106–107 version and vergence stimulation, 16–17 reading speed and ability, 106 deviation and fi xation pattern, 11 strabismus impact, 107–108 long-term maintenance, 11 recurrence, 105 muscle length adaptation, 12–13 228 Index

vergence adaptation, 12 congenital trochlear palsy, 82 Binocular vision Duane retraction syndrome, 79–81 angle of strabismus, 140–141 HGPPS, 81 at age six, 140 isolated uni-/bilateral facial palsy, 83 bilateral recession vs. unilateral recession-resection, 141 vertical retraction syndrome, 88 Blood–brain barrier, 133 Congenital esotropia, 2 Botulinum toxin A (BTXA), 197, 203, 204 Congenital exotropia, 3 Brown syndrome, 4, 203 Congenital fi brosis of the extraocular muscles (CFEOM), Brückner test 78–79 amblyopia and amblyogenic disorders, 113–114 A-pattern exotropia, 69 corneal light refl ex, 114–115 motor axonal misrouting, 67 eye movements, alternating illumination, 122 MRI, 67–68 fundus red refl ex phenotypes, 67 ametropia, 116, 118 Congenital nystagmus anisometropia, 118 clinical characteristics, 156–157 esotropia, 117–118 compensatory mechanisms foveal dimming, 117 AHP, 160–162 hypermetropia, 118 versions and vergence, 160 Mittendorf’s spot, 115 manifest latent nystagmus (MLN) optic coherence tomography, 117 clinical characteristics, 157–158 paediatric residents, 119 slow phase, 157 possibilities and limitations, 120 periodic alternating nystagmus (PAN), 158–159 pupillary constriction, 116 sensory defi cits test performance, 119–120 afferent visual defect, 155 transillumination test, 115 causes, 156 uncorrected ametropia, 118 horizontal eye movement, 154 uni-lateral astigmatismus, 119 idiopathy, 155 uni-lateral spherical ametropia, 118, 119 phenotypical characteristics, 155 pupillary light refl ex treatment eccentric vs. central illumination, 121 acupuncture, 164 iris pathology, 120 artifi cial divergence surgery, 167–168 monocular illumination, 121 botulinum toxin-A (Botox), 164 possibilities and limitations, 121–122 head tilt, 167 strabismus diagnostics, 120 horizontal AHP, 165–167 test performance, 121 medications, 162–163 prisms, 163 C refractive correction, 162 Cataract, 2, 3 retro-equatorial recession, 168–169 Child Health Promotion Program (CHPP), 96 spectacles and contact lenses (CL), 162–163 Chronic progressive external ophthalmoplegia (CPEO), surgical principles, 164–165 59–60 tenotomy procedure, 169 CNS-associated hypertropia, 4 vertical AHP, 166–167 Complete third nerve palsy hypertropia, 198–199 Congenital oculomotor (CN3) palsy, 67 Congenital cranial dysinnervation disorders (CCDDs), 66 Congenital pulley heterotopy, 62–63 brainstem and cranial nerve development, 77, 78 Congenital superior oblique paresis, 20, 21 Brown syndrome Congenital trochlear (CN4) palsy, 69 comorbidity, 85 Convergence insuffi ciency, 3 epidemiologic features, 85 Cycloplegic drug, 127 incidence and heredity, 86 Cyclovertical misalignment, 19 intra-and postoperative fi ndings, 87 laterality, 85–86 D motility fi ndings, 83–85 Diagnostic occlusion, 19 natural course, 87 Dissociated eye movements neurodevelopmental disorder, 89–90 pathogenetic role, 29 potential induction, 86–87 vergence eye movements, 25 radiologic fi ndings, 87 dissociated horizontal deviation (DHD), 25–29, 179–180 saccadic eye movements, 85 dissociated torsional deviation (DTD) sex distribution, 86 inverse and direct head tilt, 181 CFEOM, 78–79 strabismus, 180 congenital fourth nerve palsy, 82 dissociated vertical deviation (DVD) congenital monocular elevation defi ciency, 87–89 asymmetric vs. symmetric surgeries, 178 congenital ptosis, 81 bilateral, 175–176 Index 229

hypotropia, nonfi xating eye, 178–179 G IOOA and V pattern, 176–177 German Institute for Quality and Effi ciency in Healthcare SOOA and A pattern, 177–178 (IQWIG), 99 symmetric, 175 Glucocorticoids (GC), 213 Divergence paralysis esotropia, 64–65 Graves orbitopathy Double elevator palsy, 83, 87, 88 active infl ammatory phase Duane’s retraction syndrome (DRS), 69, 79–81 combined therapy, 213 Duane’s syndrome, 19 dysthyroid optic neuropathy (DON), 214 Dysthyroid optic neuropathy (DON), 214 glucocorticoids (GC), 213 immunosuppressive treatments, 213–214 E orbital radiotherapy (OR), 213 EOM surgery, 216–217 sight-threatening corneal breakdown, 214 Esotropia (ET) symptoms, 214–215 DHD, 179–180 childhood, 222 monofi xation syndrome, 35–36 classifi cation, 211–212 visual cortex mechanisms clinical assessment binocular input correlation, 50–51 activity signs, 208–209 binocular visuomotor behavior assess severity, 209–211 development, 42, 43 orbital imaging, 211 cerebral damage risk factors, 41–42 clinical characteristics, 208 cortical binocular connections, 44–46 diabetes, 222 cytotoxic insult, cerebral fi bers, 42 environmental and genetic infl uence early-onset (infantile) esotropia, 41 cigarette smoking, 221 extrastriate cortex, striate cortex, 46 susceptibility genes, 221–222 fusional vergence and innate euthyroid, 222 convergence bias, 44 Graves disease (GD), 207–208 genetic infl uence, cerebral connection, 42 inactive disease stages high-grade fusion repair, 50 extraocular muscle surgery, 216–217 inter-ocular suppression, 46–47 lid surgery, 217–220 monocular compartments, striate cortex, 44, 46 orbital decompression, 215–216 motion sensitivity and conjugate eye tracking, 44 management plan, 208, 210 naso-temporal inequalities, cortical suppression, 47 thyroid dysfunction, 220 persistent nasalward visuomotor bias, 47–50 sensorial fusion and stereopsis development, 43 H strabismic human infant repair, 50 Health-related quality of life (HRQoL), 98, 99, 106–108 Essential infantile esotropia. See Congenital esotropia Horizontal gaze palsy with progressive scoliosis Exotropia (XT) (HGPPS), 81 DHD, 179–180 Hypertropia, 3–4, 179 infantile esotropia active divergence mechanism, 26 I binocular fusion vs. dissociated esotonus, 27, 28 Immune myopathy, 60–61 clinical signs, 27 Incomplete third nerve palsy hypertropia, 199 horizontal strabismus, 28 Infantile esotropia (IE) Expected value of perfect information (EVPI), 99 defi nition and prevalence, 137 Extraocular muscle (EOM), 196, 197 dissociated eye movements Eye lid surgery pathogenetic role, 29 lower lid lengthening, 218, 219 vergence eye movements, 25 upper and lower lid blepharoplasty, 218 early vs. late infantile strabismus upper lid lengthening, 217 surgery study (ELISSS) alignment and fusion, 145 F binocular vision, 140 First Purkinje images, 114–115 horizontal angle of strabismus, 140–141 Fourth nerve palsy hypertropia methods and results, 139–140 bilateral involvement, 201 postoperative angle of strabismus, 145 congenital superior oblique palsy, 200 prospective study, 139 inferior oblique weakening procedure, 203 random-effects model, 146, 148 superior and inferior rectus recession, 209 reoperation rate, 142–143 superior oblique strengthening procedure, 209 spontaneous reduction, 146–148 superior oblique tendon laxity, 201 spontaneous resolution, 146 superior rectus contracture, 201 test-retest reliability, 144–145 surgical plan, 200 esotonus vs. convergence, 28 torsional diplopia, 202–203 exotropia 230 Index

active divergence mechanism, 26 N binocular fusion vs. dissociated esotonus, 27, 28 Neoplastic myositis, 61 clinical signs, 27 Neuroanatomical strabismus horizontal strabismus, 28 acquired motor neuropathy, 71–72 outcome parameters, 138–139 acquired pulley heterotopy, 63–64 pathogenesis, 138 congenital peripheral neuropathy sensory/motor etiology, 137–138 congenital cranial dysinnervation disorders tonus, 25–26 (CCDDs), 66 Infantile-onset image decorrelation, 38–39 congenital fi brosis of the extraocular muscles Inferior oblique (IO) palsy, 71–72 (CFEOM), 67–69 Inferior oblique overaction (IOOA), 4, 176–177 congenital oculomotor (CN3) palsy, 67 Infl ammatory myositis, 61 congenital trochlear (CN4) palsy, 69 Intermittent exotropia, 3, 4 Duane’s retraction syndrome (DRS), 69 Moebius syndrome, 70 L congenital pulley heterotopy, 62–63 Levodopa, 133 divergence paralysis esotropia, 64–65 Logistic regression analysis, 143 etiology, 59 Long-term binocular alignment control system, 14 extraocular myopathy immune myopathy, 60–61 M infl ammatory myositis, 61 Manifest latent nystagmus (MLN) neoplastic myositis, 61 Anderson–Kesternbaum surgery, 158 primary EOM myopathy, 59–60 clinical characteristics, 157–158 traumatic myopathy, 61–62 idiopathic infantile nystagmus, 158 vergence and gaze abnormalities, 72 slow phase, 157 Normal correspondence (NRC), 34 Marcus-Gunn phenomenon, 80–82, 85, 87–89 Marlow occlusion, 19 O Meta-regression model, 143 Ocular albinism (OA), 155 Microstrabismus Ocular motility disorders, CCDD number of operations brainstem and cranial nerve development, 77, 78 postoperative angle of strabismus, 145 Brown syndrome reoperation rate, 142–143 comorbidity, 85 test-retest reliability, 144–145 epidemiologic features, 85 random-eff ects model, 146, 148 incidence and heredity, 86 spontaneous reduction, 146–148 intra-and postoperative fi ndings, 87 spontaneous resolution, 146 laterality, 85–86 Mittendorf’s spot, 115 motility fi ndings, 83–85 Möbius syndrome, 83 natural course, 87 Moebius syndrome, 70 neurodevelopmental disorder, 89–90 Monofi xation syndrome (MFS) potential induction, 86–87 animal models, 37 radiologic fi ndings, 87 anisometropia, 33 saccadic eye movements, 85 bi-fi xation, 36–37 sex distribution, 86 causes, 33 CFEOM, 78–79 foveal suppression scotoma elimination, 36 congenital fourth nerve palsy, 82 manifest strabismus, 35–36 congenital monocular elevation defi ciency, 87–89 micro-esotropia congenital ptosis, 81 extrastriate cortex, 52–53 congenital trochlear palsy, 82 neural mechanism, 51 Duane retraction syndrome, 79–81 neuroanatomic fi ndings, 52, 53 HGPPS, 81 stereoscopic threshold, 52 isolated uni-/bilateral facial palsy, 83 subnormal stereopsis and motor fusion, 51 vertical retraction syndrome, 88 normal and anomalous binocular vision Ocular motor control system, 18 anomalous retinal correspondence (ARC), 34 Oculocutaneous albinism (OCA), 155 binocular correspondence, 34–35 Oculomotor palsy, 71 communication, 33 Optic neuropathy, 133–134 cortical adaptation, 34 Optical coherence tomography (OCT), 155, 156 ocular dominance column, 33, 34 Orbital radiotherapy (OR), 213 normal/near-normal fusional vergence, 37 primary MFS, 38–39 P Motor skills, 106 Paralytic strabismus Muscle length adaptation, 11–13 complete third nerve palsy, 198–199 Index 231

fourth nerve palsy hypertropia measurement technique, 188 bilateral involvement, 201 superior rectus muscle recession effects, 186–188 congenital superior oblique palsy, 200 suspension technique, 188–189 inferior oblique weakening procedure, 203 tendon incarceration syndrome, 185 superior and inferior rectus recession, 209 frenulum, 185 superior oblique strengthening procedure, 209 theoretical eff ect superior oblique tendon laxity, 201 anterior–posterior axis, 189 superior rectus contracture, 201 posterior tenectomy, 190 surgical plan, 200 SO anatomy, 190, 191 torsional diplopia, 202–203 SO tendon, 189, 192 incomplete third nerve palsy, 199 threefold function, 189 principles two-dimensional trigonometry, 192 preoperative assessment, 196–197 surgery timing, 195–196 T surgical treatment, 197–198 Th yroid-stimulating hormone receptor (TSHR), 208 sixth nerve palsy hypertropia Traumatic myopathy, 61–62 lateral and medial rectus resection, 204 Trochlear palsy, 71 medial rectus weakening, sound eye, 204–205 TSHR antibodies (TRAb), 208 Pediatric strabismus Two-dimensional trigonometry, 192 adult strabismus, 7 associated conditions, 4 U esodeviation, 1–2 Unilateral strabismus changes exodeviation, 3 cyclovertical deviation, 20, 21 hyperdeviation, 3–4 head-tilt changes, 21 surgery rates, 4 ipsilateral medial and contralateral rectus muscle, 19 worldwide incidence and prevalence, 4–7 torsional position, 20 Periodic alternating nystagmus (PAN), 158–159 vertical recordings, 21 Pharmacological occlusion, 104 Photorefractive keratectomy (PRK), 105 V Plano lens, 130 Vergence adaptation, 11, 12 Posner’s maneuver, 174 Vertical retraction syndrome, 88 Posterior partial tenectomy, 190 Visual cortex mechanisms Primary extraocular muscle (EOM) myopathy, 59–60 esotropia Primary oblique muscle overaction, 14 binocular input correlation, 50–51 Prism adaptation, 12 binocular visuomotor behavior development, 42, 43 cerebral damage risk factors, 41–42 Q cortical binocular connections, 44–46 Quality adjusted life years (QALY), 99 cytotoxic insult, cerebral fi bers, 42 early-onset (infantile) esotropia, 41 R extrastriate cortex, striate cortex, 46 Reversed fi xation test (RFT), 179 fusional vergence and innate convergence bias, 44 genetic infl uence, cerebral connection, 42 S high-grade fusion repair, 50 Sensory esotropia, 2, 3 inter-ocular suppression, 46–47 Sensory exotropia, 3 monocular compartments, striate cortex, 44, 46 Sixth nerve palsy hypertropia motion sensitivity and conjugate eye tracking, 44 lateral and medial rectus resection, 204 naso-temporal inequalities, cortical suppression, 47 medial rectus weakening, sound eye, 204–205 persistent nasalward visuomotor bias, 47–50 Stereoacuity skills, 106 sensorial fusion and stereopsis development, 43 Superior oblique overaction (SOOA), 176–177 strabismic human infant repair, 50 Superior oblique (SO) surgery micro-esotropia clinical investigation extrastriate cortex, 52–53 6–0 Polyglactin 910 sutures, 186 neural mechanism, 51 asymmetric effects, 189 neuroanatomic fi ndings, 52, 53 enucleation, 186 stereoscopic threshold, 52 Jampolsky’s recommendations, 187 subnormal stereopsis and motor fusion, 51